Schiit DACs (Bifrost and Gungnir down, one to go)? The information and anticipation thread.

[M-13]

I guess it was...

 
Quote:

 
Rather ninja like post btw.  

 

 

[M-13]

But honestly I'm confused about this whole mess. I get the part about digital signals being transmitted via analog waves and the potential for error, but I wonder how common a wave disruption is. Is it 1 bit per million or 1 bit per thousand? Wouldn't the USB transmitter/receiver play a much bigger part than the cable? Other than proper RF shielding, what else can USB cable do?
 
Also, as anyone who studied marketing knows if I create one type of shampoo it wouldn't do as well as creating six different kinds and pricing them in a ladder. The middle shampoo would be perceived as offering the highest amount of value/dollar and be the best seller.
 
Another random thought I had was that people download music from HDtracks. I'm wondering how many non-audiophile grade cables the data travels through, jumping from one server to another before finally coming home... If I store my music in an external USB hard drive do I need to get a high quality cable for that as well?

 

[grokit]

It's too bad that people using search engines to find information about the Bifrost won't find this thread, perhaps the OP should change or add Bifrost to the title of it.

 

[woody88]

hello guys and gals,
 
in a bit of a pickle here, and would like some assistance if anyone can offer any thoughts on this. I received my Lyr/Bifrost combo today. I got it hooked up. My bifrost is the non USB version. I have a Rotel CD player fed into the BF via coax, and Pure I-20 fed to BF via Toslink. Every time if I were to skip a track or forward a track, the bifrost makes a clicking sound. It is very noticeable. I kept thinking if I have a defective unit. So for the next step, I used a generic coax digital cable to use on the Rotel CD player. Same thing.
 
next, I figured why not use the pure I-20 to test it. So I used the generic cable fed to the I-20 to BF via coax, and no clicking noise when I skip tracks on the iPod.
 
weird, I thought, could it be I have a bad CD player? so I grabbed another CD player that I have laying around, but have not used in some time. It's an Arcam CD player. Used the generic cable fed via coax to BF, same clicking noise, although instead of multiple clicking noise, it's a single clicking noise. but the noise is still there. Switched back to my regular coax cable that I normally use, same thing. While I was switching from generic to aftermarket coax cable, the BF even died on me for like 3 mins. meaning there is no response of the front power light when I toggle the on/off switch in back. But as of now, it is back into working condition.
 
I don't think I have a defective CD player, as I have tried 2 of them. And don't think I have bad coax cable, I simply cannot explain this clicking noise. does anyone else that use coax from CD player to BF has this issue? any help would be greatly appreciated. Thanks in advance!

 

[Maxvla]

The clicking is normal. It's the sync lock that is clicking.

 

[ocswing]

Quote:

But honestly I'm confused about this whole mess. I get the part about digital signals being transmitted via analog waves and the potential for error, but I wonder how common a wave disruption is. Is it 1 bit per million or 1 bit per thousand? Wouldn't the USB transmitter/receiver play a much bigger part than the cable? Other than proper RF shielding, what else can USB cable do?
 
Also, as anyone who studied marketing knows if I create one type of shampoo it wouldn't do as well as creating six different kinds and pricing them in a ladder. The middle shampoo would be perceived as offering the highest amount of value/dollar and be the best seller.
 
Another random thought I had was that people download music from HDtracks. I'm wondering how many non-audiophile grade cables the data travels through, jumping from one server to another before finally coming home... If I store my music in an external USB hard drive do I need to get a high quality cable for that as well?

 
To answer your question about storage transfer I'll just point out that it's different than music playing. When transferring data over USB for storage (or downloading from the internet) it does a check to verify the integrity of data at the end of the transfer. So if a bad packet was sent then it's resent at the end so the file is whole. When using USB to a DAC for music playback it's streaming and does not have this error checking. Even if it did it wouldn't be able to send it in real time to make up for the bad packet and not lose time. Yes, the transmitter/receiver play a big part, but there is a part for the cable to play in music playback as well. Whether the differences are noticeable and how big a difference they make will vary among people, but digital playback is not as simple as some people make it out to be.
 
 

 

[M-13]

If you're talking about usb isochronous vs bulk transfer, then I'm aware of the differences. I'm still wondering about the bit disruption rate in a isochronous usb transfer and whether these differences can be attributed to poor cabling and if these errors are of a rate that can be heard by the human ear. But I guess this isn't the audio science forum. Just wanted to give people another data point since the arguments aren't black and white.

 

[woody88]

Thx for the reply. But does this happen to all DACs though? this is my first standalone dac, so I was wondering if this is something just happens to all DACs or just BF? and also excuse my lack of knowledge in all of this, but how come when I use coax from I-20 to BF, it would not make the clicking sound? I got to say the clicking sound is rather bothersome. No way around it? thx again.

 

[Maxvla]

It has to do with the stream starting and stopping. Each time it starts you hear a click and that locks the stream into a stable connection. The alternative is a solution that adapts on the fly but gives up some sound quality (most DACs). I don't know all the details, but this has been answered in this thread before by Jason Stoddard of Schiit.

 

[Defiant00]

Quote:

Thx for the reply. But does this happen to all DACs though? this is my first standalone dac, so I was wondering if this is something just happens to all DACs or just BF? and also excuse my lack of knowledge in all of this, but how come when I use coax from I-20 to BF, it would not make the clicking sound? I got to say the clicking sound is rather bothersome. No way around it? thx again.

What I recall Jason saying was that it's the muting relay for when it is not receiving a signal (certainly possible that Maxvla is correct, but that's what I seem to recall reading). I believe that some sources completely stop sending data between tracks for a split second and some keep sending something (silence) while switching, which is why for the former the Bifrost clicks twice (once when data stops and again when it restarts). Incidentally, using optical from my computer does the exact same thing.
 
As far as other DACs, well, it depends entirely on the implementation, but it is certainly normal for the Bifrost.

 

[woody88]

Quote:

What I recall Jason saying was that it's the muting relay for when it is not receiving a signal (certainly possible that Maxvla is correct, but that's what I seem to recall reading). I believe that some sources completely stop sending data between tracks for a split second and some keep sending something (silence) while switching, which is why for the former the Bifrost clicks twice (once when data stops and again when it restarts). Incidentally, using optical from my computer does the exact same thing.
 
As far as other DACs, well, it depends entirely on the implementation, but it is certainly normal for the Bifrost.

thanks for the explanation; as long as I know my unit is the not only one that does this, I can move past it and start breaking in the combo. thanks again defiant and maxvla.
 

 

[mab1376]

Quote:

What I recall Jason saying was that it's the muting relay for when it is not receiving a signal (certainly possible that Maxvla is correct, but that's what I seem to recall reading). I believe that some sources completely stop sending data between tracks for a split second and some keep sending something (silence) while switching, which is why for the former the Bifrost clicks twice (once when data stops and again when it restarts). Incidentally, using optical from my computer does the exact same thing.
 
As far as other DACs, well, it depends entirely on the implementation, but it is certainly normal for the Bifrost.

 
From: Jason Stoddard [mailto:jason@schiit.com]
Sent: Tuesday, March 06, 2012 11:56 AM
To: Mab1376
Subject: Re: Order 1366 ETA
 
Yep!
 
Jason Stoddard
Co-Founder
 
http://www.schiit.com
(323) 230-0079
 
jason@schiit.com
 
 
 
 
On Mar 6, 2012, at 7:30 AM, Mab1376 wrote:
 
when using coax I hear a relay click when I switch songs. is that normal?
 
I'm using ASIO4ALL through Foobar.

 

[IEMCrazy]

Quote:

But honestly I'm confused about this whole mess. I get the part about digital signals being transmitted via analog waves and the potential for error, but I wonder how common a wave disruption is. Is it 1 bit per million or 1 bit per thousand? Wouldn't the USB transmitter/receiver play a much bigger part than the cable? Other than proper RF shielding, what else can USB cable do?
 
Also, as anyone who studied marketing knows if I create one type of shampoo it wouldn't do as well as creating six different kinds and pricing them in a ladder. The middle shampoo would be perceived as offering the highest amount of value/dollar and be the best seller.
 
Another random thought I had was that people download music from HDtracks. I'm wondering how many non-audiophile grade cables the data travels through, jumping from one server to another before finally coming home... If I store my music in an external USB hard drive do I need to get a high quality cable for that as well?

 
When talking USB audio, there's one very, very key thing to remember about the cable that affects USB and nothing else.  Not only does it have signal and ground drain, but USB also carries power as well.   A 5v power line running inside the cable insulation in paralell to data signal cables for 100% of the length of the cable is messy indeed.   Now will all implementations be pushing power through the power pins at all times, maybe not.  For that reason, I'd say for USB the cable may actually be more important than it is for Coax.  Though the nature of optical makes the cabling very important too.  But of course if the transmitter is junk it doesn't really matter what the cable's doing!
 
Note that I'm not advocating running about buying multi-hundred dollar exotic audiophile cable (sorry, sponsors!).  But there's still differences between the cheap stuff and the good stuff up through maybe the $40-60 price points for average lengths. 
 
Even on BJC's page for the optical toslink cable they don't shamelessly tout superior audio via the cable and pretty flatly acknowledge that plastic optical is 2nd class....but at least we're selling first-rate 2nd class cable!
 
Someone else already answered, but, yes, for actual data transfer the cable is irrelevant, USB, ethernet, etc. as far as data integrity.  The protocol verifies the receiving data and endlessly re-requests it until it gets it all and it all checks out.   And out-of-order doesn't matter, it resequences it all before it gets stored and finalized.  The only causality of bad cabling there is time.  Getting all the data right once takes only the time to send all the packets.  Getting them wrong and re-sending means it takes the time to send the packets, send a re-request command, receive the packets again, etc.  So over worst-case cabling (say, sitting on top of a high voltage transformer array?) you may end up sending the entire contents of the file quite a few times before it actually all arrives in tact.  So obviously for enterprise data use, great cabling is important to save on time/bandwidth/throughput/costs. But data integrity is never an issue.
 
In audio it's all real-time.  An audio packet is sent, and audio packet is played back in a stream, like water through a hose.  It's loaded in the tube in sequence and pops out the speakers in that same sequence.  There's no time to re-request a missing packet, nor the ability to pause and wait for a missing one to arrive.  If it wasn't there when it's needed, it's just skipped and you move on. This is also how streaming video works over UDP.  A missed packet is just missed.  You'll probably see screen artifacts or color shifts as a result, but video is far, far, more complicated an affair with a lot of interpolation going on anyway.
 
The audibility of such damage is a factor of the person listening, the amount of data loss, and probably which data was lost.   The digital stream plays back, in the case of redbook CD, 44,100 16-bit samples per second.   If a bit of the 16 bit word is modified, that means the frequency peak is now different than specified.  If the packet is missing or out of order, you simply miss one of those samples entirely...silence for 1/44100th of a second.  Can you audibly detect a single oscillation variation consisting of 1/44100th of a second?  Doubtful.   Can you detect specifically an audible peak variation even 10 times a second and identify which sample was affected?  Of course not.  However if it happens with enough frequency, (consider were talking about 44,100 samples per second.  Multiply that by the number of seconds on a CD and that's a lot of packets that can end up altered or missing) the human brain/ear is likely to pick that up, at least in comparison to something without missing data, as less clear (which it literally is if it's missing data), or "muddy" (details between frequencies may be missing), or with higher noise (distortion) etc.  Meaning no brain is fast enough to identify singular missing or altered oscillations, but it can pick up a collection of non-linear audio as something artificial or muffled/damaged.  It's, like lossy compression, all a human perception game in digital.   Can a human detect 1 1/44100th of a second error?  10, 100, 1000, 10000, 10000000? At what point does the brain realize something is wrong?  Digital is all about assuming/hoping the brain doesn't notice more and more missing pieces of reality until its in the extreme.   Kind of like politics

 
But the notion of "improve your bass" or "hear clearer treble" from a digital cable is a preposterous amalgamation of analog signaling theory mixed with a little reality of human perception noticing high and low more easily than mid damage even when all are equally damaged (the tolerances of OUR internal signal bio-cable specs...).   Analog loses the high and low frequency sooner than the middle because it's at the edges of the cables current carrying conductivity spec.  Digital doesn't lose one frequency over another.  It loses entire fractional moments of audio instead!
 
My favorite was when Dolby Digital was all the rage in the late 90's for AVRs and the rise of DVD.  "Hear the audio like never before in 100% all digital quality".   Which was silly because DD AC3 is VERY lossy compression with very heavy and obvious artifacting (quantization.)  Oh you were hearing it like never before all right.... The only thing that made it seem half decent was that the noise floor was so attrocious on VHS that it burried so much audio that the artifacting still actually was more detailed than the hiss

 
 
Quote:

hello guys and gals,
 
in a bit of a pickle here, and would like some assistance if anyone can offer any thoughts on this. I received my Lyr/Bifrost combo today. I got it hooked up. My bifrost is the non USB version. I have a Rotel CD player fed into the BF via coax, and Pure I-20 fed to BF via Toslink. Every time if I were to skip a track or forward a track, the bifrost makes a clicking sound. It is very noticeable. I kept thinking if I have a defective unit. So for the next step, I used a generic coax digital cable to use on the Rotel CD player. Same thing.
 
next, I figured why not use the pure I-20 to test it. So I used the generic cable fed to the I-20 to BF via coax, and no clicking noise when I skip tracks on the iPod.
 
weird, I thought, could it be I have a bad CD player? so I grabbed another CD player that I have laying around, but have not used in some time. It's an Arcam CD player. Used the generic cable fed via coax to BF, same clicking noise, although instead of multiple clicking noise, it's a single clicking noise. but the noise is still there. Switched back to my regular coax cable that I normally use, same thing. While I was switching from generic to aftermarket coax cable, the BF even died on me for like 3 mins. meaning there is no response of the front power light when I toggle the on/off switch in back. But as of now, it is back into working condition.
 
I don't think I have a defective CD player, as I have tried 2 of them. And don't think I have bad coax cable, I simply cannot explain this clicking noise. does anyone else that use coax from CD player to BF has this issue? any help would be greatly appreciated. Thanks in advance!

Others already answered, but the gist of it (as a fellow (former) I-20 user) is it depends on the source and how it handles signals.  The Bifrost will click whenever the source changes (is added/removed) or when the sample rate changes.  You can simulate the effect with your I-20 by undocking and redocking your iDevice to it.  The Bifrost will click because it detected source disconnect/sample rate change.
 
If you have audio files of multiple sample rates, it will click between playing them (riped CDs versus high-res audio, etc.)
 
I think the general rule is if your source player/transport is equipped for gapless playback, it doesn't terminate the audio signal, no clicking.  If it doesn't handle gapless there's a good chance it starts & stops the signal and you get clicking.  It's normal, and it's ok.   I think old school (90's era) players commonly did the start & stop.  Some software players do it as well.  It's normal!  (Hint: many AV receivers will click like that as well when the source is lost.  I experience that with my HDMI switch on my Onkyo when I switch sources on the switch.
 

 

[TheGrumpyOldMan]

Can confirm this as well. When using my laptop via optical, it will click quite a lot because 1) when choosing optical as the output, everything will be routed through it, incl. system sounds 2) the laptop turns off optical when not in use to conserve power.
 
So that can mean a lot of clickityclick when just working on the laptop. The solution is obviously to listen to music

After all that's what it's about, and the continuous soundstream means no clicking when additional system sounds are played since they are overlayed on the ongoing stream.
 
Some docks keep up a continuous connection, and music also tends to be in a continuous stream (the red light at the other end stays on), so the Bifrost relay won't be triggered.

 

[Maxvla]

What I've done to limit clicking is turning off operating system sounds like popups for questions, etc.