[Nazo]

So this is kind of hard to explain. I'm not really audiophile enough to really be able to get into the exact specifics of it. It's really hard to explain, but it's like it has less detail in the sound or something of the sort. A while back I found I wasn't really enjoying my music as much as I used to but couldn't really say why. At first I sort of assumed it was my rigged up cheap setup and upgraded everything (I'd been meaning to do it anyway, so that much is ok.) No real change though. I set things up to use only my external DAC+AMP (before I was using optical out from as Asus Xonar DG2 so I'd have the supposed benefits in gaming -- but I don't really use all that extra fake processing anyway, so it didn't really make much difference in the end anyway -- while benefiting from an external DAC so now I switched to just USB straight to the DAC and took out the soundcard entirely. As I suspected, in the end there's just not much difference between 100% software and supposed hardware acceleration if you don't use sound processing effects...)

Then one day, just playing around, I connected my DAC an my Android phone. What I was missing was all there! Now, it's very important to point out here that I've disable the equalizer and I have NO external sound processing like Beats or that sort of crap. (I'm running a relatively clean CyanogenMod setup here. Not quite clear AOSP, but really close and as far as I know the only sound processing is that equalizer which, as I said, is turned off.) So in Windows I tried changing things around. I tried removing all resampling and just in case Windows was still doing some sort of processing I even switched to ASIO in my music player though I've never had tons of luck with ASIO in the past (doesn't play nice with other things, sound skips easily when I do things like loading webpages even though I have a modern multi-core CPU, etc etc.) I also tried the other options like WSAPI, but it's my understanding that ASIO completely bypasses all sound processing by Windows itself. This is with Foobar2000, though I've listened to other things via other players (such as music videos in Media Player Classic Home Cinema just for instance.) I also tried changing the Windows mixer to do 44.1KHz since most of my sources come from CD audio (I had it at 96KHz before figuring on a "highest common denominator" type of thing) but no change that my ears could detect anyway.

Now, thinking back I could ALMOST say that this all started back when I went from Windows XP to Windows 7. But I'm not sure. One thing I will be doing is trying Linux, though it isn't exactly known for ease of bit-accurate sound (I wanted to try to setup OSS as per some suggestions, but I can't seem to get the built-in OSS support to work right and if I try to install the official proprietary OSS it fails on the USB driver portion and then screws up everything such that I have no sound at all. I didn't know enough to figure out how to completely undo whatever it did either and had to actually reinstall.) I'll be doing more testing, but for now I've been feverish and stopped up with something or other and haven't really been able to listen to music very well. Still, I would swear it did seem better despite having to use ALSA. (I did disable the mixer in the ALSA plugin's settings in DeaDBeeF anyway.) If so, it's definitely something Windows is doing and maybe specific to 7 since I would just swear XP didn't do this.

 

[sonci]

Try to keep posts shorter, 
How did you connect phone to  DAC? usb or spidf out? or you are using android DAC, which generally speaking are crap, considering they have no fixed line out but only headphone out..
W 7 sound better than xp, and with asio it should sound the same, unless your card do not support asio,

 

[Nazo]

Try to keep posts shorter

No. I realize people generally have short attention spans, but I'm not leaving out information that might well prove important. I learned the hard way in the past that by doing so I ended up often spending weeks or months trying to solve a problem where providing sufficient information often resulted in someone recognizing the problem right away.

How did you connect phone to  DAC? usb or spidf out? or you are using android DAC, which generally speaking are crap, considering they have no fixed line out but only headphone out..

USB. But my point here is just that even on a completely different machine entirely I'm getting far better results and to demonstrate that even in something like that where audio is kind of an ultra-low priority it can still sound better using the exact same hardware with no changes to anything.

W 7 sound better than xp, and with asio it should sound the same, unless your card do not support asio

It's a USB device, so it works without drivers and can work with a universal ASIO driver. However, just to be certain I installed the official drivers from Schiit's website (later on I may look up C-Media's latest for this specific chipset, but frankly I can't see how it could be related to the drivers since it doesn't sound like it should even with no drivers installed.) Just to be clear, most of the differences I'm talking about should still sound pretty much just as good in even DirectSound or other such simplistic output methods. It's far more extreme than the much more subtle differences normally between them.


I've been feeling a little better today. I didn't listen much, but so far in Linux it is indeed sounding a lot better. I'd say that there's even more soundstage. Something is happening in Windows clearly. And to be clear, I'm not blaming Windows 7 itself specifically, but it could be some difference in something else installed along the way, though if so I couldn't guess what. I'll play around with it later when I can.

 

[cel4145]

No. I realize people generally have short attention spans, but I'm not leaving out information that might well prove important. I learned the hard way in the past that by doing so I ended up often spending weeks or months trying to solve a problem where providing sufficient information often resulted in someone recognizing the problem right away.

That's a mistake to assume that everyone has short attention spans. Some of us are trying to help many people, and so we have limited time to offer. So certainly, providing a good amount of information is a help, but there are superior ways of organizing it if you understand technical communication so that people don't have to wade through your text multiple times.

1) State the problem in a sentence or two.
2) Next. provide a list (short description) of the relevant equipment involved.
3) Then discuss what you have tried/not tried, and other relevant discussion.

So, what DAC/amp are you using? That could be important to tell us. (I had to read through your text more than once to figure out you didn't tell us.)

"What I was missing was all there" doesn't tell us what you were missing. Can you describe the difference?

WASAPI should work just fine for giving you a good audio signal from your PC in comparison to your phone's USB output, and is certainly a better choice than DirectSound.

Window 7's audio handling is (supposedly) superior to XP, so that shouldn't be an issue.

Be aware that if you rule everything else out, your sighted listening comparison process can be the problem. For example, even a difference of +1 db in volume can make one sound better than the other. And then once you *think* one is better, confirmation bias kicks in.

 

[Nazo]

That's a mistake to assume that everyone has short attention spans. Some of us are trying to help many people, and so we have limited time to offer. So certainly, providing a good amount of information is a help, but there are superior ways of organizing it if you understand technical communication so that people don't have to wade through your text multiple times.

You say all that and then you did just skim what I said. I specifically stated why I will not leave out information that could turn out to be relevant and important. For instance, I pointed out why I can't really test very well and explain very well right now.

So, what DAC/amp are you using? That could be important to tell us. (I had to read through your text more than once to figure out you didn't tell us.)

I didn't mention it because it is not relevant. It's the same output hardware on multiple software setups on the same system and multiple external systems. The one non-variable the entire time was the DAC/amp. I was originally using a complex setup involving SPDIF out to another external device which had previously had amazing results but which I was finding to lose that certain something more recently. In retrospect this was probably the same thing, though I don't particularly care to put the soundcard back in and test everything that way around again after I've removed it, packed it away and etc. But, since you won't be satisfied without knowing what I'm using now, it's the Schiit Modi 2U going to a Magni 2U. The problem lies elsewhere.

BTW, Android isn't the best for sound, but it's not as bad as you're making it out to be. Its chief limitation is an insistence on using 16-bit depth except in very special software with their own mechanisms to bypass this, but when the source is 16-bit, even if you output at 24-bit you're only adding null padding and placebo. It also has extreme latency, but that doesn't matter when playing music as that's essentially just a constant stream anyway.

"What I was missing was all there" doesn't tell us what you were missing. Can you describe the difference?

It's very hard to pin down as it is, but as I described I've been quite stopped up and feverish. Along this also comes a headache, so it makes it hard to do very good listening comparisons. All I can say is that the sound is less detailed with less soundstage and etc. It just sounds more dull somehow. I'm not really audiophile enough that I could probably pin it down very specifically even when I'm better though. It doesn't help that I didn't even recognize it whenever it did happen (assuming it didn't come on slowly or something.) So I've been listening that way for a long time and not understanding why I wasn't really enjoying my music much anymore. Really I was just playing around when I plugged the DAC in, fully expecting it to not be compatible as formerly Android devices rarely supported USB DACs and I was more curious just to see if it did (but now it seems 5.0 has added official USB DAC support, so no longer does one rely on whether or not the builder has actually compiled in specific modules or whatever.) It kind of surprised me that it sounded so much better, but made me suspicious enough to try Linux on the same PC only to discover that the issue is something in my Windows setup.

So the main point is I need to look at what COULD affect the sound system and it was this I was hoping to discuss, not my style of writing.

WASAPI should work just fine for giving you a good audio signal from your PC in comparison to your phone's USB output, and is certainly a better choice than DirectSound.

I'm perfectly aware of all this and that's kind of my point. The differences are so extreme that even connecting the exact same USB DAC to an Android device makes the same songs sound far far better. My point is that even with the exact same hardware it's sounding better in everything but Windows.

Actually, when I get a chance I might try setting up some VMs and testing audio in them (since it's USB I can actually direct it straight into the VM, bypassing the host OS and letting the OS inside the VM control the hardware directly.) Again, not really the best time for listening tests, but I'll give it a shot all the same as I still need some of my music fix even when sick.

Window 7's audio handling is (supposedly) superior to XP, so that shouldn't be an issue.

I didn't say Windows 7 has bad audio handling at any time. Just that the differences could have taken place along the way when I switched. That could be due to a driver or other software changes perhaps, I don't know. It's that which I'm attempting to diagnose here and which we've spent a lot of time avoiding and discussing how you don't like how long my posts are instead.

Be aware that if you rule everything else out, your sighted listening comparison process can be the problem. For example, even a difference of +1 db in volume can make one sound better than the other. And then once you *think* one is better, confirmation bias kicks in.

I expected it to sound far worse. My bias was against it. I was just playing around, fully expecting it to not be all that special, but figuring it would at least be better than the stock output path. Frankly I didn't even think I'd be doing it again since it was so inconvenient. However, this works out well enough that I may grab a Fiio E18K or something for on the go emergency uses if I can scrounge up the cash.

 

[cel4145]

You say all that and then you did just skim what I said. I specifically stated why I will not leave out information that could turn out to be relevant and important. For instance, I pointed out why I can't really test very well and explain very well right now.

You are the one who is not reading. I acknowledged that. I was just trying to point out that it was the organization of your post that is why you got the reaction you did, not the length of it. Geez.

I didn't mention it because it is not relevant. . . . But, since you won't be satisfied without knowing what I'm using now, it's the Schiit Modi 2U going to a Magni 2U.

Well, I've been involved in helping thousands of people in audio forums. You'd be surprised how often people think something doesn't help, but it does end up helping. So a basic request is always to ask what the hardware setup is. Instead of arguing about why you don't need to provide the info and then being snarky about it, just provide it. (lol)

BTW, Android isn't the best for sound, but it's not as bad as you're making it out to be. . . .

I didn't say a single word about the quality of Android's audio.

It's very hard to pin down as it is, but as I described I've been quite stopped up and feverish. Along this also comes a headache, so it makes it hard to do very good listening comparisons.

If it is that bad, then this is definitely probably not the time to be worrying about this. Being sick can definitely throw off one's hearing. I would wait until you feel better to worry about it.

I didn't say Windows 7 has bad audio handling at any time. Just that the differences could have taken place along the way when I switched.

I didn't say that you did. But sometimes people will provide a statement of fact on a forum because we don't know what other people know. We are just trying to help. If you already know it or don't find it useful, you can skip it without debating the need for it. (lol)

I expected it to sound far worse. My bias was against it. I was just playing around, fully expecting it to not be all that special, but figuring it would at least be better than the stock output path.

Reread what I said. Unless the volume levels are well-matched, then one can sound better than the other when there is effectively no difference. And then from there, confirmation bias can kick in and make one hear it again and again even if the volume levels are evenly matched in a sighted comparison. Given that your hearing is also off from being sick, this could certainly compound the problem because you likely are not hearing the frequency range the same way that you normally do.

 

[Nazo]

Nevermind. I'll figure it out on my own.

 

[yay101]

Nevermind. I'll figure it out on my own.

My SIG has a guide to setting up bit perfect audio on Linux if you want to give it a go. It really doesn't take very long and can be done on the live USB/ CD with one change.

 

[Nazo]

How strange. It seems like people were saying ALSA was definitely not bit-perfect, though it's supposed to at least not be very bad. I'll admit I was trying to make OSS work more on the principle of the thing than anything else. Anyway, I can say one thing: if you're using ALSA, it locks the device so nothing else can use it. Therefore you don't have to disable PulseAudio since it just can't access the device and outputs silence when other things try to use it. When a player uses ALSA, it locks the device to itself and then PulseAudio no longer has access for the duration. The main thing is just to disable the ALSA resampling option in the things that support it (such as DeaDBeeF) if it's even enabled at all. Really my only problem in this respect is for some reason sometimes when I turn it on my USB DAC isn't detected at all.

 

[nepherte]

There's ALSA and then there's ALSA. There is alsa code that lives directly in the linux kernel, basically a bunch of drivers. There's also alsa code living in user space. More specifically the mixer. To get bitperfect playback you only need the kernel part.
 
yay101's tutorial explains pretty well how to get bitperfect playback. I'd like to point out that by following his explanation, the device/card is grabbed exclusively by the first application that plays back via alsa. This means that you won't be able to get audio from multiple applications simultaneously. To avoid that, what I basically did is keep using pulseaudio for every application except my audio player (mpd):
 

1. Pipe all audio from alsa to pulseaudio:
    

    # /etc/asound.conf pcm.!default { type pulse fallback "sysdefault" hint { show on description "Default ALSA Output (currently PulseAudio Sound Server)" } } ctl.!default { type pulse fallback "sysdefault" }

 

1. Configure MPD to use a card/device directly:

    audio_output { type "alsa" name "my ALSA device" device "hw:0" # Depends on which card/device you want to use! }

 

[Nazo]

I guess my question here is if there is even any point in changing anything with PulseAudio. Doesn't a player like DeaDBeeF using its ALSA output go straight to ALSA? If not, would disabling (partially or not) PulseAudio really change anything?

 

[nepherte]

I guess my question here is if there is even any point in changing anything with PulseAudio. Doesn't a player like DeaDBeeF using its ALSA output go straight to ALSA? If not, would disabling (partially or not) PulseAudio really change anything?

That really depends on how that alsa output is being configured. Is alsa configured to use dmix (a mixer), go straight to pulseaudio, go straight to the device, ...
 
But let's dial back a bit, what is it you want to achieve? From your initial post I understand you were initially using Windows and tried out WASAPI, ASIO, ... Are you not happy about the results you're getting from that setup? In my experience you can get bitperfect playback in pretty much every commonly used OS (or at least I have). The end results were similar on each OS.

 

[Nazo]

Yes, I'm not satisfied with the results in Windows. It's too hard to explain (frankly I'm not audiophile enough to be able to well define it and the person above clearly didn't get this, so no matter if I'm sick or not I can't give you an exact perfect description of it) but basically the output in Windows is just plain more dull all around. My suspicion here is that something is changing it and not for the better (eg definitely not bit-perfect.) I was hoping to look into just what even could do that as, presumably, outputs like ASIO should bypass that sort of thing. In fact, even through PulseAudio it sounds a lot better in Linux, so whatever the issue is, it's not even bit-perfect being required. I believe something is changing the sound and not doing a very good job of it.

I'm almost tempted to reinstall, but that's a real to-do to say the very least. If I can more permanently switch to Linux like I hope it may not be so bad to do later on, but for now I'm not sure as usually I end up going back to Windows for a while (mostly for the gaming. People can say what they like, but WINE will never be remotely close to 100%. Just for instance, I wanted to try joining a friend on a free PSO BB server just for the fun of it only to find out that WINE couldn't handle its anti-cheat engine and despite them knowing about it for years they just blame the anti-cheat engine and refuse to work with it until it's removed rather than considering that WINE obviously needs to work with such things for... obvious reasons. Especially since the methods used by them are almost the same as DRM methods used by many more recent programs even.) I do have a suspicion though. I'm going to try uninstalling everything related to that old internal soundcard later. In fact, it uses a C-Media chipset as well like my USB DAC, so perhaps the C-Media drivers from it could be taking over and applying their mixer or something (though I had tried to disable anything that would affect sound such as equalizing and etc.) Seems worth a shot anyway.

 

[castleofargh]

 what you call generic asio is I assume, asio4all? which is IMO a bad band aid. in my experience, when you have dedicated asio drivers for a device use them, else, run away from asio with win7.
you didn't say if you tried wasapi, which is pretty well implemented when working with foobar and usually ends up actually doing the bypassing job it should.
 
as software alone(OS) seems to change things, maybe try to remove a few stuff that are even remotely related to audio, and reinstall a few drivers? I know that when I last installed viper4windows (so that I could use some convolver for all windows sound), it messed up other audio stuff(even with all effects OFF) and only after I uninstalled it, did I get my sound back as it should. so you never know(that's the kind of annoying thing with windowzzz IMO).
 
android products are often on 48khz by default instead of 44khz. no reason for that to be audible, but as we're looking for what could go wrong, I thought I'd throw that in too.
 
 
and just as a personal opinion, your answered to cel4145 is really not cool. you're the one asking for help, he comes to help and has nothing to gain from it. none of his questions were really illegitimate(remember that if call centers ask the obvious annoying questions first, that's because it's just more efficient that way). as far as I can tell I would have asked about the same and I end up giving pretty much the same wasapi advice. it didn't deserve the answer he got. it's internet, nobody cares and I'm sure he will survive, but still not super cool.

 

[Nazo]

 what you call generic asio is I assume, asio4all? which is IMO a bad band aid. in my experience, when you have dedicated asio drivers for a device use them, else, run away from asio with win7.
you didn't say if you tried wasapi, which is pretty well implemented when working with foobar and usually ends up actually doing the bypassing job it should.

I said it in the first post (you know, the one that was supposedly too thorough.) This is why I tried to provide as much relevant info as possible so that hopefully questions like those would be answered already.

As far as ASIO was concerned, I did at some time have ASIO4ALL, but ended up removing it since, as you say, it's a "bad band aid." It seems the drivers must have provided a real ASIO driver along the way though. (Though I'm not sure which set of drivers did it -- the internal soundcard was such a similar chipset that I'm not sure it couldn't have kind of killed two birds with one stone. I guess I'll find out as soon as I track down the official C-Media drivers.

as software alone(OS) seems to change things, maybe try to remove a few stuff that are even remotely related to audio, and reinstall a few drivers?

Yeah, I'm doing that now. In particular I'm wondering about that ASUS control panel. I had previously specifically set it to "HiFi" or whatever it was called that was supposed to disable ALL effects (including GX) though. Still, ASUS built a real mess of things. I was actually using the third party UNi Xonar drivers which are supposed to be cleaner, but I still had gone ahead and allowed the ASUS control panel to be installed hoping for maximum compatibility. I'm also going to install the latest official C-Media USB drivers instead of using the ones on Schiit's site just because there's no real reason the C-Media drivers on Schiit's site should be considered to be any different at all (it doesn't even brand it anywhere.) EDIT: Ugh, it looks like they only give drivers to vendors and expect them to support individual products themselves or somesuch. There ARE no official drivers from C-Media themselves for this it seems. I guess I'll stick with the drivers Schiit provides.

My biggest problem is pinning this down. While the difference is in many ways not subtle, it's still hard enough to specifically pin down that it will be hard to really tell.

I know that when I last installed viper4windows (so that I could use some convolver for all windows sound), it messed up other audio stuff(even with all effects OFF) and only after I uninstalled it, did I get my sound back as it should. so you never know(that's the kind of annoying thing with windowzzz IMO).

Yeah, that's what kind of worries me. The easiest answer would be a clean reinstall. I'd bet that would fix it. What A Pain though...

android products are often on 48khz by default instead of 44khz. no reason for that to be audible, but as we're looking for what could go wrong, I thought I'd throw that in too.

Android's system basically officially supports 44.1 and 48 with a tendency to default to 48. But things using 44.1 should still actually go to that from my understanding.

and just as a personal opinion, your answered to cel4145 is really not cool. you're the one asking for help, he comes to help and has nothing to gain from it. none of his questions were really illegitimate(remember that if call centers ask the obvious annoying questions first, that's because it's just more efficient that way). as far as I can tell I would have asked about the same and I end up giving pretty much the same wasapi advice. it didn't deserve the answer he got. it's internet, nobody cares and I'm sure he will survive, but still not super cool.

That's funny, I found his responses to be the ones that were "not cool." I asked for help with a problem and instead of looking into the specifics of the things that could potentially cause the problem I get fussed at because my writing style isn't concise enough, told I'm unable to diagnose the problem at all because of any volume differences of even a mere 1dB (basically meaning that no matter what I say it could never be diagnosed since it could always be claimed I'm off by at least 1dB and thus wrong -- this just opens the door to blame everything on confirmation bias and makes it impossible to proceed ever. Though, in retrospect, I don't think I needed to touch the volume dial. Both are digital and I was using the same song file with the same replaygain settings in players that supported replaygain, so chances are they were 0dB different, but I can't 100% prove that and thus there is no solution because everything ever is confirmation bias) and so on. He might have meant to help, but how is any of that helping with the actual problem itself? If I want to be corrected for everything I've ever done wrong, I'll go talk to my mother, not ask for help with a problem. Yes, I'm aware I write a lot. Yes, I'm aware the problem is tricky. Yes, I'm aware that I'm not audiophile enough to be able to say "in the subharmonics at frequency x there is a decreased response and the soundstage is 36% less wide" or something like that. I'm sorry, but these aren't things I can change right now. I'd rather receive no help than have another of those talks about how I'm doing everything wrong in my life thank you very much.


EDIT: It's a bit early to tell, but I do think that removing the internal soundcard's drivers did indeed fix it. Along the way I did also remove the USB audio drivers too and reinstalled them clean just to be sure. And the ASIO driver is definitely provided by the drivers themselves, not by ASIO4ALL. In fact, I had apparently forgotten to remove ASIO4ALL, so I was seeing both options (but was using the one that said "ASIO for USB" provided by the drivers, not the ASIO4ALL version.)