Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i (Read 14212 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

im hearing music in a way i have never heard it before. since i tried the above combo, i am sitting here stupid, even listening to music genres i dont even like. the sound clarity is SO beyond explanation and nuances so strong that i am blown to bits.

i have a sound blaster x-fi hd usb card (the top offering from creative) too. and even with the dsp effects (crystalizer etc) windows cannot bring comparable sound clarity and nuance to any piece of music while using x-fi hd usb.

actually, m-audio doesnt sound as good in windows as it sounds in linux too - this is the reason i am posting this topic.

despite i have tried a lot of stuff, to the extent of loading foobar2000 with vst plugins and trying all kinds of dsps from exciter to expander, i wasnt able to catch the sound quality in linux under windows, or, at least, make x-fi hd usb sound as good as m-audio fast track pro sounds in linux. i even used cross-platform player audacious (has win32 build separately) with dsps and whatnot in both platforms so i would be able to make an even handed observation.

so im thinking that, there is something in linux's sound setup, either with pulseaudio or alsa or whatever underlying technology, that brings out the max in a sound card and gives out this outstanding quality.

so if there is anyone who can explain me why this setup sounds so good under linux, i will try to tinker with it and push the envelope even further. or maybe attempt setting up the same kind of sound setup under windows.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #1 –
When you start adding effects, what sounds "better" to you may sound worse to someone else.

I think you are OK as long as you are talking about effects that you prefer.  But, please make sure you've read TOS #8[/u] before the moderators nail you for claiming that soundcard-A sounds better than soundcard-B, without conducting a blind ABX test.. 

The "traditional" idea of high-fidelity is to reproduce the original sound as accurately as possible.  With this concept, we can that one system is "better" than another.    i.e. If you are getting audible noise, distortion, or frequency response variations, we can define those as defects.

And, a purest or audiophile doesn't use any effects, except perhaps equalization to correct for speaker & room errors, and to occasionally to correct for bad recording or bad production.  (I'm not a purist, and I like to use surround effects with stereo recordings.)

If you are not adding effects or altering the  audio, the player-application or operating system should not make any difference...  You should be able to feed the same audio data into the soundcard.

If you are not hearing noise,  you are not likely to hear any difference between soundcards (assuming no effects).    And with good soundcards, you shouldn't be hearing noise from the soundcard either.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #2 –
Seems you didn’t notice #8 of our Terms of Service (edit: I didn’t notice DVDDoug’s linking it in his post; thanks!). Really, unless you perform a double-blind comparison of the two OSs/setups (perhaps by ABXing audio files recorded from the output in both cases), obtain objective evidence of the difference you’re claiming, and post it here, this topic is of no use to anyone else. You may think you’re hearing something wonderful, but all anyone else can see is an unsubstantiated anecdote.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #3 –
Quote from: DVDdoug on
If you are not adding effects or altering the  audio, the player-application or operating system should not make any difference...  You should be able to feed the same audio data into the soundcard.


In reality, unless you're feeding /dev/dsp or whatever device you have directly, the audio is always altered by software rate conversion for mixing purposes (alsa/dmix and OSS4 default to 48000). The rate conversion algorithms do vary in quality. In fact, the default algorithm used by dmix is known to horrible. That's why there is an option to use the userspace libsamplerate as an alternative:

Code: [Select]
File: /etc/asound.conf

defaults.pcm.rate_converter "samplerate_best"


On the other hand, OSS4 is known to have superior quality primarily due to using high quality rate conversion algorithms with float calculations enabled by default (not allowed in in-tree kernel code):

Quote
/*
* The src_quality setting defines the precision of the software based sample
* rate conversion algorithm used by OSS. This setting doesn't affect possible
* hardware level conversions done by the devices themselves. Using higher
* quality setting gives better quality while lower settings consume less
* CPU time.
*
* Possible values are:
*    0 - D lowest quality (normally equals to 1=low quality)
*    1 - L  low quality    (spline interpolation)
*    2 - M  medium quality (lagrange interpolation)
*    3 - H  high quality  (DEFAULT)
*    4 - HX high quality  (high quality with extra precision)
*    5 - P  production quality
*    6 - PX production quality (prod quality with extra precision)
*/


Unfortunately, OSS4 started to cause kernel crashes in my laptop. So, I had to stop using it. But I would stringly recommend it for anyone whose hardware does not reproduce the bug mine is suffering from.

pulseaudio is just another mostly useless (except for special uses like network sharing and bluetooth) layer between applications and kernel subsystems. I'm guessing that it does have its own rate conversion algorithm(s) that might be better in quality than the default offered by alsa/dmix and whatever is used in windows.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #4 –
Quote from: DVDdoug on
And, a purest or audiophile doesn't use any effects, except perhaps equalization to correct for speaker & room errors, and to occasionally to correct for bad recording or bad production.  (I'm not a purist, and I like to use surround effects with stereo recordings.)

If you are not adding effects or altering the  audio, the player-application or operating system should not make any difference...  You should be able to feed the same audio data into the soundcard.


for effects, i am concerned about dynamic range compression and removing its effects.

and for os, i am sure that there is a difference in between these two oses - especially when you ramp up the settings for pulseaudio daemon to full. which induces considerable cpu load, but makes the sound much more better.

Quote from: db1989 on
Seems you didn’t notice #8 of our Terms of Service (edit: I didn’t notice DVDDoug’s linking it in his post; thanks!). Really, unless you perform a double-blind comparison of the two OSs/setups (perhaps by ABXing audio files recorded from the output in both cases), obtain objective evidence of the difference you’re claiming, and post it here, this topic is of no use to anyone else. You may think you’re hearing something wonderful, but all anyone else can see is an unsubstantiated anecdote.


Quote from: Nezmer on
pulseaudio is just another mostly useless (except for special uses like network sharing and bluetooth) layer between applications and kernel subsystems. I'm guessing that it does have its own rate conversion algorithm(s) that might be better in quality than the default offered by alsa/dmix and whatever is used in windows.


then lets say that my experience is totally subjective. and then let me turn this into a technical question :

i have set these settings in pulseaudio's /etc/pulse/daemon.conf

resample-method = src-sinc-best-quality
high-priority = yes
default-sample-channels = 2
default-sample-format = float32le
default-sample-rate = 96000
default-fragments = 8
default-fragment-size-msec = 5

after listening to music in this state, i am unable to bear listening to any other sound that comes out of my sound cards in windows, or pulseaudio's default settings. therefore, i am wondering what the changes above do to sound signal. ill be glad if someone explains, especially sample method, sample format, and sample rate.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #5 –
Quote
i am wondering what the changes above do to sound signal


It appears to upsample the input regardless of what that input is. In other words, it does precious little except warm up your hardware, unless the default resample-method is exceptionally crappy.

The main thing to take home is that it cannot improve existing audio. Resampling a 44.1KHz file to 96KHz will not increase its quality.

My guess is that you're imagining everything. Try setting up an experiment, to be sure.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #6 –
Quote from: dhromed on
Quote
i am wondering what the changes above do to sound signal


It appears to upsample the input regardless of what that input is. In other words, it does precious little except warm up your hardware, unless the default resample-method is exceptionally crappy.


that much i was able to gleam myself. however there is quite a difference (even if subjective) in between the sound that comes in 96 khz (at least it shows as such) in my windows xp and 96 khz in my linux. i am wondering what is creating the difference. i may attempt to reproduce it in xp. or actually any potential future os i may use.

Quote
The main thing to take home is that it cannot improve existing audio. Resampling a 44.1KHz file to 96KHz will not increase its quality.


i know it cannot change the specifications audio was recorded. however, doesnt resampling rate for playback affect any kind of signal going out to analog speakers ?

Quote
My guess is that you're imagining everything. Try setting up an experiment, to be sure.


im actually quite a failsafe person. and i dont change accustomed habits or get into new things, or hell, even believe anything. as you can imagine, i actually ran approx. 2 days of listening tests with both oses, sound cards, 3 different media players and numerous plugins and setting combinations. losing 2 days of productivity in the progress.

the most important thing to notice was the 'grow on me' effect -> like the time i first bought quality speakers, at first i didnt notice an immediate difference in sound, but then increasingly i have come to not be able to bear older hardware. and in this case, older os. subjective, yes, but for someone like me, who is quite habitual in his ways, its unmistakable. i use my stuff for endless eons, unless something really unmistakably and undeniably better makes me switch. from hardware to os, from clothes to food.

so as a summary, i did my own experiments for listening, and i am here to learn what i dont know about the technical details above. i researched about snr ratios, this and that and learned a lot, but i dont know the specifics pertaining to what the above settings do in a software like pulseaudio.

96 khz upsampling, samples the sound to 96 khz samples as opposed to its original 44.1 khz. naturally, because the sound was initially recorded at 44.1 khz, this would not change the particular sound quality pertaining to sampling of the sound - it would stay at 44.1 khz, right ?

what im wondering is, does sending 96 khz analog signal and 44.1 khz analog signal to an analog speaker set, differ at all ?

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #7 –
noone ?

due to work and so on, i am not at a position to start from scratch to build good learning on audio from its basics, timewise. it would be great if someone gave me the basics of audio, audio quality, what the terms that matter mean, ie, like what difference sending a 44.1 khz and 96 khz analog signal to an analog speaker set makes and so on.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #8 –
You're not in fact sending a 44.1KHz electrical waveform to the speakers, but an analog waveform* that is reconstructed from a set of samples. The maximum frequency that can be represented by those samples is exactly half of the sampling frequency, because you need at least two samples to reconstruct a periodic signal. In other words, a sample rate of 44.1KHz can at most express a frequency of 22,050Hz.

The speaker itself is constructed in such a way that it "responds", and thus outputs, to a limited bandwidth of frequences. Since it's a physical object, it's not possible to construct a speaker that responds to all frequencies with the same output level i.e. volume. If the specs for a speaker say it responds from 20-20,000Hz, that means any input signal above 20KHz will yield significantly decreased volume. Since it's unlikely that you can adequately hear a 20KHz sine wave — let alone distinguish those frequencies in real-world music — and you'll understand why most here think 96KHz consumer audio is a waste of disk space and money.


I've probably left out some details or inadequately described some aspects. I am ready for my beating.

*) for some reason I'm not sure if it's a variation in voltage or variation in amperage, and a quick forage on wikipedia doesn't really clear that up, but eh. My electro-fu is weaker than I'd like it to be.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #9 –
so then basically, we are sending half the sampling rate we have to speakers, right. ie, if we output 44 khz signal, we can at most send 22 - 22 khz to the speakers.

and if the spekers are built to respond to this 22 khz signal, then we are well set.


isnt a speaker that would respond to 20 hz -96 khz possible ? or, an array of speakers which would respond to, say, 20hz - 20 khz, then, 20 khz - 96 khz each etc ?

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #10 –
Quote from: unity100 on
isnt a speaker that would respond to 20 hz -96 khz possible ? or, an array of speakers which would respond to, say, 20hz - 20 khz, then, 20 khz - 96 khz each etc ?

Quite possible.
Question is "why" when one can't perceive those frequencies (at safe energy levels).
Creature of habit.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #11 –
@unity100:  The reason why the bandwidth is half of the samplerate ( 44.1Khz -> 22.05Khz ), is not a consequence of the number of speakers. If you want an even simpler example, you cannot draw a line with only one dot. You need two dots and then you can draw a line between them *.

Example: \/\/\/\ . That is 7 lines. There would be 8 dots in there (because each dot has two lines, except those at the end).
A frequency is defined by a period, and in this case, the period is every two lines: 1st period: \/ 2nd period \/ , 3rd period \/ ...



There are speakers that can reproduce sounds higher than 20Khz, and amplifiers, of course. What cannot be done is what you seem to imply that one would extend the range of a speaker with another speaker. That's il·logical.

The frequency is the result of how fast the speaker vibrates. If you have two speakers that vibrate, one doesn't make the other vibrate faster.

What exists is different types of speakers, which are able to vibrate at different frequencies. That's why speakers use to have more than one cone: a big one, and several smaller and smaller, sometimes even metallic ones.





* Ok, mathematicians would say I've drawn a segment, not a line.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #12 –
Quote
There are speakers that can reproduce sounds higher than 20Khz, and amplifiers, of course. What cannot be done is what you seem to imply that one would extend the range of a speaker with another speaker. That's il·logical.


no, what i meant actually was, would there be any point to a 96 khz signal, if there were speakers that covered the entire spectrum - i said this by assuming that it would not be possible to make a single speaker which had, say a 10 hz to 96 khz response range. let me rephrase it :

lets say it was possible to make a 10 hz to 96 khz response range speaker - then would there be any point to sending a 96 khz signal out of the amp ?

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #13 –
Can you hear any frequencies above 20 kHz, as most other human beings cannot?

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #14 –
Quote from: unity100 on
lets say it was possible to make a 10 hz to 96 khz response range speaker - then would there be any point to sending a 96 khz signal out of the amp ?


If you are performing experiments with bats' echolocation, then yes.

If you are not performing these experiments, then no.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #15 –
isnt it possible that the higher vibration rates can be sensed through the body, instead of being directly heard through ears ?

http://en.wikipedia.org/wiki/Sonic_weapon

same principle.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #16 –
Quote from: unity100 on
isnt it possible that the higher vibration rates can be sensed through the body, instead of being directly heard through ears ?


Possible.
But note I explicitly used the word "perceived", not "heard", anticipating this line of argument.

Demonstrate perception of said frequencies at safe listening levels and you're in for some serious grant money!
Creature of habit.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #17 –
Quote from: unity100 on
lets say it was possible to make a 10 hz to 96 khz response range speaker - then would there be any point to sending a 96 khz signal out of the amp ?


As far as I know, the auditory range humans can perceive is 10-20,000 Hz. Any wave with frequencies outside this range may be perceived by other body parts (Example: 1 Hz soundwaves can be sensed via touch), but can't be heard with our ears.

Sending a 96kHz audio signal to a speaker is usually a waste of resources, unless you want to create an electronic mosquito repellant or dog whistle.

Wikipedia is your friend: Search  "Analog to Digital Converter", "Digital to Analog Converter" to know a little more about resolution and sampling (Example: 16/44.1)

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #18 –
Quote from: albertoh on
Sending a 96kHz audio signal to a speaker is usually a waste of resources, unless you want to create an electronic mosquito repellant or dog whistle.


A bit offtopic, but the mosquito sound is around 17Khz. Enough to be played with the standard CD-Audio range.

I used to hate it where they had it in some pubs. I had 24 years and that made me leave the place in less than 10 minutes. Guess it did its function, just at the wrong age range.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #19 –
Quote from: albertoh on
Quote from: unity100 on
lets say it was possible to make a 10 hz to 96 khz response range speaker - then would there be any point to sending a 96 khz signal out of the amp ?


As far as I know, the auditory range humans can perceive is 10-20,000 Hz. Any wave with frequencies outside this range may be perceived by other body parts (Example: 1 Hz soundwaves can be sensed via touch), but can't be heard with our ears.


that is what im saying. i am proposing the possibility of people being perceptive to frequencies they cant hear based on what we have included in this thread. is it too far fetched to say people would enjoy certain vibrations and harmonies in their sensitive body parts, and this could reflect in the enjoyment of a music piece ?

in the end, sound signals are just converted to electric signals that are sent to our brain from our ear. would it matter if the signal was sent through different channels ? we already know it is possible to invoke certain enzymes or bodily reactions by sending certain signals to brain. (actually sound weapons use that principle, even if it is generally through the hearing mechanism).

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #20 –
You post as if the perception of ultrasonics is somehow an unexplored field.  I suggest you read up on the existing research before reinventing the wheel.
Creature of habit.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #21 –
Quote from: unity100 on
is it too far fetched to say people would enjoy certain vibrations and harmonies in their sensitive body parts, and this could reflect in the enjoyment of a music piece ?

It's not, but only if you can demonstrate (under controlled conditions, like an ABX test) that the addition of inaudible frequencies can make a (no matter how) perceptible difference. AFAIK, this has not happened yet.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #22 –
Quote from: unity100 on
is it too far fetched to say people would enjoy certain vibrations and harmonies in their sensitive body parts, and this could reflect in the enjoyment of a music piece ?


And how can you be sure that this sensation is real and not a delusion?.

Quote
in the end, sound signals are just converted to electric signals that are sent to our brain from our ear. would it matter if the signal was sent through different channels?

Yes, it matter. Our ears are attuned to the "air" channel. Change the channel to "water" or "mucus" and the sounds you hear will be distorted and attenuated. The same applies to "bone", "tissue", "muscle", etc.

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #23 –
It seems to me, the discussion of possible benefits of reproducing the frequency range from 22.05kHz to 48kHz is irrelevant to the topic at hand. The resampling of CD quality material to 96Khz (as is done here) will not introduce any significant content above the original Nyquist frequency (22.05kHz) unless the process is broken. Let's talk about things that actually exist.

unity100, what are the "max settings" you refer to? Are they what you posted above or have you introduced additional processing?

Ubuntu + pulseaudio + max settings + m-audio fast track pro = beyond i

Reply #24 –
watch out Nezmer.  I got busted for making a less drastic statement than you.  I'm still at a 33% warning level because of it.  I said something like "OSS is more audiophile".  What I meant was that some audiophiles were snobbish about it.  What it was taken to mean is that I claimed it was better than ALSA.  and I didn't reference an ABX. 

so, if you have it, for your own good, change your statements to "so and so have reported OSS to be better and here's my reference to so and so." YMMV

or here's my ABX data.