LessLoss DAC - intiquing facts

ergo · 13th September 2004, 10:20 AM

I stumbled on an interesting web page
http://www.lessloss.com/main.html

It's a hand made DAC. The approach is quite interesting, but I found some real funny arguments in the text.

They claim
http://www.lessloss.com/jitter.html
".....Because LessLoss is determined to minimize jitter to the lowest of possible levels, we go the extra step to re-clock (also known as quantizing) the digital data to near-perfect timing just a few millimeters before it enters the audio domain......."

OK but then if one goes to page
http://www.lessloss.com/information.html

They write
".....In essence all you need to do is desolder and remove the quartz resonator from the CD player, and where one of its legs were, solder on the end of the middle conductor of a coaxial cable (any old coaxial cable will do). The shield of the cable should be soldered to any ground in the CD player you can find......"

This is a good example how a good presentation can be totally mocked by a few wrong sentences

********

How can one take seriously argument that few mm make a difference when on the next page they say that when feeding back the clock it does not need no care, any BS cable goes and so on.....

Ergo

ThorstenL · 13th September 2004, 10:38 AM

Konnichiwa,

Quote:

Originally posted by ergo
How can one take seriously argument that few mm make a difference when on the next page they say that when feeding back the clock it does not need no care, any BS cable goes and so on.....

Well, the wording may be extreme, HOWEVER, as long as the DAC's receiver can sync on the signal from the transport the cable quality does NOT matter (nor does any jitter introduced anywhere prior to re-clocking) as it is all removed by the reclocker. At least in theory. In practice using too grotty a cable will result in occasional "unlocking" of the connection.

Sayonara

ergo · 13th September 2004, 10:56 AM

I'm not so sure on that it is that straightforward. In theory yes, but in reality.

Kuei, as you know I'm working on improving the Behringer DCX2496 digital crossover. It has a CS8420 receiver ic in resampling/reclocking mode and at the moment I have a Tent XO2 clocking the data out of the CS8420. So theoretically the transport should not have any influence on the DAC. Still the latest test between my Sony CD with modded spdif out versus some older Technics player with stock spdif out gave a shocking result thet there was a very clear difference betwen these transports. And the difference was not subtle. Sony gave a much more relaxed and deeper soudstage....

I can not explain it yet, but I will try to somehow confirm it elsewhere with diffrent players and perhaps with measurements.

Ergo

PS. Sony player has a National LM6171 opamp as a unity gain buffer driving 75 ohm cable and using BNC connector. No connector is used on DCX end. Cable goes straight to transformer and terminated by 75ohm. For other transports RCA->BNC is used in transport end.

tschrama · 13th September 2004, 01:42 PM

hi ergo,

did you slave both the players?

ergo · 13th September 2004, 01:55 PM

No I did not slave them.

CS8420 is an asyncronous sample rate converter, thus the output should be clocked out of the chip independent of the clock on spdif line. Theoretically it should also make the dac almost immune to transport variances. In reality transport still seems to make a difference.

I know I can not compare the situation with slave configuration system 1:1, but it is somewhat similar. I would not be too surprised if the proper implementation of feeding back the clock also makes a difference there.

Ergo

rwagter · 13th September 2004, 02:24 PM

Absolutely handbuild

ergo · 13th September 2004, 02:37 PM

Yes for a kind of commercial offering it is not what people are used to I quess.

It is a fresh way of doing thing and I admire that they have done it like this and also published the photos. On the other hand if they would not publish the photos they might give a mild shock to some customers who happen to open the case I suppose

Anyway, in most part I really liked the website and the product, it's not something you see every day, but the few sentences I mentioned before managed to kill quite a bit of a good mood I had developed reading the text.

One should be precise and consistent to really do it all as good as possible if once taken that direction - at least that is how I feel.

Ergo

tschrama · 13th September 2004, 02:48 PM

but Ergo.. I totally agree with Kuei Yang Wang: they are right, the slave-clock doesn't have to be low jitter or anything, that's why you can get away with long slave-cables.. ground rerturn isn't critical anymore etc etc.. all the critical things happen right next to (few mm ?) the master-low-jitter clock..

In my view you haven't quoted any wrong sentences from their website..

Liudas · 13th September 2004, 02:54 PM

Hi, folks. Please notice that on the website under www.LessLoss.com/types.html the concept of asynchronous upsampling has been discussed, and this discussion doesn't seem to take into account the fact that the LessLoss DAC is way over and beyond that worse solution. That solution is what the DAC-1 by Benchmark uses. I don't know what it is that keeps people thinking that asynchronous sampling is the same as slaving. Totally different solutions. And about the soldering on the picture: That's just how it has to look if you are serious about keeping all crutial signal paths as short as possible. Many people might not realize that the entire picture which is posted here is only a few millimeters across. Those parts are among the smallest in existence! Looks can be deceiving.

Sorry about my tone of voice. I thought EVERYTHING was covered in the website, but it appears that still these things have to be explained again, or perhaps in a better way on the site.

ergo · 13th September 2004, 04:24 PM

I'm not saying the LessLoss guys are wrong. I just am surprised that they are willing to forget good design practise at that point.

".....The shield of the cable should be soldered to any ground in the CD player you can find......"

Lets assume a customer does it exactly like this and manages to find a ground line of motor drivers of the mechanism and stick the shield of the cable there. That ground point will fluctuate quite alot. Cable even carries some of the ground fluctuation back to DAC. It might also introduce lets say 1200ps of jitter in SPDIF going from transport to DAC. Even though the data is later reclocked I would still avoid such solution.

Might be my point of view in this matter is much stronger than you guys have. What I always shoot for is the best implementation even with smallest things. Sometimes this minor shortcoming may not be that minor on sound.

Ergo

PS. Or perhaps what they wanted to give is an impression that the slave configuration is easy to do and thus more people would be willing to try it.....

13th September 2004, 10:20 AM	#1
ergo diyAudio Member Join Date: Mar 2001 Location: Tallinn, Estonia	LessLoss DAC - intiquing facts I stumbled on an interesting web page http://www.lessloss.com/main.html It's a hand made DAC. The approach is quite interesting, but I found some real funny arguments in the text. They claim http://www.lessloss.com/jitter.html ".....Because LessLoss is determined to minimize jitter to the lowest of possible levels, we go the extra step to re-clock (also known as quantizing) the digital data to near-perfect timing just a few millimeters before it enters the audio domain......." OK but then if one goes to page http://www.lessloss.com/information.html They write ".....In essence all you need to do is desolder and remove the quartz resonator from the CD player, and where one of its legs were, solder on the end of the middle conductor of a coaxial cable (any old coaxial cable will do). The shield of the cable should be soldered to any ground in the CD player you can find......" This is a good example how a good presentation can be totally mocked by a few wrong sentences ******** How can one take seriously argument that few mm make a difference when on the next page they say that when feeding back the clock it does not need no care, any BS cable goes and so on..... Ergo

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
LessLoss Power Cable (European SCHUKO 2 meters)	fas948	Swap Meet	0	27th July 2009 01:03 PM
Facts about Fiberglass	Qub	Everything Else	5	5th February 2006 08:49 AM
Sub enclosure opinions/facts/help	drfrink24	Subwoofers	1	17th January 2005 01:45 AM
CRT Burn-In: Need the Technical Facts.	jkru	Everything Else	0	10th March 2004 06:16 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

13th September 2004, 10:56 AM	#3
ergo diyAudio Member Join Date: Mar 2001 Location: Tallinn, Estonia	I'm not so sure on that it is that straightforward. In theory yes, but in reality. Kuei, as you know I'm working on improving the Behringer DCX2496 digital crossover. It has a CS8420 receiver ic in resampling/reclocking mode and at the moment I have a Tent XO2 clocking the data out of the CS8420. So theoretically the transport should not have any influence on the DAC. Still the latest test between my Sony CD with modded spdif out versus some older Technics player with stock spdif out gave a shocking result thet there was a very clear difference betwen these transports. And the difference was not subtle. Sony gave a much more relaxed and deeper soudstage.... I can not explain it yet, but I will try to somehow confirm it elsewhere with diffrent players and perhaps with measurements. Ergo PS. Sony player has a National LM6171 opamp as a unity gain buffer driving 75 ohm cable and using BNC connector. No connector is used on DCX end. Cable goes straight to transformer and terminated by 75ohm. For other transports RCA->BNC is used in transport end.

13th September 2004, 01:42 PM	#4
tschrama diyAudio Member Join Date: Jul 2001 Location: The Netherlands	hi ergo, did you slave both the players?

13th September 2004, 01:55 PM	#5
ergo diyAudio Member Join Date: Mar 2001 Location: Tallinn, Estonia	No I did not slave them. CS8420 is an asyncronous sample rate converter, thus the output should be clocked out of the chip independent of the clock on spdif line. Theoretically it should also make the dac almost immune to transport variances. In reality transport still seems to make a difference. I know I can not compare the situation with slave configuration system 1:1, but it is somewhat similar. I would not be too surprised if the proper implementation of feeding back the clock also makes a difference there. Ergo

13th September 2004, 02:24 PM	#6
rwagter diyAudio Member Join Date: Jul 2002 Location: Netherlands	Absolutely handbuild

13th September 2004, 02:37 PM	#7
ergo diyAudio Member Join Date: Mar 2001 Location: Tallinn, Estonia	Yes for a kind of commercial offering it is not what people are used to I quess. It is a fresh way of doing thing and I admire that they have done it like this and also published the photos. On the other hand if they would not publish the photos they might give a mild shock to some customers who happen to open the case I suppose Anyway, in most part I really liked the website and the product, it's not something you see every day, but the few sentences I mentioned before managed to kill quite a bit of a good mood I had developed reading the text. One should be precise and consistent to really do it all as good as possible if once taken that direction - at least that is how I feel. Ergo

13th September 2004, 02:48 PM	#8
tschrama diyAudio Member Join Date: Jul 2001 Location: The Netherlands	but Ergo.. I totally agree with Kuei Yang Wang: they are right, the slave-clock doesn't have to be low jitter or anything, that's why you can get away with long slave-cables.. ground rerturn isn't critical anymore etc etc.. all the critical things happen right next to (few mm ?) the master-low-jitter clock.. In my view you haven't quoted any wrong sentences from their website..

13th September 2004, 02:54 PM	#9
Liudas diyAudio Member Join Date: Apr 2002 Location: Lithuania	Hi, folks. Please notice that on the website under www.LessLoss.com/types.html the concept of asynchronous upsampling has been discussed, and this discussion doesn't seem to take into account the fact that the LessLoss DAC is way over and beyond that worse solution. That solution is what the DAC-1 by Benchmark uses. I don't know what it is that keeps people thinking that asynchronous sampling is the same as slaving. Totally different solutions. And about the soldering on the picture: That's just how it has to look if you are serious about keeping all crutial signal paths as short as possible. Many people might not realize that the entire picture which is posted here is only a few millimeters across. Those parts are among the smallest in existence! Looks can be deceiving. Sorry about my tone of voice. I thought EVERYTHING was covered in the website, but it appears that still these things have to be explained again, or perhaps in a better way on the site.

13th September 2004, 04:24 PM	#10
ergo diyAudio Member Join Date: Mar 2001 Location: Tallinn, Estonia	I'm not saying the LessLoss guys are wrong. I just am surprised that they are willing to forget good design practise at that point. ".....The shield of the cable should be soldered to any ground in the CD player you can find......" Lets assume a customer does it exactly like this and manages to find a ground line of motor drivers of the mechanism and stick the shield of the cable there. That ground point will fluctuate quite alot. Cable even carries some of the ground fluctuation back to DAC. It might also introduce lets say 1200ps of jitter in SPDIF going from transport to DAC. Even though the data is later reclocked I would still avoid such solution. Might be my point of view in this matter is much stronger than you guys have. What I always shoot for is the best implementation even with smallest things. Sometimes this minor shortcoming may not be that minor on sound. Ergo PS. Or perhaps what they wanted to give is an impression that the slave configuration is easy to do and thus more people would be willing to try it.....