|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
106.51.26.237
Hi,
I use a Esi Juli@ sound card with my PC running Foobar. I take the SPDIF (coaxial) output from the ESI Juli@ to an old PS Audio Ultralink 2 DAC.
The default latency setting on the ESI Juli@ is 256 samples. I have tried both lower(48 ) and higher samples(1024).
There are no buffering issues event at the lowest (48) sample rate. However I find that there is a slight difference in sound while using the different sample rates.
At the higher sample rates I find there is slightly more detail and improved sound stage but with a slight loss in body of the mids. At the lower sample rates the midrange presentation is slightly more forward,with a loss in "air" around the instruments.
Why does changing latency rates cause these slight differences in sound?
Regards
Rajiv
Follow Ups:
I have a Juli@, but I don't normally use it much these days.
Carcass93 pretty much sums up the situation. Most people who have experimented have found that smaller buffer sizes sound better, at least up to the point where buffer underruns are happening and the sound gets broken up.
There are some measurements that show the effect of buffer sizes on noise patterns outside of a DAC, effects of digital cables on noise patterns and jitter spectra, as seen in the analog output of a DAC, etc... So this is a real phenomena. The details of exactly what happens and why it happens are only partially understood and will depend on the particular hardware and software involved, starting with the computer and including at least everything up to and including the DAC, possibly even analog equipment downstream of the DAC.
Incidentally, I don't believe the issue is latency at all, rather it's buffer size. If the number of buffers are kept fixed then there will be an inverse relationship between buffer size and latency. However, if the number of buffers is increased one can increase latency by software changes in the driver and/or player and thus it would be possible to zero in on the specific factor involved, buffer size vs. latency. The key factor appears to be the wait - busy cycle that repeats at the average buffer load rate. The situation can be more complicated because there can be multiple stages of processing in the audio pipeline and these can have different cycle times.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
my buddies are coming over Saturday night for drinks and music. We'll experiment with buffer size and see what comes of it.
What software will you be using?
Why, JRM of course!
Just want to insure you have the best sound possible.
Might I suggest listening to a bit of MJQ and try up some up-sampling with JRMC. I would love your opinion on the sound of Milt Jackson's vibes... If you know what they are and are familiar with them.
playing familiar music when auditioning/testing is important. I've got MJQ on vinyl, but not on the hard drive.
Well, Try it an see how bad the vibes sound. They will ring and sound like a bell. They will not have the proper strike tone and overtones. You will not hear the mallet. Now use something like HQplayer on a good computer and hear how much better it sounds.
And then play it on your Linn, I have one also btw. Vibes are hard to get right. Have you heard them live? Listening to the MJQ in a small club was something. They really did not speak much to each other. I remember at the Blue Note between sets Milt was just standing by himself near the men's room. They were a special group...
Could you make sure that buffer size tests are performed BEFORE that?
Rock Hill Farms. Tests before, gotcha. Things get out of control quickly with this shit.
Generally, people report improvements in SQ at lower buffer sizes.
Why it affects sound quality - there's plenty of discussion, but nothing conclusive, as far as I know.
running a Juli@ in an old XP machine, I had to run a high buffer size to get rid of drop-outs. Just couldn't get good enough sound out of it. Maybe that's why.
Interesting
effect overall sound quality. Is this process repeatable? Even after shut down/restart? After restart do you not hear a difference straight away, but do after time? Do you experience drop out's or pops? Have you run a latency checker?
Otherwise it is a flat earth statement.
TL's comments above support the contention that it isn't the file content but replay system transfer function that matters.
what does it mean exactly? I have and like the Linn LP12. Does that make me flat earth?
At what point does latency matter or become audible? Is it measured or perceived? If latency is simply delay, what part of the whole file gets delayed? We are not dealing with separate tracks (drum track, guitar track). As far as I know Midrange is not a separate track.
I can understand excessive latency causing chaos in playback such as lag, stuttering, etc.
If you don't know, then I suggest you look it up. (hint: religious conviction in the of the middle ages)
I can give you some real history references if you need them. The early church never believed in anything other than a spherical earth. This was establish well by the Greeks and was carried though the middle ages. Hopefully, you haven't taught your children (or anybody else) this misconception.
''The Flat Earth model is an archaic belief that the Earth's shape is a plane or disk. Many ancient cultures have had conceptions of a flat Earth, including Greece until the classical period, the Bronze Age and Iron Age civilizations of the Near East until the Hellenistic period, India until the Gupta period (early centuries AD) and China until the 17th century.''
It's the brainwashed and preconceived model of digital audio replay that I have been on about! This wsas how Thatcherism was called 'conviction' politics.
Your wrote "(hint: religious conviction in the of the middle ages)"Because you wrote this, I answered the way I did. If you want to forget the middle ages and find a time when the earth was thought to be flat by a majority then that is a different subject. Just try to stay on track.
"there never was a period of 'flat earth darkness' among scholars (regardless of how the public at large may have conceptualized our planet both then and now). Greek knowledge of sphericity never faded, and all major medieval scholars accepted the Earth's roundness as an established fact of cosmology."Historians of science David Lindberg and Ronald Numbers point out that "there was scarcely a Christian scholar of the Middle Ages who did not acknowledge [Earth's] sphericity and even know its approximate circumference".
Edits: 01/18/15
I have zero interest in arguing about the exact historical period. I am just interested in highlighting your refusal to acknowledge the effect of buffers on SQ, and your 'conviction' approach to PC audio.
Carry on arguing but I will not respond further.
we should get back on track.
That is only with JRiver. Otherwise there are differences in sound. It is so good a program that it is able to sound great no matter what the system or settings. It is just that good!!
Edits: 01/16/15
are you ok?
he is having a laugh!
Just need everyone to know that if they use JRiver they will have perfect sound without issue. Bis are bits. OS does not matter, settings don't matter. Latency does not matter. As long as there are no clicks or dropouts everything is good! Hope this is clear.
... is why they implement different output options, if that's the case. Smells like lack of integrity - or like "rich gassssssss" Jim H. passes, and invites everyone to smell it.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: