First up let me say that I'm still running Windows XP and I have no first hand experinece with Vista.
I've noticed that some people making claims that "audio" is much better in Vista, though what specifically they are referring to I don't know. The more outragous claims go something like "with Vista your old onboard sound will sound as good or better then the best PCI soundcard". Of course I think it's a load of bull, but what do the changes to audio in Vista really amount to?
I think that is a load of crap as well. But Vista does require significantly more computer power than XP. So it is possible that the people that are using Vista have better hardware (even onboard hardware) than the average XP user.
The only thing that might have been improved is latency, which would only affect users of Pro Audio software like Pro Tools or Cubase. Supposedly, the new audio architecture allows for all sound cards to have low latency, instead of only sound cards with ASIO drivers (typically only expensive professional sound cards).
However, pro audio software developers aren't exactly lining up to use the new Vista APIs, as ASIO has been serving them well for almost a decade.
There's certainly nothing new that would improve sound quality except for maybe a new resampling algorithm.
Resampling is also vastly improved. A side effect is that it also allows per application volume setting.
However, many gamers are complaining because several processing tasks are now handled by the CPU instead of the sound card DSP.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
If this is true, how does Vista analyze the room you are in? I mean even top notch audiophile equipment doesn't do this great feat.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
If this is true, how does Vista analyze the room you are in? I mean even top notch audiophile equipment doesn't do this great feat.
From what I've read (and I haven't used Vista since an early beta so I have no first hand experience,) this is a
one-time thing, a kind of wizard to be run, which I guess uses pink noise or similar. I may be wrong on this, though.
Audiophile equipment doesn't do it because it doesn't have a microphone or any processor to do it with, and it's also probably pretty pointless for consumer equipment...
>Audiophile equipment doesn't do it because it doesn't have a microphone or any processor to do it with
Mid to high range denons do, they have a microphone supplied with their av amps to setup the amp to match the room.
I have a (a bit old) Denon equalizer which is bundled with a microphone in order to be able to create a flat frequency response, so as Spoon pointed, audiophile equipment can do it since years.
(yes, this is pink noise that is used)
Yeah, frequency analysis of rooms is much cheaper than it used to be. A good mic, pink noise generator and an average preamp with a firewire interface can do miracles these days. Sure beats breaking out my Rane RTA (Real Time Analyzer) rack unit and doing it that way. Real Time Analyzers until recently (last 5 years) were the sole domain of pro audio gear. Now it's just plug-and-play. (I'm glad to hear it's working it's way into consumer audio receivers now) Hell, I thought it was awesome when they started incorporating speaker timings into consumer units. It's going to be an exciting future.
Anyway, from what I've read this is Microsoft's way of trying to convince audio people to dump their Macs, because you can now do "pro audio" on Windows. (Basically true. I run pro tools on a PC in my home studio) We use Macs at the larger studio I work for though. Pro Tools on both platforms is essentially identical. Of course, you don't get as many driver conflict/problems on a Mac usually.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
They use a signal somewhat like a chirp. Pink noise is spectacularly unsuited to such work.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
If this is true, how does Vista analyze the room you are in? I mean even top notch audiophile equipment doesn't do this great feat.
(if it's accepted) there will be an AES convention preprint on this in the fall AES.
They could also maybe be referring to the fact that Vista does "auto-room-adjustments."
(Analyzes the room in realtime and makes adjustments for frequency bumps)
I am not sure whether this is on by default though. And, I would assume you would need a mic. (I'm not quite sure how they would do this without pink noise either) Hey, I'm just the messenger ya know.
I will investigate this when I get home from work tonight. There has to be a way to turn it off if it exists. This would use clock cycles very unecessarily imho.
In Vista it is a one time setup, it does not run all the time.
I've noticed that some people making claims that "audio" is much better in Vista, though what specifically they are referring to I don't know.
I can think of one area that might be of interest to the HA people. They went to a 32 bit floating point "audio engine" in Vista. So any DSP that might be done by Vista is now done using 32 bit floating point math.
Of course, who wants any extra DSP done? If you want to bypass any Vista DSP, then your audio app should use the new WaveRT mode which will let it pipe data directly to the soundcard.
I will investigate this when I get home from work tonight. There has to be a way to turn it off if it exists. This would use clock cycles very unecessarily imho.
It is off until you profile a room and turn it on. In order to do a profile you need an omni or cardioid (no close-talking mikes), or even better an instrument mike (settings in UI for both).
Once you turn it on it uses a few cycles when you're doing output. It's not enormously expensive, actually. There are filter designs, and there are filter designs ...
Resampling is also vastly improved. A side effect is that it also allows per application volume setting.
The major improvement I see in this is that you can now set a "shared mode" sample rate. This makes it possible to bypass the internal resampler of the audio hardware even when using DirectSound output. I used the famous udial.wav to compare Vista's resampling quality with that of foobar2000's PPHS resampler and I couldn't hear a difference. There was clearly no siren sound in either playback method, but only a very high pitched one like some CRT TVs produce when you stand at the right (or wrong ) spot. I was using an Edirol UA-1EX set to 48 KHz.
I can think of one area that might be of interest to the HA people. They went to a 32 bit floating point "audio engine" in Vista. So any DSP that might be done by Vista is now done using 32 bit floating point math.
Combined with the above this is really something that could make kernel streaming or ASIO output superfluous for normal listening. Just set the native sample rate of your audio adapter in the shared mode configuration screen and you're done. The big advantage is that you can still hear the sound of other applications and even change their volumes in a central mixer application.
Much has been written about Audio Processing in Vista (http://www.avsforum.com/avs-vb/showthread.php?t=713073), and it is very promising. Haven't tried room correction yet, as I don't have a mic.
Hello,
I have just installed Vista dual boot with XP and from the first moment I got the impression that it sounds better than XP, more detailed highs and bass, and somewhat better dynamics.
I have a M-Audio Audiophile 2496 RCA cabled to a 50 watt NAD amplifier and stereo PSB monitors. I have tuned Vista for reduced memory and CPU consumption by only running essential services and software.
I know my claims come just by personal perception, and aren't supported or proved with ABX, but I have noticed a significant improvement and wanted to contribute to this post. I run updated drivers on both XP/Vista, and play music on foobar2000, WMP and MPClassic.
more detailed highs and bass, and somewhat better dynamics.
This forum appreciates your enthusiastic voluntarism very much, but seriously, phrases like that trigger off ToS #8. People demand statistical evidence for off-hand audiophile-speak.
I know my claims come just by personal perception, and aren't supported or proved with ABX
Then surely you know that not many will take you seriously, yes?
If audio is played with the best respective settings on Vista and XP, there will be
no distinguishable difference for identical files.
Effects like Creative's 24-bit Crystalizer are not counted.
I think that is a load of crap as well. But Vista does require significantly more computer power than XP. So it is possible that the people that are using Vista have better hardware (even onboard hardware) than the average XP user.
Since this still seems to be running...
In Vista, "kmixer" does not exist. System processing is done in 32 bit float, and digitized to the rendering resolution with an anti-clipping limiter at the end of the chain. EQ, loudness EQ, room correction, bass management, speaker fill, headphone virtualization are all done in float.
In addition, a high-quality resampler has been added for streaming audio (still a fast resampler for hihg-channel-count dsound, which may have to do 100 channels at once), including dsound (low channel count, i.e. 1 or 2 stereo, 5.1 or 7.1 streams) that is good well beyond 16 bits.
The downmix algorithm for channel folddown when nothing else is called for has been modified so that you will no longer have channel dropouts.
So, in fact, Vista audio quality is, as long as it stays in real time, considerably better in quality.
In addition, Vista has exclusive mode, so you can get exact audio in-audio out from the engine if you want that.
If audio is played with the best respective settings on Vista and XP, there will be no distinguishable difference for identical files.
Ok, please show YOUR evidence for that.
When the ABX test is set up to test for lack of differences, sure, I'll do it.
You sir, have flouted TOS #8. Where's the evidence for the discernibly better resampler in Vista? All the big numbers you throw around are useful for DSP work. For playback, a previous test conducted here has shown that most people's ears don't go beyond.... 13-bit?
And you don't seem to understand 100% accurate audio reproduction, at least, beyond the listener's ability to distinguish. If I claim that you will hear no difference between a FLAC, Wavpack and Vorbis q6 encode of the same musical file, do you want to call on me to perform a test too?
Oh my. This could be um, interesting. <gets popcorn>
When the ABX test is set up to test for lack of differences, sure, I'll do it.
You sir, have flouted TOS #8. Where's the evidence for the discernibly better resampler in Vista? All the big numbers you throw around are useful for DSP work. For playback, a previous test conducted here has shown that most people's ears don't go beyond.... 13-bit?
And you don't seem to understand 100% accurate audio reproduction, at least, beyond the listener's ability to distinguish. If I claim that you will hear no difference between a FLAC, Wavpack and Vorbis q6 encode of the same musical file, do you want to call on me to perform a test too?
Yes, I'm calling you on all of your unsupported, false claims. You have made a testable claim without any evidence. Now produce your evidence.
You are the one insisting that there is "no distinguishable difference". Prove it. Run the ABX test yourself, and don't try to shift the burden. Prove your claim. You made it, not I.
Regarding the resampler, you can measure in the resampler in Vista yourself, as I have. No, I don't have the numbers at hand presently, I measured it rather before Vista was released. If you've any experience to speak of in PC audio, you'd know that there is a substantial problem with resampling in Kmixer, and that the presence of a high-quality resampler is the solution for that problem. This is not some "new bit of news" but something that people in the field have known for years. I'm not making any new assertion, I'm stating the common understanding, and a common understanding that I have personally tested and verified, BEFORE product release.
That would go for the bass management, loudness EQ, graphic EQ (only Vista version, please!), room correction, speaker fill, headphone virtualization, and so on, as well. And I'm quite certain that all of them run in 32 bit float for calculations, as opposed to the Kmixer version of (some of them) that ran in 16 bit signed integer, including filters, etc. You are aware of the need for computation headroom in filters, are you not?
So go do the measurements. We're not talking hypotheses, guesses, or anything beyond COLD HARD FACT here, so get with the program and go make the measurements. If I tell you numbers, you obviously won't believe a word I say. (but the difference between 120dB and 50db is pretty significant, I'd say)
Evidence is commonplace that denies your "13 bit" number. Try a 13 bit piano music recording. I don't even have to run that test (again), because I know what happens. Do it in a quiet room, though. I'll cheerfully agree that in a common office, 13 bits is sufficient, but that has nothing to do with human hearing to speak of, and everything to do with how noisy the modern world is. Some of us actually listen to music in a living room instead of a computer lab.
A 13 bit INSTANTANEOUS dynamic range is about right. (meaning difference between highest energy and lowest perceptable energy across frequency at one instant). I've said this in my own tutorials and talks. However, with recorded music, in a quiet space, played back under good conditions, its very easy to show that 16 bits is marginal, ACROSS TIME, i.e. from noise floor (inaudible or not) to peak level, and if the noise floor of the 16 bit system is at all peaky at low frequencies, i.e. not flat or high-pass-like, 16 bits is not enough in a quiet environment.
That result can trivially be seen from a look at Fletcher or Stevens' work on zero loudness curves, coupled with a knowledge of modern headphone sensitivity.
I am ignoring your last bit of uninformed ad-hominem nonsense. I would suggest that you do not repeat it again.
Oh my. This could be um, interesting. <gets popcorn>
Gimme a handfull of that popcorn, will you, I missed lunch and I'm in a bad mood.
OK, just to throw a little more into this, I have a test system running 32 bit Vista Ultimate and, although I haven't and have no intention of ABXing this, the quality improvements in Vista so far as audio is concerned are sufficiently great that ABX is irrelevant. If you haven't tried this, your comments are pointless!
Don't get me wrong, I'm not a great advocate of Vista, although there are some things I do like about it, but in the audio reproduction area, great strides certainly have been made.
I agree with SLD. Examine the playback flow with an M-Audio 2496 using MA drivers (that must bypass kmixer and route to the patchbay which can enable/disable MA's monitor mixer). You have format (ogg, mp3, etc)>codec (decompress to wav pcm?)>soundcard (DAC conversion)>amp. Yes ... gross simplification. I don't know how system sounds get into the audio stream with a 2496 setup with kmixer out of the loop. I also don't know if the 2496 or it's drivers do any resampling but I would think not (at least on the primary audio channel). So ... if I basically have it right, then how could Vista sound any better than XP, or FooBar better than WinAmp or Media Center for that matter.
No intention of testing Vista ... ever. What kind of setup do you have John?
Of course I think it's a load of bull, but what do the changes to audio in Vista really amount to?
From at least some of the people I've talked to at least some of the changes in Vista amounted to "Get Creative Labs out of the Kernel". MS would never publicly admit it but the rumour is that Creative drivers are the biggest single cause of BSODs other than actually broken/failing hardware.
Sorry, I should have been more clear! I was comparing basic Vista repro with Kmixer. Using KS circumvents that issue.
Systems:
1. E4300 @ 2.4GHz, Asus P5W-DH Deluxe, 4GB generic PC2 6400, M-Audio 2496, XP Pro;
2. E6750 @ 3.4GHz, Asus P5N-E SLI, 4GB Crucial Ballistix PC2 6400, X-Fi Extreme Music, Vista Ultimate;
3. E6700 @ 3.4GHz, Asus P5N32-E SLI, 4GB OCZ Reaper PC2 6400, on-board audio, XP Pro;
4. AMD FX-57 @ 2.8GHz (stock), Asus A8R-MVP, 2GB Corsair DDR1 4400, Terratec Six-Pack, Ubuntu.
Vista is really just a testbed, although having sorted out a proper Admin account, dumped Defender and taken ownership of most of the folders, it's a lot better than it first looked!! I still wouldn't use it from choice, but I could probably live with it.
Nice overclocking there ... I'm running XP x64 and Ubuntu 7.10 x64 in a dual boot on a stock E8400 through a 2496. Both OS's sound equally groovy but pardon the subjectiveness as ABX'ing would be impossible.
Of course I think it's a load of bull, but what do the changes to audio in Vista really amount to?
From at least some of the people I've talked to at least some of the changes in Vista amounted to "Get Creative Labs out of the Kernel". MS would never publicly admit it but the rumour is that Creative drivers are the biggest single cause of BSODs other than actually broken/failing hardware.
No, the point was to get Kmixer gone. There are many issues around many drivers that cause many problems, and it would be unfair to single out any one manufacturer for this issue.
The goal in terms of Vista was 'get ALL the processing out of the kernel'. Which was successful, only the low-level drivers that talk to the hardware are in kernel. Everything else is in (high priority) user mode.
I agree with SLD. Examine the playback flow with an M-Audio 2496 using MA drivers (that must bypass kmixer and route to the patchbay which can enable/disable MA's monitor mixer).
...
I am talking about what the non-pro user would have, which would include Kmixer in XP, and the MF and WAVE stacks in Vista. This is, after all, what 99% of windows world (give or take) uses.
If you've heavily customized your system, obviously nobody but you can guess what it will do. So why would you presume anyone else would even know what you've done, or why or how it would work?
I have Vista Ultimate and a really good test microphone w/preamp (don't know the mfg - bought it from Liberty Audio). The Vista room correction is calibrated with a series of chirps, not pink noise, so I think there is some time domain correction going on as well. I was hoping to do an in-depth review of its performance by measuring steady state performance before/after, but haven't had the time to do it. The room correction can be enabled / disabled per output device, so SPDIF vs. desktop speakers can be done separately (which I did) with separate correction.
When I did the room correction on my main system (SPDIF from the PC @ 16/44.1 into a Theta Casanova digital preamp, Paradigm Active Reference 40 speakers and Velodyne HGS18 subwoofer) I noticed it sounded different in casual listening (no ABX yet). It would have been nice if MS allowed me to see what measurement and correction it performed but I can't figure out how to do that. Maybe somebody knows how to prise that out of Vista?
Having tuned systems before it sounded like the correction tended toward a technically flat response which sounded weaker in the bass than the purposely rising low end I had tuned in before. But that will have to be tested so I have some data to see what it really did.
On my desktop system (Logitec X-230, 2 satellites/woofer) it also seemed to make a noticeable difference. In this case, a big subjective improvement in clarity subject to the wrath of the microphone when I can get to it.
In general it seems to have an audible affect and I would love to get to some test data.
I have Vista Ultimate and a really good test microphone w/preamp (don't know the mfg - bought it from Liberty Audio). The Vista room correction is calibrated with a series of chirps, not pink noise, so I think there is some time domain correction going on as well. ...
In general it seems to have an audible affect and I would love to get to some test data.
http://www.aes.org/sections/pnw/ppt/room_correction.ppt (http://www.aes.org/sections/pnw/ppt/room_correction.ppt)
Is a very good description of both the design and the implimentation, at least as good as one gets without going into specific code and such.
Effects like Creative's 24-bit Crystalizer are not counted.
Hello again. Well I forgot to say that I tested both OS with same samples, using foobar2000, M-Audio ASIO output, 1024 samples latency and same volume, pure output and no DSP whatsoever.
I cannot afford an ABX at present setup and I might violate ToS#8 with my claims, but after several switchouts and paying close attention I still like better the sound in Vista ,to my ears it's smoother, less coarse than in XP.
Hang me if I sinned, but let me people contribute with my audiophile ability to assess a personal outlook on the subject.
Thx,
Let's get this straight. People here are comparing apples and oranges.
One group compares XP and Vista using methods like ASIO or kernel streaming where the goal is that *one* application can output an audio signal as unmodified as possible without any effects or mixing with other applications' audio. In that case there should really be no noticeable difference at all between XP and Vista.
The other group compares XP and Vista under more "normal" conditions like using DirectSound output and maybe using additional effects that should enhance the output subjectively. Under these conditions Vista is the clear winner because its audio engine is much more precise and even features a high quality resampler (at least udial.wav doesn't produce any audible artifacts). In Vista you can listen to high quality music while also listening to high quality game sound, for example, because every signal is treated the same (high quality) way and can even be mixed together without a problem.
So, if you are willing to sacrifice a certain amount of usability, XP can sound as good as Vista, but Vista allows for high quality audio without sacrificing anything.
Vista is really just a testbed, although having sorted out a proper Admin account, dumped Defender and taken ownership of most of the folders, it's a lot better than it first looked!! I still wouldn't use it from choice, but I could probably live with it.
I'm using it at 30% (70% XP) maybe. With Vista, you first need to turn all those annoying notifications off.
And the longer you use it, the less the constant hard disk action will be. The constant hard disk access was really annoying. It still does a lot after booting (prefetch, indexer, ...), but getting quiet later the session. the prefetch thing seems to work pretty well and it's true, applications statup times (1st non cached start) decrease to maybe 50% of how long XP needs. Also the Vista booting itself gets better.
I cannot afford an ABX at present setup and I might violate ToS#8 with my claims, but after several switchouts and paying close attention I still like better the sound in Vista ,to my ears it's smoother, less coarse than in XP.
Then one or both of your audio chains (which include the OS) can be considered as seriously broken.
Resampling is also vastly improved. A side effect is that it also allows per application volume setting.
hmm, does that mean with vista i could (finally) control web stream volume signal and my music player signal separately?
hmm, does that mean with vista i could (finally) control web stream volume signal and my music player signal separately?
Assuming that you don't use the same application for playing back the stream, yes.
One group compares XP and Vista using methods like ASIO or kernel streaming where the goal is that *one* application can output an audio signal as unmodified as possible without any effects or mixing with other applications' audio. In that case there should really be no noticeable difference at all between XP and Vista.
Yeah, I agree that it makes sense, but I still have the impression that Vista sounds better in both DSDirect and ASIO compared to whatever in XP. In Vista, like in XP, I cannot discern between DSDirect and ASIO from within foobar2000, except that in ASIO mode I get clicks when i.e. I open taskmanager and resize the window, even with 'normal' latency. It happens to me in both XP and Vista. However, with DSDirect I don't get those clicks when I do other tasks in the computer.
I cannot afford an ABX at present setup and I might violate ToS#8 with my claims, but after several switchouts and paying close attention I still like better the sound in Vista ,to my ears it's smoother, less coarse than in XP.
Then one or both of your audio chains (which include the OS) can be considered as seriously broken.
Well, I'm talking about slight subjetive improvement, I would like to do a blind test and check, it could be derived from my leaning or whatever but I have payed close attention and I got that impression right from the first time and on. I just wanted to discuss others opinions about sound in Vista.
Assuming that you don't use the same application for playing back the stream, yes.
well, 99% of the time they are flash based audio samples from my favorite record stores, and they tend to be 2X louder than my max foobar setup, so it's annoying to play with that master volume at taskbar all the time
I cannot afford an ABX at present setup and I might violate ToS#8 with my claims, but after several switchouts and paying close attention I still like better the sound in Vista ,to my ears it's smoother, less coarse than in XP.
Then one or both of your audio chains (which include the OS) can be considered as seriously broken.
So, you are arguing what? I have specifically mentioned substantial differences between XP and Vista above. What are you saying, then?
And when I say "differences" I mean testable, verifiable, ... NOT subjective.
Assuming that you don't use the same application for playing back the stream, yes.
well, 99% of the time they are flash based audio samples from my favorite record stores, and they tend to be 2X louder than my max foobar setup, so it's annoying to play with that master volume at taskbar all the time
There's also the loudness EQ in Vista, if you're concerned more with level control. It certainly does modify the signal, of course.
He says he has a 2496. Who would bother installing such a nice card then use generic win stock drivers? If he's using M-Audio drivers, then Vista CAN'T sound better than XP. I am assuming that MA Vista drivers would skirt around the OS just like they do in XP. I could damn near guarantee that you (woodinville) couldn't ABX between a properly setup XP system (kmixer or not) and that of Vista ... with just a generic soundcard using windows drivers. Seattle eh? ... your not a MS employee trying to eradicate Vista FUD from the masses, are you? Too late if so ...
He says he has a 2496. Who would bother installing such a nice card then use generic win stock drivers? If he's using M-Audio drivers, then Vista CAN'T sound better than XP. I am assuming that MA Vista drivers would skirt around the OS just like they do in XP. I could damn near guarantee that you (woodinville) couldn't ABX between a properly setup XP system (kmixer or not) and that of Vista ... with just a generic soundcard using windows drivers. Seattle eh? ... your not a MS employee trying to eradicate Vista FUD from the masses, are you? Too late if so ...
You claim that you can prove something, prove it.
You don't
know what M-audio is doing, but you assert results. You "
assume". You imply that stock drivers are somehow worse. Evidences for that, please? You assert that
one can not hear the resampling problems in Kmixer. Evidences, please? Fact: I don't know what M-audio does, either. Rather than spew supposition, why don't you try for information instead of guesswork? Do you know what the difference between stock drivers on Vista and third-party drivers on Vista gets you?
Allow me to give you a simple situation. Set your card's sampling rate to 48kHz. Use Ry Cooder's album "Jazz", the first thirty seconds of the last track. Play it through Kmixer, play it through Vista. Now explain what you heard.
Hint: Think "resampler". Go ahead, try it. Report back on YOUR findings. You want to argue, try having some facts in hand, first, ok?
You engage in ad-hominem behaviors (by the way, I do work for MS, which should be rather evident as I just talked about
testing various Vista processing before product launch, eh?) and question things that can be tested and verified by simply trying them. Why don't you try for some evidence before you make claims?
Do you have anything of FACT to say, now? How about evidence? How about doing some basic
homework like looking up "kmixer" on the net?
I would like to make sure about this: if I am using ASIO/KS drivers and compatible audio application like foobar2000 that bypass Windows Mixer, then the sound wouldn't be discernible from Vista to XP as hancoque and Xenno say, isn't it? It makes sense, because all these send audio directly to the audio card mixer for DA output. I noticed that Wave level windows monitor mixer doesn't work in this setup.
So then all Vista sound improvement would come to the internal OS kernel and mixer, DirectSound-compatible, capable of better mixing application-wide, windows sounds and Direct Sound applications like WMP, in terms of performance and stability. But speaking about audio quality, if the bits sent to the soundcard are the same, then it outputs the same signal and sounds the same, because the DA converter chip is on the soundcard.
So then I can assume equal audio quality output on XP/Vista as long as ASIO/KS drivers and compatible software are used. The question is: Does Vista DS mixer send the same or better data from the application to the soundcard as XP with the sampling untouched?
Excuse for the wordy post, hope this helps clarify..
thx in advance
>> "I just talked about testing various Vista processing before product launch, eh?) ... "
Yeah ... I missed that. But I did have a suspicion that some of your content sounded an awfully lot like an MS press release. Tester? Does that mean you don't know what's happening at the core level any better than I do?
>> "You claim that you can prove something, prove it."
I have no need to prove anything ... the burden is upon you. You say you have and the results are in ... OK ... I'll wait for un-biased 3rd party comparisons. I won't make them personally because Vista won't see the light of day here (unless they have it all straightened out by SP4).
I have no doubt that the Vista sound sub-system is better than any previous win version. Yes, kmixer apparently resamples, and Vista's "mixer" doesn't. I'm saying the improvements are much more theoretical than audible. It also doesn't apply to the OP given the soundcard he's using tho he hasn't stated yet what drivers he uses. If MA XP drivers loop around the OS and the MA Vista drivers use the system then that's great ... it achieves the same result.
FYI: I am not a MS basher with an axe to grind as I like most of what they've ever put out
The question is: Does Vista DS mixer send the same or better data from the application to the soundcard as XP with the sampling untouched?
As far as I know every application's audio data is converted to 32-bit float and resampled to a common sample rate which has been set by the user and which is also natively supported by the sound card. The different streams are then mixed and the resulting stream is quantized to the bit-depth set by the user, which is also natively supported by the sound card.
So, let's assume you have two applications running which output audio. Application A outputs data at 22050 Hz and 8-bit integer. Application B outputs data at 48000 Hz and 24-bit integer. The sound card's output is set to 44100 Hz and 16-bit. Then I would assume the signal flow is basically as follows:
Step 1 (conversion):
A [22050 Hz, 8-bit int] => A' [44100 Hz, 32-bit float]
B [48000 Hz, 24-bit int] => B' [44100 Hz, 32-bit float]
Step 2 (mixing):
A' + B' => C [44100 Hz, 32-bit float]
Step 3 (conversion):
C [44100 Hz, 32-bit float] => D [ 44100 Hz, 16-bit int]
Step 4 (output):
D => sound card
I left out any effects in this chain. But per-stream effects (like application-specific volume) should be applied to A' and B' and global effects (like EQ) should be applied to C.
If you
- don't apply any per-stream or global effects and
- only have one application outputting audio and
- the sample rate and bit-depth set for the sound card matches the material's sample rate
then there should theoretically be no difference to the original because a conversion from even 24-bit integer to 32-bit float is lossless. But I didn't verify all this. Perhaps Woodinville can shed some light on this.
>> "I just talked about testing various Vista processing before product launch, eh?) ... "
Yeah ... I missed that. But I did have a suspicion that some of your content sounded an awfully lot like an MS press release. Tester? Does that mean you don't know what's happening at the core level any better than I do?
Well, actually, I'm what they call an "architect". But that won't mean anything more to you than any other simple fact.
I have no need to prove anything ... the burden is upon you. You say you have and the results are in ... OK ... I'll wait for un-biased 3rd party comparisons. I won't make them personally because Vista won't see the light of day here (unless they have it all straightened out by SP4).
So, you made a claim, but you won't support it, and won't even run the test. What you're saying is plain as peaches in Georgia, dude.
I have no doubt that the Vista sound sub-system is better than any previous win version.
Um, then why did you just say otherwise? Sorry. I'm confused now. Is it a horse, or is it a mule?
Yes, kmixer apparently resamples, and Vista's "mixer" doesn't.
Both resample if they need to. XP in Kmixer, in the kernel, with very short filters, which you must do if you're in kernel mode. Vista in user space, with entirely better resampling for anything that's a low stream count. (there are valid reasons for a very fast resampler when you're pulling in 100 stems from a game at once)
I'm saying the improvements are much more theoretical than audible.
And I'm asking for your evidence. The issues with Kmixer resampler are well known, documented in a variety of places, etc. It's not like I'm just now stating utterly brand new news, after all.
What you, then, are saying, is that VISTA RESAMPLER IS NO BETTER THAN KMIXERS. That's what your claim boils down to.
And I'm asking you for the evidence for your positive assertion.
Where's the beef, then?
FYI: I am not a MS basher with an axe to grind as I like most of what they've ever put out
Right, like you just bashed me twice for working there. Got it, you don't bash MS, just its employees.
That's clearer now, thank you.
The question is: Does Vista DS mixer send the same or better data from the application to the soundcard as XP with the sampling untouched?
As far as I know every application's audio data is converted to 32-bit float and resampled to a common sample rate which has been set by the user and which is also natively supported by the sound card. The different streams are then mixed and the resulting stream is quantized to the bit-depth set by the user, which is also natively supported by the sound card.
So, let's assume you have two applications running which output audio. Application A outputs data at 22050 Hz and 8-bit integer. Application B outputs data at 48000 Hz and 24-bit integer. The sound card's output is set to 44100 Hz and 16-bit. Then I would assume the signal flow is basically as follows:
Step 1 (conversion):
A [22050 Hz, 8-bit int] => A' [44100 Hz, 32-bit float]
B [48000 Hz, 24-bit int] => B' [44100 Hz, 32-bit float]
Step 2 (mixing):
A' + B' => C [44100 Hz, 32-bit float]
Step 3 (conversion):
C [44100 Hz, 32-bit float] => D [ 44100 Hz, 16-bit int]
Step 4 (output):
D => sound card
I left out any effects in this chain. But per-stream effects (like application-specific volume) should be applied to A' and B' and global effects (like EQ) should be applied to C.
If you- don't apply any per-stream or global effects and
- only have one application outputting audio and
- the sample rate and bit-depth set for the sound card matches the material's sample rate
then there should theoretically be no difference to the original because a conversion from even 24-bit integer to 32-bit float is lossless. But I didn't verify all this. Perhaps Woodinville can shed some light on this.
Not quite. Since we can not assure that there was nothing added, no gain controls changed, etc, we must dither the final float->fix conversion, so you will incur one step of dithering at your card's level. As annoying as this is for gain=1 with single sources, we can't possibly single-source in general.
If you don't want even that, there is exclusive mode, which is roughly speaking a memcopy.
Here are some RMAA tests of the MAudio Audiophile 2496 in Windows XP:
http://www.daefeatures.co.uk/tests/Audiophile2496Transit.htm (http://www.daefeatures.co.uk/tests/Audiophile2496Transit.htm)
If you search the listening tests pages you'll find some samples recorded from the audio out of that card on XP. Anyone is welcome to ABX those files against the original, in Vista.
I don't believe for a second the original claim made with regards to this card is real.
Here are some RMAA tests of the MAudio Audiophile 2496 in Windows XP:
http://www.daefeatures.co.uk/tests/Audiophile2496Transit.htm (http://www.daefeatures.co.uk/tests/Audiophile2496Transit.htm)
If you search the listening tests pages you'll find some samples recorded from the audio out of that card on XP. Anyone is welcome to ABX those files against the original, in Vista.
I don't believe for a second the original claim made with regards to this card is real.
It's quite obvious that the card was set to the native rate of the input signal there. (Which you can believe I'd do as well on XP!)
Got anything from any other situation?
I want to say this first because this is from the audio programer:
"The end result of these changes should be a significant improvement in the quality of audio being rendered, especially on UAA compatible audio adapters."
I really do not understand why people do not simple go to the people who know best.. M$:
http://support.microsoft.com/kb/925901 (http://support.microsoft.com/kb/925901)
Video about the audio stack from m$:
http://channel9.msdn.com/ShowPost.aspx?PostID=145665 (http://channel9.msdn.com/ShowPost.aspx?PostID=145665)
also:
http://blogs.msdn.com/larryosterman/archiv.../19/471346.aspx (http://blogs.msdn.com/larryosterman/archive/2005/09/19/471346.aspx)
http://blogs.msdn.com/larryosterman/archiv.../20/471872.aspx (http://blogs.msdn.com/larryosterman/archive/2005/09/20/471872.aspx)
"Before Vista, the kernel audio stack set the output audio format to match the format of the audio being played. Normally, this isn't a problem, since it means that we do less DSP of the signals. Unfortunately, it can lead to some rather unanticipated consequences. For instance, if you're playing a system sound (usually stereo, 22kHz), at the same time you start playing your MP3 files, then the MP3 file rendering happens at 22kHz, which is a noticeable degradation of audio quality. Once the audio system goes quiet, the rendering format will reset to the format of the content being played, but that may be quite some time later."
really good article also linked from M$:
Vista for Audio, 1 Year Later: Talking OS Plumbing with Cakewalk’s CTO:
http://createdigitalmusic.com/2008/01/16/v...-cakewalks-cto/ (http://createdigitalmusic.com/2008/01/16/vista-for-audio-1-year-later-talking-os-plumbing-with-cakewalks-cto/)
http://search.technet.microsoft.com/Defaul...p;Brand=technet (http://search.technet.microsoft.com/Default.aspx?locale=en-gb&lang=en-gb&Brand=technet)
Jesus ... your saying the different schemes between the 2 are clearly audible and I'm saying they're not. Again ... this all sprung off claims you more or less made starting from your 1st post here that Vista will always sound better than XP .. under any circumstance. Every replyer except for yours has scoffed at that idea. Can I put it any plainer? Now whose the simpleton here that can't get the general point?
Give it up Xenno, you've been sussed!
Hardly ... I guessed you read just woodin's post's and no others.
Jesus ... your saying the different schemes between the 2 are clearly audible and I'm saying they're not.
Well, actually, it's not ME who's saying that, a whole lot of people are saying that, the same people who have been pointing out the problems with Kmixer for years. Perchance if you had some experience in the field you'd have known this. The fact you even bother to argue the situation suggests rather a lot.
Again ... this all sprung off claims you more or less made starting from your 1st post here that Vista will always sound better than XP
No, that's not my claim. I guess you couldn't win the argument using what I actually said, so you have to make up a position for me?
So perhaps you can go tilt at whatever windmill you imagined saying that. Galoop, galoop! Hi Ho Don Quixote!
.. under any circumstance. Every replyer except for yours has scoffed at that idea. Can I put it any plainer? Now whose the simpleton here that can't get the general point?
Well I could speculate who that might be, especially since rather some number of people have chimed in on the subject.
But I don't think I'll bother, perhaps you can let us know when you get tired of crying "yield" to windmills, while the knight's skewered you through and through.
Enough. One can not educate the deliberately ignorant, and he who continues to argue with ...
Vista has, as far as I know, a completely revamped audio stack which pretty much irons out all the annoyances of the old kmixer, and also sports a UNIFIED control panel (no more different control panels per audio setup - whatever those panels do, must also be supported through the regular "soundcard properties" window), better quality internal resampling, the direct mode, a bunch of built-in features for bass redirection and channel upmixing, and per-app volume control.
This latter is my favorite. You know how annoying it is when some webpage fires up a flash applet with 100% volume over-clipped over-compressed horribly sounding audio, always when you least expect it, and when you are listening to music in foobar/winamp running in the background? You can just mute your browser from the windows volume mixer now.
Well, actually, it's not ME who's saying that, a whole lot of people are saying that..
Yea I agree, I've heard this from at least 5 people now.. Woodinville1 Woodinville2 Woodinville3 Woodinville4 Woodinville5.
To cut the story short, if people by any chance and for any reason are dissatisfied with Kmixer playback in XP, I'll just ask them to switch to a FREE player called Foobar2000 that has an excellent PPHS resampler running in Kernel Streaming mode, instead of buying a NON-free copy of Vista just so they can enjoy non-discernibly better audio.
So now we have a person's untested and unverified claims that Kmixer in Vista >> Kmixer in XP (besides consistently counter-claiming and convenient references to Google which will throw up ANECDOTES on this issue). We are also amazed that some individuals don't understand that the ABX test is constructed to test for differences, not to test for a lack of difference. Perhaps they did not read the posts of the people in this forum who are trained in both audio and statistics.
One group compares XP and Vista using methods like ASIO or kernel streaming where the goal is that *one* application can output an audio signal as unmodified as possible without any effects or mixing with other applications' audio. In that case there should really be no noticeable difference at all between XP and Vista.
Considering that this thread resides on an audio forum where people learn best-practices in audio, it should have been obvious that I have been talking about this. My question now is this: anyone disagrees with the quoted assertion? And what's the implication if I cannot hear a difference between Kernel Streaming and DirectSound?
I can't be bothered with Kmixer besides the fact that having been a member of this forum for 5 years now, I tend to sniff out TOS #8 violations easily. I usually let others take on the responsibility of informing the offender, just not this time.
It's amazing how the M$ bashers appear out of the woodwork when an opportunity arises!
Personally, I find M$ a complete enigma. There is part that I admire and much that I loathe, but none of that is relevant in this thread.
The simple statement being made is that the audio reproduction through Vista, out of the box, is noticeably better than through any previous version of Windows, straight out of the box. And the simple answer is that it is; sufficiently so that any testing of this is really quite pointless. And, I am quite aged and it's still very obvious even to my old ears! Now if you haven't been in a position to put this to the test, then you really don't have a valid opinion.
I find it quite interesting that those who dig out and wield TOS #8 here are those who fall into this latter category! We are all well aware that kmixer, etc., in XP can be bypassed with an attendant jump in output quality but, again, that wasn't where this started from.
Any further discussion on this thread should be limited to those who have made the comparison and, therefore, have a valid opinion rather than those who simply want to argue a point!
Sorry!! But this is getting slightly heated and very silly.
The simple statement being made is that the audio reproduction through Vista, out of the box, is noticeably better than through any previous version of Windows, straight out of the box. And the simple answer is that it is; sufficiently so that any testing of this is really quite pointless.
The simple statement being made is that the audio reproduction through cryogenically treated solid silver speaker cables, out of the box, is noticeably better than through any other speaker cables, straight out of the box. And the simple answer is that it is; sufficiently so that any testing of this is really quite pointless.
Eh?
If you are so sure, please organize and run a test to prove your point. The whole idea of requiring tests on this forum is that people can't make unsubstantiated claims about audibility - even (or especially) in cases where it is very obvious to them. Now, I am not an MS basher, and you might very well be right - but you need to provide evidence other than an assertion.
As for me, I didn't notice a significant change in quality when installing Vista, but I don't listen to music in Windows very often - so I might well have missed it.
I'm sorry, but without two identical hardware setups, testing is an impossibility as there would be too many variables to satisfy even the most basic of testing requirements! And while I have several systems available, no one is the same as any other.
I have to say that I don't quite see where this obsession for testing is going with this topic anyway! If you use XP, then you're forced to work within what it provides, and the same goes for Vista. Is anyone really likely to make their OS choice based upon whether the audio output from the one is any better than the other? I think not. The decision, if there is a decision to be made, will be based upon other factors not the least being what may be provided with a purchased system if you're not building it yourself.
Testing, therefore, is of no more than academic interest as basically you get what you're given when you choose the OS! In reality, my, or anybody else's, opinion, one way or the other, is utterly irrelevant!
To cut the story short, if people by any chance and for any reason are dissatisfied with Kmixer playback in XP, I'll just ask them to switch to a FREE player called Foobar2000 that has an excellent PPHS resampler running in Kernel Streaming mode, instead of buying a NON-free copy of Vista just so they can enjoy non-discernibly better audio.
Oh, so that's the real reason. People don't want to switch and must find reasons why the new Windows doesn't yield any advantages. It always boils down to the sleep happy thing. And I guess the people defending Vista are the ones that already have it and therefore want to find its advantages.
Considering that this thread resides on an audio forum where people learn best-practices in audio, it should have been obvious that I have been talking about this.
That might have been best-practice in XP but in Vista this is now obsolete. Even using a simple WDM/DirectSound combination now delivers superb quality. And that you don't need this advantage doesn't mean that it isn't there. Again I read this "I don't (want to) need Vista" between the lines.
So, basically this turns out to be an XP versus Vista in general "discussion" instead of XP's audio system versus Vista's audio system, at least when looking for the possible emotional motivation that some people might have here, which would of course be no valid argument and therefore is always hidden well behind objective phrases.
good to see nobody read the articles I posted and video about the Vista audio stack which is with the two programers. They explain about the stack, wave forms, dsp, floating point etc.
frankly they have more clue then most people here but if nobody wants to watch the video fair enough flame on.
So, basically this turns out to be an XP versus Vista in general "discussion" instead of XP's audio system versus Vista's audio system, at least when looking for the possible emotional motivation that some people might have here, which would of course be no valid argument and therefore is always hidden well behind objective phrases.
Not for me. I take issue with this statement.
Hello,
I have just installed Vista dual boot with XP and from the first moment I got the impression that it sounds better than XP, more detailed highs and bass, and somewhat better dynamics.
I have a M-Audio Audiophile 2496 RCA cabled to a 50 watt NAD amplifier and stereo PSB monitors. I have tuned Vista for reduced memory and CPU consumption by only running essential services and software.
I know my claims come just by personal perception, and aren't supported or proved with ABX, but I have noticed a significant improvement and wanted to contribute to this post. I run updated drivers on both XP/Vista, and play music on foobar2000, WMP and MPClassic.
I couldn't give a crap which OS comes out top. I just want facts substantiated with hard evidence, not people's perceptions. That's what this forum is about isn't it? If I get some free time over the weekend I'll try and get a hooky copy of vista and run some RMAA tests with my two MAudio cards and realtek HD onboard.
Got anything from any other situation?
Like? Time permitting, I'll provide it. Remember, for me this is mostly about the statement made above.
I just tested Vista's resampler by recording the outgoing signal with another sound card. The sound card used for playback is an M-Audio Audiophile 24/96 and the one used for recording is an Edirol UA-1EX at 96000 Hz, 24-bit. I used a sine sweep to detect any artifacts visually. Here are the results:
44100 Hz at 44100 Hz:
(http://www.devir.de/temp/44100_at_44100.jpg)
44100 Hz at 48000 Hz:
(http://www.devir.de/temp/44100_at_48000.jpg)
48000 Hz at 44100 Hz:
(http://www.devir.de/temp/48000_at_44100.jpg)
48000 Hz at 48000 Hz:
(http://www.devir.de/temp/48000_at_48000.jpg)
There are clearly artifacts visible but they leave me a bit puzzled because they are also visible when no resampling occurs. It might be a problem related to the UA-1EX's recording process. Any conclusions from this?
So now we have a person's untested and unverified claims that Kmixer in Vista >> Kmixer in XP (besides consistently counter-claiming and convenient references to Google which will throw up ANECDOTES on this issue).
Let's start from the beginning.
As the MS documents show, there is no "kmixer" in Vista ?. So, your allegation that there is is shown by testable, verifiable evidence to be wrong.
There is no kernel-level signal processing in Vista. That is also easily determined by looking at the basic Vista documents.
It is clear to me that you are not interested in an evenhanded examination of the subject, and that furthermore you are not willing to examine publically available information, including that which somebody else (I mean, third-party, even) has cited directly from the horse's mouth.
It is time that you capitulate, and that you admit that what I'm saying is tested, verified, and factual. Your attempts to manipulate TOS into silencing your debating foes, when in fact their statements are supported by the documentation available, is transparent.
So capitulate. Admit that what I've said is testable, verifiable, and falsifiable. What's more, it's time that you admit that my comments about the new audio chain in Vista are in fact factual and material.
Now, since I've pointed out that I designed some of the stuff in the Vista audio engine already, perhaps it will dawn on you that you're arguing with the pro from Dover, here.
http://www.aes.org/sections/pnw/ppt/room_correction.ppt (http://www.aes.org/sections/pnw/ppt/room_correction.ppt)
Is a very good description of both the design and the implimentation, at least as good as one gets without going into specific code and such.
thanks for the link to a very interesting view about DRC
Is here anywhere a more complete paper on this MS room correction, I found nothing searching in the AES preprints or journal.
Thanks for infos
The simple statement being made is that the audio reproduction through Vista, out of the box, is noticeably better than through any previous version of Windows, straight out of the box. And the simple answer is that it is; sufficiently so that any testing of this is really quite pointless.
The simple statement being made is that the audio reproduction through cryogenically treated solid silver speaker cables, out of the box, is noticeably better than through any other speaker cables, straight out of the box. And the simple answer is that it is; sufficiently so that any testing of this is really quite pointless.
Lad, you need to understand. We HAVE run tests.
good to see nobody read the articles I posted and video about the Vista audio stack which is with the two programers. They explain about the stack, wave forms, dsp, floating point etc.
frankly they have more clue then most people here but if nobody wants to watch the video fair enough flame on.
Dude, make that 3 of them.
There are clearly artifacts visible but they leave me a bit puzzled because they are also visible when no resampling occurs. It might be a problem related to the UA-1EX's recording process. Any conclusions from this?
It looks like some capture artifacts, and maybe a DAC or ADC filter artifact.
The question of DAC/ADC quality is another question, worth another thread, and one in which you may experience quite some ranting from those who want BETTER HARDWARE.
Oh, you betcha...
Look at WHQL standards for audio.
Now ask "how many manufacturers went ballastic when we set them that low".
http://www.aes.org/sections/pnw/ppt/room_correction.ppt (http://www.aes.org/sections/pnw/ppt/room_correction.ppt)
Is a very good description of both the design and the implimentation, at least as good as one gets without going into specific code and such.
thanks for the link to a very interesting view about DRC
Is here anywhere a more complete paper on this MS room correction, I found nothing searching in the AES preprints or journal.
Thanks for infos
There's a paper in the last NY AES convention on the same. Authors are Johnston and Smirnov. Sorry, my disc is at home, I don't recall the paper #.
As for me, I didn't notice a significant change in quality when installing Vista, but I don't listen to music in Windows very often - so I might well have missed it.
Take Ry Cooder's album "Jazz", set your DAC to 48kHz (so you ahve resampling for sure), and play back the (.wav version) of the last track.
Listen to the last 30 seconds in both systems.
You won't be able to run a blind test. After a few runs, you'll be able to ID each one independently.
I've tried it. Now you try it.
XP Kmixer resampling is measurable, I did measure it and results are here (http://www.kikeg.arrakis.es/measurements/digital%20dsound%2016-bit%20resamp_48%20K.htm). Sorry, graphs are not visible in Firefox, the measurements page was generated with an old version of RMAA.
It doesn't seem really awful, in my opinion it would get unnoticed in most cases for most people, but it is true that this resampling could be done better, SSRC or Cool Edit Pro/Adobe Audition are clearly better in this respect. Also, this resampling only happens when the sound card playback sampling rate doesn't match the source sampling rate, for whatever reason. I think in some cards it happens always (48 KHz fixed cards), in others when there are several sounds at different sampling rates playing at same time, whatever.
Vista improvement is that it does this resampling the right way, as should have been done from the beginning (32 bit float processing vs. 16 bit integer processing) and adds many other new features. I trust 100% what Woodinville has to say about new Vista features and capabilities. It have not had the chance to measure quality of the new resampling, but if Woodinville says it's good I trust him.
(Edit: Woodinville, thanks for the nice ppt about room correction, really very interesting)
Funny that I'm accused of being an MS basher. Is Microsoft Windows XP (what I'm using) a non-MS OS?
And nice one there woodinville, you skirt my direct question and pick on my ignorance of Vista's audio stack.
Let me say it again: best practices with music playback on both XP and Vista yield no discernible difference. You'll have to convince me that 101% is better than 100%, assuming 101% exists. I don't care if resampling is involved, if Kmixer is involved, if as many as 64 bits are involved, if room correction is involved. And to set the record straight, let me state also that I now trust the claim that Vista's out-of-the-box music experience is superior to XP's.
I hope the term 'best practices' conjures up the appropriate images in people's minds for constructive discussion. If people STILL want to pick on resampling alone, I'll have to step out of this thread. Why? Because I'll truly be speechless at the inanity and obstinacy.
By the way, I live in Dover too, but my latitude is 1 degree N.
The only thing I love about Vista's audio stack is the per-application volume control. Using Foobar2000 in Vista (RTM version, ahem) was a great pleasure.
Let me say it again: best practices with music playback on both XP and Vista yield no discernible difference.
So, does best practice bypass Kmixer or not?
You'll have to convince me that 101% is better than 100%, assuming 101% exists. I don't care if resampling is involved, if Kmixer is involved, if as many as 64 bits are involved, if room correction is involved.
Room correction is irrelevant, unless you're using it, so why did you even bother to mention it. It's very clear to me that you haven't done your homework, you don't know the issues, and you're spewing what looks perilously close to absolute hatred just because I dare to dispute your religious beliefs.
Convince? No, I don't have to. You can hold to whatever myth you want, I don't do religion. I've posted measurements, measurements that make the issue clear. 50dB and 120dB are hardly the same by any standard, including perceptual. Perhaps you would like to produce some appropriately done measurements that say otherwise? Note: We're tlaking about the processing itself here, not the quality of ADC and DAC, both of which are another sore point, but one that remains. Yes, those results are the result of extreme signals, but that's life.
Ball is in your court. Can you test the resamplers? Will you do the work?
How about channel folddown? How about the quality of the graphic EQ? What about BSOD due to 3rd party driver death? Do you count these as part of quality or not?
I hope the term 'best practices' conjures up the appropriate images in people's minds for constructive discussion. If people STILL want to pick on resampling alone, I'll have to step out of this thread. Why? Because I'll truly be speechless at the inanity and obstinacy.
Well, do you have an answer for known issues like Kmixer channel folddown and Kmixer resampler? How about clipping in the equalizer due to the 16 bit arithmetic, or the SNR loss in the IIR filters in the equalizer due to the 16 bit arithmetic, or then there's the question of putting fancy arithmetic in the kernal. But you dodged the BSOD issue from 3rd party drivers entirely. Guess that's not part of quality for you.
Perhaps you've just never noticed the problems. I don't know. Maybe that's it.
ETA:KG, try the resampling in Kmixer (be sure you get that, and not the resampler in the player, which is better) with a complex signal with lots of high frequencies. Carefully.
(Edit: Woodinville, thanks for the nice ppt about room correction, really very interesting)
Thank you, it's a nice solution where you get 95% of the full effect for about 3% of the (cpu) work. I like that kind of engineering solution.
OK, I'm stepping out of this thread after all.
From 'best practices in audio playback' we move to some rambling regarding 'room correction' and then *gasp* 3rd-party driver stability. I attempt to cover all the possible loopholes a certain egotist can conjure up, and hey, he found a new one, and a very minor one at that! And when I have already set the record straight, he goes and shifts the goalposts. Now he want me to believe that the best possible setup in Vista is better than the best possible setup in XP with ZERO documented proof. Shall I bring back my lossless versus lossless example again, just for kicks?
Here's a new one: "Cool, this BSOD in Vista sounds better than the one I usually get in XP!'
And it's amazing after all that, he think I'm the one with blind faith issues. Good attempt there with the ad hominem on 'religion' and 'stranglehold' on 'science'. I guess Hydrogenaudio has been feeding me 'religion' after all, lol. I can see that he is quite challenged in the areas of imagery and vocabulary though. He cannot visualise and comprehend 'best practices'!
----
I never wanted to be nasty about this, but you leave me no choice... if Microsoft holds a public survey on 'most loyal employee', remind me to vote for you.
He cannot visualise and comprehend 'best practices'!
You have claimed many things so far without evidence. Now, without evidence, you have publically accused me of professional misconduct.
@sld: I think you've got the answers already. You want to say that using kernel Streaming, or ASIO, in XP, allows to bypass XP's audio path, and with it, getting bitperfect audio straight to the hardware.
That's really not the issue here. And that doesn't make XP better than Vista. It is true that "audio in XP can sound as faithful as in Vista". But it is true also that "Vista improves the quality of the audio path versus the path in XP".
The problems with resampling with kmixer have been known to be true, and discernable, especially with AC-97 soundcards. You've been told why this doesn't happen in Vista. If it was audible before, and now it's fixed, it is obvious that it can be ABX'd.
This is why it doesn't need more proofs.
Resuming : "Can bypass" not equal to "Works".
As a follow up to Hancoque's post (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=55459&view=findpost&p=553865) here are my results. Recorded using the MAudio Audiophile 2496 @ 32 bit float* in loopback mode on Windows XP64. Files played back in Foobar2000 and resampled with PPHS.
44.1@44.1
(http://daefeatures.music-reviewed.com/images/HA/44.1@44.1.PNG)
44.1@48
(http://daefeatures.music-reviewed.com/images/HA/44.1@48.PNG)
48@44.1
(http://daefeatures.music-reviewed.com/images/HA/48@44.1.PNG)
48@48
(http://daefeatures.music-reviewed.com/images/HA/48@48.PNG)
* I realise the soundcard is only 24bit but I didn't see a 32bit option. If this is wrong please tell me how to rectify that.
I think I now have conducted a conclusive test. Instead of recording the outgoing signal with a seperate soundcard I just recorded it with the same one using the monitor function. The disadvantage of this method is that the sample rate for recording and playback cannot differ. But that is not really a problem. I just recorded at 48 kHz and recorded sine sweeps created with sample rates of 44.1 kHz, 48 kHz and 96 kHz. This way I can measure the quality of upsampling from 44.1 to 48 kHz, 48 to 48 kHz (no resampling) and downsampling from 96 to 48 kHz.
PPHS:
(http://www.devir.de/temp/sweep_pphs.jpg)
Vista:
(http://www.devir.de/temp/sweep_vista.jpg)
PPHS (ultra mode)
(http://www.devir.de/temp/sweep_pphs_ultra.jpg)
The middle image shows gaps beetwen the samples. I guess that is because foobar2000 had to change the sample rates during playback while that wasn't the case when letting the player do the resampling. There's also a glitch visible in the first image. I recorded it about 10 times but it was always there and at the same location. So I assume it's an issue with the PPHS resampler in default mode and not related to the recording process.
My conclusion is: Vista's resampler is better than PPHS in default mode but PPHS beats Vista's resampler in ultra mode. But I think it's not about which algorithm is the absolute best in this case but if Vista's algorithm is sufficient. As it is designed solely for real-time playback I would say it's good enough. What the images don't show well is the frequency roll-off. But I hope you believe me when I say that the ranking stays the same in that regard. Vista's resampler begins to roll-off the high frequencies at about 19 kHz for the 44.1 kHz sample but doesn't reach a relevant attenuation (> 1 dB) before 20 kHz. The frequencies of the 96 kHz sample are preserved up to 21 kHz. After that a steep roll-off occurs.
.
Thanks to the folks who took the time to create the nice graphs, however – what’s needed are actual sound-files for comparison. A trio of high-fidelity clips, ‘played’ via Windows Media Player on an XP box (loaded with all updates / patches – clean), vs. the same audio clips on a cleanly loaded, fully-up to date Vista box - using the same version of WMP (ver.11), and then posted online – correctly labeled (for reference) and un-labeled (for ABX).
We can all see that a great many hours have been burnt off in these discussions thus far, by a variety of parties. If folks had spent this much time posting clips for us to ABX, along with a brief outline of the hardware involved, I suspect that this conversation would have found a conclusive result long ago.
I’ll look at this thread again Monday night, if no one has posted audible clips as per above, I'll disconnect the currently install hard disk drive and pop another identical one into the (currently), 'Vista box' that I'd recently loaded as a power-cable test-unit (its very clean & up-to-date). I’ll load the new hard disk drive with XP Pro (clean / up-to-date), and spit off a couple of clips from each O.S. (30-seconds, within 'fair-use' rules), for us to audition and ABX. Folks can then post their ABX results and we can see, once and for all if identical files, on identical machines, with identical software aside from the O.S., cleanly loaded without any 'tweaking' - are audibly different from one-another.
(Yes, I’m procrastinating on doing this myself; it’s a beautiful, sunny day here in the Okanagan today and I wanna’ go for a bike ride! Additionally, my board is beside me, absorbing the wax I just melted onto it moments ago, and I intend on riding the hell outta’ the local mountains this weekend before it closes for the season – 7 days and counting!)
Andrew D.
www.cdnav.com
.
Here's a new one: "Cool, this BSOD in Vista sounds better than the one I usually get in XP!'
I would tell you, since I'm running the exact same machine same hardware as when I had XP SP2, but alas I haven't had a single BSOD in Vista since the day I installed it many months ago.
To me, best practices includes ease of use and.. stability.
Woodinville, is anyone actually using WaveRT and/or Exclusive mode?
.
Thanks to the folks who took the time to create the nice graphs, however – what’s needed are actual sound-files for comparison. A trio of high-fidelity clips, ‘played’ via Windows Media Player on an XP box (loaded with all updates / patches – clean), vs. the same audio clips on a cleanly loaded, fully-up to date Vista box - using the same version of WMP (ver.11), and then posted online – correctly labeled (for reference) and un-labeled (for ABX).
We can all see that a great many hours have been burnt off in these discussions thus far, by a variety of parties. If folks had spent this much time posting clips for us to ABX, along with a brief outline of the hardware involved, I suspect that this conversation would have found a conclusive result long ago.
I’ll look at this thread again Monday night, if no one has posted audible clips as per above, I'll disconnect the currently install hard disk drive and pop another identical one into the (currently), 'Vista box' that I'd recently loaded as a power-cable test-unit (its very clean & up-to-date). I’ll load the new hard disk drive with XP Pro (clean / up-to-date), and spit off a couple of clips from each O.S. (30-seconds, within 'fair-use' rules), for us to audition and ABX. Folks can then post their ABX results and we can see, once and for all if identical files, on identical machines, with identical software aside from the O.S., cleanly loaded without any 'tweaking' - are audibly different from one-another.
(Yes, I’m procrastinating on doing this myself; it’s a beautiful, sunny day here in the Okanagan today and I wanna’ go for a bike ride! Additionally, my board is beside me, absorbing the wax I just melted onto it moments ago, and I intend on riding the hell outta’ the local mountains this weekend before it closes for the season – 7 days and counting!)
Andrew D.
www.cdnav.com
.
Easy way.
Fix your rendering rate to 48kHz (control panel) so that you trigger the resampler.
Even worse, fix your rendering rate to 8kHz by playing a speech file.
Add a huge quotient of silence to the speech file.
While the silence is playing, play the first 30 seconds of "We Shall be Happy" by Ry Cooder, off the album Jazz.
Do the same in Vista, only leave your rendering rate at 48kHz (for both speech and whatever).
Listen to the two.
I dare say you'll notice a touch of difference here and there. Even those above your article who would rather spout professional accusations than investigate ought to be able to notice, even if they listen on a nice PWB3.0 system.
Woodinville
Hi fellows, thank you for your input thus far; I wish to make this as realistic as possible - no fixing of rendering rates, just two clean-as-a-whistle copies of Windows head-to-head.
Its now after ski-season, and I have some time on my hands this evening and the next – thus, would like to get going on this project. I have a “nothing fancy” PC on hand, consisting of a Tyan S2495 with an Athlon 2000+ processor, one gigabyte of RAM, two identical 80-gigabyte hard drives - one loaded with Vista Business Edition and the other loaded with XP Professional, all powered by an Antec ‘SmartPower II’ 350 watt power supply. Both OSs have all service packs, updates etc. loaded and I have not 'tweaked' them in any way whatsoever - so as to mimic the experience a 'Joe Public' might have with each OS. The Tyan board is hosting an Asus Xonar PCI audio card with the latest drivers for each OS.
What I am asking for is; the process. What Open Source software should I employ for the evaluation? What steps should I take in conducting this test? That is, should I simply rip a track from CD on each OS, save it as a .wav and use that for comparison purposes? Should I play a track in real-time from the CD and record it with an Open Source application, save it as a lossless .wav and use that for comparing 'fidelity' of the OSs, or ?
I don't want to start downloading and installing AISO etc. software as I wish to keep these two setups as clean as possible, allowing for, as noted earlier, a setup that would best mimic what Joe Public might employ.
We can always do a separate ‘tweaked’ setup comparison later if we like…
Any and all suggestions welcome.
Andrew D.
www.cdnav.com
If you want to compare XP with Vista then do compare them as they are. So no additional software, absolutely no AISO drivers as these are supposed to cure some problems in Windows audio design.
Make damm sure they run on identical hardware. If you have done so, make damm sure they run on identical hardware.
I wonder if running them both on the same machine using virtualisation might be an option.
Disable network connections, spyware, anti virus etc. Personally I don't believe this will do much but there are people believing that these processes might have an impact.
Run WMP11 on both.
Use exactly the same lossless audio files on both otherwise you introduce another parameter
You might try your hand on a tool like EAC but I don't think the ripping is the problem but the way XP and Vista do the playing (the Kmixer stuff).
Another thing is the audio gear. Is it able to make subtle difference audible? I know this sounds blunt so pardon me, English is not my native tonque, but if I read _nothing fancy_ does this mean the same for the audio card, the speaker systems, etc?
Roseval
Sorry if I wasn't descriptive enough. What I have is (one) PC and two (identical) hard disk drives. One hard disk drive has Windows XP Professional loaded on it; the other has Windows Vista Business Edition installed.
I will be physically removing one drive and replacing it with another, thus the OS is the only variable.
The sound card is this unit:
http://www.asus.com/products.aspx?l1=25&am...amp;modelmenu=1 (http://www.asus.com/products.aspx?l1=25&l2=144&l3=0&l4=0&model=1769&modelmenu=1)
…it’s a good as it gets for a retail / non-studio sound card. In fact, I feel that it makes sense to employ a retail card like this, as very few out there have fifteen hundred dollar studio sound cards anyway…
I like the idea of using Windows Media Player (with all updates), as its something that everyone who runs Windows has in common.
Andrew D.
www.cdnav.com
I will be physically removing one drive and replacing it with another, thus the OS is the only variable.
That's even better than using a virtual machine, this eliminates a lot of possible intervening parameters
…it’s a good as it gets for a retail / non-studio sound card. In fact, I feel that it makes sense to employ a retail card like this, as very few out there have fifteen hundred dollar studio sound cards anyway…
You do have a point. But I do have an outboard USB-DAC matching this price tag just to overcome the limitations of a standard audio card.
It depends a bit on what one wants to establish.
If the question is "is there an audible difference between XP and Vista" then the audio part must be perfect, if not it is just another parameter fouling the experiment. (You can't judge difference in top speed between 2 cars if you stick to the speed limit.)
Another question is "will the average user using the average gear hear a difference between XP and Vista". Perfectly valid question but beside methodological problems (how to establish the average user, what is the average gear) maybe a bit dull. Nobody complains about sound quality, nobody hears a difference between any OS, only audiophile fools do.
So target your experiment at the fools. This one is very interested.
Ok guys - I need some ‘tweaker-input’ here: I've got both OS's ready to rock, Windows Media Player 11 as up-to-date as possible and ready to 'Rip' and now what...
Will the process of ripping a 30-second track from CD to a .wav file be enough or should I now play back the track on one application and record / re-save it in real-time onto another application?
What’s the next move that will be 'acceptable' in the Windows XP vs. Windows Vista 'Sound' trials...?
By the way, I'm using the first 30-seconds of Pink Floyd's 'Breathe' track for this demo. Its ripped from the 25th anniversary re-mastered CD. I picked this as it has no lack of bass and treble, along with loads of midrange vocals - and pretty much everyone on this forum, and I suppose most forums, is surely intimately familiar with this music. (PS: if you aren’t – damn man, go out and buy it today – it’s a classic among classics!)
I await your input.
Andrew D.
www.cdnav.com
It's just pointless to do this kind of comparison. Also there are two major caveats: It's not a blind test and switching between the systems takes too long. Considering the subtlety of any audible differences it's impossible to get any valuable results, especially with low-end equipment.
It's just pointless to do this kind of comparison. Also there are two major caveats: It's not a blind test and switching between the systems takes too long. Considering the subtlety of any audible differences it's impossible to get any valuable results, especially with low-end equipment.
I believe the intention is to record the output from the two configurations, then do an ABX test on a known good configuration (such as KS on XP, or whatever the equivalent is on Vista). If this isn't the intention, then no - testing won't be particularly interesting.
Perhaps the most direct method of creating test files for ABXing would be play a high quality reference file on the pc, and simultaneously record the sound mix using the pc. As this was occuring, a sound source in a different format (perhaps containing virtually no data or just intermittent data) would be introduced into the sound mix. I gather that XP may perform less well than Vista in this scenario.
If using this method, one difficulty might be how to manage to get the timing the same under both XP and Vista for introducing the supplementary sound sources. However if the extra sound source was low level or intermittent this might not matter.
Others may have a specific suggestion as to what formats to use in the mix.
That is, should I simply rip a track from CD on each OS, save it as a .wav and use that for comparison purposes?
Should I play a track in real-time from the CD and record it with an Open Source application, save it as a lossless .wav and use that for comparing 'fidelity' of the OSs, or ?
I don't think ripping is the problem, as far as I could judge there are no indications that XP or Vista differs in this matter.
If I understand correctly there are some flaws in XP design. This has a negative influence on sound quality (the Kmixer stuff).
The audio part of Vista is redisiged and is supposed to be without the flaws of XP so the claim is that it sounds better.
So this is about playing music on both OS.
I don't think it will makes a difference if one uses a CD or a WAV as long as it is a uncompressed source, so we have the best possible quality input source.
So the question is how to capture the output of both configurations in such a way that
- it is certain that nothing but nothing in the sound system of both OS is bypassed
- it can be distributed as a file
- the capture method don't have an influence ont the result.
In principe the signal right before it enters the soundcard is ideal (I2S?) but don't ask me how to do it or how to listen to it.
An AD converter might be an option but then the quality of the soundcard will have an influence and also the quality of the AD-converter.
But even if you managed to capture both sources perfectly, what are you going to listen to them on?
If you listen to them on a PC running XP then you're going to have the same XP-related problems playing back both sources. Admittedly, the problem would only be one-fold when playing back the Vista source on an XP system but would be two-fold when playing back the XP source.
Playing the two files back on a Vista-based system would still double any remaining errors normally present on the Vista-based system (unless we are happy to make the blind assumption that there are none) and give you the errors produced by the Vista-based system plus the errors from the XP-based system when playing back the XP source.
You won't get a true A-B comparison whichever way you do it.
Cheers, Slipstreem.
.
ABX is easy - as long as one is running Vista, Mac OS, Linux or perhaps UNIX.
And its only natural that you lads should have at least one Linux box running in the house :-)
Heck, the OS is free, and there isn't likely one among us who doesn't have an 'old' PC in the basement / closet / second-bedroom / den etc.
Here is your gateway to quality computing (download and enjoy today!)
www.ubuntu.com
Andrew D.
(http://digitalbackcount.setupmyblog.com/wp-content/uploads/2007/05/ubuntu_logo.png)
You could listen to them with an external DAC and a soundcard which does bit-perfect output. Another option, as digital mentioned, is to use Linux - but be sure to use the hw outputs at a rate the soundcard natively understands. With the default ALSA outputs you still end up with resampling (at least with the ALSA config that Ubuntu uses). KS on windows XP could be another option to obtain output without OS interference.
The issue is that Xp has the kmixer and Vista improved the audio stack.
But the Kmixer can be bypassed as one poster has mentioned recommending Foobar and Kernel streaming.
Why not do an apples to apples comparison and compare Vista with Xp bypassing the kmixer?
That analogy of comparing cars comes to mind, except one car has to stick to the speed limit and the other one doesn't....hardly a fair test.
The issue is that Xp has the kmixer and Vista improved the audio stack.
But the Kmixer can be bypassed as one poster has mentioned recommending Foobar and Kernel streaming.
Why not do an apples to apples comparison and compare Vista with Xp bypassing the kmixer?
That analogy of comparing cars comes to mind, except one car has to stick to the speed limit and the other one doesn't....hardly a fair test.
So, what you're saying is test one as tweaked, and one as standard package? Compare the amp'ed to the normal?
I guess you have to stretch a bit to explain that one.
Using a tweaked XP and a stock Vista is like comparing an apple to a grapefruit. It's silly, and meaningless. And it better not be much different then, eh?
Use it like any standard user would use it. I doubt if any here of us use anything quite like the standard user.
.
I’ve been spinning this one around and around in my head for some time, and of course, the solution appears to be the simplest – always overlooked one:
What if I just run a quality-built cable from the line out jack, right back into the line in jack while playing the file from CD in Windows Media Player and recording it with Audacity. I don't know if this is a common ability with soundcards, but it certainly is with the Asus Xonar I'm using for these evaluations.
I’ve tried it and the file is reproduced faithfully, allowing me to capture and save a lossless .wav file.
Does anyone here have an issue with this method? It appears to me to be the only way to capture the ‘played’ file so that we can all give it a listen.
Would really like to hear from any Audacity developers if any exist on this forum…
All of the hardware is ready to go, both drives are loaded and fully up-to-date (no modifications or tweaks applied), and I’ve picked the music: Tracy Chapman’s HDCD recording of ‘Fast Car’.
Andrew D.
PS: Those wanting to try the audition on Ubuntu might be interested to learn that the latest version of the OS allows you to do a 'virutal install', which doesn't modify any existing hard disk drive files and can be 'uninstalled' without leaving a trace!
Learn more here:
http://lifehacker.com/software/ubuntu/inst...dows-228956.php (http://lifehacker.com/software/ubuntu/install-and-run-ubuntu-without-disturbing-windows-228956.php)
If you're stuck on Windows love - there is always this (free) option:
http://www.microsoft.com/windows/products/...ault.mspxAndrew (http://www.microsoft.com/windows/products/winfamily/virtualpc/default.mspxAndrew) D.
www.cdnav.com
.
Ok boys - here she is: One PC, two hard disk drives.
Drive one installed and loaded with Windows XP Pro / all updates / patches installed / fully defragmented, rebooted clean – no tweaks applied.
Drive two installed and loaded with Windows Vista Business / all updates / patches installed / fully defragmented, rebooted clean – no tweaks applied.
The drives were physically installed / un-installed independent of each other; no ‘dual-boot’ or ‘Boot-order in the BIOS’ tricks here.
After throwing out a number of queries, looking for suggestions on how I might go about this process, I decided to conduct it as follows: Each file was ripped with Windows Media Player (Version 11.0.5721.5230 on XP --- 11.0.6000.6344 on Vista), saved to hard disk drive (identical model drives as well), and played back with WMP 11 while being simultaneously recorded using Audacity (Version 1.2.6).
I ran a high-quality, shielded line from the line-out, right back into the line-in on the Asus Xonar audio card (running driver version 6.12.8.1731) in order to create an environment where the file was ‘played and captured’. I don’t know if this is a common feature with audio cards, but the Asus allows it – I believe its part of a little-publicised ‘feature’ called ALT – on the Xonar cards – offering a way to circumnavigate DRM encrusted files…
Thus, original CD playback / ripping / saving / playback and subsequent re-recording was accomplished on two completely separate operating systems and the resultant files from each are supplied here for your musical enjoyment and comparison. The rest of the PC hardware is detailed earlier on in this thread – suffice to say, it’s all kick-ass / high-quality and as electrically isolated / shielded as possible.
ABX the living hell outta’ them and decide if indeed there is an audible difference between two ‘stock’ editions of Windows XP / Windows Vista.
Warning: those with dial-up modems might not wish to play this game: each file is exactly 10 megabytes in size.
_______________________________________________________________________________________
I'll have to post the 'answer' I've been promising for so long on my own forum as the length of the sample files are over the 30-second limit as stated in the Terms Of Service HA. I've recorded a full one-minute of each track (on XP and Vista) in order to assist in concentrating on the differences, if any, between the files.
If this (redirect) is still against the rules, I ask that the Mods' here please PM me and I'll (sigh), cut them down to 30-seconds... Here is a link to my small, non-profit discussion forum, along with said files (http://cdnav.com/cdnav/viewtopic.php?p=7879#7879)
Andrew D.
...
Were you careful to ensure that in each case, you had a different rate stream running before you started the main signal?
It can be a zero signal.
No tweaking / software adjustments - this was a 'factory-settings' comparison. The Operating Systems / Applications were 'clean' - no mods' or 'improvements' whatsoever.
The whole plan was to see what 'Joe Normal' would find with an 'Fresh Out Of The Box' experience.
Andrew D.
PS: Anyone feel they hear any differences?
So, what you're saying is test one as tweaked, and one as standard package? Compare the amp'ed to the normal?
I guess you have to stretch a bit to explain that one.
Using a tweaked XP and a stock Vista is like comparing an apple to a grapefruit. It's silly, and meaningless. And it better not be much different then, eh?
Use it like any standard user would use it. I doubt if any here of us use anything quite like the standard user.
Did you read your last sentences Wood??? They contradict each other. I agree with the latter. Few here are standard users. So why test it out of the box then??
All I am saying is that Vista supposedly did away with the Kmixer. But if you can bypass the Kmixer in XP that is what you should compare with Vista...which one allows you a truer and better signal.
Test Vista with Direct sound Vs XP with ASIO or Kernel streaming...no other tweaks. Or if you like, use ASIO or Kernel streaming in Vista if you want.
No tweaking / software adjustments - this was a 'factory-settings' comparison. The Operating Systems / Applications were 'clean' - no mods' or 'improvements' whatsoever.
The whole plan was to see what 'Joe Normal' would find with an 'Fresh Out Of The Box' experience.
Andrew D.
PS: Anyone feel they hear any differences?
Exactly. And what Joe Normal hears is streams being mixed together.
So have you constructed a topical experiment or not? Sounds like not.
Supporting confirmation bias rarely shows anything interesting.
Exactly. And what Joe Normal hears is streams being mixed together.
So have you constructed a topical experiment or not? Sounds like not.
Supporting confirmation bias rarely shows anything interesting.
I don't know if that is fair. The use case that this tests is, in my opinion, the most common - the user boots up, starts a player (iTunes, WMP, etc) and listens to their MP3s. Sure, there are use cases where your criticism is fair, but I think this one is the most obvious to test.
I think this is a very interesting test, thanks for organising it, digital.
Is someone changing the goal posts again?
If audio is played with the best respective settings on Vista and XP, there will be no distinguishable difference for identical files.
Ok, please show YOUR evidence for that.
Is someone changing the goal posts again?
If audio is played with the best respective settings on Vista and XP, there will be no distinguishable difference for identical files.
Ok, please show YOUR evidence for that.
You know, it's rude and deceptive to pick quotes out of context and represent them as something they weren't.
Woodinville
Correct me if I’m wrong, but I think what you are suggesting is that I should be tweaking the O.S.’s before the 'test' files are created. I don’t feel that this would represent an honest comparison of the two Operating Systems ‘Out Of The Box’ experience.
That said – I am prepared to tweak the living heck out of each O.S. and re-do the files. If some of the more ‘tweaker’ members would present me with a succinct / detailed strategy for said ‘re-do’, I would be more than happy to oblige.
I don’t mean to be rude, but if you look at the original posters query is was:
First up let me say that I'm still running Windows XP and I have no first hand experience with Vista.
I've noticed that some people making claims that "audio" is much better in Vista, though what specifically they are referring to I don't know. The more outrageous claims go something like "with Vista your old onboard sound will sound as good or better then the best PCI soundcard". Of course I think it's a load of bull, but what do the changes to audio in Vista really amount to?
There was no mention whatsoever of tweaking.
With regards to personal bias. I truly don’t have any with regards to audio on either O.S. I’m a Microsoft Certified Software Technician (‘Sales & Support of 32-Bit Operating Systems’), and a Computer Trade Industry Association Certified Hardware Technician.
Andrew D.
www.cdnav.com
Well, I think I know what will be the outcome of the test.
If you are just playing one sound alone, with the sound card clock set at the source sample rate (44.1 KHz), which I guess is the default behaviour of the sound card, both systems will sound the same.
But if you play simultaneously another stream at different sample rate, or the sound card is not set at the source sample rate, XP may sound different (worse), depending on the source material used and the difference on sample rate on the streams played.
So, the question is, ¿which of both cases is "Joe Normal" case? I guess most times is the first case, but the second case will happen too sometimes.
So, the question is, ¿which of both cases is "Joe Normal" case? I guess most times is the first case, but the second case will happen too sometimes.
Then testing the second case would be a good idea too. Digital, would it be possible for you to test that? One of the members who knows more about Windows than me could probably advise you on a simple way (which could happen to "Joe Normal") which would force Windows to resample.
Studied the topic a lot. A summary of what I've gleaned from multiple sources, including Microsoft tech' sites is as follows:
The issue with KMixer only occurs when one triggers a 'secondary' sound event - such as one of the Windows System Sounds - while a 'primary' audio file is being played in the foreground. KMixer then resets itself to the 'correct' sample rate within moments of the (in our example) System Sound ending.
Thus, I'm not sure it’s truly a problem. For all of you audio geeks out there (myself included, not a slight), simply turn off the annoying System Sounds in the first place (Control Panel / Sounds & Audio Devices / Sounds Tab / Select No Sounds). They are kinda’ childish and a waste of resources anyway…
I'm not sure that a scenario of individuals playing multiple audio files occurs very often (System Sounds aside). I've been a Windows user since Windows Version 3.0 (all hail i8088 / i8086!), and don't find that this occurs very darn often. Even when it does, the 'reset' generally occurs quickly.
However, that said, give me a few weeks and I'll try it out as the basis for another evaluation - just too far over my head in clients at the moment. How shall I do it? What immediately comes to mind is recording a long .wav file of ‘dead-air’ and saving it, then assigning it a function in System Sounds. I’d follow by triggering said System Sound in the background while simultaneously playing and recording a high-quality file in the foreground, following the same recording procedure as the last round. What say ye?
To be honest, I’d say we’re getting a little far-fetched with our scenarios here, as this particular event would never occur in ‘real-life’…
Andrew D.
www.cdnav.com
I'm sure I've experienced this problem when listening to a BBC stream and playing Foobar2000 (I'd switch between the two). Sometimes the stream sample rate quality would drop and it would screw the rest of the sounds in Windows. It only ever happened a handful of times and it was sometime ago now.
The quality degradation was no slightly less bass and less bright highs kind of thing. It sounded awful.
You know, I'm going to bail on this thread. I have described, repeatedly, a common occurance for average listeners. Yes, I do know that's the facts.
I've been accused by SLD of being an incompetant non-scientist, and I'm waiting for him to support those extremely serious professional accusations so that we can take them forward to an ethics hearing, since his accusations are fully serious enough to require exact formal treatment. SLD, however, simply appears unwilling to back up his serious accusations, has shouted his accusations, and then run into hiding.
On top of all that, as you can see by my profile, I've left MS for greener pastures, and I really don't think it's appropriate to be discussing their issues any longer.
Ergo, I am walking on this thread.
Woodinville
I was a bit puzzled about what could possibly be the differences between XP and Vista.
Your contribution to this thread made matters much clearer to me.
Thanks
All
What I understand up to now.
If you play a single stream at its original sampling rate, XP and Vista sounds the same
If you play a single stream and apply upsampling there might be a difference as Vista uses float and XP integer
If you play 2 (or more) streams and these streams have a different bitrate, XP will play all streams at a the lowest bit rate, Vista will play each stream with its own bit rate
If you play a single stream at its original sampling rate, XP and Vista sounds the same
Not always. Your soundcard needs to support the sample rate of the sound you are playing, and some other conditions need to hold. This is in contrast to what happens on some linux distros (like Ubuntu), where resampling is always used.
If you play a single stream and apply upsampling there might be a difference as Vista uses float and XP integer
It's deeper than float vs. integer. Vista uses a higher quality algorithm, at the expense of CPU cycles. As far as I am concerned, this is a sensible thing to do.
For the record: it looks like Woodinville knows his stuff, check out his BIO:
http://www.neuralaudio.com/pr2008_0407.html (http://www.neuralaudio.com/pr2008_0407.html)
Andrew D.
Is someone changing the goal posts again?
If audio is played with the best respective settings on Vista and XP, there will be no distinguishable difference for identical files.
Ok, please show YOUR evidence for that.
You know, it's rude and deceptive to pick quotes out of context and represent them as something they weren't.
Hear, hear!
Wasn't my final post in this thread enough, by the way? Was it a must to get me to clarify myself in a PM when my firm stand and conclusions have all been stated in this thread?
Gentlemen (and ladies, if any), I refuse to budge from my stand as quoted above, because it will be nonsensical to ABX lack of difference, and we all know the power of placebo in wanting to hear differences. In addition, Hancoque has provided visuals in post #72, with a corresponding final paragraph. His final paragraph essentially states that both Vista's default resampling algorithm and PPHS Ultra are so good, a human can't hear a difference between them. A
liar would state that Vista's default is clearly better than PPHS Ultra, and a deluded audiophile would state that PPHS Ultra is clearly better than Vista's default. I hope this thread doesn't make a parrot out of me.
For the record, here are my best practices on XP: Foobar2000 0.9.5.2 beta, Kernel Streaming, PPHS Ultra.
These were my best practices on Vista: Foobar2000 0.8.x, DirectSound.
With regards to Woodinville's stand, please refer to post #66. And oh, the words are in bold. I don't know why I bolded the words... perhaps I wanted people to actually notice them?
With regards to Woodinville's stand, please refer to post #66. And oh, the words are in bold. I don't know why I bolded the words... perhaps I wanted people to actually notice them?
Your failure to reply substantively is noted.
Allow me to quote the offending claim:
I can see that he is quite challenged in the areas of imagery and vocabulary though. He cannot visualise and comprehend 'best practices'!
----
I never wanted to be nasty about this, but you leave me no choice... if Microsoft holds a public survey on 'most loyal employee', remind me to vote for you.
Any professional engineer (I use the term in the degreed or highly experienced, technical, rather than artistic sense here) must show superior ability in imagery and vocabulary. First off, you accuse me of incompetence in that sentence alone. Then, you say "best practices" which is a direct reference to how one proceeds to run, document, and write up experiments. The term "best practices" is not defined by a few guys on a bulletin board,
it is a formal term describing how one carries out work professionally.
Finally, with your crack of "most loyal employee" you insinuate that I am lying in order to demonstrate some kind of loyalty or something of that sort. Given the situation at the time I wrote my previous replies, that particular accusation quite ridiculous. In some ways, that accusation is the most insulting, and given that you stipulate by your own words that you state it by malice, perhaps the worst of your misconduct.
Show your evidence.
Any professional engineer (I use the term in the degreed or highly experienced, technical, rather than artistic sense here) must show superior ability in imagery and vocabulary. First off, you accuse me of incompetence in that sentence alone. Then, you say "best practices" which is a direct reference to how one proceeds to run, document, and write up experiments. The term "best practices" is not defined by a few guys on a bulletin board, it is a formal term describing how one carries out work professionally.
I'm a layman and I can damn well define words temporarily to suit my case and push my point as and when necessary. I treat my music professionally, hence the term 'best practices' is valid as used by me. I perform the same repeatable setup on all virgin installations of Windows XP after thoroughly researching the use of bit depth, resampling and output options on Foobar2000. Are you accusing me of not taking my music playback seriously?
As a professional, the onus is on YOU to correct me, the ignorant layman, as to the correct usage of words in your industry. If you fail to do that early, the fault lies with you and on you for not passing on the requisite knowledge.
Finally, with your crack of "most loyal employee" you insinuate that I am lying in order to demonstrate some kind of loyalty or something of that sort. Given the situation at the time I wrote my previous replies, that particular accusation quite ridiculous. In some ways, that accusation is the most insulting, and given that you stipulate by your own words that you state it by malice, perhaps the worst of your misconduct.
Show your evidence.
Given the situation? Given the situation, you were insisting that you could hear a difference between the best possible playback setups on XP and Vista.
With no evidence. With CONTRARY evidence (post #72). You call that professional?
Your obstinacy would naturally call into question your integrity, whether your loyalty lies with your profession, or your employer. Perhaps you will want to step back from this petty debate and read your own words all over again. Instead of the one party being malicious, it may well be the other party being a jerk without realising it.
This thread has gone way off-topic because of an individual's refusal to accept the facts (I do not dare use the word 'truth' lest he play the religion card again).
This thread has gone way off-topic because of an individual's refusal to accept the facts (I do not dare use the word 'truth' lest he play the religion card again).
I implore you to look at the first post. It was your insistence on a special case that drove this thread off topic.
It's pretty fitting that you started your reply saying that you can bend things to suit your fancy and that is just what you have done with this topic.
I'm a layman and I can damn well define words temporarily to suit my case and push my point as and when necessary.
Yes, you've used the words in a way to deceive everyone here. Effectively quitting my interest from hearing more from you.
I'm not going to recheck what did everyone said, or why we've ended here, but I've learned enough: Vista default audio quality is better than XP default audio quality,
especially specifically playing 44Khz audio (pretty much everything feed for quite some years already) with AC-97 soundcards.
I don't bother to hear anything else.
[Edit -> especially -> specifically ]
Given the situation? Given the situation, you were insisting that you could hear a difference between the best possible playback setups on XP and Vista.
Once again, you show contempt for the facts of the case. You haven't read the various user experiences reported with XP, I have.
You insist on your own special case for this and that, I refer, and have made it clear that I refer, to the out-of-the-box user experience, which is what the layman (which is what you have just called yourself) generally experiences.
The conclusions?
1) You aren't a layman. You insist on defining "best practices". That's not a layman's activity. Yet you wish to be instructed. As evidence shows, you reject instruction. This is, at best, intentional ignorance.
2) You reject "out of the box". Yes, that's a wise thing to do, but that's not what the LAYMAN does. so your actions, as well as your posing and words, demonstrate that you wish to be treated as a layman, while making substantive pronouncements about professionals.
3) You persist in making unsubstantiated accusations, over and over and over, and you persist in trying to twist the context to fit your agenda, rather than accept the context of the comments.
4) You threaten people with TOS #8 and try to extort their silence, and are profoundly angry and upset when they do not kowtow to your extortion.
5) You refuse to retract your accusations even when you yourself have stipulated malice in making them.
I think that you wish to deny some obvious facts, that you have trouble with authority, and that you really, unwisely, do not know when to admit that you're just full of yourself.
It's telling that you accuse me of egoism. I've earned my position, and I back my words with experience and testable, verifiable experiments. (which nobody has bothered to do, which is not my problem, of course, most informed people are well aware of the vagarities of kmixer) You, on the other hand, define "best practice" but insist that you're a layman. You jump into a discussion of out-of-the-box layman usage, and demand to hold the layman to professional standards, but all the time passing off your "layman" behavior as needing instruction.
It would be impolitic for me to suggest the kind of instruction that you need.
Jaz,
With all due respect, I don't feel that your statement:
“but I've learned enough: Vista default audio quality is better than XP default audio quality”…
...is rational / valid, unless you can back it up with posted ABX results. If you look at my post number 91, you will find a link which will allow you to download / compare / post ABX results using two very carefully created files, 'no-tweaks-applied' / ‘clean install out of the box’; one on Windows XP, the other Windows Vista.
From what I can gather [with clean installs of the two operating systems]; as long as there are no other sound-effects or audio files playing the background, there is no discernable difference in the reproduction of digital files under either of the two Windows Operating Systems.
Andrew D.
www.cdnav.com
Is it not possible for us all to agree that the two arguments of technical superiority and a perceived improvement in audio quality don't necessarily have to go hand-in-hand?
We don't all have ears like bats and it can't automatically be assumed that everyone (or maybe even anyone) will hear a difference between the two systems if they both achieve near-perceptual transparency under single-source listening conditions which is, after all, how the vast majority of users use their systems for the vast majority of the time.
I'd like to respectfully propose that this thread be locked as I don't think it's doing the usually very high reputation of the HA Forums any good. It's just descending into a slanging match with no further technical information of any value coming to the fore.
Cheers, Slipstreem.
Is it not possible for us all to agree that the two arguments of technical superiority and a perceived improvement in audio quality don't necessarily have to go hand-in-hand?
On the other hand, it does not take peer-reviewed large-group ABX testing for people.. even the layman.. to pronounce correctly that HD Radio FM sounds better than legacy AM.. and the layman would be correct. Technical superiority
does improve audio quality quite often.
1) You aren't a layman. You insist on defining "best practices". That's not a layman's activity. Yet you wish to be instructed. As evidence shows, you reject instruction. This is, at best, intentional ignorance.
What's to stop me from taking standard English words and defining them clearly for my own purposes until someone comes along to correct me? If the words have long been used for another meaning in the industry, I stand corrected. But if that someone intentionally ignores my definition up to the point of correction... I call that malice.
2) You reject "out of the box".
I
rejected "out of the box"? Did you read post #66?
3) You persist in making unsubstantiated accusations, over and over and over, and you persist in trying to twist the context to fit your agenda, rather than accept the context of the comments.
I am sorry, for a long time now you have been the one with the agenda. I sense that it is not victory enough for you to have everybody, including me, acknowledge that out-of-the-box Vista is better than out-of-the-box XP. You are trying to get everybody, including me, to acknowledge that out-of-the-box Vista is better than best-possible XP, even in the face of post #72.
Notice that we haven't even gotten started on audible superiority, just technical superiority.
4) You threaten people with TOS #8 and try to extort their silence, and are profoundly angry and upset when they do not kowtow to your extortion.
ToS #8 keeps people on their toes. Some continually defy ToS #8 and live to regret it. If ToS #8 has not been violated, the thread proceeds on.
5) You refuse to retract your accusations even when you yourself have stipulated malice in making them.
First, live up to your reputation as an objective professional. Acknowledge post #72. If HA.org is an objective forum then all the more I tend to suspect commercial infiltration. You have to forgive my petty paranoia.
I implore you to look at the first post. It was your insistence on a special case that drove this thread off topic.
I'm sorry... my special case is invalid, then, no? I did persist for a long time, but come on, are you telling me that there are people in here holding out against the facts for as long as I have persisted? There is post #72 for everybody to read... why am I still being harassed into providing evidence for my special case, when all the time the onus is on Woodinville? As of this post, he is still beating a dead horse, because post #72 exists.
When Woodinville acknowledges post #72, further progress can be made in this thread. The superiority of default Vista over default XP has long been acknowledged by everybody in here,
including me (post #66). Even then, the quality improvements may only be audible under certain technical conditions. I'm surprised that I have been the only one claiming ToS #8 over audible claims, and even more surprised it took us so long for a post to mention the technical conditions.
Yes, you've used the words in a way to deceive everyone here. Effectively quitting my interest from hearing more from you.
I'd like to respectfully propose that this thread be locked as I don't think it's doing the usually very high reputation of the HA Forums any good. It's just descending into a slanging match with no further technical information of any value coming to the fore.
Cheers, Slipstreem.
Seconded.
If people cannot understand my words, then I have failed to communicate my ideas across as well. I'll wait for the day when fellow English forumers can understand standard English that has been defined clearly (not some profession's definition of certain terms). In the meantime, I will like to request for this thread to be locked too.
On the other hand, it does not take peer-reviewed large-group ABX testing for people.. even the layman.. to pronounce correctly that HD Radio FM sounds better than legacy AM.. and the layman would be correct. Technical superiority does improve audio quality quite often.
Not in the realm of diminishing returns. What was the last conclusion of 24-bit vs 16-bit for music playback?
Gentlemen;
You’ve all been on H.A. long enough to know that the way we solve ‘the unknown’ around here is with ABX. Up to this point all you’ve done is made the pages of H.A. look like the insane asylum; or should I say ‘The Audio Asylum’ – where difficult concepts are never resolved, instead – they spiral out of control until the thread is locked by a moderator.
Is that what you hope to accomplish here?, cause’ you’re sure headed in that direction on an out of control freight train! Ranting, raving, sniping at one another until none of us learn anything because the thread is locked has never – in the history of A/V discussion forums, solved anything.
Now climb off your high horses and create some evaluation files for us to audition or use the ones I’ve supplied.
Have some respect for the excellent – progressive and realistic forum this is – don’t try to turn it into one of the 100+ A/V forums full of audiofools that revel in spewing disinformation and madness…
Andrew D.
www.cdnav.com
Jaz,
With all due respect, I don't feel that your statement:
“but I've learned enough: Vista default audio quality is better than XP default audio quality”…
...is rational / valid, unless you can back it up with posted ABX results.
I should have written "specifically", not "especially". I have a Vista here with an HD Audio soundcard, so i cannot verify that claim right now via ABX, but note that your soundcard didn't meet that scenario either.
Also, it has to do with the default resampling method of windows XP. (Placed in audio properties.. can't remember the exact place now) At some point, that was changed from "fast" to "accurate". Specifically with "fast",
it was bad, as in causing aliasing. it caused aliasing (of course, in samples that aliasing could be heard)
Andrew, I don't understand.
We have already resolved everything, now it is up to the individual reader to find them... or really have we? We have Woodinville stating that default Vista triumphs over default XP technically
when there is mixing and resampling involved with a few others backing him up. I finally understand the technical details (better resampler, no more lowest common denominator, etc). But as for audible differences, all we have is Woodinville's citation of user anecdotes.
Gentlemen, some are quick to retort that if there are audible differences, there is no need to perform an ABX test to confirm them. However, if there ARE audible differences, wouldn't an ABX test be exceedingly easy to pass?In other words, "come on, you have better ears, can't you just indulge the eyes of your fellow forumers in the name of science?"
Also, Woodinville has been quick to assure us that even without the mixing and resampling, default Vista still trumps default XP, citing user anecdotes, including post #18 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=55459&view=findpost&p=553424).
There has been no ABX tests performed so far.
We have Hancoque's graphs (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=55459&view=findpost&p=554264) detailing the slight
technical superiority of an optimised XP setup over default Vista. I stated that this is as good as saying no one can hear a difference, and Woodinville disputes me without explicitly stating the counter-case.
There has been no ABX tests performed so far.
I am afraid that some people simply refuse to read for themselves. That said, I previously mentioned that the best selling point of Vista's audio system for me was the per-application control. I am also relieved that this independence extends to the fundamental way Vista treats audio streams now. This is undoubtedly thanks to Woodinville's efforts. That does not mean, however, that he can throw his weight (http://www.neuralaudio.com/pr2008_0407.html) around on a forum where scientific principles are adhered to. By no means does a validation of the default case imply that the 'special case' obeys the same conclusion.
By the way, I'm still surprised no one else bothered to confront the 'audiophile speak' in post #18 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=55459&view=findpost&p=553424). I'm also surprised that up till this point, Woodinville has chosen to go on a witch-hunt instead of categorically backing up all his assertions with concrete evidence, in attempting to salvage his reputation from my (harsh, but up till now still justified) comment about his previous employer.
Jaz, the default resampling setting in XP (at the least, SP2 for sure) is set to 'Best', not 'Fast'. I believe that when comparisons were made, SP2 was the XP version involved.
Have some respect for the excellent – progressive and realistic forum this is – don’t try to turn it into one of the 100+ A/V forums full of audiofools that revel in spewing disinformation and madness…
We will all do well to keep this in mind. I have seen my fair share of unqualified morons taint this forum. I certainly expect qualified individuals to live up to their qualifications, as well as intellectual superiority. Intellectual superiority carries (or at least is supposed to carry) with it a heightened sense of
objectivity.
--------------------------------------------------
Andrew has already set up test files (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=55459&view=findpost&p=558870) for any and everybody to crack their ears on.
--------------------------------------------------
foo_abx 1.3.3 report
foobar2000 v0.9.5.2 beta 2
2008/04/19 19:24:34
File A: C:\Documents and Settings\Shaun\desktop\First Minute - Exactly at 60 WinVista Business.wav
File B: C:\Documents and Settings\Shaun\desktop\First Minute - Exactly at 60 WinXP Pro.wav
19:24:34 : Test started.
19:27:29 : 00/01 100.0%
19:27:48 : 01/02 75.0%
19:28:07 : Trial reset.
19:32:24 : 01/01 50.0%
19:33:07 : 01/02 75.0%
19:34:30 : 01/03 87.5%
19:36:32 : 01/04 93.8%
19:37:54 : 02/05 81.3%
19:38:27 : 03/06 65.6%
19:39:07 : 04/07 50.0%
19:39:54 : 04/08 63.7%
19:40:44 : 05/09 50.0%
19:41:26 : 05/10 62.3%
19:41:57 : 06/11 50.0%
19:42:28 : 07/12 38.7%
19:43:07 : 08/13 29.1%
19:45:26 : 09/14 21.2%
19:46:25 : 10/15 15.1%
19:46:58 : 11/16 10.5%
19:47:23 : 11/17 16.6%
19:48:59 : 12/18 11.9%
19:49:20 : 13/19 8.4%
19:50:12 : 13/20 13.2%
19:52:34 : Test finished.
----------
Total: 14/22 (14.3%)
No ReplayGain (can't replaygain wav files, but track scan shows identical volume), DSP settings were Convert Mono to Stereo, PPHS Resampler (Ultra).
Equipment: Mylarone X3i
Output: First ~15 trials DirectSound, next ~7 trials Kernel Streaming (sorry, I know this is bullcrap methodology, let me explain)
I actually heard the difference more with DS than with KS (hence the shrinking p-value). But I had to crank up the volume to an uncomfortable level (especially for an in-ear), and the difference was so slight I didn't manage to push the p-value below 0.01 (1%). I'll try this later at night when ambient noise is lower.
The part that I was analysing was 0:27, the cymbals on the opening of the next stanza, right on the bass guitar note that succeeded a string of bass notes and a pause. Sorry for my music illiteracy...
I listened to samples made by Digital.
I couldn’t detect any difference.
Option 1: there are no differences between XP and Vista. I have heard it my self! Anybody who hears a differences is a fool ( or tries to sell a product)
Option 2: On my equipment, using the samples supplied, my hearing is not able to detect a difference.
Option 1 seems to be the more popular one, option 2 is the valid one.
Does Option 2 apply to Joe Normal?
Does it apply to you?
According to my ABX results, yup. I probably need funding for better equipment, too. That makes me Joe Normal x 2. And if the majority is Joe Normal... Option 1 is true too, without the straw men.
Come on guys, we are 29 short of the sample size needed for a normal distribution... ok, half-jesting.
The next set to ABX must trigger the mixing/resampling capabilities of both XP and Vista.
The last but not least set to ABX pits the best-possible XP playback mode against default Vista playback mode.
As a relative newcomer to this forum, I do not know some of the basic do's and dont's about successful use of XP or Vista for playback or for recording. I haven't discovered anything in the FAQs. Is anyone prepared to start a new thread along the following lines?:
[blockquote] Tips for maximising recording or playback audio quality when running (a) Windows XP, (b) Vista, and © any other operating platform.[/blockquote]
I note that, particularly for recording purposes, some people out of curiousity or an abundance of caution would like to know technically superior settings, even if such settings might not make any ABXable audible diifference.
But if a particular setting or technique could actually make an ABXable audible difference, then that obviously would be of considerable interest and importance.
My own experience is quite limited. However, I recently had occasion to use an Audigy 4 hub to make recordings of a stereo signal. I found I got a lower noise when recording if I used a driver called 'ASIO for all' than the standard multi-channel Creative Audigy 24/96 driver. This may be a somewhat specialised situation.
I'm sure a number of members would have tips that would be of general interest and practical use. Does anyone wish to start the ball rolling?
Gentlemen; thank you very much for 'righting the ship'. I know I came across as ‘and who made you god?’ on that last post; its just that I am so damned happy with the high-reality-content on this particular forum compared to the vast majority of A/V discussion forums (I have a list of well over a hundred on my site – and at least 95% of them are chalk-full of excrement artists…).
I’ve also ABX’d my files and cannot discern with more than a hair over 50% confidence which is which. What I had expected to hear were amplitude differences, but no-go.
With regards to MLXXX’s query (BTW ML’ – welcome to the Internet’s most intelligent A/V discussion forum!) - Check out this link:
http://extra.benchmarkmedia.com/wiki/index...k_-_Setup_Guide (http://extra.benchmarkmedia.com/wiki/index.php/Computer_Audio_Playback_-_Setup_Guide)
You’ll come across some pretty crazy suggestions on the Net’ regarding tweaking PCs’ for audio, including nut-cases who tell you to strip out RAM, mess with BIOS settings and under-clock your CPU… but the link supplied is by far the best / most intelligent / logical and easy to follow suggestions you’re likely to find anywhere. Kudos to Benchmark Audio.
Andrew D.
www.cdnav.com
I listened to samples made by Digital.
I couldn’t detect any difference.
Option 1: there are no differences between XP and Vista. I have heard it my self! Anybody who hears a differences is a fool ( or tries to sell a product)
Option 2: On my equipment, using the samples supplied, my hearing is not able to detect a difference.
Option 1 seems to be the more popular one, option 2 is the valid one.
Option 3, you did not create a situation commonly experienced by out-of-the-box users on XP.
That's the facts.
Anyone demanding an ABX test is arguing that one can not ABX 8kHz sampling vs. 44.1 kHz sampling with a full-band 44.1 signal as input.
Who here is arguing that, please?
That, in a nutshell, is what it all comes down to.
Gentlemen; thank you very much for 'righting the ship'. I know I came across as ‘and who made you god?’ on that last post; its just that I am so damned happy with the high-reality-content on this particular forum compared to the vast majority of A/V discussion forums (I have a list of well over a hundred on my site – and at least 95% of them are chalk-full of excrement artists…).
I’ve also ABX’d my files and cannot discern with more than a hair over 50% confidence which is which. What I had expected to hear were amplitude differences, but no-go.
With regards to MLXXX’s query (BTW ML’ – welcome to the Internet’s most intelligent A/V discussion forum!) - Check out this link:
http://extra.benchmarkmedia.com/wiki/index...k_-_Setup_Guide (http://extra.benchmarkmedia.com/wiki/index.php/Computer_Audio_Playback_-_Setup_Guide)
You’ll come across some pretty crazy suggestions on the Net’ regarding tweaking PCs’ for audio, including nut-cases who tell you to strip out RAM, mess with BIOS settings and under-clock your CPU… but the link supplied is by far the best / most intelligent / logical and easy to follow suggestions you’re likely to find anywhere. Kudos to Benchmark Audio.
Andrew D.
www.cdnav.com
Now, create some samples that reflect real usage on XP. Otherwise, like I've said, you're arguing that there is no audible difference between 8kHz sampling rate and 44.1 kHz sampling rate. Are you?
I note that, particularly for recording purposes, some people out of curiousity or an abundance of caution would like to know technically superior settings, even if such settings might not make any ABXable audible diifference.
I do not suggest that a TWEAKED XP system will be any different, assuming that you have only one input at a time, or that the inputs matched in sampling rate.
But if a particular setting or technique could actually make an ABXable audible difference, then that obviously would be of considerable interest and importance.
Well, IP telephony is particularly bad about this...
With regards to MLXXX’s query (BTW ML’ – welcome to the Internet’s most intelligent A/V discussion forum!) - Check out this link:
http://extra.benchmarkmedia.com/wiki/index...k_-_Setup_Guide (http://extra.benchmarkmedia.com/wiki/index.php/Computer_Audio_Playback_-_Setup_Guide)
Thx Andrew D.
I do not suggest that a TWEAKED XP system will be any different, assuming that you have only one input at a time, or that the inputs matched in sampling rate.
o.k.
The next set to ABX must trigger the mixing/resampling capabilities of both XP and Vista.
Yes, I think this is the important one. But, as woodinville has said, this happens to many people and in most cases (like downsampling to 11kHz) is not going to require ABX, because we know that a 5kHz lowpass is pretty audible, even if the resampler is perfect
Anyone demanding an ABX test is arguing that one can not ABX 8kHz sampling vs. 44.1 kHz sampling with a full-band 44.1 signal as input.
Who here is arguing that, please?
That, in a nutshell, is what it all comes down to.
What about 44.1kHz and 48kHz? I think this would be the interesting one to ABX (or at least make samples for) because, as you point out - 44.1kHz versus 11kHz or 22kHz is pretty trivial.
It is pretty clear, though, that as soon as resampling is triggered (which happens very easily), Vista is better than XP - purely because it behaves much more sensibly. I think another interesting test (but also flame-war prone) would be Vista's resampler versus dmix on Linux.
Woodinville,
Reading post 20 or 23, technical improvements have been made in many area’s, not only the K-mixer problem has been tackled.
So mixing a low and a high bitrate stream, as you suggest, will probably demonstrate the difference loud and clear but I wouldn’t be surprised at all that experiments in other areas might also yield an audible difference.
As the improvements in the design are not limited to 1 specific point, I have the feeling that if we want to compare the differences between XP and Vista properly, we are in need of more than 1 experiment targeting a well known (and solved) problem.
Woodinville
Be cool man – I’m on your side here, in fact, I have all the respect in the world for your wealth of knowledge and experience in the field. However, I firmly disagree that a ‘standard scenario’ with Windows XP sees folks deploying multiple audio applications simultaneously.
I too speak from a well of experience: I’m also a Microsoft Certified Technician, one with nearly 15 years of experience in the field, and have worked on thousands of computer systems, with just as many clients. I own / operate a computer sales / support / training business called Digital Frontiers Ltd in the city in which I reside. Additionally, I’m a certified instructor who has spent years in classrooms of computer users - thousands upon thousands of hours in total.
It’s [very] rare that I personally would run simultaneous audio applications, and I am what you might, for all intents and purposes, refer to as a ‘power-user’. Neither do I see my clients (well over a thousand users - split roughly 60/40 between ‘Joe / Jane Normal’ and corporate clients), employing their computer systems in a manner where they multitask audio applications.
Ninety-nine percent of them run one basic audio application such as WinAmp / Windows Media Player / iTunes or the Real Audio Player at a time. It just doesn’t make sense that anyone would want to listen to simultaneous audio streams – try it, the sound is just a mumbo-jumbo…
Now, I may be way off base here, and perhaps completely misreading your assertions – and I apologize if I have in fact been doing so; but from what I gather – your argument is wholly based on the declaration that Windows XP and Windows Vista ‘sound different’ in an out-of-the-box state, when in the hands of a general client - due to the fact that they will trigger multiple audio applications simultaneously.
Please correct me if I’m wrong. If not, I urge you to clearly assess the probability that your argument is valid, given the percentage of time that clients triggering multiple audio events actually occurs – and given the fact that even if said scenario actually occurs, the audio stream is quickly switched back to full bandwidth in a rapid manner once the secondary audio event is complete.
With all due respect Woodinville, I’m not seeking an argument here; just a clear-headed evaluation of the reality of the situation.
Andrew D.
www.cdnav.com
I had assumed kmixer forces a resampling to a lower sample rate, until I saw the following explanation about the kernel mixer in Windows XP, on a Benchmark Media Systems webpage: http://extra.benchmarkmedia.com/wiki/index.php/Kmixer (http://extra.benchmarkmedia.com/wiki/index.php/Kmixer) :-
[blockquote]
Example:
Mixing 44.1-kHz stream 'X' + 48-kHz stream 'Y' + 96-kHz stream 'Z'
will result in the sample rates of 'X' and 'Y' being converted to 96 kHz.
[/blockquote]
If this is true, then multiple audio streams in XP do not result in a resampling to the
lowest sampling rate, but rather to the
highest sampling rate. On that basis, a low sampling rate "system sound" should not disturb the sampling rate of other [higher sampling rate] audio streams.
___________________
EDIT:
Microsoft appear to confirm the above, if I am correctly interpreting the Microsoft Developer Network webpage at http://msdn2.microsoft.com/en-us/library/ms790027.aspx (http://msdn2.microsoft.com/en-us/library/ms790027.aspx)
An abridged extract of the Microsoft webpage follows:-
[blockquote]
Connecting Additional Audio StreamsFollowing connection of the first input stream, KMixer applies a different policy to the creation of connections by clients to any additional input streams. The following table describes the KMixer policy on rate conversion and mixing for these additional streams. The table lists the tests in the order in which KMixer performs them.
New incoming rate is equal to the current output rate. Perform mixing only. New incoming rate is equal to the rate of another input stream. Mix the new input stream with the other input stream at the same rate to take advantage of the existing SRC from input to output. New incoming rate is lower than the current maximum input rate. Do SRC to the current output rate and mix.* New incoming rate is higher that the current maximum input rate.[blockquote]
If the hardware supports the new input rate:[blockquote]Change the current maximum output rate and the hardware rate to the new incoming rate.
Do SRC on all other streams to the new maximum output rate and mix them with the new stream. (When converting to the new maximum output rate, KMixer mixes together any input streams that happen to share the same sample rate before doing SRC on those streams.)[/blockquote]If the hardware does not support the new input rate, maintain the current output rate and do SRC on the new incoming stream to the current maximum input rate.[/blockquote][/blockquote][/color]
__________
*I presume this means "apply SRC to the new incoming stream, to take it up to the current output rate".
Will the udial audio sample be of good use in testing comparisons?
Please correct me if I’m wrong. If not, I urge you to clearly assess the probability that your argument is valid, given the percentage of time that clients triggering multiple audio events actually occurs – and given the fact that even if said scenario actually occurs, the audio stream is quickly switched back to full bandwidth in a rapid manner once the secondary audio event is complete.
My observation is based on the number of complaints recieved.
It is hard to know exactly what that makes the probablity, but there are a (*&(&* lot of complaints.
Also, to the other fellow reading the MS web site, if a signal is already running at some sampling rate, it will not change the card's rate, at least under some (many?) conditions. So if you start at 48 kHz, you're cool, except for the birdies in your low-rate signal. (which will also be low-qualiity). If you start at a lower rate, say 22 or 16 or 8, (and this depends on 3rd party drivers, too, so I can't make any generic statement) it may hold the rate at the low rate. (if it doens't, you get a monstrous "click", which is a different problem) (The text above is right for codecs that allow changing the rate, but then we get complaints about the clicks. Win some, ...)
When that happens (holds at the low rate) we get lots of complaints.
And while I can't speak for current apps, some VOIP programs HOLD THE FREAKIN' thing open, even while not sending data. This holds the sampling rate for those drivers at whatever it was, for better or worse.
And that leads to what I can best describe as fury on the part of some users.
Mmmn, VOIP, a method of making a phone call that often involves delays and echoes, but which is very cheap!
It seems that a lot of these kmixer problems would relate to the particular sound card (or motherboard sound chip?), and the associated drivers.
Is there any indication that Windows's resampler works or not at the given moment?
Is there any indication that Windows's resampler works or not at the given moment?
You mean a little 'tray utility' that informs a user that resampling has kicked in / out? Not that (I) know of, but this is certainly the place to drop a lot of hints so that one of the many programmers that haunt the place will write said utility... :-)
Andrew D.
www.cdnav.com
Mmmn, VOIP, a method of making a phone call that often involves delays and echoes, but which is very cheap!
It seems that a lot of these kmixer problems would relate to the particular sound card (or motherboard sound chip?), and the associated drivers.
Basically, your last sentence nails it. The idea is that you can change sampling rate any time, but people (*&*(& to holy (*&( about the clicks that causes, so a lot of drivers lock up changes while anything is open to the DAC.
Some only have one or two sampling rates.
All depends on the device, the driver, and if there is a class driver than can replace what the manufacturer provided.
'twood be nice if it was better. I know.
Believe me, I know.
jj
Hi All,
I fully agree with Zenno, he describes my setup, and I cannot see how an additional overhead in Vista can be an improvement of the sound! Ok, it does it in 32bit, but still it interferes (and mostlikely changes) the original sound, best option it is the same quality, but better?
I guess it maybe all in the description of "better"? If I hoop up the EQ, nice solid bass, is it then better? In my book, the best signal is the unchanged signal, not the one with DSPs or Vista room enhancements.
So how can Vista make things better to the unchanged signal proces as I have it in XP?
Cheerz,
Eric
Hi All,
I fully agree with Zenno, he describes my setup, and I cannot see how an additional overhead in Vista can be an improvement of the sound! Ok, it does it in 32bit, but still it interferes (and mostlikely changes) the original sound, best option it is the same quality, but better?
I guess it maybe all in the description of "better"? If I hoop up the EQ, nice solid bass, is it then better? In my book, the best signal is the unchanged signal, not the one with DSPs or Vista room enhancements.
So how can Vista make things better to the unchanged signal proces as I have it in XP?
Cheerz,
Eric
XP isn't perfectly unchanged either - Vista is more mathematically correct in this regard. But the difference should be unaudible, for most common uses.
I prefer Vista because it has better functionality (per-app volume control), and all driver features are finally global (you can edit most features from Windows itself, without having to rely on third party control panels that come with the driver). And, finally they put some thought into the audio setup, including HTPC uses, whereas XP was more in the line of "lol sound".
I guess it maybe all in the description of "better"? If I hoop up the EQ, nice solid bass, is it then better? In my book, the best signal is the unchanged signal, not the one with DSPs or Vista room enhancements.
How can (good) room correction be considered bad? Canceling the effect of the room on music/movies is a really good way of getting better sound in a space which isn't ideal for it. Same with EQ - my PC speakers are a little bright (due to baffle step), so I dial in some EQ to get a flat frequency response. My living room is very bass heavy around 45Hz, so I dial in a little EQ to integrate the subwoofer better.
EQ and room correction, when done well, are good - they can improve your enjoyment of your music, make it sound more (not less) like it was intended to, make bad equipment sound much more acceptable, and have a whole pile of other benefits. Now, I don't know if Vista's room correction is good (but from their technical info it sounds well designed), but saying that sound is better without room correction just isn't true.
Also, how do you propose your operating system mix multiple sounds of different sample rates together if it doesn't do any processing? Would you rather your browser locked up if it tried to play a sound while listening to music? Would you rather only the current application could play sound? Sure - these things make sense for some uses - but in almost all cases resampling and mixing is The Right Thing for an OS to do.
I fully agree with Zenno, he describes my setup, and I cannot see how an additional overhead in Vista can be an improvement of the sound! Ok, it does it in 32bit, but still it interferes (and mostlikely changes) the original sound, best option it is the same quality, but better?
Do you assert that you are, as well, the average user, who gets the average experience from an average, untweaked system?
Hi All,
Thangs for your replies,
So here my reply, I wasn't actually saying which one was better, I was trying to make the point, that a definition of "better" is required. For me the only thing that I am concerned with is an unchanged signal. My room doesnt need adjustments and my D/A, Amps and speakers for sure not. So my requirement, if I read a 24bit signal I want it as is fed into my D/A. If I achieve that, then I have my better signal.
However I understand that if one doesn't have such a setup, that the requirement might change, for instance for small speakers, I want some extra boom, or in my crappy listening room, I want to run vistas room enhancer.
In the 6 pages of discussion I havent seen anyone asking the guy what kind of speakers he has, he might as wel be playing from a laptop onto his laptop speakers in his bathroom!!!
Well, lets assume this is the case, how much of the above discusions is actually still relevant?
Cheerz,
Eric
Btw, my setup uses a 2496 M-Audio with drivers that bypasses the operating systems mixers, hence my requirement. And No, I am not an average user, but is any of us?