I do not wish to enter into the argument over whether or not the computer can influence the sound quality, but I offer this data point. I rip my CD's to WAV using EAC or sometimes Windows Media Player. I have determined that both create the same data files when using a new unscratched CD.
Why then does Foobar sound different than Windows Media Player on the same WAV files through the same USB DAC?
On a different note, ever see the little "Auto Rip" icon on most CD's sold by Amazon.
This means that you can go to Amazon's MP3 store and buy an album for $10 to $15, or buy individual songs for $1 to $1.29 each.
OR
You can go to Amazon's CD store and buy the CD for $10 to $15. Sometimes the CD will actually be cheaper than the same music on Digital Download.....When you but an Auto Rip CD, you ALSO GET THE MP3's for FREE! Why don't people do this? I but the CD and the MP3 files magically appear in my phone, my Google tablet, and my iPad. I could get the MP3's on my PC's too, but I will rip the CD myself into WAV, and then put it in a box on the shelf with about 600 other CD's.
Want to play CD's in the car, just pick a dozen or so songs from your PC's library and ask WMP to burn you a CD. Leave it in the car until it gets scratched up and dies, or suffers an overdose of UV and dies, then toss it and make a new one.
Why then does Foobar sound different than Windows Media Player on the same WAV files through the same USB DAC?
On a different note, ever see the little "Auto Rip" icon on most CD's sold by Amazon.
This means that you can go to Amazon's MP3 store and buy an album for $10 to $15, or buy individual songs for $1 to $1.29 each.
OR
You can go to Amazon's CD store and buy the CD for $10 to $15. Sometimes the CD will actually be cheaper than the same music on Digital Download.....When you but an Auto Rip CD, you ALSO GET THE MP3's for FREE! Why don't people do this? I but the CD and the MP3 files magically appear in my phone, my Google tablet, and my iPad. I could get the MP3's on my PC's too, but I will rip the CD myself into WAV, and then put it in a box on the shelf with about 600 other CD's.
Want to play CD's in the car, just pick a dozen or so songs from your PC's library and ask WMP to burn you a CD. Leave it in the car until it gets scratched up and dies, or suffers an overdose of UV and dies, then toss it and make a new one.
If different, is it just "different" - or would you say one player is "better" than the other?
Foobar seems more dynamic and realistic on some music, and indistinguishable from WMP on other music.
The "DAC" is a MOTU Micro Book II, which is actually a pro audio piece intended for use in a personal recording studio. I have used it in my studio on a Core I7 PC, with a tube amp and Yamaha NS-10M Studio Monitor speakers. I have also used it on this laptop, a budget W8.1 machine with a core I7, a cheap class D amp and some DIY speakers, for field recording.
Both machines can also use the DAW (Digital Audio Workstation) software (Cakewalk Sonar) for playback.
About a dozen of so musicians have played with all 3 and the consensus is that it's WMP that's different.
The "DAC" is a MOTU Micro Book II, which is actually a pro audio piece intended for use in a personal recording studio. I have used it in my studio on a Core I7 PC, with a tube amp and Yamaha NS-10M Studio Monitor speakers. I have also used it on this laptop, a budget W8.1 machine with a core I7, a cheap class D amp and some DIY speakers, for field recording.
Both machines can also use the DAW (Digital Audio Workstation) software (Cakewalk Sonar) for playback.
About a dozen of so musicians have played with all 3 and the consensus is that it's WMP that's different.
Windows media player is a corporate digital rights management "service" , not a
real audio centric, user oriented program. 🙄
Foobar winamp , BSplayer ... all use a standard output decoder (below).
Output is standard PCM CD 44.1 khz stereo. this can also be written to a
uncompressed file in <10 seconds (in the same format) using the exact same
output decoder.
foobar will output the same uncompressed file as the other 2. The internal or
external dac will take any of these decoders output and shake the house
with the same definition. NO difference , they all use the same standards.
Some work more efficiently with the CPU , but that's about it.
PS - windows media player runs the output through another driver (DRM) ...
uses much more processor , all to protect IP. I just run direct to hardware (DAC).
real audio centric, user oriented program. 🙄
Foobar winamp , BSplayer ... all use a standard output decoder (below).
Output is standard PCM CD 44.1 khz stereo. this can also be written to a
uncompressed file in <10 seconds (in the same format) using the exact same
output decoder.
foobar will output the same uncompressed file as the other 2. The internal or
external dac will take any of these decoders output and shake the house
with the same definition. NO difference , they all use the same standards.
Some work more efficiently with the CPU , but that's about it.
PS - windows media player runs the output through another driver (DRM) ...
uses much more processor , all to protect IP. I just run direct to hardware (DAC).
Attachments
On most versions of windows, the default behaviour of applications (especially native windows applications such as WMP) is to output audio to the windows sound system. This sound system is a layer which sits between the application and the driver/hardware. It's purpose is to allow multiple applications to output audio simultaneously, and also provide features such as volume control, mixer, effects, etc. In order for this to function, it is a big help for all of the sounds from various applications to be running at the same sample rate. The default for windows is 48khz. So any sound file that does not have a native sample rate of 48khz will be converted, resampled. There are many ways to perform this operation. Some are much better than others. The better the resampling algorythm, the more CPU intensive it is. The default is to use a crappy but fast method, so it works on any computer.
More on this later, as the wife needs me right now...
Last edited:
No need for more, you've just reminded us all of us (who kind of knew this but somehow forgot) why Windows sucks so badly with sound.
Even "to allow multiple applications to output audio simultaneously" means the several sources have to be reduced in volume so their sum doesn't clip the output. This may not be as damaging as sample rate conversion, but it does "mess with the bits."
Even "to allow multiple applications to output audio simultaneously" means the several sources have to be reduced in volume so their sum doesn't clip the output. This may not be as damaging as sample rate conversion, but it does "mess with the bits."
Computers must be the worst way to store and playback audio if you are looking for quality.
When you start unravelling the layers of hardware and software data processing though the system the simple CD player starts looking a lot better.
When you start unravelling the layers of hardware and software data processing though the system the simple CD player starts looking a lot better.
In just raw performance , I'm way ahead of ANY CD player.
A cd player with it's missing data (aging), and no CRC is more like vinyl than digital.
My oldest 17 year old audio file is the same file it was 17 years ago.
The final check of the data at the DAC (amp) is compared to the file it is converted
or decompressed from , any "layers" of hardware/software are quite transparent.
From a convenience standpoint , playing a CD is a"ritual" I would rather avoid.
(like cleaning a record).
OS
For me and I think for most users the path is anything but transparent.
There just doesn't seem to be the clear and direct path between storage and output that is present in a CD player.
No arguments on the convenience or other advantages of the PC.
There just doesn't seem to be the clear and direct path between storage and output that is present in a CD player.
No arguments on the convenience or other advantages of the PC.
This is why most serious musicians use Apple hardware for their DAW systems. I however will not pay the price premium to use Apple, since all of my systems are DIY with reasonably priced components.
The trick to getting decent quality sound from Windows is to use a DAW and a sound system that bypasses the Windows layers, and also avoid any that add "DSP features" in the drivers unless they can be truly disabled. Often this is not in the manual, and you must wade through pages of reviews from users who don't understand why they get pops and clicks to find one where someone explains how to remove these "features".
This was relatively easy with XP and PCI (not PCIe) bus, but you are limited by computer noise. I had an old Emu (Creative) system that used a digital card in the PCI bus that got the bits out of the PC to an external 24 bit DAC box that had all the hardware including a phono stage safely outside the PC box. Creative killed all support for the PCI bus shortly after I bought the $500 box, so they are off my list of suppliers forever.
I am in the process of building a new home studio and need to find a new multi channel interface for a budget price.......
Anyone out there have good experiences with Firewire on a PC? If so, what components are you using?
But picking what to play can take longer with a NAS setup. If you come in and say ' I want X' then the NAS wins (assuming your control device is ready to go). If you come in and say 'What shall I listen to...' it's easier to scan a wall of CDs than navigate a disk. As a music collection grows the problem gets bigger.
maybe one day I will own a car new enough to appear as a mount point on my network so I can copy music across to it.
Yes (used Echo Audiofire), but Firewire is a dead issue now. The key to making Windows work properly is ASIO.
I have not had problems with noise using a PCI card, but again, PCI is a fairly dead issue now, what with the good ones (M-Audio 192, Juli@, et al) now being discontinued.
This is a specific example of where I was going next with the discussion. First I should note to benb that my point is not to say Windows sucks. There are many other better reasons to complain about Windows, but I will not get into that now. Also, most of what I say about Windows audio is from memory, as I have not used Windows as a dedicated audio platform in about 15 years.
The default behaviour of the sound system - the resampling to a common sample rate - is a necessary evil and is actually ideal for most typical computer uses. In fact, this is the default behaviour of the various Linux sound systems (ALSA, PulseAudio, etc.) as well. So to say this is a reason why Windows sound sucks is a bit erroneous. I am not sure about Mac OSX, but I would assume that the same is true for it; in order to mix and simultaneously play back audio from multiple applications, a common sample rate is needed.
This brings us to SY's comment about ASIO. ASIO is one of the ways to bypass the windows audio abstraction layer and gain direct access to the audio hardware. I don't know the details over every version of Windows, but I recall that somewhere around Windows Vista an option was added somewhere in the settings to allow direct hardware access, but I think the application needs to support that. Again, it has been a long time since I have messed with Windows audio. Suffice it to say that there are several ways to obtain "bit perfect" playback from Windows (one of them being ASIO - perhaps the easiest and best?)
On Linux, there are several ways as well and it depends on what sound system is being used and what application is being used. In some cases it can be as simple as installing "DEADBEEF" (as an example) and then going into the settings and selecting the hardware audio device from the drop-down menu. In other cases it requires one to uninstall "PulseAudio" (which is the default sound system on many distros) and then edit configuration files and finally configure the player to use the "ALSA audio sink" as the output device.
This can seem daunting to some, regardless of OS, but it really is as simple as a quick google search and then following some steps (in most cases). The one thing that sets Linux apart (for me) is how easy it is to verify that the system is doing what I want it to do. A few simple commands can tell me exactly what formats are supported by the hardware, and exactly what format is currently being dumped to the audio device (in my case an external USB/SPDIF converter and on to a DAC). In Windows, the methods I remember using did not really provide a lot of confidence.
So if some (more or less) simple steps are taken, one can completely bypass a PC's sound system (regardless of OS) and have the application access the hardware directly (via the driver). This does not necessarily "guarantee" bit-perfect playback, but it gets us a hell of a lot closer. The choice of application still matters, because applications can mess with the bits before they are dumped to the hardware.
More on this later, as my breakfast is ready. 🙂
Last edited:
But when it doesn't 'just work' Linux can be a pain. debugging is a nightmare unless you can find someone who had the same problem with the same version of ALSA as you. I found this swapping from a USB DAC to a USB to SPDIF converter. Finding the one parameter that was wrong or knowing what order to kill and restart for it to pick it up reminded my of windows 95 plug and pray. Took about 3 hours all told.
But yes once working I can check and see the bitrate that is going out and if it has done anything nefarious with my precious bits.
I am sure I can also do some EQ on the server, but it works so I am leaving it well alone.
That's pretty much SOP these days. You can use, for example, foobar or EAC or JRiver, and they'll all sound identical (if you don't peek).
SOP maybe, but I suspect that there are many people who overlook this and then come up with questions about why "digital audio" doesn't sound so good. I suspect that many of the believers in "upsampling" are people who are not bypassing the windows sound system. Time and time again I see people saying things such as "when I upsample to 96k it sounds better to me". Well, guess what? Upsampling to 96k in software (typically done with high quality, CPU-intensive algorithms) and then dumping it to the windows sound system will sound better than dumping 44.1k into the sound system. Why? Because a fast, crappy resampling algorithms used in OS sound systems can do a much cleaner job of downsampling 96k to 48k than it can upsample from 44.1k to 48k. Mathematically, going from 96 to 48 is far simpler. Either way, the DAC is getting 48k and the user often has no clue.
But lets get back to George's question; why does WMP sound different than Foobar on his system? Well, that really does depend on many factors: what version of Windows? Is there an option for direct hardware output, and is it selected? If not, is there any other effort being made to bypass the windows audio system? etc. For example, if Foobar is set up to use ASIO and WMP is set at default settings, then the mystery is over. Without knowing every detail, it is impossible to tell George why Foobar sounds better.
The playback application itself can have many features which mess with the bits; volume control, gapless playback, EQ, etc. Generally, these features will be disabled if the application is configured to directly output "perfect" bits to the hardware. For many casual listeners, on PCs that are used for many tasks and audio is not a priority, these are essential features. For the discerning listener, it is worth it to sacrifice all these features in favour of "bit perfect" hardware output.
When this is achieved, the player has a few simple jobs; read the data from the disk, decompress and/or format to a raw PCM bit stream, and fire it at the output hardware. All players that only do these three things will sound identical.
Furthermore, a computer that is configured in such a way is every bit as simple as a CD player with respect to the "chain".
Getting this right has much more impact on one's impression of "digital audio quality" than what DAC chip you are running or what type of computer you are using.
But lets get back to George's question; why does WMP sound different than Foobar on his system? Well, that really does depend on many factors: what version of Windows? Is there an option for direct hardware output, and is it selected? If not, is there any other effort being made to bypass the windows audio system? etc. For example, if Foobar is set up to use ASIO and WMP is set at default settings, then the mystery is over. Without knowing every detail, it is impossible to tell George why Foobar sounds better.
The playback application itself can have many features which mess with the bits; volume control, gapless playback, EQ, etc. Generally, these features will be disabled if the application is configured to directly output "perfect" bits to the hardware. For many casual listeners, on PCs that are used for many tasks and audio is not a priority, these are essential features. For the discerning listener, it is worth it to sacrifice all these features in favour of "bit perfect" hardware output.
When this is achieved, the player has a few simple jobs; read the data from the disk, decompress and/or format to a raw PCM bit stream, and fire it at the output hardware. All players that only do these three things will sound identical.
Furthermore, a computer that is configured in such a way is every bit as simple as a CD player with respect to the "chain".
Getting this right has much more impact on one's impression of "digital audio quality" than what DAC chip you are running or what type of computer you are using.
Last edited:
This depends entirely on the player software. Lately I switched from Daphile (which is a proprietary Linux-based OS dedicated to audio playback) to MPD (media player daemon). The player PC (Asus EEEBox) is headless, and I use an Android app to control the player (via wi-fi).
This is incredibly convenient. I can sort by artists, albums, genres, etc. instantly at the touch of a button and another touch gets the music playing. The app (MPDroid) is far quicker and more user friendly than a typical folder/file browser, etc.
Still, I love sifting through my vinyl, because I love the smell of the album covers. 🙂
PC is a "dirty" environment ....
For analog , that is.
A good trend is to get the analog OUT of the PC.
You can have your 300$ fancy sound card. But it will still be powered by a SMPS
and lamblasted by RF.
As long as you can get that PCM stream out of the PC with
USB or SPDIF , this opens the options for much better D/A performance.
OS
For analog , that is.
A good trend is to get the analog OUT of the PC.
You can have your 300$ fancy sound card. But it will still be powered by a SMPS
and lamblasted by RF.
As long as you can get that PCM stream out of the PC with
USB or SPDIF , this opens the options for much better D/A performance.
OS
Using Windows 8, on a recently acquired, old laptop the software package Media Monkey (MM) is winning for me - plenty of output options to choose from, and I still haven't finished finding if I can squeeze some more quality out of it. WMP is awful, and default foobar is also dreadful - I've started cleaning up the latter, it may be possible to get it as good as MM, more investigating is needed ...
Last edited:
- Status
- Not open for further replies.