|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
207.237.13.151
AllDoes anyone know the specifics of the new DAC-1 USB? Does it convert the USB to i2s?
Follow Ups:
I am looking to make a few changes to my PC playback. Its a bit all over the place at the moment.
I currently output from whatever Media player I am using, but mainly Wavelab in ASIO, to an M-Audio Firewire 410...which converts the signals to SPDIF and then feeds that to a Musical Fidelity X-Dac V3 and then onto an X-Cans V3, both on a V3 PSU.
This sounds pretty good but lots of cables and cases... so would the Benchmark fed by USB and using its onboard headphone amp be an improvement sonically.
I also like the idea of the balanced output, I could then chuck a feed to my main Hi Fi which I don't do on my current set up. Might finally be able to integrate all sources.
I could flog off the MF stuff and pretty much cover the cost of the new unit.... in an ideal world.
Anybody have any ideas on this, better, worse or no idea....
I thought there was one sure-fire test to test if an audio stream was resampled... the DTS test.You simply play back a dts track through a device that has a SPDIF output. If your receiver is set to "audio" it WILL auto-detect the DTS format. If the stream was resampled, you'll get a nice blast of pink noise.
Anyone care to confirm or deny this *seemingly sensible* test?
I have Holst - The Planets in DTS.
Guess what I am going to try tonight...
I will set my default Windows Audio Device to my soundcard and get SPDIF working using waveout or directsound. Yeah. THAT should get the old kmixer working.
I think you *can* get 44.1 OUT of the kmixer. But *internally* the kmixer does it's MIXING at 48khz. How could you *mix* different sample rates and bitdepths? I thought you needed to have identical sample rate and bitdepth at SOME point to mix digitally...
The article claims no "resampling" is done. I think kmixer might resample BACK to the input sample rate... but only after resampling to the default 48khz at SOME POINT in the stream.
Another question is: are we talking kimxer under AC97? Or the new Intel HD-Audio spec? Not the same thing!
I dunno. Those "answers" by the engineer seem just too good to be true based on a collective "experience" with kmixer.
I don't care if it just *works* - I am not a gamer who want to hear my friends sign in and out of chat over my game sounds! We need to be able to account for each step in the process - and know what each is doing to the stream along the way. If you're doing NO DSP crossover or room correction processing, you should be demanding bit-perfect playback.
Otherwise why use a PC at all?
< < you should be demanding bit-perfect playback. Otherwise why use a PC at all? > >If that's why you are using a PC, you are spending a lot of time worrying about nothing. The cheapest stand-alone CD player will deliver "bit perfect" playback.
The only reason programs like EAC need to re-read the data multiple times is because they are trying to read at 52x (or whatever). Then you will get read errors. But a normal CD player will never have read errors unless the disc is physically damaged.
For more info on this, go to the Stereophile website and look at some of the old articles written by Robert Harley in the early '90s.
Hey Presto - I tried the DTS test one time with a Sony Vaio laptop and an M-Audio Transit (hooked up to a Denon receiver). I played a DTS file with WMP 10 and lo & behold it worked. Just for fun I found out what happens when I drop the volume to less than 100% - that blast of pink noise you were talking about. So when you test, make sure all your pc volume controls are set to MAX.So you *can* get 16/44 out of kmixer. I think it is a false rumor that the kmixer always resamples to 16/48. But it probably depends on what device you are using.
BTW, the Sony laptop probably had the Intel HD-Audio spec, but I'm not sure what difference that makes since I was not using onboard audio.
The Transit is not using the system provided USB audio driver and the custom M-audio driver is known to bypass kmixer.Cheers
nt
I've encountered a similar situation. I was able to get HDCD playback with MP10 and also Itunes (6.0 ?) With each player, I lost HDCD lock when lowering the media players volume. This was with my old Terretec 2496. Can't make the same happen anymore with my EMU 0404.Greg
Apparently if your device driver advertises itself as having a "hardware mixer" and promises to be compliant to AC97, Windows will let it do the mixing instead of using kmixer.For example, my Audiotrack Prodigy 7.1 will play back DTS files fine, because it *has* a hardware mixer. I think this is also the case for some of the other soundcards based on the Via Envy chipset.
The reason why this discussion is interesting is because the Benchmark USB DAC claims to use generic XP USB drivers, which (I think) do not register as having a hardware mixer to the OS.
PS - a clever idea would be to claim to have a hardware mixer even if you don't, and implement software mixing in the device driver.
thread is going on right now about DAC-1 USB. A Benchmark engineer Elias is answering some interesting questions.
Bit-perfect this, bit-perfect that. It's jitter that we are interested in.This guy was describing isochronous or adaptive USB mode, if I am not mistaken. He didn't clearly say that it was asynchronous.
Talk to me later if and when you can say this.
always the same blahblah bit perfect , kmixer ...
listen more music htat 's the way to go with or without kmixer...
But if avoiding needless resampling is just a whimsical concept it still makes sense to me. Since I have gone with kernel streaming and ASIO methods I have never looked back.Then again, I have some pretty darned revealing (Focal) speakers that tend to make these efforts seem worthwhile...
But as always, each to his own.
Like Thomas, I am skeptical of the claims of kmixer being bit perfect - that is not my experience, using exactly the same sort of tests that Thomas describes in his posts.Also the Windows system sounds are 22kHz, not 44.1kHz as claimed. This can be verified very easily by inspecting the properties of the actual wave files on any Windows system.
So something is fishy here, but I'm sure Thomas will get to the bottom of it and we'll find out the answer.
PS - Having said the above, there are also claims on the net about specific motherboards with specific chipsets being bit perfect via SPDIF even through kmixer. Perhaps there is more to this than meets the eye.
I can speak to the motherboards since I have one in one of my test machines. Certain older realtek HD-audio drivers for XP hapen to work bit perfect. However, none of the current ones do neither are the Realtek drivers for Vista.The specific machine has an AMD motherboard from Shuttle with an ULI Southbridge that supports HD audio. This was an early HD audio solutuion for AMD CPUs. The codec is a Realtek ALC880 and the S/PDIF worked bit perfect for me when I ran XP on it.
However, I also managed to get the Realtek in my Acer TM 8200 to work bit perfect on the optical output with that old driver version under XP. Demonstrating that little oddity is probably the only reason I can still dual boot that machine.Cheers
> Certain older Realtek HD-audio drivers for XP happen
> to work bit perfect. However, none of the current ones do;
> neither [do] the Realtek drivers for Vista.Would you be willing to reveal where to lay hands on these
earlier Realtek drivers (and/or what versions they are)?I put together a PC with an Asus (P5WD2-E Premium)
motherboard a few months ago (Intel 975X chipset,
Realtek ALC882D HD audio codec), and was disappointed
to discover that the "Realtek HD Audio Manager"
(that came with driver package Release 1.51, for XP)
doesn't even allow you to select 44.1 as the
sample rate.P.S. -- I found your post at
http://www.head-fi.org/forums/showthread.php?p=2592777 .So it appears I need to find driver
5.10.0.5221, eh?P.P.S -- Found a Czech site with earlier Realtek
drivers:
http://www.bgslevi.cz/soubor.php?produkt=ALC8xx&typ=o
driver 5.10.00.5221 seems to be part of Release 1.31
(wdm_r131.exe). Claimed to support ALC882. I
guess I'll find out.
It's a 23MB driver package (Realtek WDM_R131.exe). Search shows a few hits on that.Cheers
Thomas
I couldn't remember the motherboard and/or chip.I'm curious as to how the bits get passed "unmolested". Doesn't kmixer multiply everything by 0.999? Or is that a myth?
Does the bit perfect work ALL the time or "most" of the time? Does it work on both HDCD and DTS or only DTS?
If it's only DTS I'm wondering whether it's due to the same mechanism that enables passing DVD DD/DTS directly to SPDIF unmolested.
It just works for all PCM streams 16/44.1. To be honest I have not investigated why.As a guess I assume the driver registers itself in a way that it performs all mixing itself so kmixer is not in the picture. You can hear multiple streams so some code or firmware is doing mixing.
Cheers
Thomas
Might be worthwhile investigating this, since the Benchmark USB DAC could potentially be doing the same thing?I wonder if it's registering itself as having a hardware accelerated AC97 mixer supporting hardware resampling to 48kHz (but then not actually doing the resampling)?
Hmm,probably worth checking but when the driver/hardware does its own mixing, kmixer is not in the loop.
Cheers
This post is made possible by the generous support of people like you and our sponsors: