smoothdancer,
my congratulations... I felt free to post your fantastic results with your DDDAC1794 over on the Killerdac-site... if you might want to have a look, please go here:
DDDAC 1794-NOS 24-192
especially: #295 and #296
No problem, this great dac deserve every word. Maybe 8 deck will be next upgrade, but not shure. It play so great with 4 for now, time will tell.
Thx
yes, you are right, and I know how you feel now... well, in April this year we had some different DDDAC1794 here with us, including the ones from James and Supersurfer and the personal 8 deck DAC from Doede himself...
well, I have to tell, the spectacular amelioration in sound from 4 to eight decks is the same going from 60 to 120 chips with the DDDAC1543.... it is that unbelievable better, there is definitely no way back...
I will go for more than 8 decks, that's for sure... you might want to read the experiences that Chanh has made with his 11 decks...
enjoy...
Last edited:
Hermann, I'm sure you will get more response if you post this on ACKO's thread. I have a S03 to build so would like to see a reply too.
It surprises me that 4 to 8 decks makes such an apparent difference. There must be other differences. Something wrong somewhere.
Also on another aspect why does Fikus Lamp.... say TDA1541 is better as a single dac, another current out DAC chip. We cannot have our cake and eat it. Ask Doede to investigate. He is very clear in what is good,too good and not so good.
We should only be expecting moderate gains with any well developed set up and 4 decks has bettered 1 deck to all accounts, better than 1 if it is all matched and clocked the same. If an individual is getting big gains it may mean they have merely moved up a ladder from a lower position, and that is good, because this type of improvement happens to us all, as we keep changing things and get a Eurika moment
So you think your are superior then the rest with both a setup and ears set? May I ask, what is your setup's configuration? e.g; interconnects, speakers' cables, preamp type and your most important Speakers and poweramps?
Have you wonder your position/experience in audio prior came to a conclusion that your experience deem realistic and others are not?
Happy New Year by the way!
Good NewYear to you.Why the big difference in SQ with 8 over 4 decks. I was just asking why. There must be an overlooked technical reason. I started with ELS57 speakers now DHT and solid state power amps own TQWT s. I can hear minute differences in SQ.
It is not going to be lost bits going from 4 to 8 decks if they are matched and fast clocked in synchrony. It could be one of many things, so where to start. As it is Doede's blog and his excellent work, I will not give detailed response other than say any comparison like this needs a careful review. My impression from Doede was that the first four boards make a big difference with his design, gets less with more boards. And the increasing current can now better feed into i/v transformer or cap output. This would help but with diminishing improvements as you add more boards.
And the transfomer will take a small amount of SQ but not necessarily
enough to detract from the overall SQ improvements. This will probably be subjective
Hi,
Lets not start a fight or do a show-off of audio gear
It is apparant that most of the discussed changes in this thread lead to better sound.
For me it is more interesting to discover the beste price/improvement ratio for all these modifications.
I would not go for a high dac stack like Chanh has, that has to do with two reasons: I like my dac to be modestly compact, and my amps do not need all the extra current.
The latter matters for most, as we all use different amps some will benefit by a larger stack and some won't ( and of course; using cables with low capacity helps here)
In order of sound improvement and costs I would choose:
-direct balanced out (without capacitor)
- tantalum IV resistors
- silmic and oscon capacitors on the dac boards
-shunt regulators on the dac boards analog side
- unregulated choke power supply with a large transformer
-I2S connected raspberry pi or bbb
-acko S03 reclocker board ( or to be released botic cape from TP)
- shunt regulators on digital supplies
Pin 20 modification to be put in the above order somewhere, but I have not made this so can't judge the effect.
The explanation from rhlauranna is also not complete; during a meet last year with him and James we found that the 1 dac board from james with the shunts sounded better than the standard 4 dac board.
A happy new year to all!
Kind regards,
Lets not start a fight or do a show-off of audio gear
It is apparant that most of the discussed changes in this thread lead to better sound.
For me it is more interesting to discover the beste price/improvement ratio for all these modifications.
I would not go for a high dac stack like Chanh has, that has to do with two reasons: I like my dac to be modestly compact, and my amps do not need all the extra current.
The latter matters for most, as we all use different amps some will benefit by a larger stack and some won't ( and of course; using cables with low capacity helps here)
In order of sound improvement and costs I would choose:
-direct balanced out (without capacitor)
- tantalum IV resistors
- silmic and oscon capacitors on the dac boards
-shunt regulators on the dac boards analog side
- unregulated choke power supply with a large transformer
-I2S connected raspberry pi or bbb
-acko S03 reclocker board ( or to be released botic cape from TP)
- shunt regulators on digital supplies
Pin 20 modification to be put in the above order somewhere, but I have not made this so can't judge the effect.
The explanation from rhlauranna is also not complete; during a meet last year with him and James we found that the 1 dac board from james with the shunts sounded better than the standard 4 dac board.
A happy new year to all!
Kind regards,
Exactly. Also, we haven't had a direct side by side comparison of a well modified single deck dddac using a jfet buffer vs a well modified multi deck stack yet. But soon we will...Hi,
Lets not start a fight or do a show-off of audio gear
Happy new year all
Efficient measurment setup?
Hi,
After finishing the first stage of my DDDAC (see below) and which sounds excellent, I have some thoughts to share.
Following recent discussions along with own comparisons it seems to me that the DDDAC project lacks some kind of "common denominator" which everyone can rely and relate to in order to have a base from which people can advance their builds. As this is impossible with audible comparisons alone, as they are first of all very subjective and second can't be compared as we can't hear the builds of each other, I wonder if it would not be the next step in the community to agree on a cost-effective and efficient measurement setup. This way, everyone could test his build against this setup and cross-check that the basic build performs the very same as e.g. the initial build of Doede.
I wonder if it would be possible to carefully choose measurement instruments, e.g. an efficient external USB soundcard along with some PC-based software which costs ~ $150 and is thus affordable for everyone. Paired with selected audio signals (e.g. FLAC of Sinus waves in various resolutions) and test patterns it should be possible to compare builds on a measurement base.
One reason for this are for instance my findings and "improvements" by reclocking the I2S signal of the RPi. I personally could not hear any audible differences between the DIYINKH isolator and the S03 reclocker, although I was personally sure that reclocking means lower jitter and thus better sound, moreover there were many who wrote that the audible improvements were significant. After doing some DSO measurements, I found out that the reclocker actually added ~ 8ns jitter instead of reducing it. So - what does this mean? Do others simply like a more jittered signal? Or was the setup before they implemented the reclocker somehow flawed and e.g. the better isolator on the board improves the sound? Or are there simply no audible differences and people "like what they want to hear"?
Or, take another example: Let's assume one deck in my 4-deck setup has some flaw due to some soldering error. It may well be that I like this better although the signal is in reality quite distorted, however, no one will be able to reproduce this, thus everything stays very subjective.
So, what is your opinion on this? Would a carefully selected measurement setup and test patterns form a common denominator? Do you think this could improve the project? If yes, which hardware and test patterns would you recommend?
Best Regards,
Hermann
Hi,
After finishing the first stage of my DDDAC (see below) and which sounds excellent, I have some thoughts to share.
Following recent discussions along with own comparisons it seems to me that the DDDAC project lacks some kind of "common denominator" which everyone can rely and relate to in order to have a base from which people can advance their builds. As this is impossible with audible comparisons alone, as they are first of all very subjective and second can't be compared as we can't hear the builds of each other, I wonder if it would not be the next step in the community to agree on a cost-effective and efficient measurement setup. This way, everyone could test his build against this setup and cross-check that the basic build performs the very same as e.g. the initial build of Doede.
I wonder if it would be possible to carefully choose measurement instruments, e.g. an efficient external USB soundcard along with some PC-based software which costs ~ $150 and is thus affordable for everyone. Paired with selected audio signals (e.g. FLAC of Sinus waves in various resolutions) and test patterns it should be possible to compare builds on a measurement base.
One reason for this are for instance my findings and "improvements" by reclocking the I2S signal of the RPi. I personally could not hear any audible differences between the DIYINKH isolator and the S03 reclocker, although I was personally sure that reclocking means lower jitter and thus better sound, moreover there were many who wrote that the audible improvements were significant. After doing some DSO measurements, I found out that the reclocker actually added ~ 8ns jitter instead of reducing it. So - what does this mean? Do others simply like a more jittered signal? Or was the setup before they implemented the reclocker somehow flawed and e.g. the better isolator on the board improves the sound? Or are there simply no audible differences and people "like what they want to hear"?
Or, take another example: Let's assume one deck in my 4-deck setup has some flaw due to some soldering error. It may well be that I like this better although the signal is in reality quite distorted, however, no one will be able to reproduce this, thus everything stays very subjective.
So, what is your opinion on this? Would a carefully selected measurement setup and test patterns form a common denominator? Do you think this could improve the project? If yes, which hardware and test patterns would you recommend?
Best Regards,
Hermann
happy new year first of all ! to all !
yes, you are right, supersurfer, that was indeed spectacular to find out that the 1 dac board from James with shunts sounded better than the standard 4 dac board... simply incredible...
but, one has to keep in mind that the multiplying of the chips with the former 16 bit DDDAC1543 up to 240 chips in parallel and now the multiplying of the decks with the DDDAC1794 starting from 1 deck, to 4 decks and then Doede's own 8 deck - and here is the point - without any DIY tweakings at all - brought a really immense effect in amelioration in sound that way that for us here there is no way back...
I severely wish having already had all the necessary listening experiences with the (best) ameliorations implemented within a multiplied amount of boards... Doede has already described on his site the positive effects in progressively reducing jitter...
and from detailed listening tests with 1, 2, 4 and eight deck DDDAC1794 during December 2013 till march 2014 with all kinds of different power supplies I cannot emphazise enough the gain in amelioration in sound...
it is clear, this all has to be checked out further, if only Doede had more time...
resistor
Hi Simon,
You could parallel two 271ohm resitors. 271 or 267 are standards in the E96 range.
Basicly it should be 134ohm. So 135,5 or 133,5 I asume is fine...
Hi Simon,
You could parallel two 271ohm resitors. 271 or 267 are standards in the E96 range.
Basicly it should be 134ohm. So 135,5 or 133,5 I asume is fine...
Last edited:
You response helps to clarify the increasing DACs issue and Hermanns comments are spot on.
Now has anybody tried the SRPP or any other valve buffer amp directly connnected from the I/V to the first buffer valve grid , with or without decoupling cap and compared the sound, with the i/v transformer Cemag or other transformer.
This should be compared if possible even if the power amp is solid state.
In term of technical, I certainly am not qualified and will not pretend to be but rather leaving it to the more competent audiophiles or perhaps Doede is to address that espect.
In term of judging by ears, if that is any credential, I can share with you my practical experience with one board, 4-boards, 8-boards, and now 11-boards over the course of 6-Months, 2-3 hours Listenning session per day. Otherwise, it would have been such a foolish for I with 11-boards all in, include having the Tent Lab Shunts. Fyi, I use Accuphase ADL interconnects @ AUD$500 a pair, Speakers Cable is Cable Reasearch Lab @AUD$3000 a pair, ......., and certainly not the least a delicate 20A audio grade powerline direct from main switchboard to my AUD$7000 power conditioner PS Audio Perfectwave Powerplant P10. The infor give to provide you the setup means and hence my observations with every increment increases in SQ with additional DAC boards. In term of cost vs performance, probably not wise but as audiophile I seek and appreciate any noticable improvement despite a small incremental improvement.