...
Most of my MP3's are @ 44.1 kHz or 48; my mixer is capable of 96 kHz. Do you think that putting the sampling frequency at 96 kiloHertz will make a better sound?
...
In many cases the native samplerate of "sound card" is 48kHz and resampling 48->44.1kHz is poorly done. Many audio interfaces might not give flat frequency response (range 20Hz-20kHz) for lower (44.1-48kHz) samplerates but does for 96kHz (as for an example).
Many audio interfaces might not give flat frequency response (range 20Hz-20kHz) for lower (44.1-48kHz) samplerates but does for 96kHz
Many?
(as for an example)
I don't see an example.
In many cases the native samplerate of "sound card" is 48kHz
All the cheap & on-board soundcards I've owned (manufactured 2002-2011) can only sample at 44.1 or 48, but I don't know enough about ADCs to say for sure that one rate is actually "native" and that the other is mathematically converted from it.
and resampling 48->44.1kHz is poorly done. Many audio interfaces might not give flat frequency response (range 20Hz-20kHz) for lower (44.1-48kHz) samplerates but does for 96kHz (as for an example).
Sounds plausible, and I have probably wondered the same things aloud in the past... but this being Hydrogenaudio, I really wish we could see some actual evidence of a soundcard recording more poorly at one rate than another, and having different frequency response curves for different playback rates. What would be a simple way to test?
and resampling 48->44.1kHz is poorly done. Many audio interfaces might not give flat frequency response (range 20Hz-20kHz) for lower (44.1-48kHz) samplerates but does for 96kHz (as for an example).
Sounds plausible, and I have probably wondered the same things aloud in the past... but this being Hydrogenaudio, I really wish we could see some actual evidence of a soundcard recording more poorly at one rate than another, and having different frequency response curves for different playback rates. What would be a simple way to test?
The simplest way would be an RMAA test into another sound card that is known to have a flat frequency response.
In many cases the native samplerate of "sound card" is 48kHz
All the cheap & on-board soundcards I've owned (manufactured 2002-2011) can only sample at 44.1 or 48, but I don't know enough about ADCs to say for sure that one rate is actually "native" and that the other is mathematically converted from it.
and resampling 48->44.1kHz is poorly done. Many audio interfaces might not give flat frequency response (range 20Hz-20kHz) for lower (44.1-48kHz) samplerates but does for 96kHz (as for an example).
Sounds plausible, and I have probably wondered the same things aloud in the past... but this being Hydrogenaudio, I really wish we could see some actual evidence of a soundcard recording more poorly at one rate than another, and having different frequency response curves for different playback rates. What would be a simple way to test?
In fact the ear is well known to be very insensitive to even a complete loss of any sound reproduction at all (AKA a brick wall filter) starting as low as 16 KHz. All this wailing and gnashing of teeth about minor difficulties in the 20-22 KHz range is therefore pretty strange.
Right, Juha needs to provide proof of his claim that 96 KHz playback is flat and sub-96 isn't, or that soundcards have a "native" sample rate and that they mathematically resample, perhaps badly, to other rates, affecting the audible spectrum. I have tried to be constructive and solicit guidance for him (+ whoever) toward a way to test. It's not like we can just say ABX it in foobar2000.
Right, Juha needs to provide proof of his claim that 96 KHz playback is flat and sub-96 isn't, or that soundcards have a "native" sample rate and that they mathematically resample, perhaps badly, to other rates, affecting the audible spectrum. ยด
..-
? You don't remember this?
16/44.1: http://ixbtlabs.com/articles2/creative-aud...m-ex/index.html (http://ixbtlabs.com/articles2/creative-audigy2-platinum-ex/index.html) (resampled)
16/48: http://www.ixbt.com/multimedia/creative/au.../1648-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/1648-out.shtml) (native)
14/96: http://www.ixbt.com/multimedia/creative/au.../2496-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/2496-out.shtml)
http://www.ixbt.com/multimedia/audiotrak/m...-pro-2496.shtml (http://www.ixbt.com/multimedia/audiotrak/maya44-mkii/audigy2-zs-pro-2496.shtml)
I believe this still can be true when we are speaking of Creatives sound card models which are not equipped with genuine X-Fi DSP (if one has these newer models, just RMAA it to find it out).
In fact the ear is well known to be very insensitive to even a complete loss of any sound reproduction at all (AKA a brick wall filter) starting as low as 16 KHz. All this wailing and gnashing of teeth about minor difficulties in the 20-22 KHz range is therefore pretty strange.
You must mean 20kHz-22kHz range?
IMO, flat frequency response in range 20Hz-20kHz is still one of the factors which makes the product status (pro/.../crapware).
On the topic of "native" sample rates of sound cards, many boards based on the C-Media CMI8788 or similar chips, such as the ASUS Xonar family, have worse noise performance at 44.1, 88.2, or 176.4 kHz sample rate than at 48, 96, or 192, because of the re-clocking needed by the former. According to my tests, the line output SNR of the Xonar D1 and Essence STX dropped from 116 to 110 dB, and 118 to 111 dB when switching the sample rate from 48 to 44.1 kHz, respectively. While those numbers are still plenty good enough, and other parameters like frequency response and distortion are not adversely affected, the increased noise floor can be a problem when using the built-in headphone amplifier of the STX (which increases the 0 dBFS level to about 7 Vrms, and has no analog gain or volume control) with very sensitive headphones.
However, in most cases with modern hardware, there is no real benefit from playing 44.1 kHz audio upsampled in software. Even with the above mentioned Xonar STX, the noise is not audible with an external amplifier that has an analog volume control, and even with the built-in amplifier only affects some headphones and mainly IEMs.
16/48: http://www.ixbt.com/multimedia/creative/au.../1648-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/1648-out.shtml) (native)
14/96: http://www.ixbt.com/multimedia/creative/au.../2496-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/2496-out.shtml)
So? I thought you wanted to prove your point, not negate it.
Nowadays even onboard HD audio interface can have very flat 44.1k response
http://www.hydrogenaudio.org/forums/index....st&p=856574 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=104390&view=findpost&p=856574)
Furthermore, Win7 (maybe Vista and 8 too?)'s resampling quality are not only transparent, but also looks very clean in spectrum graphs. Audible resampling artifacts are only possible if the driver of the audio interface is forced to use its own resampling algorithm and bypass the one offered by Windows. Win7 users just remember to install KB2653312 to have great OS level resampling quality.
16/48: http://www.ixbt.com/multimedia/creative/au.../1648-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/1648-out.shtml) (native) 14/96: http://www.ixbt.com/multimedia/creative/au.../2496-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/2496-out.shtml)
So? I thought you wanted to prove your point, not negate it.
LOL
16/48: http://www.ixbt.com/multimedia/creative/au.../1648-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/1648-out.shtml) (native) 14/96: http://www.ixbt.com/multimedia/creative/au.../2496-out.shtml (http://www.ixbt.com/multimedia/creative/audigy2-platinum-ex/2496-out.shtml)
So? I thought you wanted to prove your point, not negate it.
LOL
Ohh, I see, you meant the frequency response at 44kHz in the first link? Mea culpa. Still, it's questionable whether that is audible.
I don't see how that proves that 96k is better than 48k. It used to be common knowledge here 10 years ago that you avoided creative cards because only 48k was supported and they resamples poorly. If you dig up the really old cards you will see that everything but 48k is distorted.
So, how exactly does this relate back to the topic's original post bearing the unsubstantiated claim?
Are these old cards being used in a DAW? (
EDIT: Topic was split)
Still, it's questionable whether that is audible.
Indeed!
BTW, those plots of the Platinum eX do not show much (if any!) difference in frequency response from 20 to 20k between 48k and 96k sample rates. Also, there is only one plot of the Audigy 2 ZS, not that we accept plots as proof of differences in sound quality.
I don't see how that proves that 96k is better than 48k. It used to be common knowledge here 10 years ago that you avoided creative cards because only 48k was supported and they resamples poorly. If you dig up the really old cards you will see that everything but 48k is distorted.
Hmm... why should it be (better) flatter? 48kHz stream goes through DSP (fixed for 48kHz) w/o need for SRC and the 96kHz stream bypasses the DSP (in my example the 96kHz link was just to prove that it's flat compared to 44.1kHz one).
So, how exactly does this relate back to the topic's original post bearing the unsubstantiated claim? Are these old cards being used in a DAW? Still, it's questionable whether that is audible.
Indeed! BTW, those plots of the Platinum eX do not show much (if any!) difference in frequency response from 20 to 20k between 48k and 96k sample rates. Also, there is only one plot of the Audigy 2 ZS, not that we accept plots as proof of differences in sound quality.
Graphics for ZS is quite same as for A2Plat. I couldn't see the graphics for 96kHz (they were there but totally mess) behind the link given on that ixbtlabs review page (yes, there are links for detailed RMAA results) so to be sure the graphs can be seen I linked the A2ZS graphs. My intention wasn't to compare 48 vs 96 because of diferent path the streams goes (see. my previous post).
I don't see how that proves that 96k is better than 48k. It used to be common knowledge here 10 years ago that you avoided creative cards because only 48k was supported and they resamples poorly. If you dig up the really old cards you will see that everything but 48k is distorted.
Hmm... why should it be (better) flatter? 48kHz stream goes through DSP (fixed for 48kHz) w/o need for SRC and the 96kHz stream bypasses the DSP (in my example the 96kHz link was just to prove that it's flat compared to 44.1kHz one).
Some can bypass 96k, some cannot.
I don't think anyone doubt's that bad resampling can perform badly. But I doubt that higher is necessarily (if ever) better. Cards certainly exist that resample higher rates down for instance. I'm not aware of any that force resampling to 96k though.
If you talk about old Creative cards, SB Live (10k1) series only support up to 16-bit 48k and they have neither visible nor audible resampling artifacts at 48k. The SPDIF result (not bypassing DSP) proved this point.
SB Live (CT4830) recorded by X-Fi XtremeMusic with digital I/O module, analog results also included:
http://forums.dearhoney.idv.tw/download/file.php?id=650 (http://forums.dearhoney.idv.tw/download/file.php?id=650)
As an anecdote.
A long time ago (2005), I was attempting a 96kHz vs 44.1kHz ABX test on Windows XP and a Creative SB Audigy2 NX sound card manually set to 96kHz in the configuration application. To my surprise, I heard (and proved with PCABX) a difference. But one WAV file was 44.1khz and one was 96khz. To eliminate the effect of the sampler, I resampled the 44.1kHz test track to 96kHz. I could no longer hear a difference. The conclusion? The resampler had audible flaws.
Creative Live/Audigy series are very complicated. In Win2000/XP, some of them use EMU DSPs (10k1/10k2) to resample audio and synthesize soundfonts (so-called EMU patented 8-point interpolation). This resampler has audible resampling artifacts at least in some test signals such as RMAA's IMD swept test and old RMAA's 19/20khz IMD test.
http://audio.rightmark.org/test/audigy/cre...udigy-1644.html (http://audio.rightmark.org/test/audigy/creative-audigy-1644.html)
Just copy the pictures and open in other photo viewers if the graphs can't be displayed correctly.
Some cheaper Live/Audigy models or USB-based models use different chips which are hated by gamers because they have no hardware EAX support, and hated by home musicians because they are not compatible with kX drivers. Such cards MAY make use of Windows' internal resampler to resample audio. Win2000/XP's resampling quality can be manually configured, if it is not set to best, resampling artifacts can be audible.
http://www.geocities.co.jp/anothergs/kXTut/1st-step.html (http://www.geocities.co.jp/anothergs/kXTut/1st-step.html)
The "Good" quality should be something similar to linear interpolation.
http://forums.dearhoney.idv.tw/download/fi...fac62a5b9b46e78 (http://forums.dearhoney.idv.tw/download/file.php?id=306&sid=6eb7234620f9fe6d7fac62a5b9b46e78)
In "Best" setting I could not hear any artifacts even when listening to test signals, although it is not "graphically" as good as the resampler in Win7.
If you talk about old Creative cards, SB Live (10k1) series only support up to 16-bit 48k and they have neither visible nor audible resampling artifacts at 48k. The SPDIF result (not bypassing DSP) proved this point.
SB Live (CT4830) recorded by X-Fi XtremeMusic with digital I/O module, analog results also included:
http://forums.dearhoney.idv.tw/download/file.php?id=650 (http://forums.dearhoney.idv.tw/download/file.php?id=650)
EDIT: Please copy dearhoney's link and paste on the browser's address bar to download, if still unsuccessful, visit http://forums.dearhoney.idv.tw (http://forums.dearhoney.idv.tw) and keep that window open while downloading.
Yawn.
Yes, bad resampling can result in audible artifacts. Search the forum for udial.
EDIT: I only mention this for noobs will otherwise see this discussion and wonder if we're insane and for placebophiles who will otherwise see this discussion and be reaffirmed that we are insane.
Wow, has it been 10 years already?
On the topic of older Creative EMU cards and audio-routing-only offshoots (like P17V), these typically required explicitly setting up hardware sample rate via Creative's control panel app, at least as far as Windows 2000/XP was concerned. A SB Live! 24-Bit (P17V/CA0106) gave options of 48 and 96 kHz, plus 44.1 for SPDIF only. (I think I tricked it into analog operation at 44.1 once, which gave something like 8-bit performance.) Aside from the hassle, not a bad little card if you wanted 24/96 recording on the cheap.
I have a Vista machine, and what I've seen of (what I presume was) its resampler performance didn't look too exciting. Win7 up should be better.
While we're at performance at various sample rates, 24/44 output used to be broken in Realtek HDA drivers until not that long ago (some time in '12 or '13) and would set up the hardware all wrong.
A lot of posting to get to a simple resolution. 44.1 or 48 upsampled will not give you "better". The source is the source. If your audio device outputs a different frequency response at different sampling rates, it needs to be replaced NOW. See how easy and painless that was?
If I have a sound card that gives me adequate performance and suits my purposes using a software-based resampler because its on-board resampler isn't up to snuff, why should I replace it?
That seems wasteful.
In order not to give a false impression that native 44/48k could not make a flat response, Arnold's old website has a PC soundcard benchmark database. Just don't use today's standard to judge the results because those cards are about 10-20 years old, and some of them don't even support 48k.
http://web.archive.org/web/20061205221243/...mpare/index.htm (http://web.archive.org/web/20061205221243/http://www.pcavtech.com/soundcards/compare/index.htm)
Also attached my VIA HD Audio's 44/48/96 results to show that it performs similarly in different sample rates. (Recorded by X-Fi Titanium HD)
http://www.hydrogenaudio.org/forums/index....ost&id=7804 (http://www.hydrogenaudio.org/forums/index.php?act=attach&type=post&id=7804)
If I have a sound card that gives me adequate performance and suits my purposes using a software-based resampler because its on-board resampler isn't up to snuff, why should I replace it? That seems wasteful.
I can agree that (I still have Audigy 2 in daily use in one of my PC's ... mostly because of its AUD_EXT connector http://pinouts.ru/Multimedia/creative_int_pinout.shtml (http://pinouts.ru/Multimedia/creative_int_pinout.shtml) ).
...
Also attached my VIA HD Audio's 44/48/96 results to show that it performs similarly in different sample rates. (Recorded by X-Fi Titanium HD)
Yes, frequency response is quite flat on some of those onboard sound codecs. I have measured the (Front L/R) output from VIA VT1828S chip (found in ASUS board) (VIA->E-MU0404USB). Frequency response (20Hz-20kHz) is quite OK for 44.1kHz and 48kHz. Hoping they fix the bit-resolution part next (for 1828S I could get for it's Front L/R output (all modes) only 87dB which means ~15-bit).
Is that 87 dB due to actual noise or due to ground loop related crap in the test setup? According to test results on the web, the chip itself should be able to hit the 90s in practice. (The spec even claims up to 110 dB, and 100 dB for the ADC - try a loopback test.)
I have recently uploaded loopback test results for the Audigy FX (http://stephan.win31.de/audfxrmaa.zip) (which employs a Realtek ALC898), rear Front L/R to Line-In. While 44.1 and 88.2 are affected by what seems to be a stupid driver bug limiting recording to 16 bit, 48 and 96 show rather respectable dynamic range approaching the 104 dB(A) ADC spec. (ADC passband ripple isn't so hot at about 0.04 dB p-p, but they've got to cut corners somewhere, right?)
...
Also attached my VIA HD Audio's 44/48/96 results to show that it performs similarly in different sample rates. (Recorded by X-Fi Titanium HD)
Yes, frequency response is quite flat on some of those onboard sound codecs. I have measured the (Front L/R) output from VIA VT1828S chip (found in ASUS board) (VIA->E-MU0404USB). Frequency response (20Hz-20kHz) is quite OK for 44.1kHz and 48kHz. Hoping they fix the bit-resolution part next (for 1828S I could get for it's Front L/R output (all modes) only 87dB which means ~15-bit).
Probably the SNR has little to do with the DAC and instead is just the board its put in.
...
Also attached my VIA HD Audio's 44/48/96 results to show that it performs similarly in different sample rates. (Recorded by X-Fi Titanium HD)
Yes, frequency response is quite flat on some of those onboard sound codecs. I have measured the (Front L/R) output from VIA VT1828S chip (found in ASUS board) (VIA->E-MU0404USB). Frequency response (20Hz-20kHz) is quite OK for 44.1kHz and 48kHz. Hoping they fix the bit-resolution part next (for 1828S I could get for it's Front L/R output (all modes) only 87dB which means ~15-bit).
Probably the SNR has little to do with the DAC and instead is just the board its put in.
IME a godly number of chips have asymmetrical performance, and the only way to test the full potential of the DAC is to use the ADC on a board with higher, symmetrical performance such as most of the pro cards.
Also a very good point, but I think the E-MU0404USB he used was probably good enough.
The more "external" you get, the bigger the ground loops though, hence my question. Dedicated audio analyzers have galvanically isolated inputs and outputs for good reason. The 0404USB itself gets 113 dB(A) in loopback, that ought to do. One may just have to include an isolation transformer or determine output impedance and build a pseudo-balanced adapter that reroutes output ground to the "cold" input via a resistor of determined value (per channel).
Of course some ground loops or other performance-degrading issues may already be present on the board itself. It may also be beneficial to disconnect the front panel audio if that poses ground loop problems.
A member used 0404USB to test Fuze+ and got -98dB in RMAA so I think 0404USB should be good enough.
http://www.hydrogenaudio.org/forums/index....=0&p=833017 (http://www.hydrogenaudio.org/forums/index.php?showtopic=100657&start=0&p=833017)
0404USB spec:
http://www.creative.com/emu/products/product.aspx?pid=15185 (http://www.creative.com/emu/products/product.aspx?pid=15185)
- Input Impedance: 1Mohm
- Max Level: +12dBV (14.2dBu)
- Dynamic Range (A-weighted, 1kHz, min gain): 113dB
- Signal-to-Noise Ratio (A-weighted, min gain): 113dB
- THD+N (1kHz at -1dBFS, min gain): -101dB (.0009%)
And this is the self loop result of my X-Fi Titanium HD, different sample rates have similar results.
http://www.hydrogenaudio.org/forums/index....howtopic=100481 (http://www.hydrogenaudio.org/forums/index.php?showtopic=100481)
I don't know the exact formula to convert dBV to Vrms but X-Fi series' I/O are at 2Vrms maximum, 5dB digital gain is needed to match my VIA HD audio's RMAA test signal which is not shown in my reports because I adjusted the digital gain in my X-Fi's mixer already.
Is that 87 dB due to actual noise or due to ground loop related crap in the test setup? According to test results on the web, the chip itself should be able to hit the 90s in practice. (The spec even claims up to 110 dB, and 100 dB for the ADC - try a loopback test.) I have recently uploaded loopback test results for the Audigy FX (http://stephan.win31.de/audfxrmaa.zip) (which employs a Realtek ALC898), rear Front L/R to Line-In. While 44.1 and 88.2 are affected by what seems to be a stupid driver bug limiting recording to 16 bit, 48 and 96 show rather respectable dynamic range approaching the 104 dB(A) ADC spec. (ADC passband ripple isn't so hot at about 0.04 dB p-p, but they've got to cut corners somewhere, right?)
Actually, as I got about same DR results for ESI DuaFire I took a closer look after possible reason for this. Finally got the DR improved for both by placing the E-MU as far away from PC gear and power leads as possible (no cable crossings, etc). The DR resuts for VIA codec stayed <93dB for 16-bit modes and <98dB for 24-bit modes.
Probably the SNR has little to do with the DAC and instead is just the board its put in.
- yes, from me it was just a side note.
Assuming the external device is USB powered off the same PC, a ground loop is probably not too serious of a problem.
Assuming the external device is USB powered off the same PC, a ground loop is probably not too serious of a problem.
Ground loops are possible even between different parts of the same PC (for example, a USB port and an internal sound card). This can be seen on these (http://www.head-fi.org/t/500369/lightbox/post/8406949/id/610989/user/190229) graphs that show all possible loopbacks between two sound cards installed in the same computer. The cases when the ADC and the DAC are not on the same card are significantly noisier. Adding a simple differential amplifier (http://cdn.head-fi.org/0/0d/0daf9f91_diff_amp.png) to the loopback fixed the ground noise in another test (http://cdn.head-fi.org/a/a9/a94d86b9_rmaa1_a.png).
Assuming the external device is USB powered off the same PC, a ground loop is probably not too serious of a problem.
Ground loops are possible even between different parts of the same PC (for example, a USB port and an internal sound card). This can be seen on these (http://www.head-fi.org/t/500369/lightbox/post/8406949/id/610989/user/190229) graphs that show all possible loopbacks between two sound cards installed in the same computer. The cases when the ADC and the DAC are not on the same card are significantly noisier. Adding a simple differential amplifier (http://cdn.head-fi.org/0/0d/0daf9f91_diff_amp.png) to the loopback fixed the ground noise in another test (http://cdn.head-fi.org/a/a9/a94d86b9_rmaa1_a.png).
Interesting. I suppose the most likely explanation is that one of those cards uses a virtual audio ground that is significantly different than the other. I've seen this in car stereos with MP3 players where the stereo virtual ground is several volts different than the virtual ground generated from the low voltage battery regulator, resulting in extreme interference.
From a quick test with a multimeter, the ground of the input and output jacks of both cards seems to be connected to the chassis of the PC. But many other devices (motherboard, graphics card, power supply, etc.) are also connected to the chassis, which is why it is noisy.
Just curious and installed my old X-Fi XtremeMusic then recorded by X-Fi Titanium and got -109dB which is same as the advertised spec. Creative is really quite honest (in some aspects )
http://www.hydrogenaudio.org/forums/index....st&p=858334 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=104556&view=findpost&p=858334)
Also a very good point, but I think the E-MU0404USB he used was probably good enough.
The 0404s can be very good indeed, for both recording and playback.
From a quick test with a multimeter, the ground of the input and output jacks of both cards seems to be connected to the chassis of the PC. But many other devices (motherboard, graphics card, power supply, etc.) are also connected to the chassis, which is why it is noisy.
Common grounding does not necessitate noise. How the common grounding is accomplished, IOW the exact routing and sizing of circuit board traces can matter a great deal.
There is an area of electrical engineering called
mixed signal design which addresses this among a great number of other things. It is composed of both art and science.
I've improved the SNR of a piece of commercial audio gear by up to 20 dB by redoing a very few land patterns on a certain circuit card, and this isn't even my thing!
My Xonar DS has terrible ultrasonic performance, I get an audible tone near the 4-5k area if I play a ~44kHz sinewave (http://i.imgur.com/zu9Kevt.png)
The internal DSP (all settings, 44.1/48/96/192 same result) must lack any sort of anti-aliasing on its LPF, though why it applies a LPF in the first place, I have no clue.
I make sure I have Windows set to 48kHz as it has a functional LPF for the rare occasion I ever come across high sample rate content.
Terrible. Are you getting the same result in different APIs (ASIO/KS/MME/DS/WASAPI)?
I'm recording with sox, but played the file back with Foobar2000 (WASAPI?) and sox(MME?), both produce the same spectrogram, play back of a 44kHz tone with foobar is also audible.
Have you applied the Windows 7 MME resampling patch? http://support.microsoft.com/kb/2653312 (http://support.microsoft.com/kb/2653312) (of course, If you are using Windows 7..)
I suspect it is some kind of Windows sample rate conversion issue, which can be difficult to avoid or fix sometimes, depending on the drivers. Although I only have the Xonar D1 as a reference, which is a one step higher model, the hardware is similar other than the better DAC, and it performs much better (http://www.head-fi.org/t/629729/xonar-d1-measurements) (tested on Linux). Also, I have seen RMAA results of the cheap Xonar DG with flatter frequency response. Even onboard codecs should be better than that. It is almost certainly a software problem.
I was making some tests, and there are quite a few places where things can go wrong. (Edit: this is a realtek ALC268 HDAudio integrated laptop soundcard)
I made a sweep signal with audacity from 100Hz to 47999Hz , logaritmic, 10 seconds, amplitude 0.97 (-0.3dBFS approx). The project setting was set to 96Khz.
I played it inside audacity using the WASAPI driver. (audacity 2.0.5)
I recorded with sox with the line: sox -d -c 2 -b 24 -r 96000 record.wav (sox 14.4.1)
The default recording device was set to stereo mix. The stereo mix was set to 44Khz 16bits. The Windows 7 MME patch is applied.
Sox decided to record at 48Khz 16bits.
So the audio did 96Khz->44Khz->48Khz->96Khz. It had some aliasing, but it did not bounce back.
When I realized that, I changed the stereo mix to 96Khz.
And then the audio did 96Khz->96Khz->48Khz->96Khz (yes, sox still decided to record at 48Khz). Now, the aliasing was almost nonexistant and of course it did not bounce back either.
And that is without even adding the DAC and ADC into play... go see...
Edit: Ok, I realized the way to specify the samplerate of the wave input is adding it before the -d switch: sox -r96000 -b24 -d -c 2 -b 24 -r 96000 record96-96-96-stereo_mix.wav Now i had quite a clean signal with a slight lowpass filtering from a bit before 40Khz. (slight as in I see the waveform amplitude decreasing. It decreases in 10dBs at 48Khz)
Edit2: Seems audacity has a bug and stops the audio too soon. I had to add a second of silence at the end because else the sweep was cut at 45Khz or so.
Also, I've finally tried with a loopback cable and it acts like the stereo mix. In fact, even better, because the stereo mix had noise shaped dither.
The image doesn't look like Win7's bad MME resampling because linear interpolation before applying the hotfix will generate reflections all over the graph like this
http://src.infinitewave.ca/images/Sweep/SRC_Lin.png (http://src.infinitewave.ca/images/Sweep/SRC_Lin.png)
I guess it is caused by Asus' driver/software issues. Try to use ASIO4all to see if there are any improvement or not. See
http://www.hydrogenaudio.org/forums/index....st&p=829916 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=92856&view=findpost&p=829916)
How much of these plots are due to the unrealistic use of FS signals?
My Xonar DS has terrible ultrasonic performance, I get an audible tone near the 4-5k area if I play a ~44kHz sinewave http://i.imgur.com/zu9Kevt.png (http://i.imgur.com/zu9Kevt.png)
The internal DSP (all settings, 44.1/48/96/192 same result) must lack any sort of anti-aliasing on its LPF, though why it applies a LPF in the first place, I have no clue.
I make sure I have Windows set to 48kHz as it has a functional LPF for the rare occasion I ever come across high sample rate content.
I think your card is probably running at 48KHz in that picture (note that it aliases about 48k/2Hz), and what you are seeing is just a resampler from whatever sampling rate your sweep was at to your cards actual sampling rate.
FWIW, I don't think that plot looks all that bad. Playing a pure, full scale ultrasound tone just a hair below the sampling rate of a sound card will generally introduce artifacts unless the resampler is very, very slow. Its almost impossible to +100dB of rejection very close to the Nyquist rate, so a pure tone at that point will generally be audible. Doesn't matter in practice though, since no one masters real audio huge massive amounts of power 40KHz+. You'd risk blowing out people's tweeters.
What is with this latest trend of using absurd signals in order to demonstrate something is "broken?"
Yes indeed, one can always pick a signal to break a well engineered system, and over-engineering systems to handle irrelevant or absurd circumstances is generally not a great idea.
FS sweep detection with an output that says the user is an idiot shouldn't be too hard to code.
I can hear it now: "This doesn't sound like your everyday jitter. Oh dear, my signal generator must be broken!"
I would argue that RMAA is essentially that. A tool that can give you what you want to know without forcing you to have a detailed understanding of the underlying theory and hardware design.
YellowOnion's picture is quite similar to MadTracker 2.5
http://www.maz-sound.com/index.php?show=mp...&page_id=62 (http://www.maz-sound.com/index.php?show=mpcs&id=&mpcg=&mpc_id=34&page_id=62)
Audio samples can be downloaded by clicking the speaker icon at the top right corner of each software, which include a tambourine sample.
greynol: Sweeps are the natural tool in this case to see if there is any serious issue in the playback chain. And in this case there is one.
You might argue that it doesn't matter, but if you do so, then you say that is perfectly fine that something that can play 96Ksamples signals should be previously filtered at around 24Khz.
I am with saratoga in this. There are several resamplers going on in there. It is really suspicious that it mirrors in the middle of the frequency range. That's why I started to make the tests with my soundcard in the attempt to find possible ways to reproduce something similar to that.
Edit:
Mm... Yup... seems the only plausible explanation is 96Khz-> Resampled to 48Khz with a sinc that does not filter (to create the bounce back), and then resampled back to 96Khz, applying a filter at 24Khz frequency.
In case of trackers like MadTracker, this way of working is by having a precalculated table, which filters at the input samplerate frequency. When upsampling (going down in the note scale) this is ok, but when downsampling (going up in the note scale) this doesn't work (it alias without any filtering).
You might argue that it doesn't matter, but if you do so, then you say that is perfectly fine that something that can play at 96Ksamples signal should be previously filtered at around 24Khz.
No, just that they don't contain very high power ultrasonic tones, which is very reasonable as no real music will contain those. If they do, you should configure a slower resampler or else figure out how to put that card into whatever sample rate you're actually using.
I am with saratoga in this. There are several resamplers going on in there.
No, I think its just one resampler, from whatever the source is to 48k.
The posted image is at 96Khz, and contains aliasing above 24Khz, that's why i talked about two resamplers resampler passes.
Of course, the problematic one is the first one. (i edited my post with what I think that could happen)
I have no problem with sweeps. I have a problem with sweeps using a full scale signal with the expectation that nothing goes awry when the signal approaches Nyquist.
My bad if that isn't what is going on here.
...else figure out how to put that card into whatever sample rate you're actually using.
As I mentioned, running my card at 192 or 96kHz doesn't work, this test was in fact run with 96kHz set on Windows "output", the soundcard DSP and the Loopback input, and yet this tone is audible (even over SPDIF at a native 96/24), not much below the test signal volume, it doesn't install much confidence in their product if they can't even achieve the same performance as the Windows resampler, though in most real world situations outside audio engineering and dog research I don't see this being an issue.
There are other issues I've had with the card, like the soundcard list taking 30-60s to open, and the driver install process failing and causing my system to load the kernel at 100% CPU, ironically the Linux drivers are awesome, I can get jack running down to 1.5ms, and I even got the dev (Clemens Ladisch) of the driver to fix a polarity issue with the sub-channel causing a very large lack of bass.
On a side note I'm running Windows 8, I don't think that hotfix applies (nor did it matter in the Foobar test).
I can provide the sox script if anyone is interested.
...else figure out how to put that card into whatever sample rate you're actually using.
As I mentioned, running my card at 192 or 96kHz doesn't work
What do you mean "doesn't work"? Looking online theres RMAA plots for the Xonar DS in 96k mode, and it seems to work fine: http://www.elitebastards.com/index.php?opt...mp;limitstart=4 (http://www.elitebastards.com/index.php?option=com_content&id=729&Itemid=27&limitstart=4)
What do you mean "doesn't work"? Looking online theres RMAA plots for the Xonar DS in 96k mode
Sorry I mean it doesn't fix the issue, RMAA can use ASIO, Foobar with ASIO output does not produce any audible tone, I guess something could be said for testing common playback APIs over Audio Engineering APIs, I have absolutely no clue how to disable the onboard DSP.
Also greynol, is it common for someone to not read axis? it is not a fullscale signal, its -6dB, as none of it is white like shown on the side indicator, I appreciate if I wasn't called an idiot for your own mistakes, or assumptions.
I don't know about Windows 8 specifically, but I don't think you should need to use ASIO to get 96k playback, or at least get better resampling. Are you sure you're setting everything correctly?
Sorry, it wasn't directed at anyone personally.
About my mistake, I suppose I should apologize for not being an expert at deciphering subtle shades of yellow, even if one of the graphs submitted looked pretty darn white. Whatever.
I don't know about Windows 8 specifically, but I don't think you should need to use ASIO to get 96k playback, or at least get better resampling. Are you sure you're setting everything correctly?
I've doubled checked this a bunch of times, I'll include screenshots to clarify my setup.
If I set Windows soundcard settings to 48kHz I get a nice clean signal with a proper LPF/Resampler, if I set my entire chain (Windows speaker soundcard settings, the onboard DSP, and the mic soundcard settings) to 96kHz the onboard DSP applies some form of LPF.
My batch script is below, though I've also tried outputting the same content in to a file and playing it back in Foobar2000 to eliminate any possible MME issues.
set SOX=H:\\Apps\sox.exe
start %SOX% -r 96k -b 32 -t waveaudio "Wave (Asus Xonar DS Audio Devic" -n trim 0 31 spectrogram -w Kaiser
%SOX% -r 96k -n -t waveaudio synth 30 sine 10-44000 gain -6
(http://i.imgur.com/Qr0zoEa.png)
What do those options on the screenshots exactly do? Is it the mixer samplerate, or output samplerate? What does the bitwidth actually mean? According to the linux driver xonar supports only 16 and 32 bits http://git.kernel.org/cgit/linux/kernel/gi...xygen_pcm.c#n44 (http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/tree/sound/pci/oxygen/oxygen_pcm.c#n44) , what are those 24bits?
Typical questions about the closed-source black box, unlikely to get answers. I personally would not waste time with this glass ball guessing.
Foobar with ASIO output does not produce any audible tone, I guess something could be said for testing common playback APIs over Audio Engineering APIs, I have absolutely no clue how to disable the onboard DSP.
Did you mean no sound at all or no audible aliasing when using ASIO? If the latter then it is obviously Asus' fault and have nothing to do with Windows.
According to the linux driver xonar supports only 16 and 32 bits
The 32-bit recording mode is just 24 bits effective resolution with left shifting and zero padding by 8 bits. Playback probably does the opposite (throw away the lowest 8 bits).
@greynol: Please, look at the graph again, you interpreted it incorrectly. There is a signal sampled at 96Ksamples with a sweep that should go from 10Hz to 44Khz (as said in the original post and by the sox line provided), but bounces back at 24Khz. So it is not near Nyquist (48-44=4) and the problem is not subtle (there is a strong signal at 4Khz).
From here comes the several comments that I've made about two resampling passes, a sinc resampling that doesn't adapt the filter frequency, and an attempt to locate which part of the system could be working at 48Khz even when the rest is set at 96Khz.
@YellowOnion: i tried your sox lines. I've found that the signal generated by audacity (2.0.5) is of slightly better quality than the one generated by the synth effect of sox (sox's one has aliasing below -100 dBFS and some DC.).
Now, one question: is the "Wave" recording device that you use, the equivalent of "what you hear"/"Stereo mix"? Could it be that you changed the settings to the incorrect element?
I did several more tests to find out more use cases. With the 96khz wave file, and recording in sox at 96khz, the playback and recording devices also have their own samplerate, and I tried several configurations.
The most similar graphic to that of yours was when i configured both, playback and recording to 48Khz.
With playback 96Khz and recording 48Khz, i had the aliasing at 48Khz but the bounce back was filtered in my case
With playback 48Khz and recording 96Khz, i didn't have the aliasing at 48Khz but the bounce back was just a bit filtered (around -60dB down)
With playback 48Khz and recording 48Khz, i had the aliasing and the bounce back a bit filtered (around -60dB down)
(I add just the first one as an image and the rest as urls to not fill the screen)
In my case, the recording device only accepts 16bits. This helps to see where the aliasing comes from
(http://psycle.free.fr/josepma/soundcard/spectrogram4848.png)
48-96 (http://psycle.free.fr/josepma/soundcard/spectrogram4896.png)
96-48 (http://psycle.free.fr/josepma/soundcard/spectrogram9648.png)
96-96 (http://psycle.free.fr/josepma/soundcard/spectrogram9696DC.png) I have some DC offset in my line in that can be seen here. In the other captures, i activated the DC offset cancellation. Also, there's noise shaping, contrary to what i said in a previous post. In that case i was mislead because of different volume level.
I also added 44-44 (http://psycle.free.fr/josepma/soundcard/spectrogram4444.png), for fun. Quite some amount of aliasing, but most below 70dB.
In all them, it is present one obvious defect of my soundcard (at least as implemented in this laptop), that constant signal above 15Khz (with aliasing in some cases).
I would say that the bounce back is added by the playback chain at one point resampling to 48Khz, and the mirrored aliasing is added (by windows?) when converting from the recording device samplerate, to the program's asked samplerate (in this case, sox at 96Khz from the 48Khz input).
samplerate of original signal -> samplerate in which the output device is opened by the program -> samplerate at which the output device is configured to work (windows setting) -> (optional) samplerate at which the output device own's control panel (hardware) is configured to work -> (optional) samplerate at which the input device own's control panel (hardware) is configured to work -> samplerate at which the input device is configured to work (windows setting) -> samplerate in which the input device is opened by the program -> recorded file samplerate.
Supposedly, the ASIO scenario avoids the windows setting stages, and forces the program and the hardware to work at the same value. WASAPI exclusive should do so as well, depending on the hardware drivers' goodwill.
Samplerate of the file and the program's samplerate are supposed to be the same (but in sox it can be different)
Edit: serveral edits to clarify and change sentences.
depending on the hardware drivers' goodwill.
Agree. My X-Fi's WASAPI exclusive mode sample rate can be set differently from the driver's hardware sample rate, but it is not possible with ASIO. However, ASIO is also not 100% secure in revealing the true sample rate, for example, kX driver can use 44k ASIO in my old SB Live CT4830 but the card itself only support 48k. When using 44k ASIO aliasing artifacts can be easily measured.
Also YellowOnion, did you disable SVN and enable HF DSP mode in your Xonar control panel and try 2 speakers mode instead of headphone mode?
depending on the hardware drivers' goodwill.
Agree. My X-Fi's WASAPI exclusive mode sample rate can be set differently from the driver's hardware sample rate, but it is not possible with ASIO. However, ASIO is also not 100% secure in revealing the true sample rate, for example, kX driver can use 44k ASIO in my old SB Live CT4830 but the card itself only support 48k. When using 44k ASIO aliasing artifacts can be easily measured.
Hmm... Before X-Fi serie cards, Creative's ASIO was fixed for 48kHz because of the DSP (10k serie) which is fixed for 48kHz (this way the ASIO outputs can be routed through DSP effects w/o additional SRC). IIRC, E-MU series cards which has a 10k series DSP onboard supports DSP effects but only when SR = 48kHz.
There's a note on Creative's ASIO driver: "For use
with ctaud9x.vxd, ctprxynt.sys and ctprxy2k.sys" (this information is taken from driver .dll (ctasio.dll) installed for Audigy 2 series PCI card in Vista 32-bit). What could this mean? BTW, Creative's ASIO driver bundled with X-Fi and E-MU PCI cards worked well with Audigy series card on XP installation (tried these about 2 years ago).
SB Live can use X-Fi's (normal Windows DS/MME/soundfont) driver and it is also possible to use SB Live and X-Fi together with the same driver. However SB Live could not use Creative's ASIO driver because ASIO is not an advertised function for SB Live and Creative explictly blocked it. Hacked APS driver and kX driver enable ASIO for SB Live.
IIRC, kX's 44k ASIO for SB Live have some limitations. Firstly, only playback is supported. Secondly, only two channel mode is supported. In 48k mode kX ASIO supports up 16in/16out full duplex multiclient operation for my SB Live.
All of the above were tested in Win XP in about 2005-2007
Frequency response for ESI DuaFire (24-bit, various samplerates).
http://i60.tinypic.com/34hapvb.png (http://i60.tinypic.com/34hapvb.png)
Measured using path ESI [Direct Sound] --> E-MU 0404 USB [ASIO]. Samplerate settings 1:1.
Does responses look OK?