or Connect
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › USB to SPDIF converters shoot-out : EMU 0404 USB vs. Musiland Monitor 01 USD vs. Teralink-x vs. M2Tech hiFace
New Posts  All Forums:Forum Nav:

USB to SPDIF converters shoot-out : EMU 0404 USB vs. Musiland Monitor 01 USD vs. Teralink-x vs. M2Tech hiFace - Page 110  

post #1636 of 1712

I posted how to do the mods a long time ago n another forum & have repeated the post here with annotated pics: http://www.head-fi.org/forum/newestpost/498151 for all those wishing to DIY it. Nobody' is affiliated with me

 

Regal has done some of these mods too :)!

post #1637 of 1712
Quote:
Originally Posted by noinimod View Post

Eh? He made it very clear - "modified Hiface". Besides, it's not going to take much to find out only jkeny does the mods, at least for now. Not sure if you're affiliated with jkeny by any chance? lol


Wow, since when has jkeny become persona non grata around here.....to the extent that the mere mention of his name brings up shilling accusations?

 

USG

post #1638 of 1712

Long time lurker, I have a question about the HiFace even though this may come a little late.

According the PDF FAQ on the M2Tech website, this device can perform bit-perfect data transmission without resampling.

 

Quote:

29) Is hiFace a bit-perfect device?

It depends on the mode it is used: in Direct Sound mode it can’t be bit-perfect, as the kernel mixer is on the samples way from the player to the interface and produces

alterations which are not due to both player and hiFace, but that nevertheless change their value. But when Kernel Streaming mode is used, a bit-perfect transfer is ensured, because samples are directly delivered by the player to the interface, without any change. 

 

 

30) Does hiFace perform data resampling?

No, hiFace can’t do any other conversion operation than USB-S/PDIF translation. What comes out of it is always same as what comes from the player. Differencies between the original value of data and that of data coming out of hiFace is due to either the player or the kernel mixer (or core audio on Mac computers).

 

However, on page 100 I read this:

 

Quote:
Heres what hface said about why it only show 32bits and why theres a lag

 

yes, hiFace only works with 32 bits samples. When 24 or 16 bits files are played, the driver performs the bit stuffing with trailing zeroes to reach the 32bit depth.

the delay is inherent and it's due to the size of the buffer hiFace uses to operate. Maybe your A/V player has a setting to fix this problem, please check.

So does this device perform sampling modifications to the audio stream or not? If so, then wouldn't that violate the principle of bit-perfect playback by padding the data from the PC before it is fed to an outboard receiver such as a DAC? Since Red Book audio is 16-Bit and even HDTracks are 24-Bit at most this is troubling me. Thanks for your help.

post #1639 of 1712
Quote:
Originally Posted by grokit View Post




AJ here again, bumping my original questions to see if I can get any more specific answers. Soulrider, are you using the HiFace with your Matrix? I can't really tell from your response. The Matrix does have 75 ohm input impedance, does that means it need resistance (how much?) or just a 75ohm adapter?


Yes using the bnc hiface to matrix, i just bought a male to male bnc adapter on ebay - worked amazingly well, in other words with my speakers, very noticeable.

post #1640 of 1712

Pardon one final interruption, folks......

 

As much as I feud with a certain person, I can honestly say he has learned something from his encounters with me. So, I do not feel my time was a total waste.

 

Look, guys.........it is real simple: You can stick an attenuator on your SPDIF doo-hickey, and see if it works. If it does, fine. Might be the best $12 you have spent.

 

If you can't hear a difference, well, it might mean you can not hear the difference. I'm sure someone will be willing to take it off of your hands.

 

Most of you will spend more money, on dubious stuff, that has no basis in physics, and won't help you get better sound.

 

Now, one thing he asked, and felt he did not get a truthful answer, had to do with a direct connection. I did answer it, but in lieu of actually having one of these units, I can not give a 100% certain answer.

 

Here is the problem:

 

If you stick a short cable, and by short, I mean......oh, say.............9" long, in between your doo-hickey, and your DAC. This would be a very bad idea.

 

Why, you ask?

Simple.

 

There are going to be reflections. All transmission lines have reflections. Because of the way the clock is built into the SPDIF stream, its recovery is affected by reflections.

 

So, how does the cable length, enter into it? Well, explaining it, without visual aids, might be tricky. But, try to hang in there with me.

 

Rise time is the amount of time the waveform takes to go from 20% to 80%. If the reflection arrives during that 20% to 80% window, it is going to skew the waveform. So, what happens with a very short cable, is that reflection is going to land right dab smack somewhere in that part of the waveform, and the rise time will actually increase!  As you lengthen the cable, that reflection arrives later and later, in the waveform, until it starts arriving after the waveform reaches its maximum level.

 

So, as you lengthen the cable, the rise time decreases. Which is good. As you further increase it, the rise time starts to increase, as pulse dispersion slows down the HF components of the waveform. (Cable attenuation goes up, as frequency increases.)

 

Now, how long, or how short can the cable be?

Well, without knowing the rise time of the circuit, you can not know. But, let's say it is 5 nSec or so. In which case, it might need to be 4-5' or so.

 

But, all of this is dependent on the rise time, the  electrical length of the circuit up to the connector, how big the reflection from the far end is, etc., etc., and more etc.

 

(If you have really poor impedance control, have a so-so output stage, that is further crippled, to lower EMI, which is what almost of you have, even if you do not realize it, what works best is a long cable, that has low BW. This is what some of you call "Jocko's cable". I should know: I designed it in '92, for just those situations.)

 

But, if you can stick a 10 (or 20 dB!) pad in between the two doo-hickeys.............hey, it might work really good. Really.

 

Only costs $12 to find out. In lieu of having one in my hand, trying it is the best advice I can offer. If I had one, I could measure it and know for sure.

 

Now, coming full circle to the top of this post.........

 

For those of you with as depraved a sense of humor as I have, "our buddy" JK is having an exchange over on DIYAudio with someone who portrays himself as someone who has designed equipment, up to 2.4 GHz.

 

If he worked for me, and I read what he has posted, he would be fired on the spot. No questions asked. Folks, as much as JK gets under my skin, he is a lot smarter than this other guy. I do not see any way someone who claims to have his experience, could write such nonsense.

 

Only goes to prove, what I say, all the time...........

 

Don't automatically believe everything you read on the 'Net. A lot of it is just flat out wrong. Don't believe what I write!!!!!!!!!! I want you to do some work on your own, to find out if I am right. Or just BS'ing you, for my private amusement. (No, that is my buddy Phred, who does that. Not me.) Hate me if you like, just learn something from our exchange. (Actually, Phred thinks you should learn from his abuse, but he won't care if you don't. It's the German genes!)

 

Just don't ask someone on some other forum if I am right or not! 98% of them are worse off than you are. Trust me.....................

 

Jocko

post #1641 of 1712

@Jocko Homo

 

Thanks for all this precious info. The S/PDIF signal is a very fragile one and I think that we should get righ of S/PDIF cables and directly incorporate the hiface design into DACs or just use a simple T-Pad attenuator to link both the hiface and the DAC without any S/PDIF cables. However, there are no attenuators that have Male to Male plugs on the market as of yet. Probably someone could DIY one but I think that it would be very hard to do because it requires a lot of precision to have perfect symetry and the right measures to in order to insure a true 75ohms connection.


Edited by punk_guy182 - 6/20/10 at 8:24pm
post #1642 of 1712

So someone with an RCA input DAC and RCA hiFace is doomed? Even with an RCA-RCA male converter instead of cable. The heavenly 75Ω is lightyears away from these poor bastards (like me).

post #1643 of 1712
Thread Starter 
Quote:
Originally Posted by Pacha View Post

I must say that the Teralink-X with the FUN don't sound very good. But also I never got used to it after having tried the DAC19DSP/C2 with modded HiFace at your place since the last time, everything sounded totally anemic after having listened to your setup. I have no interest in buying the DAC19DSP/C2 right now as I'm leaving soon until mid august but as soon as I come back I will place an order on the couple, listening to the FUN is such a pain when you know how the DAC19DSP/C2 sound.

Hopefully we won't torture ourselves in the same way if we could listen to the Ref1 with Phoenix, as it is hard to find. Good thing for our wallets, at least for the near future


Sorry for spoiling the FUN for you

 

As for me, I have been very tempted to try some of those new usb converters ... but I keep telling myself that I don't need it. I guess that not being able to test for oneself all the gear we are interested in can be a blessing sometimes ... for our wallets

 

Quote:
Originally Posted by upstateguy View Post

@slim.a

 

Just to be clear, your talking about jkeny's HiFace and not a stock model.

 

USG


Yes, I was talking about jkeny's modified Hiface. By the way, it is clearly stated in my signature which shows what hiface I am using and with the link to the review (http://www.head-fi.org/forum/thread/483900/review-jkeny-s-modified-hiface). I don't have a stock hiface anymore.

 

Quote:
Originally Posted by Sebhelyesfarku View Post

So someone with an RCA input DAC and RCA hiFace is doomed? Even with an RCA-RCA male converter instead of cable. The heavenly 75Ω is lightyears away from these poor bastards (like me).


Knowing the theory is good as it helps avoid making bad purchase decisions. However, if you are satisfied with you current RCA to RCA cable, there is no reason not to enjoy it even if the theory says otherwise.

On my previous DAC, I had 2 inputs: one RCA input and one 75 ohm canare BNC input. I personally heard more differences between different digital cables than between the different inputs. The 75 ohms BNC input was better but the RCA was almost as good. I am also pretty sure that if people were not told which one was the "true 75 ohms" they might have preferred the RCA input.

So in my opinion, you are not doomed :)

post #1644 of 1712

Jocko,

I appreciate your post above for the information that you have given in it. I know we have some "history" in this cable issue but it arose from a need on my part to learn about SPDIF cable length & I'm always willing to learn more.

 

At the risk of starting another war can I ask some questions about your statements on short cables:

 

- you say that " So, what happens with a very short cable, is that reflection is going to land right dab smack somewhere in that part of the waveform, and the rise time will actually increase!"

 

- Can I ask you this:

 

There is a length below which the cable will not show "transmission line" behaviour.
The TL effects come from the delayed return of energy.
If the delay in the cable is so short that the returned energy is returned much faster than the signal rises, them we may reasonable conclude that no transmission line effects occur.
 
As many common cmos outputs have a risetime of around 10 - 20nS we can actually get away okay with a cable of around 30cm length as having no significant TL effects?
 
Can I also ask in relation to long cables:
- as you lengthen the cable there will be certain lengths where the reflections fall outside the rise-time window BUT lengthen it further & this reflection could well now hit the next rise-time window - so there are only discrete lengths that will work which have to be calculated?
 
Also this length will change depending on the datarate so it will be different for 44.1, 88,2, 96,176, 192KHz speeds, no? Doe sthis mean we have to have different length cables for different speeds?  Swapping cables when listening to music at different rates would not be fun!
 
I'm not trying to pick a fight or get a dig into you, I respect your knowledge & experience in this field & simply want to tease out these questions that I have. Please don't go off on a rant!

Edited by jkeny - 6/21/10 at 4:32am
post #1645 of 1712

 

Quote:
Originally Posted by Jocko Homo View Post

Pardon one final interruption, folks......

 

Jocko


Jocko no offense to you, but do you think it is a nice thing to use the Duce as your avatar? Is it hidden fascism panegyric or some kind of jibe through this picture? I don't think it'd be nice if people around here began to use Hitler, Klaus Barbie or other similar person as one's avatar anyway. No taboo here, just some kind of obvious ethical behaviour within an international community like Head-Fi.


Edited by Pacha - 6/21/10 at 3:16am
post #1646 of 1712
Thread Starter 
Quote:
Originally Posted by Trioa View Post

Long time lurker, I have a question about the HiFace even though this may come a little late.

According the PDF FAQ on the M2Tech website, this device can perform bit-perfect data transmission without resampling.

 

 

However, on page 100 I read this:

 

So does this device perform sampling modifications to the audio stream or not? If so, then wouldn't that violate the principle of bit-perfect playback by padding the data from the PC before it is fed to an outboard receiver such as a DAC? Since Red Book audio is 16-Bit and even HDTracks are 24-Bit at most this is troubling me. Thanks for your help.


When going through the Hiface, your data whether it is 16bits or 24bits goes through it unharmed. The "bit stuffing" does not harm your data. It is not like it is truncating or resampling anything. Some DAC chips can actually behave better when fed 24 bits data instead of 16 bits even if the additional bits are only stuffed zeros.

 

What the Hiface does to the data is totally different from something like the Teralink X2 that resamples (by the driver) 88.2 to 96. In that case, there is no bit perfectness and there is damage to the data. In the case of the Hiface, the additional zeros do not change anything to your actual data.

post #1647 of 1712

Thanks slim, I just wanted to be sure.  Was it the self charging boxed version or the standard version with the external battery pack? (sigs are not always up to date)

 

USG
 

Quote:
Originally Posted by slim.a View Post

Yes, I was talking about jkeny's modified Hiface. By the way, it is clearly stated in my signature which shows what hiface I am using and with the link to the review (http://www.head-fi.org/forum/thread/483900/review-jkeny-s-modified-hiface). I don't have a stock hiface anymore.
 

post #1648 of 1712
Quote:
Originally Posted by upstateguy View Post

Thanks slim, I just wanted to be sure.  Was it the self charging boxed version or the standard version with the external battery pack? (sigs are not always up to date)

 

USG
 

It doesn't really matter as far as the sound goes - they are both equivalent!
 

post #1649 of 1712

Jkeny - read this about cable length.  It includes some analysis:

 

http://www.positive-feedback.com/Issue14/spdif.htm

 

Steve N.

Empirical Audio

post #1650 of 1712

 

Quote:
Originally Posted by audioengr View Post

Jkeny - read this about cable length.  It includes some analysis:

 

http://www.positive-feedback.com/Issue14/spdif.htm

 

Steve N.

Empirical Audio



Steve, what do you think about using an RCA to RCA male adapter plugged directly into the DAC instead of using 1.3m good RCA cable like the Oyaide DR-510?

I quickly went through your link and will read it thouroughly.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
This thread is locked  
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › USB to SPDIF converters shoot-out : EMU 0404 USB vs. Musiland Monitor 01 USD vs. Teralink-x vs. M2Tech hiFace