Audio Processing in Vista Explained - Page 6

why did you you choose not to implement exclusive mode in in WMP11 and media center ?
If you would I'm sure many more people would use media center (I would).

... with which you say that it would come to this in the XP environment, or the other way around : in Vista this would work out the same as in XP, in non-exclusive mode.

I can't help myself ... FYI :

I have a soundcard with non-certified drivers. Lucky me, because they bypass kmixer.
I've written a dedicated DirectSound player for that (so this is no KS and no ASIO), with a DMA buffer in the soundcard of 112 samples. The souncard won't allow me to go lower ...
The audible result is amazing, mind you, in my system (which is not an el cheapo system).

When I apply all to have -as you say- "an essentially direct path" via the certified drivers of the same soundcard (which do not bypass kmixer), the sound is still amazingly good because of the way I process the data (with the same very low latency), but it is different. It is -if you like- more crispy. Is crispy better ?

No, it psychologically can't be better, because of the stupid reason I can measure it's not fed with a 1:1 bit perfect stream. The differences are marginal, but, all over the place. Looking at the waves (not the data therein) it looks like the least significant bit wobbling to the wrong direction.

Since the theory is -indeed- that when all is applied to let kmixer not interfere (no resampling "needed" etc.), it still touches the sound somehow. This is stupid and unnecessary, once the "rules" have been applied.

What I understand from you, Amir, is that around this phenomenon nothing will change in Vista, as long as non-exclusive mode is applied. Okay, depending on what causes the stream not being bit perfect, it can be that my "all over the place" turns into "here and there" because of the better processing in Vista. However, there's still no reason to touch the sound, again, once the rules are applied.

Please note what this is about : when all is done like it *can* be done, no player needs to be re-written (towards exclusive mode etc.) to achieve bit perfect playback. Also, I don't think it is that easy to have a player working in exclusive mode, because of all the environmental constraints that would apply, thinking of it causing deadlocks within itself, at e.g. capturing sound at the same time.
Just a thought.



I think this is a good example of how far people can be off track from eachother;
No matter how well you mean all, the time that I "allow" my harddisk to be accessed during playing has long gone. Don't think I'm alone ! Just think of by 112 sample buffer ... it couldn't be done ...
Oh well ...




I know, this is in "your" context (of the general audience ?). But with 192KHz sampling it would be quite nothing. In fact, it should be infinite, because of being in the digital domain without D/A conversion, yet. Again, it will be in your context, and I just place it in "our" context ...

Lastly, please note that I am (and not alone again) in the leage of eliminating the $$.$$$ boxed CD player, which should be an easy task because of, say, the PC principles. And no, it's not about saving the $$.$$$, but about achieving much better playback. That too was achieved long gone by many, but it's a struggle. Today it needs the (by MS) not advised DirectKS hence instable driver-systems, or the workaround with ASIO which someone like me wouldn't even begin thinking of writing my own. Add to that the way better results with DirectSound, once you own a driver that is not certified ...
And please don't say that there can't be a difference once there's bit perfect playback, because there's always the (very audible) jitter phenomenon, which in the end comes to latency ...

Thanks.

"(ie X-Fi converts 44.1kHz to 48kHz unless you use bit-perfect mode)"

yes it does.... but does your following audio gear (pre-amp, power amp and we never mention speakers) introduce any distortion/artifacts that are greater than -104db ?? Be honest.....

Sorry guys, with all this discussion of 2-channel PCM, analog outputs, etc., my less-than-audio-savvy head is spinning a little.

I think Amir gave an answer to this question, but I'm not sure I understood it: Will it be possible to pipe multi-channel audio to an external receiver digitally (via optical), and have it do the decoding?

If so, where does Vista's audio processing capability come in? Would it take the audio off of my DVD or music or whatever, "process" it and make it sound better or whatever, and then send it to my sound card, which would compress the stream and send it to my receiver digitally, for glorious audio playback?

If you want the benefits of Vista's audio processing (room correction, bass management, etc.) and the ability to output a multichannel digital stream to your receiver, here's how I interpret it working:

Your player application (WinDVD, TheaterTek, PowerDVD, MCE/WMP) will play (decode) the DVD's lossy compressed DD/DTS audio. The decoding will convert the audio into multichannel PCM - uncompressed audio data. Vista will then process the multichannel PCM using their new DSP features. Next, using a soundcard capable of encoding DD/DTS, you will re-encode the multichannel PCM into lossy compressed DD or DTS audio. Then your soundcard will output the DD/DTS digital stream to your receiver using the S/PDIF connection, where the receive will finally convert the digital stream into multiple analog waveforms, etc, etc.

This means the audio will go through 2 passes of lossy compression. Lossy compression removes audio "data" to reduce the size of the file/stream. Using psychoacoutic modeling, they are able to remove audio data without dramatically degrading the perceived quality of sound. Decoding to PCM doesn't restore any lost audio. Re-encoding the audio back again to the lossy-compressed formats results in more lost data. Repeated decompressions and compressions will result in generational loss that will effect the sound quality.

Or you could just use a really good soundcard (internal or external) with excellent DACs and analog circuitry connected to really great sounding amps.

Correct, but unless your PC is your only audio source (or can process external sources) you will still need a pre/pro or receiver.

I look forward to the day when I can do everything on a PC, but we are (or at least I am) not there yet.

Thanks guys, it all makes sense now

Hey Amir,

How about some love for us guys with home recording studios?

How long till Microsoft starts making magic rocks I can place in my speakers so they sound better?

J/K

Glad to see the high end gaming computers I build myself might be more useful in the not too distance future...

Where is the room calibrator, I can't find it in 5536

Hello Amir,
does Vista support multichannel audio over firewire (IEC 61883-6) including rate adaption (Sony HATS, Pio PQLS).

Two quick comments here. First, the decision whether to decode in the application or pass through to a S/PDIF digital out is a decision made by the application. Certain apps will include a software decoder and decode to PCM. Other apps will choose to output the undecoded signal to the S/PDIF digital out and then rely on an external receiver to do the decoding. Guess what? This requires bit perfect transfer. That leads to the second comment. The only way to achieve bit perfect transfer is to use exclusive mode in WASAPI. As a result, WMP11 (and MCE) on Vista will use exclusive mode to output encoded (compressed) samples to a S/PDIF digital out if you configure it that way. No lossy anything. Bit perfect out. This is the only scenario (compressed encoded content) where WMP and MCE will use WASAPI exclusive mode. The PCM (uncompressed) side, I think we have beat to death here.

If WMP11 (and MCE) on Vista are using exclusive mode to output to S/PDIF, will you be able to take advantage of the new DSP features like room correction and bass management?

My assumption was if you want to use the new DSP features, then you will have to go through a decode/re-encode cycle on audio that is already lossy compressed. This additional cycle will make the audio even more lossy (lossier).

exclusive mode is used for spdif passthrough only, not for pcm over spdif

Eho or Amir :
did you also work on improving the synchronisation of audio and video for multimedia playback ( a la reclock ) ?

Sorry for my lack of understanding...

Lets say I listen to music on my computer using my soundcard (not SPDIF to receiver)

If I use this new "Exclusive Mode" for bit perfect audio playback, will I hear the music (such as 16bit 44.1kHz 2 channel) exactly as the source stream is encoded? Essentially an untouched stream?

or

Does "Exclusive Mode" only mean that the untouched source stream will pass through SPDIF to the receiver untouched?

Again, as a purist, I am trying to see if it is possible to listen to a source stream exactly as it is encoded. Is this possible (assuming the sound card driver also has a bit perfect option so it does not upsample 44.1 to 48kHz)?

if you want to have bit perfect playback you'll have to use an application that can provide that for pcm, foobar does that using asio or kernel streaming. WMP11 or media center will not do it, they will only provide a bitperfect output on spdif with and compressed format (let's say AC3 or DTS).

Sorry if this is a stupid question, but... I currently have a lot of recordings in WMA Lossless, using WMP 10.0. Would I get a better recording if I re-recorded them under WMP 11/Vista? Have you enhanced the Codecs at all?

Joe-

The short answer is no.

By definition the WMA lossless codec will be bit-identical with the original uncompressed audio data after it is decoded. This is the same for all lossless codecs like FLAC, Apple Lossless, APE, etc. The only difference between the lossless codecs is their efficiency and speed of encoding and decoding.

Eho, Amir,

I am still unclear about the PCM playback scenalio when all volume sliders are at their top position, and no other sounds are on. Let's call this "an audiophile use case"(AUC). I understand that WMP and similar apps will do the playback through a filter graph. I also understand that for various reasons, this graph cannot be reduced to 2 pins (a source filter and an audio port pin). But why should intermediate filters do any steam alterations in the AUC? PeterSt observed LSB changes; this is probably due to rounding errors. If you do int->float conversion, then float multiplication, float division and then float->int conversion, it seems plausible you can get them. If this indeed happens as the sound stream goes though the filter graph, it is "just a matter of programming" for the filters to detect the AUC use case and implement a separate code branch where the bits would simply be passed through. Perhaps your PM can include the AUC support into a post-Vista requirement set?

Are the SysFx DSPs being implemented on the Xbox 360?

Whereby Vista could share it's configuration with a 360 acting as a UPnP device or Media Center extender. I'm guessing "no" because it might require Vista to support multiple zones which has already been stated as not being available yet.
That is unless the 360 (via a dashboard update) could provide the means to calibrate a room with a usb microphone and store the configuration locally.

Thanks Amir et al for your contributions to the forum.

Just to clarify, is the room response done once and then the settings stored?

I wont need a mic always attached at the listening position, correct?

This is great news for HTPC owners. I'm currently using a high-end laptop with Realtek ALC880 chipset which is Intel Azalea HD audio compliant. I basically use it's S/P DIF output to drive my Integra Research RDC7 which is a high-end processor which does like bass/management and pro-logic 2 decoding on a matrixed signal even at 24bits 96Khz.

In pure stereo the HTPC even rivals my CD players S/P DIF output which is in every way modified (2 high-end LC Audio clocks, modifications in >100 places). The secret rabbit code really is the best resampler code it's GPL so you could use this. I compared this with many other resamplers and even the hardware ones don't rival the secret rabbits code sound quality.

What would be nice is being able to use a minimalistic & puristic player like foobar, it's secret rabbit code to make it 24/96, then send it's output to the SysFx engine for room correction. Calculating DSP on an upsampled signal just has much more margin as you have more headroom for calculations to be done. If you do this on 16/44.1 you basically lose information. That's why dolby pro-logic 2 music mode on an upsamped 24/96 sounds completely different than on 16/44.1. The first works the last sounds artificial as you drop information because of no extra headroom for calculations.

If foobar has a line input plugin Vista could even be a room corrector for a CD player that you connect to a soundcard with S/P DIF input.

What I wonder is how this will handle a matrixed signal.

You basically do room correction on a 2 channel stream which contains some surround information. By trying to compensate room issues, will this alter the surround information which is encoded as an oposite phased mono signal in the left and right channel.

Does the SysFx engine decode a 2 channel stream that contains matrixed audio internally to a 4.1 or 5.1 data, process those channels individually with room correction, and then re-encode this as a matrixed stream which can be decoded by any pro-logic (2) receiver, or does it encode the stream as dolby digital ?

Another solution would be like this :

- input : 2 ch matrixed audio (these days most cd's even contain out-of-phase information typically in the reverb)
- decode it with SysFx to individual channels, then apply room correction on those channels
- provide this as raw pcm to a virtual multi-channel soundcard like ac3filter which does live re-encoding to ac3 and send this over a high-end S/P DIF interface like the ones from empirical audio

I'm trying to build a chain of high-end & puristic components including a HTPC for some serious processing.



Basically the AC3 option would be equivalent of what ac3filter in software and nvidia nforce chipsets can provide.

So, why not on PCM?

Call me stupid, but is this a technical difficulty related to WMP, or is there some logical reason which escapes me, for not supporting bit-perfect PCM in WMP (and by extension, in MCE?)

Here's a question:

It appears that the sound support of Vista is very, very good. Let's say I wanted to output analog from the HTPC and just run it to amps. How do the DACs from good sound cards compare to something like a HK645 receiver?

Thanks,
Chris

http://theinquirer.net/default.aspx?article=34094

Amir, is this true? I am **very** unhappy with this if this is accurate and there's no way to turn off start up sound!

Please tell me this ain't so...

Edit - just to clarify, I don't care whether the start up sound is customizable or not, as long as I can disable *all* windows sounds.