Topic: Wasapi, Asus Xonar cards and resampling... (Read 41394 times)

Mikkel · « **on:** February 17, 2013, 12:04:22 pm »

Hi,

Some of you may know I'm struggling a bit with ASIO on my Asus Essence ST card. So for the moment I'm using Wasapi - Event Style.
However, I'm not entirely sure if wasapi bypasses the ASUS Control Panel (which resamples to what-ever sample rate is set in the control panel).

Do any one know if this is true, and whether the control panel can be disabled to allow the card to pick the right sample rate for the audio file that is playing?

Best regards,
Mikkel

InflatableMouse · « **Reply #1 on:** February 17, 2013, 12:16:27 pm »

I believe this is not the case with the Uni drivers, but I have no idea how to check it to be honest.

mwillems · « **Reply #2 on:** February 17, 2013, 04:06:59 pm »

Quote from: Mikkel on February 17, 2013, 12:04:22 pm

Hi,

Some of you may know I'm struggling a bit with ASIO on my Asus Essence ST card. So for the moment I'm using Wasapi - Event Style.
However, I'm not entirely sure if wasapi bypasses the ASUS Control Panel (which resamples to what-ever sample rate is set in the control panel).

Do any one know if this is true, and whether the control panel can be disabled to allow the card to pick the right sample rate for the audio file that is playing?

Best regards,
Mikkel

If you have the Unified low latency driver, the Asus control panel doesn't load at startup and doesn't work its tragic latency-producing magic. But even though it's not running, the settings you made in it will continue to apply until changed.

You seem to be trying to get things set up so that JRiver can automatically match the sampling rate the card is expecting to the media you're playing. My understanding is that this isn't possible for these cards with WASAPI, only in ASIO. The wiki (here: http://wiki.jriver.com/index.php/Audio_Output_Modes ) says that "[t]he only way to prevent a Creative Labs X-Fi based card or Asus Xonar card from resampling all incoming audio is to use ASIO and the driver that came with the card. With these cards, WASAPI exclusive will not change the master clock of the card."

I'm sorry ASIO is still giving you grief :-(

InflatableMouse · « **Reply #3 on:** February 18, 2013, 06:14:59 am »

Quote from: mwillems on February 17, 2013, 04:06:59 pm

The wiki (here: http://wiki.jriver.com/index.php/Audio_Output_Modes ) says that "[t]he only way to prevent a Creative Labs X-Fi based card or Asus Xonar card from resampling all incoming audio is to use ASIO and the driver that came with the card. With these cards, WASAPI exclusive will not change the master clock of the card."

Does that count for the uni drivers as well then? I was under the impression it wasn't but I don't know for sure, I have no means to check it as I'm using the analog outputs (ie, no leds or display on the receiver to tell me what its receiving).

nwboater · « **Reply #4 on:** February 18, 2013, 10:26:20 am »

I just posted the following at the Unified Drivers Blog http://brainbit.wordpress.com/2010/07/19/asus-xonar-unified-drivers/comment-page-45/#comment-10145 My post may not appear for a little while as it's waiting for moderation.

"Some of us on the JRiver Forum have questions about upsampling:
1. We believe that when using ASIO with the Asus Essence ST/STX upsampling is bypassed because ASIO communicates directly with the soundcard. Do you know if this is true?
2. We are wondering the same thing when using WASAPI Exclusive?

In addition I have the CPanel version of your drivers installed. When I am using the soundcard are there any differences between running with the CPanel open or closed? I don’t have any effects or DSP turned on in the panel.Again it would be interesting to know the same thing for both ASIO and WASAPI Exclusive.

Thanks very much for whatever information you can provide. Have been using your drivers for quite awhile and really appreciate what you have done for us.

Cheers,
Rod"

InflatableMouse · « **Reply #5 on:** February 18, 2013, 11:08:21 am »

Thanks Rod. Lets see what that brings.

mwillems · « **Reply #6 on:** February 18, 2013, 12:02:33 pm »

Quote from: InflatableMouse on February 18, 2013, 06:14:59 am

Does that count for the uni drivers as well then? I was under the impression it wasn't but I don't know for sure, I have no means to check it as I'm using the analog outputs (ie, no leds or display on the receiver to tell me what its receiving).

I'm also using the analog outs so have no proof, but here's what I do know:

1) I previously used a xonar DX and there was a relay "click" whenever it changed sample rates, i.e. if I went into the control panel and changed the sample rate there was an audible click every time. When I used JRiver's ASIO output mode with it, it would perform the same click whenever I changed to a media type with a different sample rate. It was quite loud so I had several opportunities to observe it. I did not ever hear that "click" when using WASAPI mode which suggests that there was no sample rate changing going on (on the card) when using WASAPI. This was with the unified drivers.

2) I just performed a simple listening test with 44 KHZ 16 bit source material on my Essence ST. The ST was initially set in the windows playback device properties panel (and in the Asus control panel) to use 24-bit, 196 KHz. I tried playback to the card with WASAPI event style and with ASIO with all other settings identical. JRiver shows no difference in the sample rate of the output format or the audio path, but it sounds really, drastically different. I then tried setting the card's output settings to 16-bit 44KHz and tried the same comparison. They didn't sound identical, but it was much, much closer, close enough that, while I thought I could hear a difference, I'm not sure I could reliably tell which was which.

Obviously I'll be very interested to hear what the folks behind the Unified Drivers have to say, but if anyone is curious, you can try the same experiment for yourself and see what you think. Not the most scientific of tests, but you can see for yourself in about five minutes (for what it's worth). If you do try it, I would recommend closing JRiver while changing the sample rates and reopening afterwards, as I had some instability while hot-swapping the rates, which is perhaps to be expected.

InflatableMouse · « **Reply #7 on:** February 18, 2013, 12:12:03 pm »

Cool. I hadn't noticed the clicks, only when changing outputs. I'll try this right now.

Brb

mwillems · « **Reply #8 on:** February 18, 2013, 12:23:48 pm »

Quote from: InflatableMouse on February 18, 2013, 12:12:03 pm

Cool. I hadn't noticed the clicks, only when changing outputs. I'll try this right now.

Brb

Just so there's no misunderstanding: the output change clicks don't happen with the ST, only with my old Xonar DX. The listening test in part 2) of my post was done with an ST/H6 because I couldn't use the clicks to diagnose it (unlike the DX, which makes it more obvious). Sorry if I wasn't clear about that.

InflatableMouse · « **Reply #9 on:** February 18, 2013, 01:29:33 pm »

Oke, I did misunderstood the clicking thing but I did figure out it IS resampling using Wasapi.

Although the relays don't click when changing the samplerate on Essence ST with H6 and they only click when changing the output, there's a very obvious chirp in the audio playback when changing the samplerate in the control panel during playback using Wasapi(-eventstyle).

When I'm playing back 24/96 and I change it to 44.1 in the control panel I can also hear some subtle differences, although I must admit its subtle and the quick test I did doesn't really prove anything of course.

But the fact that it chirps on every change must mean its resampling, why else would it chirp like that? So it accepts whatever MC outputs and resamples it to whatever is configured in that Control Panel.

ASIO completely bypasses that control panel. I've also come to believe during this simple, unconclusive test that ASIO actually sounds better HAHA

(I wish ASIO would work properly with video playback but it doesn't, sound is like half a second ahead of the movie and I can't seem to figure out how to get that right).

mwillems · « **Reply #10 on:** February 18, 2013, 02:16:49 pm »

Quote from: InflatableMouse on February 18, 2013, 01:29:33 pm

Oke, I did misunderstood the clicking thing but I did figure out it IS resampling using Wasapi.

Although the relays don't click when changing the samplerate on Essence ST with H6 and they only click when changing the output, there's a very obvious chirp in the audio playback when changing the samplerate in the control panel during playback using Wasapi(-eventstyle).

When I'm playing back 24/96 and I change it to 44.1 in the control panel I can also hear some subtle differences, although I must admit its subtle and the quick test I did doesn't really prove anything of course.

But the fact that it chirps on every change must mean its resampling, why else would it chirp like that? So it accepts whatever MC outputs and resamples it to whatever is configured in that Control Panel.

ASIO completely bypasses that control panel. I've also come to believe during this simple, unconclusive test that ASIO actually sounds better HAHA

(I wish ASIO would work properly with video playback but it doesn't, sound is like half a second ahead of the movie and I can't seem to figure out how to get that right).

I personally think ASIO sounds better as well (for whatever reason). ASIO works fine for video playback for me, but obviously everyone's set up is different. I assume you've tried videoclock and manually changing the AVsync settings under the advanced pane in video options?

I also improved my video syncing/latency by disabling large hardware buffers in output mode settings, but in your case it sounds like the sound is actually leading the video, which I've never experienced.

dean70 · « **Reply #11 on:** February 18, 2013, 04:28:32 pm »

anything other than ASIO (WASAPI or Kernel) with go through the Windows mixer & be resampled to the control panel setting - you can verify by adjusting the system volume level, even if using the Unified drivers.

nwboater · « **Reply #12 on:** February 19, 2013, 09:31:49 am »

Quote from: nwboater on February 18, 2013, 10:26:20 am

I just posted the following at the Unified Drivers Blog http://brainbit.wordpress.com/2010/07/19/asus-xonar-unified-drivers/comment-page-45/#comment-10145 My post may not appear for a little while as it's waiting for moderation.

   "Some of us on the JRiver Forum have questions about upsampling:
   1. We believe that when using ASIO with the Asus Essence ST/STX upsampling is bypassed because ASIO communicates directly with the soundcard. Do you know if this is true?
   2. We are wondering the same thing when using WASAPI Exclusive?

   In addition I have the CPanel version of your drivers installed. When I am using the soundcard are there any differences between running with the CPanel open or closed? I don’t have any effects or DSP turned on in the panel.Again it would be interesting to know the same thing for both ASIO and WASAPI Exclusive.

   Thanks very much for whatever information you can provide. Have been using your drivers for quite awhile and really appreciate what you have done for us.

   Cheers,
   Rod"

And here is the answer about the Unified Drivers from their author:

1. Yes, the ASIO component is detecting and changing the sample rate according to its source.
You also should hear a pop when you change to music with different sample rate.
2. Don’t know about this one.

I am not sure what you are referring as open, as in running or as in having the windowed opened. I would say that is does not matter if the C-media Panel is running or not. While there are some claims that C-Media Panel offers a better quality compared to Asus one, I am suspecting that Asus Panel is crippling the sound quality, and not that running C-Media Panel is improving it. So not running is just as fine.
Panel options don’t affect ASIO playback. Again I am not sure about WASAPI Exclusive.

InflatableMouse · « **Reply #13 on:** February 19, 2013, 09:56:09 am »

Thanks Nwboater. It's good to get that confirmed.

I've done another comparison between ASIO and Wasapi and that resampling makes some stuff sound mushy.

InflatableMouse · « **Reply #14 on:** February 19, 2013, 09:58:38 am »

Quote from: mwillems on February 18, 2013, 02:16:49 pm

I personally think ASIO sounds better as well (for whatever reason). ASIO works fine for video playback for me, but obviously everyone's set up is different. I assume you've tried videoclock and manually changing the AVsync settings under the advanced pane in video options?

I also improved my video syncing/latency by disabling large hardware buffers in output mode settings, but in your case it sounds like the sound is actually leading the video, which I've never experienced.

Yes, sound is ahead. I need to set A/V sync to 300 for it to be in sync.

nwboater · « **Reply #15 on:** February 19, 2013, 10:01:42 am »

Quote from: InflatableMouse on February 19, 2013, 09:56:09 am

Thanks Nwboater. It's good to get that confirmed.

I've done another comparison between ASIO and Wasapi and that resampling makes some stuff sound mushy.

I'm glad too to have the confirmation.

I've never been a fan of upsampling, or any kind of resampling. I think though that if someone wants to upsample, the results should be much better having MC do it. A year or two ago I tried it in MC and had mixed feelings. Left things setup so it would use native resolution.

Cheers,
Rod

Mikkel · « **Reply #16 on:** February 20, 2013, 03:04:56 am »

Strange why Asus made the cards that way but thanks for the effort in investigating the matter.

Anyway, I have tried testing whether resampling affects the sound. I must confess that when I *think* I can hear a change in sound I often fail in hearing this change when listening again.

I think mainly it is the thought that perhaps the sound is degraded that makes us prefer ASIO rather than whether we can or cannot hear a difference. Just my opinion, though.

Best regards,
Mikkel

mojave · « **Reply #17 on:** February 21, 2013, 12:48:01 pm »

Quote from: Mikkel on February 20, 2013, 03:04:56 am

Strange why Asus made the cards that way but thanks for the effort in investigating the matter.

Since the Asus and Creative cards are intended for the general public, they have to be made that way so that a user can listen to YouTube and still hear his system sounds. Everything needs to be resampled to the same rate in order for the sounds to be mixed/played at the same time. Even if a user pauses the sound in one program, playback can only occur if the same sample rate is used for the sound in another program. To alleviate any playback issues, Asus/Creative chose to resample everything to the same sample rate.

Since ASIO users are more "advanced" and understand the limitation, they can use this mode with the understanding that they can't use the master volume control and can't use programs simultaneously.

InflatableMouse · « **Reply #18 on:** February 21, 2013, 02:19:13 pm »

I understand those points Mojave, but that's why there's exclusive mode. When MC claims exclusive mode over the device, nothing else can play. At that point, it should not resample and just accept the samplerate that comes in (within its specifications of course).

Belarathon · « **Reply #19 on:** March 10, 2013, 07:38:45 pm »

So the bottom line is, the card is re-sampling when using WASAPI? That's a bummer if true. I have a Xonar ST, and it definitely clicks when switching sampling rates with ASIO. How many years has bit been since they introduced this card? You'd think they could come up with a driver that would allow bypassing internal re-sampling without having to resort to ASIO.

Incidentally, my M-Audio 24/96 clicked when changing sampling rates, but my M-Audio Audiophile 192 does not.

Mikkel · « **Reply #20 on:** March 11, 2013, 02:09:35 am »

Quote from: Belarathon on March 10, 2013, 07:38:45 pm

So the bottom line is, the card is re-sampling when using WASAPI? That's a bummer if true. I have a Xonar ST, and it definitely clicks when switching sampling rates with ASIO. How many years has bit been since they introduced this card? You'd think they could come up with a driver that would allow bypassing internal re-sampling without having to resort to ASIO.

Incidentally, my M-Audio 24/96 clicked when changing sampling rates, but my M-Audio Audiophile 192 does not.

I agree it is annoying (especially since asio does not work very well on my system). But then, upsampling anything to above its native sample rate is not going to make a difference to the sound quality, since nothing is added to the original signal. Hence, I have decided to just forget about it and just set the ASUS Control Panel to a fixed sample rate of 192khz.

Best regards,
Mikkel

InflatableMouse · « **Reply #21 on:** March 11, 2013, 02:48:46 am »

I would suggest that if you're going to use upsampling to let JRiver do the work for you, and set the Asus control panel to the sample rate you've configured JRiver upsamples to. This is to prevent the Asus driver to resample because JRiver does a much better job IMO.

There is (to me) a descernable difference when Asus is resampling compared to using ASIO and no resampling. I'm glad I got ASIO to work perfectly with video as well now that I'm back on Windows 7.

Belarathon · « **Reply #22 on:** March 12, 2013, 10:32:12 am »

For me the key to getting ASIO working was to set latency at 40ms, rather than the default 10ms. I guess I can deal with the clicking between sampling rate switching, as the alternative is to use my M-Audio and listen to the hard drive spin down LOUDLY when I shut my PC down. The audio relays are a nice engineering touch on the ST.

raul · « **Reply #23 on:** March 12, 2013, 03:17:34 pm »

Hi guys,

just to clarify something that's in my mind: asio/wasapi/wasapi exclusive mode/kernel streaming are all the same thing in terms of it's overall function, correct? one should NOT be hearing any quality differences between them. all of these output methods are essentially "streaming", correct? they are all bit-perfect reproduction.

i read on one of the "audio" jriver pages that one should just leave the output settings on the highest sample rate that a DAC can reproduce such as 24-bit, as there is no need to change it for each source, as a 16-bit source being upsampled to 24-bit won't degrade sound quality.

here is the quote from the page:

"Output Bitdepth

It is recommended to output to your soundcard or DAC using the highest bitdepth that the hardware supports. This is 24bit for most high-end DACs.
If you play 16bit input, you might feel inclined to output 16bit data even though your DAC is 24bit. This will at best sound the same as outputting 24bit. But it has two important drawbacks:
Transitioning between 16-bit and 24-bit source material will require reopening the audio hardware (so make gapless transitions impossible)
If you apply any digital processing, including volume, the sound quality will be worse"

SURELY though, if one is feeding an external DAC (in my case, headroom's ultra desktop amplifier) a signal/source, that signal/source SHOULD be UNTOUCHED completely for the DAC? i use an S/PDIF optical output from my laptop (dell xps 15) to the ultra desktop, so say for instance, i am playing a song which is 16/44.1... then i should feed my optical cable 16/44.1, NOT 24/44.1 or 24/48, right? since ANY CHANGES to the source before it reaches my external DAC means it is not bit-perfect anymore.

[additional information: the soundcard in my laptop is "realtek high definition audio", however when i went to 'control panel/programs, i saw "advanced audio FX engine" listed on there as well... & the publisher is "creative technology".]

after reading this:

"Notes

The only way to prevent a Creative Labs X-Fi based card or Asus Xonar card from resampling all incoming audio is to use ASIO and the driver that came with the card. With these cards, WASAPI exclusive will not change the master clock of the card."

i am now a bit worried... is my source, unbeknownst to me, being altered? however when i play music using wasapi, the audio path tells me it is DIRECT.

will the audio path feature in jriver immediately notify a user when a signal is NOT bit-perfect, even if wasapi is used?

thanks for reading.

Raul

dean70 · « **Reply #24 on:** March 12, 2013, 05:50:53 pm »

WASAPI bypassing the Windows Mixer seems to be driver specific, eg: I can set a HDMI out to WASAPI and the Windows volume has no effect on output, but another PC with Realtek onboard audio and also the Essense ST resamples with WASAPI.

ASIO seems to be the only interface that works consistently without having Windows resample or control the volume.

Mikkel · « **Reply #25 on:** March 13, 2013, 03:54:58 am »

@Raul:
As Dean70 mentions if you use any Asus card the only way to get bit-perfect is by using ASIO and full line output voltage (volume) to the DAC (and no DSP... I think?). If you turn volume down the noise floor will rise - whether you do it through digital or analog volume control.

If your goal is to bypass the windows mixer, then for any card you can use Wasapi, kernel streaming or ASIO.

I would not worry about bit perfect playback, though. If you think about it: adding more bits and more samples does absolutely nothing to degrade the original sound in terms of what you can hear. It does not make it better either. So whether it is bit perfect or not really means nothing in terms of audible improvements/degredations as long as the bit depth is reasonable and the sample rate is not below 44.1 khz, of course. You can pay hydrogenaudio a visit if you would like some double blind listening tests to confirm this.

So to sum up: don't worry... at least not about this :-)

Best regards,
Mikkel

mojave · « **Reply #26 on:** March 13, 2013, 11:19:03 am »

Quote from: raul on March 12, 2013, 03:17:34 pm

i am now a bit worried... is my source, unbeknownst to me, being altered? however when i play music using wasapi, the audio path tells me it is DIRECT.

will the audio path feature in jriver immediately notify a user when a signal is NOT bit-perfect, even if wasapi is used?

thanks for reading.

Raul

JRiver can't know what your audio card does after it receives the audio. For example, my Creative card will resample if I use WASAPI. JRiver will show it as bit-perfect, but it is still being resampled. What dean70 says is correct.

raul · « **Reply #27 on:** March 13, 2013, 04:27:13 pm »

thanks guys.

there's just a few questions that i posted that weren't clarified/answered for me:
i"ll repost them:
"just to clarify something that's in my mind: asio/wasapi/wasapi exclusive mode/kernel streaming are all the same thing in terms of it's overall function, correct? one should NOT be hearing any quality differences between them. all of these output methods are essentially "streaming", correct? they are all bit-perfect reproduction."

thanks.

in terms of what i'm trying to achieve, i am wanting to get an unaltered signal from the source within the player sent DIRECTLY UNTOUCHED [streamed] {completely by-passing the onboard soundcard} to my S/PDIF cable (itz called 'realtek digital output' on my system).
WASAPI/WASAPI event-style should achieve this quite easily correct?
there is nothing else that i should b worried about in terms of anything else interfering with the signal, correct?

on a side note: from what i'm hearing, it DEFINITELY sounds like MC is streaming it unaltered.

but in the future, i am keen to go for an ASUS desktop or ASUS laptop... i just hope i'm not gonna run into any problems. but itz not like i'm gonna have any use for the onboard soundcard on that machine either... everything via S/PDIF output.

thanks again

DarkPenguin · « **Reply #28 on:** March 14, 2013, 07:33:36 pm »

The spdif output is an output of your sound card.

Mikkel · « **Reply #29 on:** March 15, 2013, 12:39:05 am »

@Raul: It is not possible to say without knowing your soundcard. If it is an onboard chip my guess is it will resample in its own control panel but I really don't know.

raul · « **Reply #30 on:** March 15, 2013, 12:16:38 pm »

Quote from: DarkPenguin on March 14, 2013, 07:33:36 pm

The spdif output is an output of your sound card.

hhhmmmm, but wud u still say that the best way to output the source transparently [to the S/PDIF port -> external DAC] is WASAPI/WASAPI event-style? at the end of the day, NO audiophile wants a bit-altered signal sent to their outboard DAC.

Quote from: Mikkel on March 15, 2013, 12:39:05 am

@Raul: It is not possible to say without knowing your soundcard. If it is an onboard chip my guess is it will resample in its own control panel but I really don't know.

i have a realtek high definition audio card inside my laptop.

i apologise for going around in circles with this, a signal is either bit-transparent or bit-altered. i ALWAYZ want bit-transparent a does everybody who uses JRiver i assume.

thanks

transmaster · « **Reply #31 on:** March 16, 2013, 01:18:05 am »

How about using the DLNA option JRiver has and as I do feeding my DLNA capable Onkyo AVR via my network. I can tell you the sound is astonishingly better then the SB Recon3D THX feeding the same AVR via Toslink. I was going to purchase an Asus Zonar Essence but was put off by the lack of drivers for Windows 8 after reading through this thread I am glad I didn't. I purchased the Recon instead for much cheaper, and just for the Toslink output.

Mikkel · « **Reply #32 on:** March 16, 2013, 06:42:28 am »

Quote from: transmaster on March 16, 2013, 01:18:05 am

How about using the DLNA option JRiver has and as I do feeding my DLNA capable Onkyo AVR via my network. I can tell you the sound is astonishingly better then the SB Recon3D THX feeding the same AVR via Toslink. I was going to purchase an Asus Zonar Essence but was put off by the lack of drivers for Windows 8 after reading through this thread I am glad I didn't. I purchased the Recon instead for much cheaper, and just for the Toslink output.

There should be no difference in audio quality between DLNA and digital output of your soundcard. Regarding the Asus Xonar Essence there is nothing in this thread about its audio quality capabilities that should stop you from buying it... except it is a show stopper not having a driver for it in windows 8. I would gladly show you my RMAA tests of the analog outs to proove that this card delivers supreme (analog) sound quality. But if you output digital audio any card is probably as good as the other since your AVR will then be the main influence of D/A conversion of the audio signal.

Best regards,
Mikkel

transmaster · « **Reply #33 on:** March 16, 2013, 11:46:25 am »

Quote from: Mikkel on March 16, 2013, 06:42:28 am

There should be no difference in audio quality between DLNA and digital output of your soundcard. Regarding the Asus Xonar Essence there is nothing in this thread about its audio quality capabilities that should stop you from buying it... except it is a show stopper not having a driver for it in windows 8. I would gladly show you my RMAA tests of the analog outs to proove that this card delivers supreme (analog) sound quality. But if you output digital audio any card is probably as good as the other since your AVR will then be the main influence of D/A conversion of the audio signal.

Best regards,
Mikkel

There is a considerable difference, at least in the digital area. I agree the ASUS on the analog side is better and I do plan on using one on the next PC I am going to assemble in the next couple of months. I have always used ASUS motherboards so to me an ASUS soundcard makes since. The network connection streams to the DAC's on the Onkyo TX-NR709 and not through the sound card. I am a musician, which in some corners condemns me because according to them I have a brain that can fill in the blanks. The Onkyo has a setting called Pure Audio with a push of a button I can turn the Onkyo into a integrated Stereo Amp everything not associated with powering the front right and left speakers is shut down, everything, equalization, Audyssey, HDMI board, even the display goes on standby. I am very careful to make all sound enhancements are disabled upstream as well. So when I do an A/B comparison between network stream and the TOSlink feed feed from Recon card the difference is evident. I am told the reason for this is Windows 8 has a completely new WASAPI and how the sound card vs the Onkyo work with it. Your thoughts have given me an idea. I have an older and very good Sound Blaster Audigy2 Z5 soundcard, which when I purchased it was the best SC they made. I am going to install this card and using Blue Jeans Cable interconnects use it's analog outs to feed the Onkyo it strikes me that it is a duplication to use two digital feeds for the same thing. I will let you know how it turns out. I am having a great deal of fun with JRiver I have just short of 6 terabytes of music and video files it is a real treat to hear the Onkyo singing through my totally rebuilt Polk Monitor 10B's via the my network. JRiver's transcoder is a good one.

Mikkel · « **Reply #34 on:** March 16, 2013, 12:39:33 pm »

Quote from: transmaster on March 16, 2013, 11:46:25 am

There is a considerable difference, at least in the digital area. I agree the ASUS on the analog side is better and I do plan on using one on the next PC I am going to assemble in the next couple of months. I have always used ASUS motherboards so to me an ASUS soundcard makes since. The network connection streams to the DAC's on the Onkyo TX-NR709 and not through the sound card. I am a musician, which in some corners condemns me because according to them I have a brain that can fill in the blanks. The Onkyo has a setting called Pure Audio with a push of a button I can turn the Onkyo into a integrated Stereo Amp everything not associated with powering the front right and left speakers is shut down, everything, equalization, Audyssey, HDMI board, even the display goes on standby. I am very careful to make all sound enhancements are disabled upstream as well. So when I do an A/B comparison between network stream and the TOSlink feed feed from Recon card the difference is evident. I am told the reason for this is Windows 8 has a completely new WASAPI and how the sound card vs the Onkyo work with it. Your thoughts have given me an idea. I have an older and very good Sound Blaster Audigy2 Z5 soundcard, which when I purchased it was the best SC they made. I am going to install this card and using Blue Jeans Cable interconnects use it's analog outs to feed the Onkyo it strikes me that it is a duplication to use two digital feeds for the same thing. I will let you know how it turns out. I am having a great deal of fun with JRiver I have just short of 6 terabytes of music and video files it is a real treat to hear the Onkyo singing through my totally rebuilt Polk Monitor 10B's via the my network. JRiver's transcoder is a good one.

Looking forward to hear your impressions

. Digital signals really should be transfered unaltered... I will think a bit more about it, though, since I'm certainly no expert in the audio field, to say the least!! You could probably test whether there is a difference by making sure that:source (e.g. music) = windows mixer sample rate = sound card control panel sample rate.

No resampling would be made then and the sound of DLNA vs the soundcard can be compared (except if some unknown mixer or DSP is interferring).

Best regards,
Mikkel

Best regards,
Mikkel

DarkPenguin · « **Reply #35 on:** March 16, 2013, 04:17:33 pm »

Why don't you just use something like a vlink or hiface if you want to guarantee delivery to your DAC?

raul · « **Reply #36 on:** March 17, 2013, 06:54:08 pm »

Quote from: DarkPenguin on March 16, 2013, 04:17:33 pm

Why don't you just use something like a vlink or hiface if you want to guarantee delivery to your DAC?

after doing more research, i realized i'm having doubts for no reason really.
i have never heard of vlink or hiface.
my DAC has 3 input options: USB, S/PDIF optical, & S/PDIF coaxial.
the best out of all these options & most commonly used on hi-end audio equipment is S/PDIF optical. but even coaxial is great.
i specifically bought the Dell XPS L502x laptop for it's high quality optical output.
my aim: the audio stream must remain pure & untouched & be fed directly unaltered {i am a hardcore audiophile without a doubt}. http://en.wikipedia.org/wiki/Bit-perfect

from what i have been hearing from WASAPI/WASAPI event-style, i have been achieving the correct results.

my very first post in this thread was just me in a "state of confusion" u could say, but more research & assurance has helped me out.

if anybody would like to add more regarding the computer audiophile world & bit-perfect goals,
i'm happy to learn more. (as long as u r stating FACTS

)

Thanks

DarkPenguin · « **Reply #37 on:** March 17, 2013, 07:50:53 pm »

Belarathon · « **Reply #38 on:** June 05, 2013, 12:09:30 pm »

If I understand this correctly, when the Asus up-samples, it is merely padding with zeroes. So if I set the Xonar to 192Khz, anything played below that sample rate would not actually be degraded by the upsampling. Am I correct in this assumption?

Edit:

Ah, well, I already know I am wrong, and I'd delete this if could. Nevermind.

Matt · « **Reply #39 on:** June 05, 2013, 12:22:57 pm »

Changing the sample rate is lossy, changing the bitdepth (to a larger value) is lossless.

To avoid lossy resampling with a Xonar, you can use ASIO and enable bit-matched playback. A little more here:
http://wiki.jriver.com/index.php/Audio_Output_Modes#Notes

Belarathon · « **Reply #40 on:** June 05, 2013, 12:24:38 pm »

Quote from: Matt on June 05, 2013, 12:22:57 pm

Changing the sample rate is lossy, changing the bitdepth (to a large value) is lossless.

To avoid lossy resampling with a Xonar, you can use ASIO and enable bit-matched playback. A little more here:
http://wiki.jriver.com/index.php/Audio_Output_Modes#Notes

Thanks Matt. I'll read up.

Mikkel · « **Reply #41 on:** June 05, 2013, 12:56:44 pm »

Are you sure upsampling is lossy? Down sampling would be but I can't see what is lost from upsampling.

Best regards
Mikkel

Matt · « **Reply #42 on:** June 05, 2013, 01:14:02 pm »

Quote from: Mikkel on June 05, 2013, 12:56:44 pm

Are you sure upsampling is lossy? Down sampling would be but I can't see what is lost from upsampling.

Best regards
Mikkel

I suppose it's a question of semantics.

You haven't lost frequency information, but you have introduced a layer of processing. A good resampler (like the one in MC) will be audibly transparent. A bad resampler (like a linear interpolator) can make a mess of the sound.

In general, I wouldn't upsample unless you have a good reason. One good reason is to give other digital processing frequency headroom (ie. avoid Nyquist issues near the threshold of hearing) for things like Convolution. Another good reason is to avoid a low quality resampler used by hardware or the Windows mixer.

mojave · « **Reply #43 on:** June 05, 2013, 01:22:33 pm »

Quote from: Matt on June 05, 2013, 01:14:02 pm

One good reason is to give other digital processing frequency headroom (ie. avoid Nyquist issues near the threshold of hearing) for things like Convolution.

Interesting. Would you recommend something like 88.2 or 96 kHz?

Matt · « **Reply #44 on:** June 05, 2013, 01:55:31 pm »

Quote from: mojave on June 05, 2013, 01:22:33 pm

Interesting. Would you recommend something like 88.2 or 96 kHz?

The conventional logic is that whole ratio resampling is best, like 44.1kHz to 88.2kHz. I haven't instrument tested our resampler enough to know if this actually matters.

As for boundary issues on other processing, I'm only speaking theoretically. It would be interesting to get some of the Convolution authors to comment on if it might help them.

Belarathon · « **Reply #45 on:** June 05, 2013, 02:31:53 pm »

Quote from: Matt on June 05, 2013, 01:14:02 pm

In general, I wouldn't upsample unless you have a good reason.

Ah, well I was hoping to use WASAPI for better transitions between various bitrate source material. Dang Xonar; Cirrus Logic anyhow.

Mikkel · « **Reply #46 on:** June 06, 2013, 02:38:59 am »

Quote from: Belarathon on June 05, 2013, 02:31:53 pm

Ah, well I was hoping to use WASAPI for better transitions between various bitrate source material. Dang Xonar; Cirrus Logic anyhow.

I never did any thorough tests of ABXing 44.1 vs. hardware upsampling to e.g. 96khz with the Xonar. I did try, of course, to AB native sample rate with upsampling 44.1 material and downsampling hi-res material. And while initially thinking I heard a difference, reverting from upsampled to native sampled left out any signs of a (positiv or negative) effect. Same for downsampling of hi-res audio to 44.1. I did it over and over again and ended up concluding that it made no difference to my ears in my setup. My guess is I probably wouldn't hear an effect even if I performed an ABX test, which would be quite in line with blind-tests made on the subject.

Anyway, since ASIO works really poorly on my system (for reasons I don't know) I let the Xonar resample to 192khz using Wasapi (exclusive mode/the old event-style).

@Matt: I do use convolution but I resample after convolution has been applied. I could try and use the JRiver upsampler to provide you with some feedback. I will let you know how that went if I get to it.

Best regards,
Mikkel

Belarathon · « **Reply #47 on:** June 06, 2013, 10:40:32 am »

Asus Xonar owners using ASIO:

I was able to get rid of the clicking between sample rate changes by ticking "play silence" in the ASIO options, and also under track change, untick "Do not play silence." As far as I can tell, that's taken care of it. While I had the ASIO option ticked, I think the option was not being honored because of the track change option.

Mikkel · « **Reply #48 on:** June 07, 2013, 05:38:47 am »

Quote from: Belarathon on June 06, 2013, 10:40:32 am

Asus Xonar owners using ASIO:

I was able to get rid of the clicking between sample rate changes by ticking "play silence" in the ASIO options, and also under track change, untick "Do not play silence." As far as I can tell, that's taken care of it. While I had the ASIO option ticked, I think the option was not being honored because of the track change option.

Could I ask what your latency settings are? I have occasional stutter with 80 ms buffer size in the asio CP an 100ms in jriver bufer settings.

Best regards,
Mikkel

InflatableMouse · « **Reply #49 on:** June 07, 2013, 07:04:02 am »

I get clicks and ticks too if I click aroudn and use windows explorer, outlook or some other programs. ASIO seems very sensitive. I've not heard it during song transition with different bitrate/samplerates, mostly because I tend to play an entire album instead of shuffled playlists.

I'll try it out see what happens.

INTERACT FORUM

Author Topic: Wasapi, Asus Xonar cards and resampling... (Read 41394 times)