Audio-gd Digital Interface
Jul 23, 2010 at 1:55 AM Post #241 of 4,156
You are on the right track but remember in the digital domain as long as bitperfectness is maintained the content is the same.  As an example I could hook a bitperfect transport up to a digital input on my freinds computer,  he could then send it to china and back,  and its the same.  Its only when the digital is converted to to analog that the transports "sound signature" becomes appearant,  this is due to jitter from the transport.  Its a difficult concept to explain,  but if you google jitter and study you will figure it out.
 
Jul 23, 2010 at 5:45 AM Post #245 of 4,156
why dont the DAC design include such thingy as error correction. When all the bits are captured by a chip next to DAC (or DAC itself) with sufficient buffer then only it is processed. I m off track of course, because such chip would not be musical like PCM1704u-k
 
Jul 23, 2010 at 6:02 AM Post #246 of 4,156
The DAC by design just turns the presented data into analog signals. That's why everyone desires bit perfect output from their computer and USB> SPDIF converter. Jitter is the main concern of this process. Many of these devices are bit perfect. Jitter levels vary.
 
Jul 23, 2010 at 6:09 AM Post #247 of 4,156


Kind of off topic but the issue is that a clock in the DAC is not going to match the clock in the transport exactly,  eventually the buffer would over-run or run out,  so a PLL is sometimes used with a variable clock in the DAC to match the incoming clock.  No one has come up with a way to eliminate jitter's affect on a DAC,  it can be minimized with expensive schemes but IMHO its best to just start with a good transport.   You will also find that DAC designers that design great sound and top notch analog design (ex AudioGD) aren't typically good at the digital side and vise-versa (Arye.)  
Quote:
why dont the DAC design include such thingy as error correction. When all the bits are captured by a chip next to DAC (or DAC itself) with sufficient buffer then only it is processed. I m off track of course, because such chip would not be musical like PCM1704u-k



 
Jul 23, 2010 at 7:03 AM Post #249 of 4,156
 
Quote:
...One thing that bothers me though is that an increasing number of mannufacturers use the Tenor chips (Teradak, Audio-gd, Firestone...) and it has one major flaw (in my opinion) as it is not capable of passing 24/88 through. As far as I know (correct me if I am wrong), they can only do 44.1, 48, 96. Most people won't mind but for those who got some 24/88 files, it can be an issue.


I agree.  The Tenor chip seems to be a low cost solution.  I'm curious how it performs compared to the TI USB chips even for standard resolution 16/44 files.  I'm not a technical person so this may be a complete non-issue.  I've never heard of Galaxy Far East Corporation, though, and Texas Instruments is a reputable company.
 
Jul 23, 2010 at 8:44 AM Post #250 of 4,156


Quote:
 
 
the TI USB chips sounds pretty bad...PCM2902, PCM2702E. Surely, you can find better adaptive chips.
 


I agree. But how do those chips compare to the Tenor?  Also, the CEntrance engineered solutions used in Lavry, Benchmark, etc products utilize TI TAS1020B chips as I understand.  These might be better solutions than the Tenor based ones?  
 
Jul 23, 2010 at 9:43 AM Post #251 of 4,156


Quote:
 many people who like R2R DACs have owned many sigma-delta DACs prior to that. The only reason sigma-delta DACs are here was to cut costs.
 


Agreed!
 
R2R DACs are good.
 
But... out of all the sigma delta DACs - to me the AK4393 is the best sounding of the bunch - never heard the AK4396 (actually i have but not in a situation which allows me to comment) ......
 
Jul 23, 2010 at 10:32 AM Post #252 of 4,156
My first dac was a ak4396 ht omega claro and claro halo sound card, I still have it and it's sometimes fun to listen to it, like looking at a reel of film to watch a movie, very objective with little room for misinterpretation, but very 2d hehe.
 
Jul 23, 2010 at 3:54 PM Post #253 of 4,156
As of now there's a big discount on ap-810 power cables here: http://www.mcmelectronics.com/product/58-13630
 
I purchased 6 of them at 3.09 each and $9 shipping, so like $27.50 total. They are 14x3 awg, shielded and ferrite bead, and gold-plated power cables. After 15 hours of initial use they seem to be quite good when used on my ac 215 power conditioner and computer connected to it and outputting usb, better highs, light sweet mids, quick dynamics, and coherent imaging compared to my stock unshielded 14 awg power cables, on headphone amp and receiver it has less black background though, and not too full-bodied lows and mids.
 
Maybe it will improve with more burn-in, but if not, I still think it's a great power cable and if anyone here has yet to get power cables definitely a good opportunity to get enough cables for your audio system and computer etc.
 
On the same topic, how many of you would be interested if audio-gd made $20-30 14 awg power cables? Their $75/4.5ft or $60/3ft power cables are a bit pricey, especially when you have to buy like 6 for your audio system. I am thinking about asking him but maybe they already offer too much products, and heh.. maybe liquidation prices on audiophile cables may be more common than I think.
 
Jul 23, 2010 at 9:03 PM Post #255 of 4,156


Quote:
I agree. But how do those chips compare to the Tenor?  Also, the CEntrance engineered solutions used in Lavry, Benchmark, etc products utilize TI TAS1020B chips as I understand.  These might be better solutions than the Tenor based ones?  


Not to forget that some chip designs require licensing fees and custom drivers........to keep an interface affordable and do it's job well is not an easy prospect. In many ways designing a cost no object piece is much easier.
 
Excellent points Regal ! I have more to add (about transports) but this isn't the thread for it.
 
Peete.
 

Users who are viewing this thread

Back
Top