• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

ASIO, WASAPI, Direct Sound... is there any difference in sound quality?

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
I registered just to weigh in on this one.. ASIO (using ASIO4All) sounds completely different than WASAPI and DirectSound. In my experience this is regardless of what hardware you’re using, and it’s easy to test for yourself. Don’t even bother trying to find ASIO drivers, just use ASIO4All as it works with everything, including built in sound cards in ten year old laptops. The biggest material difference right off the bat is gain. If I play the same FLAC file through Foobar using WASAPI/DirectSound, the volume is consistent. Flip output to ASIO using ASIO4ALL, 9-12db louder.

i don’t know why this is, but it’s consistent whether I’m using my main system (Computer —> Wadia 121 —> Classe CAP-151), desktop (PC —> Musiland Monitor 02US —> Klipsch THX Promedia) or any combination of headphones and miscellaneous USB dacs I’ve collected. I prefer the ASIO sound, not sure if getting the extra gain up front is why or not. I’d actually love for someone to replicate my findings and help explain them so that it wasn’t a mystery to me. I’d also like to add a +1 to the person who said that when using WASAPI/DirectSound that the bit and sampling rates seem to stay consistent (which implies resampling ((not bit perfect)). Whereas when I use ASIO, if I play a 24/192 file or just use a Foobar add-on to upsample manually I’m met with a satisfying CLICK on my Wadia indicating that it’s locked into the elevated file.

For consistencies sake, if you’re gonna try this, make sure you have track and album gain disabled in Foobar.

A true Scooby Doo mystery imo, Amir?
 

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
9-12dB can be easily measured with cheap or even onboard Realtek ADCs, no need to use AP to verify the difference.

The ASIO vs (something else) thing just repeats itself. Just like in the Win2k to XP era some people believed only KS guarantees "unmolested" audio. Of course things can vary with different hardware/software combinations. Read this drama:
https://hydrogenaud.io/index.php?topic=92856.msg828666#msg828666

According to Amir his AP does not support WASAPI, at most cases he either used ASIO4all or native ASIO drivers from the manufacturers. In some cases (e.g. testing smartphones), the phone doesn't run Windows so he could only copy some test files to the phone to test it.

In many cases ASIO just worked fine, but he also encountered issues like 16-bit truncation and so on. My X-Fi Titanium HD's ADC (TI PCM4220) for example is capable of identifying truncation of my mainboard's ALC892's analog output:
https://www.audiosciencereview.com/...measurements-of-aune-s8-dac.10486/post-289595
 

MC_RME

Addicted to Fun and Learning
Technical Expert
Audio Company
Joined
May 15, 2019
Messages
854
Likes
3,564
There seems to be a big misconception of ASIO and its purpose. ASIO stands for Audio Stream Input/Output, and basically creates a direct connection with audio hardware and the processing software. This is a Windows protocol created (and still licensed) by Steinberg for recording purposes to reduce latency, since in the 80's and 90's, it wasn't unusual to have over 100ms latency when processing audio streams in and out through the OS, which would obviously be a huge issue when trying to sing or play with already recorded tracks, or even monitor your recording live.

Besides latency ASIO solved multiple issues of Windows audio in those times:

- Bit perfect at 24 and 32 bit

- synchronized start/stop of recording and playback (very important for overdubs)

- higher sample rates

- higher channel counts

Things you need when trying to replace the 24 track tape machine with a computer. Microsoft only cared (and cares) about games and home entertainment. For everything else ASIO is a solution that will stay.
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
There's just no pleasing some people! :facepalm:

Uhh, my DAC has separate LEDs for each combination of word length and sampling rate, if 24/96 or 24/192 doesn’t light up, that means it’s being downsampled. This isn’t me being difficult or obtuse, it’s stating the obvious.

It borders on comical that so many “Audiophiles” on ASR seem to hang their hat on the fact that WASAPI includes a mode called “Exclusive Mode” (Omg!) and that there’s a check box that says it’s doing what it says it’s doing, rather than doing any additional verification or digging.

For me, the fact remains that when playing back high bitrate FLAC files, the only one of the three options that actually seems to pass the file unmolested to any of my DACs is ASIO4ALL. Like I said, that’s across multiple setups. The extra gain that I reported is a side effect I’ve noticed over the years that I was curious about and thought someone might have an explanation for.
 
Last edited:

Pluto

Addicted to Fun and Learning
Forum Donor
Joined
Sep 2, 2018
Messages
990
Likes
1,631
Location
Harrow, UK
When you said earlier…
I’m met with a satisfying CLICK
…it's in fascinating contrast to many recent posters here who have moaned about a (barely audible) pop when changing sample rate or from PCM to DSD and have been threatening to take their DAC manufacturer to the Court of Human Rights for this horrendous shortcoming ;)
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
When you said earlier…

…it's in fascinating contrast to many recent posters here who have moaned about a (barely audible) pop when changing sample rate or from PCM to DSD and have been threatening to take their DAC manufacturer to the Court of Human Rights for this horrendous shortcoming ;)

In a world where you are often expected to just trust that things are working as intended, a tactile click accompanied by the aforementioned LED’s is a small but welcome comfort. This isn’t a cheap hobby after all..
 

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
@Dougey_Jones
Could you show the properties pages of your DACs showed volume differences with ASIO4all?
Capture.PNG
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
@Dougey_Jones
Could you show the properties pages of your DACs showed volume differences with ASIO4all?
View attachment 65771

I’m @ work at the moment but will try to put that together later when I get home. Your question doesn’t really make sense though. The volume difference I’m referring to isn’t because I’ve adjusted the volume. The windows master volume and Foobar2000 volume slider remain the same when switching between output modes. As I noted previously, DirectSound and WASAPI are idéntical, while switching to ASIO4ALL requires that I roll the volume way back so I don’t blow my ears out.

Judging by the responses thus far, it doesn’t seem like a lot of people use ASIO4ALL regularly, maybe it would be more helpful for me to do a step by step tutorial that reflects my test setup so that it can be easily replicated.

It’s worth noting that this anomaly persists if I switch to a different player, like KODI for instance, whose developers have trash talked other devs as being incapable of correctly writing a WASAPI driver. Meanwhile their own software does a piss poor job of properly outputting an unmolested multi-channel bitstream compared to something like PotPlayer, which I have nothing but nice things to say about.
 
Last edited:

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
I’m @ work at the moment but will try to put that together later when I get home. Your question doesn’t really make sense though. The volume difference I’m referring to isn’t because I’ve adjusted the volume.
That's exactly what I meant and I didn't misunderstand you. I did not say the volume difference is caused by foobar's volume control. I just want to see your screenshots.
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
That's exactly what I meant and I didn't misunderstand you. I did not say the volume difference is caused by foobar's volume control. I just want to see your screenshots.

I’ll provide the screenshots, but I’m still curious what you’re hoping to glean from a basic properties screen. Driver revision? Whether I’ve installed drivers at all? Please, do expound.
 

scott wurcer

Major Contributor
Audio Luminary
Technical Expert
Joined
Apr 24, 2019
Messages
1,501
Likes
2,821
Besides latency ASIO solved multiple issues of Windows audio in those times:

- Bit perfect at 24 and 32 bit

- synchronized start/stop of recording and playback (very important for overdubs)

Right on that, with my work Dell XP laptop I could use an M-audio USB DUO essentially as a lab instrument (albeit at the time only 24/48) almost two decades ago. Two instances of CoolEdit 2000 would play and record simultaneously perfectly synchronized with no glitches at all. I used the setup to do non-audio R&D saving the hassle of justifying another $20k National Instruments/Labview license and avoiding the learning curve when all I wanted was a simple data acquisition system.
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,168
Likes
3,715
Play a raw DTS or AC3 (Dolby Digital) file, making sure that decoding is done in your connected hardware (AVR) not by the player software.

If it decodes, congrats, your system is capable of bit perfect output 'as is'. If you get white noise, something is messing with your bits before they go out to your cable. E.g., volume not set to 100%; replaygain, EQ, or some other DSP engaged; windows mixer is resampling; etc;
 

hyperplanar

Senior Member
Joined
Jan 31, 2020
Messages
301
Likes
581
Location
Los Angeles
The volume difference I’m referring to isn’t because I’ve adjusted the volume. The windows master volume and Foobar2000 volume slider remain the same when switching between output modes. As I noted previously, DirectSound and WASAPI are idéntical, while switching to ASIO4ALL requires that I roll the volume way back so I don’t blow my ears out.
Are you comparing the two with the Windows volume set to 100? ASIO’s output is independent of the Windows volume, unless your DAC supports hardware volume control through the Windows volume control.
 

Dougey_Jones

Addicted to Fun and Learning
Forum Donor
Joined
May 27, 2020
Messages
552
Likes
460
unless your DAC supports hardware volume control through the Windows volume control.

Download ASIO4ALL and configure it for whatever DAC or built in sound card you have and you'll be able to control the volume using Windows master volume. I haven't found a sound card or DAC that isn't compatible with ASIO4ALL with some tweaking of the buffer sliders as long as you don't have an ancient CPU.

As for Bennetng's screen shots, here you are:

Wadia USB Device.PNG

Wadia USB 2.PNG

Foobar 2.PNG

Foobar 1.PNG

ASIO.PNG
 

MC_RME

Addicted to Fun and Learning
Technical Expert
Audio Company
Joined
May 15, 2019
Messages
854
Likes
3,564
Play a raw DTS or AC3 (Dolby Digital) file, making sure that decoding is done in your connected hardware (AVR) not by the player software.

If it decodes, congrats, your system is capable of bit perfect output 'as is'. If you get white noise, something is messing with your bits before they go out to your cable. E.g., volume not set to 100%; replaygain, EQ, or some other DSP engaged; windows mixer is resampling; etc;

Correct. Let me add that this is a 16 bit perfect test. Dolby/DTS is transmitted via the upper 16 bits only.
 
Last edited:

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,692
Thanks. So no miracle here, just a Microsoft class compliant driver. ASIO4all did not replace it.
https://docs.microsoft.com/en-us/windows-hardware/drivers/gettingstarted/user-mode-and-kernel-mode
userandkernelmode01.png


A dedicated driver provides hardware acceleration via ASIO running in kernel mode. ASIO4all is a user mode driver. The Roland interface I showed in my previous post has a physical sample rate button, and it is the only way to change the unit's hardware sample rate. The Windows audio properties only shows a single sample rate, depending on the button's position.
roland 44.png

roland 48.png


With ASIO4all and the physical button at 48kHz, 44.1k files still play, and that means resampling.
asio4all resample.png


With the manufacturer provided driver with kernel mode support, it showed an error, because this dedicated driver works at a lower level, it really knows what the hardware is doing.
kernel asio error.png


Do the same thing using WASAPI exclusive mode. Unplayable, but the error message is misleading.
wasapi.png


Finally, no matter how I fiddled with ASIO4all, there is no difference in volume, including my Roland USB interface, Creative soundcard and Realtek codec, when compared to other APIs. I don't know why you have such a difference.
 
Last edited:

Aerith Gainsborough

Addicted to Fun and Learning
Joined
May 4, 2020
Messages
853
Likes
1,280
It borders on comical that so many “Audiophiles” on ASR seem to hang their hat on the fact that WASAPI includes a mode called “Exclusive Mode” (Omg!) and that there’s a check box that says it’s doing what it says it’s doing, rather than doing any additional verification or digging.
I am curious: What exactly do you expect me to dig for?

1) My AVR displays the sample/bitrate of the file in question in the "signal info" panel when I use WASAPI-Exclusive. It does not do that when using Direct Sound.
2) I get a rather annoying error message by other programs that want to use the audio device, some even fail to start or lock-up completely.

The only thing that would deviate from "bit perfect" is the fact, that foobars volume control still works (though it should be bit perfect at 100% output). The same thing would be done by the digital volume control of my AVR. I'd rather use foobar or even windows for convenience sake.

Seems to me, that WASAPI exclusive is working as advertised. Can't hear a lick of difference between the APIs anyway, so I take my AVRs word for it. BTW: no odd volume discrepancies between the formats here.
Hardware: X-Fi Titanium-HD for headphone use | Yamaha RX-V377 for speaker/multichannel use ran via optical.

Disclaimer:
I keep around a few "Hi-Res" files for testing purposes but I do not believe in any benefit of listening to them over bog standard 44.1/16 Redbook a.k.a. "CD quality". I also cannot hear any degradation in the sound when resampling under windows (believe me I've tried). Measure via RMAA or Music Scope? Sure. Hear? Nope.
My system is typically the standard 48/24. It's basically a "fire and forget" thing. I only avoid resampling when using VSTi because of latency concerns.
My card has native ASIO drivers, so I never used ASIO4ALL.
 
Top Bottom