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

[mmerrill99]

Quote:

Originally Posted by slim.a /img/forum/go_quote.gif Jkeny, the only 100ps figure I have read about the Musiland was a theoretical figure guessed from the specs of the Xilinx FPGA used in the Musiland. That guess doesn't take into account the actual jitter at the spdif output which would have accounted for the layout of the PCB, the quality of the clocks and PLLs, the transformers ... You claimed the same thing about the 100ps figure on the Computer Audio Asylum without providing your source.

OK, I haven't found the reference to the 100ps yet but here's John Swenson's analysis of how the Musiland works with theoretical jitter figures Computer Audio Asylum: RE: Musiland 01 USD at 192k by audioengr


Quote:

Also, while I haven't seen the graph, the comments made by the person who measured the Musiland talk about 20db of additional noise. Then he says that since the Musiland was measured at 192k, the jitter at 48k might be lower. But I don't read anywhere where he talks about 2-5 db. Do you ?

Here's the graph attached - you can see this is where I derive the 2-5dB figure from! Edit: Can't seem to get the graph looking any bigger - it's fine on my computer?

Quote:

As for the SYS2722 used by stereophile, the lowest digital jitter they could get was 1.75ns (1750ps) with a Boulder 1021 disc player that costs $24,000. Do you really believe that the Musiland measures 17 times better than one of the best measuring CD players that exist ? Also, I have yet to find a jitter digital measuring device that can go below 0.5ns (500ps).

The graph shows the Musiland Vs the SYS2722 - you make up your own mind & then answer your question

Quote:

Could you please give a link to your sources ? It would be greatly appeciated. Also, could you please copy the graph and post it in this thread so that the readers could see by themselves ?

Search on Audioasylum for author fmak & musiland - he has done a lot of testing of the musiland

 

[slim.a]

Quote:

Originally Posted by jkeny /img/forum/go_quote.gif OK, I haven't found the reference to the 100ps yet but here's John Swenson's analysis of how the Musiland works with theoretical jitter figures Computer Audio Asylum: RE: Musiland 01 USD at 192k by audioengr

Jkeny, the 300 ps is a best case theoretical figure given at the pin out of the FPGA. It doesn't say much about the real jitter of the unit.

As for the link you gave me, Steve Nugent (audioengr / the designer of the Emirical off Ramp and a respected engineer) disagrees with your conclusions. He has already used FPGAs in his designs but only in non critical parts.

Below is a quote from an answer he gave you :

RE: Musiland 01 USD at 192k - audioengr 11:44:21 09/05/09 (3)

In Reply to: RE: Musiland 01 USD at 192k posted by jkeny on September 01, 2009 at 13:12:31

If you simply look at the output buffers on Xilinx PGA's and other programmable parts, you will find that they are generally a compromise. Really large buffers are required with dedicated power pins in order to get clean, fast, noise-free edges. This is what is needed for low jitter. These parts with hundreds of pins usually have maybe 4 power and 6-8 ground pins. The buffers cannot drive 50 or 75 ohms loads directly. There is no way that this can compete with simpler more discrete logic with more power pins and less sharing of the power currents. Highly integrated parts often have these shortcomings. Lots of functionality, but the signal integrity and drive may be issues in your circuit.

Jkeny, I really don't understand why you spread false information about the Musiland and discard everything that is being said to you. The thread is not named Musiland 01 USD at 192k by audioengr but Musiland 01 USD at 192k y the way. And when you read the thread, you understand that audioengr disagrees with you. What did you feel the need to rename the link ?
I am not saying the Musiland is bad. It is a very good unit for $75. But it is bad to spread false information and each time I tried to verify your sources, there were inconsistencies.

 

[mmerrill99]

Ok, I just uploaded the graph to here: 4shared - free file sharing and storage - share folder - Musiland

Jeez, the things I do for you guys

 

[slim.a]

Quote:

Originally Posted by jkeny /img/forum/go_quote.gif Here's the graph attached - you can see this is where I derive the 2-5dB figure from! Edit: Can't seem to get the graph looking any bigger - it's fine on my computer?

Your graph seems to contradict what the author of the measurement is saying in the thread here and here

 

[mmerrill99]

Listen, Slim.a you seem to have some beef with me & I don't know what it is? Have I upset you somewhere? I'm just giving the information fairly as I know it & not trying to give any mis-information whatsoever! Let people read those threads & make up their own minds. Audioengr has a business selling expensive USB units that the Musiland threatens - I do not - you figure who is likely to have a biased opinion

I'm not going to get into the same argument with you as I did with him but just read the whole thread for a fair report on it not your skewed one-sided reporting of just his comments.

Quote:

Originally Posted by slim.a /img/forum/go_quote.gif Jkeny, the 300 ps is a best case theoretical figure given at the pin out of the FPGA. It doesn't say much about the real jitter of the unit.

These are theoretical maxima, slim - stop trying to make the words fit your world view. EDIT: So in fact it is a worse case figureYes, I agree the real jitter of the unit isn't mentioned but look at the graphs & do a search on audioasylum under fmak as I suggested.

Quote:

As for the link you gave me, Steve Nugent (audioengr / the designer of the Emirical off Ramp and a respected engineer) disagrees with your conclusions. He has already used FPGAs in his designs but only in non critical parts. Below is a quote from an answer he gave you : RE: Musiland 01 USD at 192k - audioengr 11:44:21 09/05/09 (3) In Reply to: RE: Musiland 01 USD at 192k posted by jkeny on September 01, 2009 at 13:12:31 If you simply look at the output buffers on Xilinx PGA's and other programmable parts, you will find that they are generally a compromise. Really large buffers are required with dedicated power pins in order to get clean, fast, noise-free edges. This is what is needed for low jitter. These parts with hundreds of pins usually have maybe 4 power and 6-8 ground pins. The buffers cannot drive 50 or 75 ohms loads directly. There is no way that this can compete with simpler more discrete logic with more power pins and less sharing of the power currents. Highly integrated parts often have these shortcomings. Lots of functionality, but the signal integrity and drive may be issues in your circuit.

If you are going to quote him why don't you also quote my reply to him & his final answer - it seems to me that you are keen to mis-represent me?
Here's what I said in response:

Quote:

I don't know about other FPGAs but this particular FPGA has 13 supply pins of which 6 (Vcco) are dedicated to output buffer stages. John Swenson did a good theoretical analysis of the potential jitter. Until someone measures the jitter accurately we are just guessing, however.

To which he replied with just an opinion & nothing to back it up Quote:

I use Xilinx and I have not been impressed with the performance of their output buffers.

Listen, I'm not going to get into this s**t - you've got your agenda - I'm just representing my research on the Musiland - take it or leave it - I'm not interested in revisiting old ground!

I suggest you read John Swenson's views, an unbiased non-manufacturer with technical no-how & be careful of sources with vested interests!

Quote:

Jkeny, I really don't understand why you spread false information about the Musiland and discard everything that is being said to you.

I don't discard anything that is being said to me about the Musiland - I tanked John Swenson for his fair & frank summary of the unit in which he said 300ps is not good performance in his eyes!
Quote:

The thread is not named Musiland 01 USD at 192k by audioengr but Musiland 01 USD at 192k y the way. And when you read the thread, you understand that audioengr disagrees with you. What did you feel the need to rename the link ?

Jeesus, slim, I didn't rename the link I simply copied it & that is what it came over as OK, christ you have got some bee up your arse, haven't you?


Quote:

I am not saying the Musiland is bad. It is a very good unit for $75. But it is bad to spread false information and each time I tried to verify your sources, there were inconsistencies.

Don't accuse me of spreading false information - you haven't shown any inconsistencies with my sources - you have just quoted snippets out of context for your own ends - so who is spreading the false info?

I will say it again, I'm a fan of the Musiland - I like the sound - I have reported as much of the technical information as I feel is appropriate to this forum (I have read the Xilinx datasheet & background papers about the jitter performance of their FPGAs so I know a lot more about it than you have demonstrated you know about the working of the M2tech). If I have made any mistakes, I apologise but don't accuse me of spreading mis-information - I have absolutely no need to do this.

I & others who wanted to look at & discuss the technology behind the Musiland, got a lot of stick on Diyhifi & in the thread you mentioned from the owners of companies who purvey expensive USB solutions. So a lot of mis-information was spread by them about this dirt cheap USB asynchronous 24/192 Chinese "piece of crap" according to them. You tell me who is mis-representing the facts?

 

[mmerrill99]

Quote:

Originally Posted by slim.a /img/forum/go_quote.gif Your graph seems to contradict what the author of the measurement is saying in the thread here and here

What does the author say that I'm contradicting?
BTW, it's not my graph - it's simmconn's - the author's graph

EDit: I'll be gone from the computer for some hours but no doubt you will have some cheery message awaiting me when I get back

 

[mmerrill99]

OH, btw, look at this post (posted before your diatribe/rant) - do I take on-board what is being said about the Musiland? http://www.head-fi.org/forums/6112309-post546.html

 

[slim.a]

Quote:

Originally Posted by jkeny /img/forum/go_quote.gif 300ps jitter is probably consider good (but not excellent) compared to the 970ps being reported for the M2Tech. This 300ps is also the max value that it could be with all gates in the FPGA firing, etc.

Wow ! I can't help but being impressed by you once again! You are comparing a theoretical figure (based on specs alone) at the FPGA pins with the measurement of a device at the end of the spdif line and it doesn't bother you the least ? I am truly impressed

This is the last comment I will make on your posts and I won't waste any more time on you. I hope you get paid by Musiland at the end of the month because you seem to be very passionnate about them.

I suggest that this discussion about the Musiland be continued in the thread you already started here

I don't know if you read my review, but it is mainly about how those units sound not about discussing about XILINX specs sheets and theories of jitter. So please continue spreading the Musiland gospel on your own thread and every thread you can find on the internet. As for here, it is not needed anymore, I have finally understood that the Musiland is the best creation ever. Thanks so much!

 

[DarKu]

common guys, chill out, relax it's a disscusion forum

My PC has a LOT of coaxial noise (when i click or move my mouse i hear cracklings at highest volume on my amp) and i know it is because of my computer.
I'm trying to decide what to get to lower the noise, the Musiland, Hiface or the teralink
Now you confusing me even more...

 

[tosehee]

Quote:

Originally Posted by DarKu /img/forum/go_quote.gif common guys, chill out, relax it's a disscusion forum My PC has a LOT of coaxial noise (when i click or move my mouse i hear cracklings at highest volume on my amp) and i know it is because of my computer. I'm trying to decide what to get to lower the noise, the Musiland, Hiface or the teralink Now you confusing me even more...

Any of these will be better than your internal coaxial, and most likely fix your cracking noise.

I personally went with hiface, but if you have the skills and time to mod like jkenny, you might be better off with musilad 01 us and mod away.

 

[slim.a]

Quote:

Originally Posted by DarKu /img/forum/go_quote.gif common guys, chill out, relax it's a disscusion forum My PC has a LOT of coaxial noise (when i click or move my mouse i hear cracklings at highest volume on my amp) and i know it is because of my computer. I'm trying to decide what to get to lower the noise, the Musiland, Hiface or the teralink Now you confusing even more...

You are right, we should relax. I won't comment anymore on jkeny

I listened extensively to the coaxial/BNC output of the 3 units. You can read my review here.
All three units are very good. My prefrence go for the hiface and Teralink. However the Teralink -x is limited to 16/48. So if you need 24/96 or 24/192, your choice is limited between the hiface and the musiland. All three units should isolate your computer from your dac.

 

[DarKu]

slim.a i'm using about the same characteristic DAC as you use, a very good DAC, but it is very sensible to what i feed at it, when i'll go home i will test USB input and Optic out of my PC to my DAC, to see if the noise goes away ( but i have a very small hope that it will go away)
I've allready red your review and got an ideea about this devices, thank you for that.
Anyway i'm leaning towards the M2Tech because of the simplicity and the size, i hope they have black colour hiFace too ...

 

[slim.a]

Quote:

Originally Posted by DarKu /img/forum/go_quote.gif slim.a i'm using about the same characteristic DAC as you use, a very good DAC, but it is very sensible to what i feed at it, when i'll go home i will test USB input and Optic out of my PC to my DAC, to see if the noise goes away ( but i have a very small hope that it will go away) I've allready red your review and got an ideea about this devices, thank you for that. Anyway i'm leaning towards the M2Tech because of the simplicity and the size, i hope they have black colour hiFace too ...

if the optical connection doesn't cure the noise, maybe it is due to the buffer size in your media player or your latency settings. I remember than when I used to use software upsampling, I would have glitches when the latency settings were set too low or the buffer size in the media player (foobar in my case) was set too low.

 

[rosgr63]

slim.a great review thanks!
One question please sometimes my DAC momentarily looses the USB signal from my Mac. Have you an idea what it might be?

 

[ReopyFototbom]

As the internet is comely a biggest role of our always lives, it is chic easier to download movies to ipod nano newly released movies online. There are all sorts of benefits to watching movies online as opposed to present to the cinema or find ***** movies to buy shin-plasters that you haven't got at the restricted blockbuster store.
Firstly - you can note newly released movies online in the cheer of your own home and you don't tease to pass on a penny on a cinema ticket! Secondly - some video stores are in this day charging up to $10 a twilight to downloaded tv shows illegal. Why would you spend that lolly when you can watch movies online instead of free?
When you look for list of latest hindi movies in every way your home radiogram or satellite carton there are usually extra fees you bear to pay repayment for those movie, film or “specialty” channels. Watching feee divx full movies download absolutely eliminates those fees with a click of your mouse button.
No proceeding which party you download movies warez, clever the argument between a collectible cinema broadsheet and a reprint is important. After all, if you buy movies zune ready to attain an filler that intent enhance in value, it must be an source print. If, on the other hand, you solely demand to elegant up your off-campus apartment, there’s no perception spending liquid assets on a pricier collectible when an equally appealing reprint would tv shows canadian downloads.
What makes a layer print collectible? How is the quality of a phrasing determined? How can I sanctuary a collectible once I’ve obtained one? If you are new to the collecting position, you indubitably be undergoing uncountable questions that you latest movies at the cinema need answered prior to you make your premier purchase. Underneath, you’ll diva divx movies a two tips to labourers get your gleaning started on the right foot.