adhoc said:
What are you trying to tell us?
I discussed it at JRMC. They are telling me that they don't resample the stream unless you configure it!
If you run 44,1 16bit as source signal it just passes "untouched",
when it comes to resampling/upsampling.
With foobar afaik you need to get the resampling module in
to do resampling, right?
There is still the USB-Audio ASIO driver left. The guys over there
told me they wouldn't manipulate the source data either.
If they do some kind of resampling or rather reclocking or buffering when it comes to (latency) jitter reduction or
elimination , I don't have a clue.
rossco_50 said:
Ross. Very interesting information.
It seems that more Linux lovers respond to this thread until now.
What I miss is some kind of comparision statement Linux vs XP.
Before I start digging more into Unix, because that'll will absorbe
quite some time, I'd like to post the question:
Can I expect better audio results by using Linux? Why would that
be?
Better process/interrupt handling and real realtime operation as well as minimalistic configurations are some arguments I am aware of. What else?
Still - I don't have a feeling if that's gonna improve the audio quality.
soundcheck,
thank you for your reply.
first off, i'm using foobar2k and asio4all - but all this time i've never been able to verify or hear from an external source that i'm doing things 'correctly' and not (unknowingly) resampling or altering the digital sound stream.
right now, i'm streaming USB to a HagUSB usb-to-spdif convertor and then connecting that to to a DAC. the HagUSB uses a PCM2704 to do the conversion, and accepts 44.1/48kHz - right now i'm still in a little fluster as i still have no idea regarding whether my audio is somehow being resampled!
i wasn't trying to 'tell anyone anything', but i was definitely a little unclear - apologies for that!
thank you for your reply.
first off, i'm using foobar2k and asio4all - but all this time i've never been able to verify or hear from an external source that i'm doing things 'correctly' and not (unknowingly) resampling or altering the digital sound stream.
right now, i'm streaming USB to a HagUSB usb-to-spdif convertor and then connecting that to to a DAC. the HagUSB uses a PCM2704 to do the conversion, and accepts 44.1/48kHz - right now i'm still in a little fluster as i still have no idea regarding whether my audio is somehow being resampled!
i wasn't trying to 'tell anyone anything', but i was definitely a little unclear - apologies for that!
thomaseliot said:
Look at these threads. The audio is almost an afterthought. It is more like a dog and pony show for computer weenies. All this wrestling with drivers and refrains of " I had to uninstall Umbongo 4 and reinstall Umbongo 3.8 with the 12 Monkeys patch otherwise it wouldn't run with Django 7". Who else but computer people would think USB is a good way of moving multiple channels or the inside of a computer is the ideal home for a ADC. Compared to the pro world some have used to proclaim the superiority of the PC, it is about as user unfriendly as it could be.
To make that comparison I feel I would need a decent soundcard that works just as well on both platforms. My audigy is great for its dsp through the Kxproject drivers on windows, but Ive never considered it for use in my listening system and it doesnt sound as good as my cd player. RME would be ideal as they attempt to work just as well on both, but I am not willing to spend this much. Maybe the TI chips work with ALSA, but I have yet to come across anyone doing this.
I dont actually think there is any reason to believe you get better sound quality from linux, it is just the ability to do whatever you want if you have the skills, and a often a helpful community with similar goals. There isnt as much choice as in windows, so you would be through checking out all versions of umbungo quite quickly.
Perhaps rfbrw is right, if a little firey - but soundcheck did start the thread based on listening impressions, and it is me who is getting a bit caught up in the software side of things without offering much in terms of listening impressions. It is a useful thread.
Your experience with J river ramdisk is certainly of interest to me, and perhaps the most useful starting point is to identify problems in your current setup. Perhaps the areas to look at are power supply and grounding rather than software.
But if you want to conduct listening tests, download the demudi live cd, if your usb dac is supported it will be recognised upon boot. No cost and only the time to download and burn the image.
Personally Im not particualrly sold on usb - but have yet to try it. I have never heard anyone say the dddac sounds bad, but there are critics of the design on the forum. Perhaps looking at different DACs in comparision with the DDDac could be a next step. spdif vs usb as well.
Cheers,
Ross
I dont actually think there is any reason to believe you get better sound quality from linux, it is just the ability to do whatever you want if you have the skills, and a often a helpful community with similar goals. There isnt as much choice as in windows, so you would be through checking out all versions of umbungo quite quickly.
Perhaps rfbrw is right, if a little firey - but soundcheck did start the thread based on listening impressions, and it is me who is getting a bit caught up in the software side of things without offering much in terms of listening impressions. It is a useful thread.
Your experience with J river ramdisk is certainly of interest to me, and perhaps the most useful starting point is to identify problems in your current setup. Perhaps the areas to look at are power supply and grounding rather than software.
But if you want to conduct listening tests, download the demudi live cd, if your usb dac is supported it will be recognised upon boot. No cost and only the time to download and burn the image.
Personally Im not particualrly sold on usb - but have yet to try it. I have never heard anyone say the dddac sounds bad, but there are critics of the design on the forum. Perhaps looking at different DACs in comparision with the DDDac could be a next step. spdif vs usb as well.
Cheers,
Ross
adhoc said:
Yes. The above mentioned driver replaces the usb-audio.sys driver of Microsoft.
If you use ASIO4ALL you'll still have the poor MS driver in the stream!
The demo is FOC. Strongly recommended!
Another point is the USB cable.
Get yourself a Monster Ultimate performace USB cable. Its far better shielded than other cables.
Beside that some high performance Belkins would be a good choice. Both cables cost close to nothing. Just give it a try.
Please investigate some of the available lossless audio compression codecs. I like flac because it is free to use and open-source. It is also supported in various hardware devices and programs (via plug-ins). WAV files compressed with flac take up only about 60% of the space of the original WAV (and of course they decompress back to the original WAV -- identical bit-for-bit).So, unless you own stock in Western Digital or Seagate, check out http://flac.sourceforge.net/faq.html.
rfbrw said:
While I am somewhat "computer friendly" and comfortable with PCs, I'm not enough of a "weenie" to be running Linux. More hassle and cross platform problems than it is worth, for me anyhow, althought I've tried it couple of times_grin_Leaving the issue of Linux aside (hopefully) -
Having recently pulled much of my system together, I started thinking about upgrading my somewhat neglected front end. After looking around, IMHO, it appears reasonably obvious that now or "soon" it wll be time to go hard drive based playback.
Hard drive based playback is to a large extent still in it's infancy yet offers tremendous advantages.
Several reasons -
1. much higher quality sound for the dollars invested
2. flexibiity of storage as a user once ripped, and ability to manage your collection easily and play your collection in playlists.
Easy transfer of bulk lossless compressions is also not to be ignored as a major issue.
Will it require learning more? sure.
Will it be an initial hassle to many that make the transition to it? Probably.
Will it be worth it to many? you bet!
Will everybody do it? Nope.
Should everybody do it? Nope.
Once they make a successful transition to it will almost everybody that did so be glad they did? Probably.
Several years ago, some friends of mine heard a VRS system at VSAC. Couldn't believe how good it sounded. "vinyl on steriods" was one of the comments I remember. These guys loved the sound and they are hard core analog.
A post at AA where "Tuckers" comments that he dumped his $6500 DAC for a $675 card because it sounded better
http://www.audioasylum.com/forums/pcaudio/messages/15745.html
Will I go through a certain amount of BS and hassle to get that kinda sound for that kinda money?
Darn right!
munchkin said:
So, unless you own stock in Western Digital or Seagate, check out http://flac.sourceforge.net/faq.html.
I (we) did analyse it and discussed it with some other audio freaks.
Conclusion:
Yes - flac and other lossless companions are lossless, when it comes to the bit perfection issue.
Still they have to be unpacked in realtime by a codec, which is not manageable. You' ll hear audible differences.
We concluded every time you're touching, delaying the base data you are catching other audible differences, which most probably has to do, with the extra processing time and routines you gotta run. Even if everbody claims "its all bit perfect", it doesn't mean the timing is perfect!
More routines/processes causing more interrupts and buffer issues. Take each of the task priorities into account and you can be sure that you catch some latency jitter. Read explanation below:
http://www.rme-audio.com/english/techinfo/lola_latec.htm
Still -- if your sound card has a great reclocking and buffering
you might have the chance that these effects are less critical.
I am running the PCM2707 as USB receiver . It'll mainly receive the data as they are processed by the processor. That's why its
so important for my setup to have a good usb audio driver, as the last instance to get the stream properly out.
Of course there are more issue to consider on the PC:
Here some XP tuning Tips:
http://www.musicxp.net/tuning_tips.php
Especially the first tip is a must . It reduces XP latencies heavily!
Ken L said:
I fully agree, though I'd like to extend your statement.
My experience:
HD beats CD in every aspect. RAM beats HD in every aspect.
If you play full tracks - No streaming!! out of RAM you' ll get
the best sound. Fatiguing effects (most probably jitter based) well known from cheap cd drives are also audible if you play
from HD. If you play from RAM, they are gone!!
Low level details and seperation easily to verify on complex classical orchestral music are just getting perfect.
If that has something to do with physical aspects in the PC or Harddisk and/or software I do not have a clue.
soundcheck said:
Please do not confuse jitter and latency.
You've already got YEARS of latency/delay in the recording as it was recorded a long time before you play it.
This 'latency jitter' (apart from being mostly a madeup marketing term) has nothing to do with single track playback. don't get suckered in by jargon.
rossco_50 said:
Doede started with SPDIF on his DDDAC, he was running it with precision Tent clock. It gave good results.
Afaik, he has thrown SPDIF out and is using USB only, because on his rig the USB sound quality is superior to SPDIF!
At that time he hadn't even introduced my recommended tweaks!
Again, neither USB is perfect nor SPDIF, I think that's common sense.
Still I have a good feeling just to have a USB to I2S conversion
ongoing. I think you can't get a more minimalist solution, delivering such a great sound quality.
Perhaps there is an alternative solution.
Take an internal soundcard. Get I2S off it and connect it right to the DAC. (some people have done it)
This way you'd mainly have PCI to I2S, without SPDIF or USB in between.
It has to be a quite good card though, with precision clock and at least without upsampling to 48khz, low noise supply asf!.
I guess such a solution means much more hassle.
And if it's better in the end - who knows?
Soundcheck,
Have you confirmed that latency has an effect upon sound quality during playback of a file?
Latency only ever seems to be mentioned where audio must be synced with video, or where an external midi source is used (this is what the rme page seems to be about as well). I have always just accepted that latency only means a slight delay (acceptable) in playback rather than having any detrimental effect on quality.
The RME page is interesting, but I dont think latency has been neglected in the context they are talking about it in; its effect on musician timing. Is there any confirmation that there is something non-linear about the delays caused by latency and that can cause jitter in the playback of a prerecorded file? As far as I know the latency was only an issue for the artist at the time of recording (if they were even using a computer).
Yes I have heard many times that usb is an improvement - so it is worth a try. I actually have some ti samples so will do this. There is a very interesting thread at diyhifi on usb asio jitter with measurements and it is considerably worse than spdif - However I know this does not mean it necessarily sounds worse. Would be interesting to see how your optimised setup measures, you may have side stepped the driver problems that were considered to be causing the distortion.
Have you confirmed that latency has an effect upon sound quality during playback of a file?
Latency only ever seems to be mentioned where audio must be synced with video, or where an external midi source is used (this is what the rme page seems to be about as well). I have always just accepted that latency only means a slight delay (acceptable) in playback rather than having any detrimental effect on quality.
The RME page is interesting, but I dont think latency has been neglected in the context they are talking about it in; its effect on musician timing. Is there any confirmation that there is something non-linear about the delays caused by latency and that can cause jitter in the playback of a prerecorded file? As far as I know the latency was only an issue for the artist at the time of recording (if they were even using a computer).
Yes I have heard many times that usb is an improvement - so it is worth a try. I actually have some ti samples so will do this. There is a very interesting thread at diyhifi on usb asio jitter with measurements and it is considerably worse than spdif - However I know this does not mean it necessarily sounds worse. Would be interesting to see how your optimised setup measures, you may have side stepped the driver problems that were considered to be causing the distortion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.