IPB

Welcome Guest ( Log In | Register )

2 Pages V   1 2 >  
Reply to this topicStart new topic
Playing 44.1 music at 44.1 vs 96, What's the best sampling rate to use to play 44.1 music?
Jp4ragon
post Apr 13 2013, 15:13
Post #1





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



So my speakers (Logitech z5500) and my soundcard (Xonar DG) allow me to select the PCM sampling rate I can listen to music at.

I notice most music is 44.1, but it sounds different (almost better, I'd even say) when I oversample and listen to it at 96. But, does oversampling distort the sound/affect the quality? Should 44.1 music always be played at 44.1?

I'm aiming for the truest sound, and I've read that when oversampling, it can distort the way the music was supposed to sound due to interpolation. Any thoughts or suggestions?
Go to the top of the page
+Quote Post
SonicBooom!
post Apr 13 2013, 15:25
Post #2





Group: Members
Posts: 118
Joined: 16-February 11
Member No.: 88210



What you might be hearing can be just expectation bias so always do some ABX first. IMHO, I don't really think oversampling can actually improve the music and if it did, it may be so subtle. And also, "the truest sound" can be very difficult and can be expensive to achieve, as it include, perhaps a studio-grade equipments to sound almost true.


--------------------
sin(α) = v sound/v object = Mach No.
Go to the top of the page
+Quote Post
greynol
post Apr 13 2013, 15:34
Post #3





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



The lion's share of the expense need only be in the speakers with careful consideration payed to the listening environment.

PS: I'm leaving the discussion open so that we may talk about how the OP might test the validity of his claim and then discuss the results of such a test.

This post has been edited by greynol: Apr 13 2013, 15:39


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 13 2013, 15:36
Post #4





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



QUOTE (SonicBooom! @ Apr 13 2013, 16:25) *
What you might be hearing can be just expectation bias so always do some ABX first. IMHO, I don't really think oversampling can actually improve the music and if it did, it may be so subtle. And also, "the truest sound" can be very difficult and can be expensive to achieve, as it include, perhaps a studio-grade equipments to sound almost true.
It's a pretty definitive difference. I have someone change the sampling rate from 44.1 to 96 and vice versa a few times and I can differentiate between the two. As for ABXing, to prove this, how can you ABX based on sampling rates? I don't think that's possible, as the only way to change the output is to manually adjust it in my soundcard settings, as I have been doing.
Go to the top of the page
+Quote Post
greynol
post Apr 13 2013, 15:42
Post #5





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



You already violated TOS8 once; now twice. Please don't do it again.

This post has been edited by greynol: Apr 13 2013, 15:44


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 13 2013, 16:31
Post #6





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



QUOTE (greynol @ Apr 13 2013, 16:42) *
You already violated TOS8 once; now twice. Please don't do it again.

So if I were to simply, without inserting my differentiation claims, ask what sampling rate should I be playing 44.1 music with--44.1 or 96, is that acceptable?

If that is an appropriate question, what would the answer be?

EDIT: also, I apologize for inserting that claim into the original post. This is the place to be precise and I should have been more careful--so that's my bad.

This post has been edited by Jp4ragon: Apr 13 2013, 16:37
Go to the top of the page
+Quote Post
SonicBooom!
post Apr 13 2013, 17:30
Post #7





Group: Members
Posts: 118
Joined: 16-February 11
Member No.: 88210



First of all, what format do you use? I hope it's lossless, because playing a lossy MP3 in an upsampled frequency of 96 kHz is somewhat feels like nonsense to me.

Second, because every one of us can be very different in the perception of sound, every one might have their own meaning of "true" or perfect sound. As I have said, this can be very difficult to achieve, why not aim for a better sound? After all, it is for your own enjoyment. If it sounds fine to you (meaning at this level, you cannot tell a lossless from the lossy) then you do not have to do this oversampling stuff. In short, "what is better" can vary from person-to-person, equipment, and environment so if you think that playing a 44.1 music in a 96 output is good enough at least for you, then go with it.

This is only my opinion, no bashing needed smile.gif


--------------------
sin(α) = v sound/v object = Mach No.
Go to the top of the page
+Quote Post
Apesbrain
post Apr 13 2013, 17:55
Post #8





Group: Members
Posts: 490
Joined: 3-January 04
Member No.: 10915



As SonicBoom said there is no "best" sampling rate. Technically, there is no reason to play 44.1 files at a rate other than 44.1, but if another rate sounds better to you then use it. There is no real harm; you're not altering the original files, just playing them back.

You may be "hearing things" due to expectation bias or maybe there is some peculiarity of your Xonar card or your Logitech speaker system -- or of the way they interact -- that alters the sound when different sampling rates are used. If you're that interested, you might be able to measure any difference using an application like Audio Diffmaker. Just curious, is your soundcard connected to the z5500 controller via digital cable or analog?
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 13 2013, 19:04
Post #9





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



Sonic,
I play all my music in FLAC. What you're saying makes sense tho. Maybe I am overcomplicating it.



Apesbrain,
Would audio diff maker help run a double blind like that? And the z5500 is connected to that Xonar with an optical cable, SPDIF.
Go to the top of the page
+Quote Post
Fabith
post Apr 13 2013, 20:28
Post #10





Group: Members
Posts: 1
Joined: 8-April 13
Member No.: 107571



Make a Oversampling have no sense... Music producers mixing down 24/96 to 16/44.1 with technologies that allow you to lose the least possible audio quality. So the difference between 16/44.1 audio and another 24/96 is almost imperceptible.

PD:Sorry for my poor english :S
Go to the top of the page
+Quote Post
Destroid
post Apr 13 2013, 20:54
Post #11





Group: Members
Posts: 550
Joined: 4-June 02
Member No.: 2220



That said above goes with the argument that up-sampling can not improve sound quality but may change the sound of the play back.

Perhaps one of those old AC97 sound cards with fixed 48KHz sample rate cards might lessen the mangling of the sound with an "easy" 2x re-sampling (not very likely unless the sound card is really old, i.e. PCI), or- as already mentioned- the "expectation bias."

OP: check out the Xiph article on Neil Young and high-resolution, it covers a lot.


--------------------
"Something bothering you, Mister Spock?"
Go to the top of the page
+Quote Post
greynol
post Apr 13 2013, 21:16
Post #12





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Of course, one might think resampling issues we resolved over a decade ago. Nope.
http://www.hydrogenaudio.org/forums/index....showtopic=86676

We also had a recent discussion where samples were submitted that showed changes in equalization, thought I don't know that this outcome was independently verified.


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
MikeFord
post Apr 14 2013, 05:53
Post #13





Group: Members
Posts: 132
Joined: 22-May 04
Member No.: 14241



Many technical points do not have logical answers when it comes to PC audio. Keep it simple, do what works best with your system and drivers. If possible use a direct mode like ASIO or WASAPI, IF it works well on your system.
Go to the top of the page
+Quote Post
sinnet3000
post Apr 14 2013, 06:11
Post #14





Group: Members
Posts: 1
Joined: 14-April 13
Member No.: 107659



One problem that has not been mentioned is that maybe your Xonar soundcard works with 96 khz audio internally, and the upsampler has some flaw. I doubt it, but still is a real possibility.

QUOTE (MikeFord @ Apr 13 2013, 21:53) *
Many technical points do not have logical answers when it comes to PC audio. Keep it simple, do what works best with your system and drivers. If possible use a direct mode like ASIO or WASAPI, IF it works well on your system.


I agree with this. You didn't mention how you are interfacing with the Xonar soundcard, if you are not using a direct mode, Windows resamples all your audio to a common sample rate to mix audio for different applications. Anyway, as Apesbrain mentioned, it if feels better for you why not use it?? Your audio is still intact. I feel there is not need to argue about that.

This post has been edited by sinnet3000: Apr 14 2013, 06:21
Go to the top of the page
+Quote Post
MikeFord
post Apr 14 2013, 12:41
Post #15





Group: Members
Posts: 132
Joined: 22-May 04
Member No.: 14241



I didn't notice until now you are using a Xonar sound card, you have my sympathy as I just put a Xonar DS in my media system.

I have also noticed a sonic difference between 44.1k and 96k, sometimes I get pitch shift with about 10% faster playback. I suspect what happens is that 44.1k is shifted to 88.2k then without correction played at 96k.

I wish I could call it better, but the sensation I get with many settings using this card is that simply something is wrong, distortion of some kind is being introduced. To make any serious progress I suggest you list out every element of your system and connection type as well as settings.
Go to the top of the page
+Quote Post
Propheticus
post Apr 14 2013, 13:14
Post #16





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



For all the people with Xonar cards that are experiencing buggy behaviour: try the UNi Xonar drivers. I'd recommend choosing the low-latency C-media panel version.

Some notes: front panel switching will no longer be automatic (need the little app fpswitch.exe provided on the website) and stereo upmixing to 5.1/7.1 is disabled by default but you can re-enable it.

Since the Xonar drivers are somewhat strange, WASAPI (even exclusive) will not guarantee up-sampling will be skipped and device samplerate will not be matched to the output rate. Instead the preferred samplerate you have selected is used. The only way to be sure no resampling is done is by using the Xonar ASIO drivers. These do switch the card's clock to whatever the media player's output samplerate is (as long as it's supported..if not you'll get an error of course.) Resampling, if any is needed, should be performed by Foobar2000 (SoX resampler for instance) as the windows mixer is bypassed entirely. Also be aware this means that the windows volume control will not affect the volume, and output will be 100%! To limit volume use Foobar's volume control or an external analog volume knob.

This post has been edited by Propheticus: Apr 14 2013, 13:15
Go to the top of the page
+Quote Post
stephan_g
post Apr 14 2013, 16:07
Post #17





Group: Members
Posts: 197
Joined: 6-June 10
From: Bavaria
Member No.: 81240



Will we ever get to the point where soundcard drivers just work? *sigh*
Well, I s'pose you could make a card that's UAA compliant, and hope that OS manufacturers eventually get the remaining kinks in their standard drivers worked out.

For now, I run the codec at 44.1 kHz and have Foobar resample everything to that, as suggested above. Assuming DAC interpolation filters are resembling something half-decent, that generally gets the job done. When using an NOS DAC, I'd feel compelled to upsample to the highest sample rate it supports. (Which essentially amounts to oversampling in software, but you don't need to tell anyone. wink.gif Plus, the SoX resampler in highest quality is beyond any doubt.)

A catch when using resampling: Intersample-overs may occur, just as during D/A. A need for 1..2 dB of extra headroom is not that uncommon. The most elegant fix is simply using ReplayGain, as correspondingly "hot" recordings are always brought down in level (IME), and it makes life with a CD collection covering the last 3 decades that much easier anyway.
Go to the top of the page
+Quote Post
Propheticus
post Apr 14 2013, 16:13
Post #18





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



I see no use in upsampling... it isn't recovering/creating info that isn't there in the lower samplerate source. The D->A process is a form of (extreme) upsampling, from discrete to continuous. A software up/oversampling step in the middle does not make sense i.m.h.o. Except when your soundcard or DAC really has significantly better (as in audible...which is unlikely) audio specs when you feed it a particular sample rate.
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 14 2013, 16:15
Post #19





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



Thanks for the replies everyone, this has been really helpful.


QUOTE (MikeFord @ Apr 14 2013, 06:53) *
If possible use a direct mode like ASIO or WASAPI, IF it works well on your system.
I had never heard of this before so I did some research. I use Foobar2000 and just now installed the WASAPI input component into Foobar.

What do I do/say if I perceive a change in the audio quality without violating ToS 8 again? For example, IF I were to say that there was a definite change after I switched the playback option in Foobar from "DS: Primary Sound Driver" to "WASAPI (event): S/PDIF pass-through device (ASUS Xonar DG audio device)" and wanted to thank you guys for enlightening me to the notion of interacting directly with my sound card with WASAPI, how would I do that? Would there be a way to ABX this?

This post has been edited by Jp4ragon: Apr 14 2013, 16:18
Go to the top of the page
+Quote Post
Propheticus
post Apr 14 2013, 16:20
Post #20





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



You don't, because the Xonar drivers are not bypassing the windows mixer in WASAPI mode like they should. DS hooks into WASAPI as well, so it's unlikely there's a difference... if there is, you might be bypassing detrimental sound 'enhancement' settings in the xonar control panel.
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 14 2013, 18:16
Post #21





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



QUOTE (Propheticus @ Apr 14 2013, 17:20) *
You don't, because the Xonar drivers are not bypassing the windows mixer in WASAPI mode like they should. DS hooks into WASAPI as well, so it's unlikely there's a difference... if there is, you might be bypassing detrimental sound 'enhancement' settings in the xonar control panel.
Okay I noticed the difference with the WASAPI is due to a change in volume. What is happening is that the volume can then only be controlled by Foobar which was at a higher setting than my main controls and switching to the WASAPI output then dictates that the volume is controlled only through Foobar. And when I then try to control the volume through the main volume control nothing happens.

Why is that?
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 14 2013, 19:00
Post #22





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



Propheticus,
Would I need to change my drivers then for the Xonar to ASIO to ensure no resampling is done?
Go to the top of the page
+Quote Post
Propheticus
post Apr 14 2013, 19:12
Post #23





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



Well, the fact that your volume control does not work might mean the windows mixer is actually bypassed like it should. My info might be outdated...and perhaps the latest xonar drivers do play nicely with wasapi.
All you should need to use Asio is the Foobar component for Asio output. But if playback is working now I don't see a reason to switch from WASAPI.
Go to the top of the page
+Quote Post
Jp4ragon
post Apr 15 2013, 05:07
Post #24





Group: Members
Posts: 13
Joined: 26-December 12
Member No.: 105401



Ok well there's a new problem--when using WASAPI, I get a bunch of intermittent clicks and pops during playing songs. That doesn't happen when I use DS: Primary sound driver. I tried installing the ASIO Foobar component to see if I heard clicks or pops with that, and ASIO seems to work fine. I guess it doesn't matter as it seems people think ASIO is just as good.

However, I cannot control/adjust the volume or any of the soundcard settings through the Xonar DG audio center (accessed through control panel) when I'm playing music in Foobar using the ASIO or WASAPI component. Why is this--is there a way around this?

I understand that ASIO allows you to have a direct connection with your soundcard, but then how do I adjust my soundcard settings when using ASIO?

Also, does using ASIO using result in an audible difference when compared to using DS: Primary sound driver? Is this something I should be worried about? The only thing I can tell is a volume difference when switching back and forth between the two....

This post has been edited by Jp4ragon: Apr 15 2013, 05:09
Go to the top of the page
+Quote Post
Propheticus
post Apr 15 2013, 09:24
Post #25





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



Volume control in windows is bypassed when using either asio or wasapi. This is normal. If DS works properly, just use that. You probably wont be able to tell an audible difference.
Go to the top of the page
+Quote Post

2 Pages V   1 2 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 22nd September 2014 - 01:17