irgendjemand,
Look, I have never said that I obtained those Y's. You didn't read me carefully enough, but it's absolutey normal. Because I'm sure they're not selected, I don't feel like paying the extra asked for them. I have stated clearly, on more occasions, that I have : original P's in my rotel player, and had shown here the test result; original BBjapan K's in my tent dac, and an extra pair of korean standard P-s - which are measuring the best from the bunch..
On the other hand, for a better understanding of the situation, it would be interesting for Yourself, on the occasion of meeting your technical expert, to try and repeat these same tests by yourself, on your dac. [and communicate or show the result]. It doesn't matter if the chips are soldered in place: for these tests you don't have to remove them. Also, before or after the listening tests, try & measure those chips & dacs, also. And please tell us if You have found a clear correlation between measured grade levels and the stamps on the chips; and the correlation between sound & and measured grade levels.
Bernhard, I think you are right, probably the chips are changing. Also, what about power requirements and low level distortion?
Ciao, George
Look, I have never said that I obtained those Y's. You didn't read me carefully enough, but it's absolutey normal. Because I'm sure they're not selected, I don't feel like paying the extra asked for them. I have stated clearly, on more occasions, that I have : original P's in my rotel player, and had shown here the test result; original BBjapan K's in my tent dac, and an extra pair of korean standard P-s - which are measuring the best from the bunch..
On the other hand, for a better understanding of the situation, it would be interesting for Yourself, on the occasion of meeting your technical expert, to try and repeat these same tests by yourself, on your dac. [and communicate or show the result]. It doesn't matter if the chips are soldered in place: for these tests you don't have to remove them. Also, before or after the listening tests, try & measure those chips & dacs, also. And please tell us if You have found a clear correlation between measured grade levels and the stamps on the chips; and the correlation between sound & and measured grade levels.
Bernhard, I think you are right, probably the chips are changing. Also, what about power requirements and low level distortion?
Ciao, George
irgendjemand,
No, no hierarchy, please. I only and exclusively trust the information published here, on these pages, that is, Bernhard, Paul, PA0SU, Spencer, myself. Because I could see in my own tests the same things proven by the others here.
For this reason, I question strongly the approach of Finney. I find that elitist ad conservative. He claims that the stamped chips are better quality, in a qualitative way, because they were tested such in factory. THEN they should be still better, in a repeated test, and those test are not a mistery, they are very well defined and public, let's show us the repeated results!! There is an ever growing mole of evidence on the contrary, so prove me otherwise!
Then, exactly, I hear differences after each changes, one of the strongest was just changing the way of the screening on the DIR chip, the 8412... So there are many ways of influencing the sound, Why should I care about this elitist approach of a bloody stamp fabricated by somebody on top of a chip, if it does not have a proven connection to any kind of quality? If not by suggestion?
Ciao, George
No, no hierarchy, please. I only and exclusively trust the information published here, on these pages, that is, Bernhard, Paul, PA0SU, Spencer, myself. Because I could see in my own tests the same things proven by the others here.
For this reason, I question strongly the approach of Finney. I find that elitist ad conservative. He claims that the stamped chips are better quality, in a qualitative way, because they were tested such in factory. THEN they should be still better, in a repeated test, and those test are not a mistery, they are very well defined and public, let's show us the repeated results!! There is an ever growing mole of evidence on the contrary, so prove me otherwise!
Then, exactly, I hear differences after each changes, one of the strongest was just changing the way of the screening on the DIR chip, the 8412... So there are many ways of influencing the sound, Why should I care about this elitist approach of a bloody stamp fabricated by somebody on top of a chip, if it does not have a proven connection to any kind of quality? If not by suggestion?
Ciao, George
Intermodulation in my op amp I/V
Is there still somebody interrested in measurements?
With my challenged op amp I/V-convertor with second best PCM63-K's (I sorted out from a batch of K's and Y's from A'af + a number of myself) in my TentDAC, I tested with two tones: 19 kHz and 20 kHz - 6.4 dB so that the sum became -0.3 dB on the testdisk.
This measurement have been executed (or simulated?) by Hawksford in his article we quoted before.
Any comments?
Again I begin to think that I can only trust myself, my measurements, my ears, and let others shoot the breeze, including Hawksford.....
Is there still somebody interrested in measurements?
With my challenged op amp I/V-convertor with second best PCM63-K's (I sorted out from a batch of K's and Y's from A'af + a number of myself) in my TentDAC, I tested with two tones: 19 kHz and 20 kHz - 6.4 dB so that the sum became -0.3 dB on the testdisk.
This measurement have been executed (or simulated?) by Hawksford in his article we quoted before.
Any comments?
Again I begin to think that I can only trust myself, my measurements, my ears, and let others shoot the breeze, including Hawksford.....
Attachments
At this point it would be nice to see some IMD test for one of the jfet IV circuits..
Though I should add that the original tent configuration, the one that I [still, or yet..] use, would give ugly results in this test. That is, the resistor [60ohm] IV with the zero feedback tube buffer. Which fulfills all the requisites posed by Finney & Hawksford, stable resistive IV impedance up until a resistor can do it, and no NFB.
Still, I've just started to think to use something more performant - after that in some recent tests I have touched the limits. That is, my shameful 6172 based IV seemingly resolves situations with ease, where that tube gets harsh and screechy..
Ciao, George
Though I should add that the original tent configuration, the one that I [still, or yet..] use, would give ugly results in this test. That is, the resistor [60ohm] IV with the zero feedback tube buffer. Which fulfills all the requisites posed by Finney & Hawksford, stable resistive IV impedance up until a resistor can do it, and no NFB.
Still, I've just started to think to use something more performant - after that in some recent tests I have touched the limits. That is, my shameful 6172 based IV seemingly resolves situations with ease, where that tube gets harsh and screechy..
Ciao, George
Joseph K said:
To avoid misunderstanding: My TentDAC is the original DAC except the PLL, the VCXO and the IV-convertor! The print layout of the master board, with the reclocking, is excellent.
The IV-conv. in use, is the one on my web site with three times OPA134.
Cant talk much as I am really busy today.
As I have said in previous messages, people have measured K2 and found them to have less noises. One guy bought a batch of Ks, measured their noises, kept the best ones to himself then sold the rest cheap to other DIYers.
So the ways to measure, to pick up better PCM63s do exist. It's just nobody know the real picture on how K2, KY, and Y were picked.
Listening tests also show that K2, Y, etc do sound better than plain K. Certain Ks are almost as good as K2. Clear enough?
BTW, yes, DAC chips do aging. It's more or less like a burn-in process. Before EM effect kicks in, every DAC may perform differently already. So if you want to figure out how BB picked K2 and Y, you will need a big lot of K2s and Ys to eliminate the individual variances. It's possible for K2 yet unfortunately not for Y as it's impossible to get enough samples of Ys.
Also as a side note, Spencer and I are working on a PCM1794 based DAC. So far, even with a JFET based I/V, the sound is still a bit thin on the bottom, a typical symptom of 1794 DACs. One trick is to put a small RC network *before* the I/V yet I think we will try transistor based I/V first.
-finney
As I have said in previous messages, people have measured K2 and found them to have less noises. One guy bought a batch of Ks, measured their noises, kept the best ones to himself then sold the rest cheap to other DIYers.
So the ways to measure, to pick up better PCM63s do exist. It's just nobody know the real picture on how K2, KY, and Y were picked.
Listening tests also show that K2, Y, etc do sound better than plain K. Certain Ks are almost as good as K2. Clear enough?
BTW, yes, DAC chips do aging. It's more or less like a burn-in process. Before EM effect kicks in, every DAC may perform differently already. So if you want to figure out how BB picked K2 and Y, you will need a big lot of K2s and Ys to eliminate the individual variances. It's possible for K2 yet unfortunately not for Y as it's impossible to get enough samples of Ys.
Also as a side note, Spencer and I are working on a PCM1794 based DAC. So far, even with a JFET based I/V, the sound is still a bit thin on the bottom, a typical symptom of 1794 DACs. One trick is to put a small RC network *before* the I/V yet I think we will try transistor based I/V first.
-finney
finneybear said:
The individual variances are larger than the grade variances so that you will never find out how BB picked whatever, if they picked out anyway.......
A risitor after a current source has no sence...😡
Late answer to George
Joseph,
As you said once that you are waiting to the "Y" measurements results of a mamber here; I was assuming that you were also waiting a shipment yourself. Sorry.
So let's go now through the topics: For me there is no any reason why BB should have put a selection's-level stamp on a chip "just like this". If SONY or YAMAHA are using a special PCM63 grade in their Professional Gears (very expansive, very good reputation, very good engineering) I do trust them that they know the reason for it.
I certainly don’t think that the individual variances are larger than the grade variances (also in responding to Herb).
One for sure: To know the true, we should have got Chips from similar production time and fabricant. We did not come to this: There are MANY flaw chips around, “rest post” as we call it, and this is the situation IN EACH GRADE.
There are true wonderful efforts being done here by our superb members; this is certainly not contradictory to the above, neither contra-productive. I never implied this. What people are doing here is excellent and your own work is as well admirable. It certainly gives this forum a lot of inspiration.
You can much help us with your own TENT (which as you said, with the K inside, it sounds much better then your Rotel (which have the "P" that measured almost the same)). Can you tell us how the sound changes if you do put your excellent pair of "P" instead of the "K" in the Tant? This should be quite easy to do, and please correct me if I missed or mixed something in this particular point.
I will surely report on what will be found. However, because of true shortage in time, we will mostly go after listening. We might measure one or too Chips behaviours as well.
We are having some other tasks to complete, all of them have to do with the topics appeared here: Different Clocks (at least Tent and Burson); different Capacitors between DAC “I“ and “I/V” (better controlling the DAC’s output, according to Herb, NP V1D3, etc. I mean the by-pass cap. I am afraid that there might be decreasing in High-Frequencies information!!). Eventualy we will try playing around the I/V OpAmp (I am using the last generation of Burson’s Discrate and they are excellent).
What I am not sure about is that what you wrote about not removing the chips while doing the tests. I want the results to be 100% accurate. EVERY slight change in my DAC had showed (at least until now) HUGH impact on the sound. I remain very careful about this.
Last yet important: I don't find Finney reaction elitist. He is just putting the facts "on the ground", at least as known to him and to many of us here. Once we pre-select the best chips in each grade, we hear the differences (I agree with him). May be his answer to Herb was quite harsh 🙁 Well, Herb insists that by using a reckloking inside the DAC and using a by-pass cup (etc.,), the differences between the P/K/Y grades disappear....
Joseph K said:
Joseph,
As you said once that you are waiting to the "Y" measurements results of a mamber here; I was assuming that you were also waiting a shipment yourself. Sorry.
So let's go now through the topics: For me there is no any reason why BB should have put a selection's-level stamp on a chip "just like this". If SONY or YAMAHA are using a special PCM63 grade in their Professional Gears (very expansive, very good reputation, very good engineering) I do trust them that they know the reason for it.
I certainly don’t think that the individual variances are larger than the grade variances (also in responding to Herb).
One for sure: To know the true, we should have got Chips from similar production time and fabricant. We did not come to this: There are MANY flaw chips around, “rest post” as we call it, and this is the situation IN EACH GRADE.
There are true wonderful efforts being done here by our superb members; this is certainly not contradictory to the above, neither contra-productive. I never implied this. What people are doing here is excellent and your own work is as well admirable. It certainly gives this forum a lot of inspiration.
You can much help us with your own TENT (which as you said, with the K inside, it sounds much better then your Rotel (which have the "P" that measured almost the same)). Can you tell us how the sound changes if you do put your excellent pair of "P" instead of the "K" in the Tant? This should be quite easy to do, and please correct me if I missed or mixed something in this particular point.
I will surely report on what will be found. However, because of true shortage in time, we will mostly go after listening. We might measure one or too Chips behaviours as well.
We are having some other tasks to complete, all of them have to do with the topics appeared here: Different Clocks (at least Tent and Burson); different Capacitors between DAC “I“ and “I/V” (better controlling the DAC’s output, according to Herb, NP V1D3, etc. I mean the by-pass cap. I am afraid that there might be decreasing in High-Frequencies information!!). Eventualy we will try playing around the I/V OpAmp (I am using the last generation of Burson’s Discrate and they are excellent).
What I am not sure about is that what you wrote about not removing the chips while doing the tests. I want the results to be 100% accurate. EVERY slight change in my DAC had showed (at least until now) HUGH impact on the sound. I remain very careful about this.
Last yet important: I don't find Finney reaction elitist. He is just putting the facts "on the ground", at least as known to him and to many of us here. Once we pre-select the best chips in each grade, we hear the differences (I agree with him). May be his answer to Herb was quite harsh 🙁 Well, Herb insists that by using a reckloking inside the DAC and using a by-pass cup (etc.,), the differences between the P/K/Y grades disappear....
Re: Late answer to George
I prooved it with measurements as I reported here.
I could not say it better🙂 🙂 🙂
irgendjemand said:
I prooved it with measurements as I reported here.
Well, Herb insists that by using a reckloking inside the DAC and using a by-pass cup (etc.,), the differences between the P/K/Y grades disappear....
I could not say it better🙂 🙂 🙂
molto difficile - sehr problematisch....
Let me put it slightly differently then, Herb.
To know for sure that the measurements are right and considering again all other factors, includes aging, moulding influances, etc., one need to have the different chips & grades from the same time period and same fabricant, AT FIRST.
Then, one has to choose matched pairs: It is quite depressing to learn that BB left such big margin (tolerant) inside each grade. Therefore, each of them has to be sorted, in each grade seperately, very unfortunately.
My point is: Only at the very END of the process, when all existing differences in each of the grade has already been sorted out, we would really know if there is an hierarchy P-J-K-K2-Y or not.
All what I am saying besides this, is based on my own LISTENING experience. My tests has been done in worth conditions that the above described "ideal case", of course. My personal conclusion: There must be differences also between grades, and this is for sure NOT auto-suggestion.
Like Spencer, Paul and Finny, I hear the differences. However, I can't tell what does is mean in a pure measurement terminology. No one can do the measurement in this "perfect way" and it might be too late for it. We will need a big amount of P-J-K-K2-Y, let’s say all of them from same machine (your point), Japan, 3rd week of April 1992... In realistic terms, we can’t find them.
We could also look upon this differently: The best clarinet-reads manufactory (Vandoren), who has the absolute best machines, knowledge, skills, wills and everything one can only think about, can not supply you with 2 identical reads, even if being taken from the same tree. They cut it in an unbelievable preciseness, with very high precision (of .0000), but you will always hear the difference between reeds, that it - the wood is "working", the wood "lines" are always different, etc.
So I wonder what is exactly "that" is "working" in our DACs. MAY BE – it the simple result of the combination of multiplication & subtraction of frequencies under the noise level + above 20,000 Hz, that what “kills” us… This is no new news and yes, it can’t be Voodoo.
I am thinking about it again and again: The engineers of the professional gears have their reasons for choosing what they choose. It must have been the absolute right implementation in their fantastic gears. Therefore, I start wondering again (sorry about this guys!) if all PCM63s were indeed done in the way describes here since years, when the major grade-selection is being done at the end of the process….
What are the chances that BB produces K2, KY and Y in an entirely separate process, let’s say - just for the very big: SONY, YAMAHA & Co.?
PA0SU said:
Let me put it slightly differently then, Herb.
To know for sure that the measurements are right and considering again all other factors, includes aging, moulding influances, etc., one need to have the different chips & grades from the same time period and same fabricant, AT FIRST.
Then, one has to choose matched pairs: It is quite depressing to learn that BB left such big margin (tolerant) inside each grade. Therefore, each of them has to be sorted, in each grade seperately, very unfortunately.
My point is: Only at the very END of the process, when all existing differences in each of the grade has already been sorted out, we would really know if there is an hierarchy P-J-K-K2-Y or not.
All what I am saying besides this, is based on my own LISTENING experience. My tests has been done in worth conditions that the above described "ideal case", of course. My personal conclusion: There must be differences also between grades, and this is for sure NOT auto-suggestion.
Like Spencer, Paul and Finny, I hear the differences. However, I can't tell what does is mean in a pure measurement terminology. No one can do the measurement in this "perfect way" and it might be too late for it. We will need a big amount of P-J-K-K2-Y, let’s say all of them from same machine (your point), Japan, 3rd week of April 1992... In realistic terms, we can’t find them.
We could also look upon this differently: The best clarinet-reads manufactory (Vandoren), who has the absolute best machines, knowledge, skills, wills and everything one can only think about, can not supply you with 2 identical reads, even if being taken from the same tree. They cut it in an unbelievable preciseness, with very high precision (of .0000), but you will always hear the difference between reeds, that it - the wood is "working", the wood "lines" are always different, etc.
So I wonder what is exactly "that" is "working" in our DACs. MAY BE – it the simple result of the combination of multiplication & subtraction of frequencies under the noise level + above 20,000 Hz, that what “kills” us… This is no new news and yes, it can’t be Voodoo.
I am thinking about it again and again: The engineers of the professional gears have their reasons for choosing what they choose. It must have been the absolute right implementation in their fantastic gears. Therefore, I start wondering again (sorry about this guys!) if all PCM63s were indeed done in the way describes here since years, when the major grade-selection is being done at the end of the process….
What are the chances that BB produces K2, KY and Y in an entirely separate process, let’s say - just for the very big: SONY, YAMAHA & Co.?
I wonder if there is Y in 2003 or 2004 date code, where are those P, K, K2 or KY in the same period. May be the so call licenced manufacturer has optimised the process to produce the best PCM63 grade. Or after they sorted the Y, the fall out chips is dump into the salt water sea? Moreover why such production is in such a small scale which make most diyer cannot buy from the normal distribution channel. Any one ever see any machine is using a 03/04 date code chip inside? May be all the equipment using these chips sound too good and the owner or studio never give them up and sell as scrap....
In electronics, there no ONE measurement can tell the whole story and we cannot just judge by one measurement to correlate to sonic performance. Somemore measurement is by sine wave but music is dynamic signals.
Is Sony and Yamaha just mark the chips for fun or as a selling point in the pro gears? If this is a selling point, why they dont label the equipment outside to emphasis that and tell the whole world best chips is used inside? Why they pay extra money for the extra markings?
Sorry I hv too many why questions ... ...
In electronics, there no ONE measurement can tell the whole story and we cannot just judge by one measurement to correlate to sonic performance. Somemore measurement is by sine wave but music is dynamic signals.
Is Sony and Yamaha just mark the chips for fun or as a selling point in the pro gears? If this is a selling point, why they dont label the equipment outside to emphasis that and tell the whole world best chips is used inside? Why they pay extra money for the extra markings?
Sorry I hv too many why questions ... ...
Hopping to hear that someone here (diyaudio member) has a personal & direct contact to some of BB’s retired employee, or SONY’s Professionals employee who was working there - let’s say - about 10 years ago.
Or: Chat/forum/internet pages/addresses where these person(s) are available (any known name?)
BTW, once SONY moved from PCM63 to PCM1702 (remember that SONY recording machine/desks?) the PCM63KY/Y might already lost part of its “glory”. We need a person from that time period...
Sony and Yamaha either mark the chips for fun nor as a selling point in the pro gears: Both companies are well known for their own chip-developments, using "only" their own chips (see Yamaha DIR for example), so using BB is not a selling point. Of course, they don’t label the equipment outside with BB (or any other either), as they will not wish to emphasis = tell the whole world that the real best chips used inside are BB. But, they surely had to pay extra money for the extra quality. Not a big deal if you are SONY and want to be the leader in recordings studios & keep the market on your terms.
By the way, I was very surprised to learn through this forum (mainly thanks to Paul’s researches) that SONY was using a BB PCM in that SONY PCM 3348 HR (see picture...). Quite an overwhelming news if you know how SONY standard(ed) the industry.
Of course, just open a SONY consumer DAT, CD Player, etc... There is no BB inside. But, the professional gears always had different demands and standards: Once they need an excellent PCM – and if they can’t make it themselves to that extent, then they have to go to some other manufactory. It is a s simply as this.
What happened much later? Hard to say: All other "Y" s (2003, 4) might well been licensed... OK, a retired person from the licenses department could help us as well.
Meanwhile all we can trust is our ears: This should not be put so much aside...
Or: Chat/forum/internet pages/addresses where these person(s) are available (any known name?)
BTW, once SONY moved from PCM63 to PCM1702 (remember that SONY recording machine/desks?) the PCM63KY/Y might already lost part of its “glory”. We need a person from that time period...
Sony and Yamaha either mark the chips for fun nor as a selling point in the pro gears: Both companies are well known for their own chip-developments, using "only" their own chips (see Yamaha DIR for example), so using BB is not a selling point. Of course, they don’t label the equipment outside with BB (or any other either), as they will not wish to emphasis = tell the whole world that the real best chips used inside are BB. But, they surely had to pay extra money for the extra quality. Not a big deal if you are SONY and want to be the leader in recordings studios & keep the market on your terms.
By the way, I was very surprised to learn through this forum (mainly thanks to Paul’s researches) that SONY was using a BB PCM in that SONY PCM 3348 HR (see picture...). Quite an overwhelming news if you know how SONY standard(ed) the industry.
Of course, just open a SONY consumer DAT, CD Player, etc... There is no BB inside. But, the professional gears always had different demands and standards: Once they need an excellent PCM – and if they can’t make it themselves to that extent, then they have to go to some other manufactory. It is a s simply as this.
What happened much later? Hard to say: All other "Y" s (2003, 4) might well been licensed... OK, a retired person from the licenses department could help us as well.
Meanwhile all we can trust is our ears: This should not be put so much aside...
Attachments
To Irgendjemand and Spencer
Al the listening tests on the different grades of the PCM63 have been done in different environments. I do not speak about the PA and loudspeakers. They are not changed during the listening sessions.
Here a piece of system theory:
Each complex system can only be compared if the pre-conditions are equal. The problem is that the system itself is part of these pre-conditions if it has not been isolated from its environment. In other words 'the precondition must be made true' in system theory terms.
Par exemple: if we connect a DAC directly to a digital filter, the output signal of the filter (= the input signal of the DAC) will depend on the DAC behind it, so the precondition for different DAC's is NOT fulfilled if we exchange them.
In my opinion the only way to isolate the filter from the DAC is: reclocking. (Reclocking should always be done because of less jitter, but this is another subject).
The remaining question is: how to build a reclocking circuit which correctly isolates the DAC from its input source.
The TentDAC-boys made a very good beginning in an excellent print layout (= very important!!!). Henk ten Pierick fulfilled the specs as formulated before.
This theory is overlooked very often, just like on this thread.
The statement I make is that the audio performance of the PCM63's becomes much less dependent of its grade or whatever difference in production if the precondition is fulfilled.
In other words, in the listening tests performed by a lot of people here, the differences in audio performance logged here are more dependent on the filter-behaviour and on the IV-behaviour than on the DACs itself.
Al the listening tests on the different grades of the PCM63 have been done in different environments. I do not speak about the PA and loudspeakers. They are not changed during the listening sessions.
Here a piece of system theory:
Each complex system can only be compared if the pre-conditions are equal. The problem is that the system itself is part of these pre-conditions if it has not been isolated from its environment. In other words 'the precondition must be made true' in system theory terms.
Par exemple: if we connect a DAC directly to a digital filter, the output signal of the filter (= the input signal of the DAC) will depend on the DAC behind it, so the precondition for different DAC's is NOT fulfilled if we exchange them.
In my opinion the only way to isolate the filter from the DAC is: reclocking. (Reclocking should always be done because of less jitter, but this is another subject).
The remaining question is: how to build a reclocking circuit which correctly isolates the DAC from its input source.
The TentDAC-boys made a very good beginning in an excellent print layout (= very important!!!). Henk ten Pierick fulfilled the specs as formulated before.
This theory is overlooked very often, just like on this thread.
The statement I make is that the audio performance of the PCM63's becomes much less dependent of its grade or whatever difference in production if the precondition is fulfilled.
In other words, in the listening tests performed by a lot of people here, the differences in audio performance logged here are more dependent on the filter-behaviour and on the IV-behaviour than on the DACs itself.
Thank you Herb for this excellent explanation and your much invested energy here.
Let's agree at first that most of us have gathered some serious experience with Digital Filters and its advantageous, the same is true concerning I/Vs behaviour.
Assuming that each of us already came to what seemed to him to be the best suitable/possible Filter for his configuration, we can take this fact as a proven and non-variable pre-condition.
Same about the basic in the I/V. Otherwise, nothing is to be agreed upon. Here it might be even worth: Changing the all I/V seemed to me a bit "off": A lot of work and not clear results - as we have different experiences and tasts.
Of course, we can go to the different improvements, but which of them?
Take for example the clock and re-clocking: I am going to check this very slowly... My personal wish is not to "dry out" the sound and the natural acoustics-vibrations on the recording, while clocking it might have some negative influance exactly on this aspect. I would also not like to lose high-frequencies information when a cup is coming between DAC and I/V. AND - I have diferent lay-out of course.
So there is a certain limit when following your advices, like any other good advices. For example, I do want to be sure that I have the control and the know-how about the all process.
After all this, an exchange of a PCM in a socket will be a small child-play. Not hearing the sound differences by PCM's change will be surely reported once occurs. I am by far not there yet.
Let's agree at first that most of us have gathered some serious experience with Digital Filters and its advantageous, the same is true concerning I/Vs behaviour.
Assuming that each of us already came to what seemed to him to be the best suitable/possible Filter for his configuration, we can take this fact as a proven and non-variable pre-condition.
Same about the basic in the I/V. Otherwise, nothing is to be agreed upon. Here it might be even worth: Changing the all I/V seemed to me a bit "off": A lot of work and not clear results - as we have different experiences and tasts.
Of course, we can go to the different improvements, but which of them?
Take for example the clock and re-clocking: I am going to check this very slowly... My personal wish is not to "dry out" the sound and the natural acoustics-vibrations on the recording, while clocking it might have some negative influance exactly on this aspect. I would also not like to lose high-frequencies information when a cup is coming between DAC and I/V. AND - I have diferent lay-out of course.
So there is a certain limit when following your advices, like any other good advices. For example, I do want to be sure that I have the control and the know-how about the all process.
After all this, an exchange of a PCM in a socket will be a small child-play. Not hearing the sound differences by PCM's change will be surely reported once occurs. I am by far not there yet.
irgendjemand said:
You missed the essence of my plea: a digital filter never will be a precondition! Its performance is influanced by the following DAC. The only findings you will get is eg. that after filter-A a PCM63-Y will sound best and after filter-B a PCM63-K2 performs better. You cannot do any predicate about the PCM63's itself!
An electronic circuit is not a musical instrument! You can't "dry out" the sound. An electronic circuit "does not know" that it has been used to (re)produce music let alone that it "knows" what 'natural accoustic-vibrations' are! It should execute its electronic function as good as possible! In the technique a complex function has to be splitted up into smaller manageable steps, which are to be investigated independently.
When you, as a musician, have to select an musical instrument, say a piano, you do not listen to them in different rooms with different artists. You know as nobody else here what a great difference there is between performers and acoustics. For a selection you put all the pianos in one an the same room with one and the same performer. These are the preconditions.
Reclocking brings the music back into your room. I promise you, but this is for another reason: reclocking is as the function of a conductor: the orkestra gets pitch, the tempi are predictable, the instruments start the measures at the same time.
The same with a reclocked DAC: the Letch comes mutch more reliable at the right stroke, so that the conversion from digital to analog takes place with less distortion.
which is only possible if the preconditions are fulfilled.
This is the trap: it is so easy...... but it does bring you nothing.
Thanks for all this, Herb!
Filter:
Might be that you missed the essence of my plea two: If I can't change to a different digital filter (I can use indeed only the DF1704 or the old DF1700 which is surely NOT good!), then I call it "precondition" ... What I mean is that there are limitations for each configuration. Of course, in your theoretical argument here, you are right.
So here is a question to you: Would you say that you have a total flexibility to implament whatever Filter you want? Let's say: SM5847, DF1704, PMD100? Is this realistic? AND: Isn't it possible that there is one of them which is the best one? I trully wonder!.
Clock:
I experienced once a CD-Clock that make the sound appearing to be so direct, that I felt that the music became "brainy" and "dry". I know what you mean here as well. I simply assume that by a true right approach to the Clock, the overtones “sit” a bit different, better of course, “in ordnung" – but that every small change in this point, will change the all impression of the sound-stage & feelings.
Of course - the clock does not change anything! It is just “asking” for several other changes in other places in the HiFi system.. For example – the loudspeaker position (unexpected but true). I can take them 2 meter from the wall as you once suggested, and the highs will sit again "right" in the room. There might be less midrange, as the bass reflexes becomes less imposing the sound due to the new space. etc. etc. etc. It is only that what I meant before. Once again, technically I do agree with you also here.
PCM:
Changing PCM as a trap? As I said, we still have time for this… your argument have yet to be confirmed by several members (you might not like this idea..
).
But mostly, I admit that all the rest (re-clocking inside the DAC, I/V, de-cup etc. etc) are MUCH more interesting topics then the PCM exchanging! But, can we give up on this? We will see. May be, may be not... 😉 may be yes and no...
Once again, many thanks for your support. BTW, I am awaiting the AVM DAC 1.2 plan ("Schalt/Liege-Plan") and I will contact you upon receiving it. With my technician
, I am about to work on several other topics (I/V, eventually a new Clock & PSU for my CD Player, some cups-changes, Digital-Out, etc) but this special topic of re-clocking inside the DAC will remain yours, if you will agree, of course.
I am Looking forward to meet you then.
Greetings!
Filter:
Might be that you missed the essence of my plea two: If I can't change to a different digital filter (I can use indeed only the DF1704 or the old DF1700 which is surely NOT good!), then I call it "precondition" ... What I mean is that there are limitations for each configuration. Of course, in your theoretical argument here, you are right.
So here is a question to you: Would you say that you have a total flexibility to implament whatever Filter you want? Let's say: SM5847, DF1704, PMD100? Is this realistic? AND: Isn't it possible that there is one of them which is the best one? I trully wonder!.
Clock:
I experienced once a CD-Clock that make the sound appearing to be so direct, that I felt that the music became "brainy" and "dry". I know what you mean here as well. I simply assume that by a true right approach to the Clock, the overtones “sit” a bit different, better of course, “in ordnung" – but that every small change in this point, will change the all impression of the sound-stage & feelings.
Of course - the clock does not change anything! It is just “asking” for several other changes in other places in the HiFi system.. For example – the loudspeaker position (unexpected but true). I can take them 2 meter from the wall as you once suggested, and the highs will sit again "right" in the room. There might be less midrange, as the bass reflexes becomes less imposing the sound due to the new space. etc. etc. etc. It is only that what I meant before. Once again, technically I do agree with you also here.
PCM:
Changing PCM as a trap? As I said, we still have time for this… your argument have yet to be confirmed by several members (you might not like this idea..
).But mostly, I admit that all the rest (re-clocking inside the DAC, I/V, de-cup etc. etc) are MUCH more interesting topics then the PCM exchanging! But, can we give up on this? We will see. May be, may be not... 😉 may be yes and no...
Once again, many thanks for your support. BTW, I am awaiting the AVM DAC 1.2 plan ("Schalt/Liege-Plan") and I will contact you upon receiving it. With my technician
, I am about to work on several other topics (I/V, eventually a new Clock & PSU for my CD Player, some cups-changes, Digital-Out, etc) but this special topic of re-clocking inside the DAC will remain yours, if you will agree, of course. I am Looking forward to meet you then.
Greetings!
Mutual isolation of the circuits
Sure, if they have the same pinning, yes! Guido Tent made a little board on which an SMD SM5847 has been assembled. First I used the SM5842. You can hear the difference between them. But, again, you should do such tests in an environment in which the IC's are totally isolated as on the TentDAC board. I come to this subject later.
However the Bartok-recording (Sonata for 2 pianos & percussion, on: DECCA 443 894-2) is verry sensitive to jitter as the recording of Varèse's "Tuning Up" (on DECCA 460 208-2: 'The complete works' with KCO & Riccardo Chailly).
There is something different on isolation (which is my absolute precondition). The lay out of the printed circuit board (PCB) should be designed for optimal isolation for digital signals between the various IC's. Guido Tent is an expert on this subject. He worked with Philips in a design-group for DVD I think. He wrote a chapter on this subject here:
http://www.tentlabs.com/InfoSupport/page35/files/Supply_decoupling.pdf
How can I explain this very technical subject in your terms? I'll give it a try:
Bartok prescribed that the 2 pianos in his 'Sonata for 2 pianos & percussion' should be placed on a large distance of each other. If these two pianos are on a wooden stage with its own resonances, the isolation between the legs of the pianos and the stage (in Dutch we call this 'isolatie voor contactgeluid') becomes important outherwise the stage will be a Schallboden for both pianos so that the sound will blur and the separation in the sound image is gone!
Replacing one or both pianos will not help as long as the legs have not been isolated from the stage.
If you want to make a good start, order for the main board of the TentDAC. The isolation on this PCB is two-fold: the right lay out for mutual isolation of the IC's from supply and ground loop currents ,and the isolation by reclocking just before the DAC . Moreover the reclocking circuit is on the right place on the board.
All these aspects are important. Reclocking is not just an 'add on'.
You are wellcome of course, at least for listening to my demonstrations........
Greetings!
irgendjemand said:
Sure, if they have the same pinning, yes! Guido Tent made a little board on which an SMD SM5847 has been assembled. First I used the SM5842. You can hear the difference between them. But, again, you should do such tests in an environment in which the IC's are totally isolated as on the TentDAC board. I come to this subject later.
One of the preconditions here is that the recording you are listening to, is perfect. I have some (poor) recordings which sound better on the unmodified CD624.......... I have an excellent recording (Virgin Classics: Schubert String Quintet Quartettsatz with the Artemis Quartet) which I use to judge loudspeakers and amplifiers. This recording is not so sensitive to clock-jitter!
However the Bartok-recording (Sonata for 2 pianos & percussion, on: DECCA 443 894-2) is verry sensitive to jitter as the recording of Varèse's "Tuning Up" (on DECCA 460 208-2: 'The complete works' with KCO & Riccardo Chailly).
No, not the changing is the trap, the trap is that you change them in a not isolated environment 'because it is so easy'PCM:
Changing PCM as a trap? As I said, we still have time for this… your argument have yet to be confirmed by several members (you might not like this idea..
Once again, many thanks for your support. BTW, I am awaiting the AVM DAC 1.2 plan ("Schalt/Liege-Plan") and I will contact you upon receiving it. With my technician
There is something different on isolation (which is my absolute precondition). The lay out of the printed circuit board (PCB) should be designed for optimal isolation for digital signals between the various IC's. Guido Tent is an expert on this subject. He worked with Philips in a design-group for DVD I think. He wrote a chapter on this subject here:
http://www.tentlabs.com/InfoSupport/page35/files/Supply_decoupling.pdf
How can I explain this very technical subject in your terms? I'll give it a try:
Bartok prescribed that the 2 pianos in his 'Sonata for 2 pianos & percussion' should be placed on a large distance of each other. If these two pianos are on a wooden stage with its own resonances, the isolation between the legs of the pianos and the stage (in Dutch we call this 'isolatie voor contactgeluid') becomes important outherwise the stage will be a Schallboden for both pianos so that the sound will blur and the separation in the sound image is gone!
Replacing one or both pianos will not help as long as the legs have not been isolated from the stage.
If you want to make a good start, order for the main board of the TentDAC. The isolation on this PCB is two-fold: the right lay out for mutual isolation of the IC's from supply and ground loop currents ,and the isolation by reclocking just before the DAC . Moreover the reclocking circuit is on the right place on the board.
All these aspects are important. Reclocking is not just an 'add on'.
You are wellcome of course, at least for listening to my demonstrations........
Greetings!
New tread on IV's with op amps...
To avoid to mutch polution here, I started a new thread:
I'm not yet ready with the subject, so:
Welcome there.....
To avoid to mutch polution here, I started a new thread:
I'm not yet ready with the subject, so:
Welcome there.....
PCM1704 versus PCM63
Investigation of four PCM1704's on a altmustech plugin.
I measured the distortion of a 997 Hz signal with levels of 0 dB down to -78 dB in 6 dB-steps, just as I did with fourteen PCM63's reported earlier here.
The 1704's measure as the average PCM63.
I also measured the intermodulation distortion with two signals of - 6dB, 19 and 20 kHz, so that their sum is 0 dB in the peaks.
The intermod seemed to be better, but in fact the intermod of the IV-conv is measured. Because the output current of the 1704 is 1.2 mA in stead of 2.0 mA (with the 63) the intermod should be less, and indeed there is no intermod noticable with my opamp IV-conv. The 63 gives the same pictures with an input - 4dB which correspondents with 1.2 mA.
With recordings that do not exceed the -6 dB, I could not hear remarcable differencies between 1704 and 63.
Investigation of four PCM1704's on a altmustech plugin.
I measured the distortion of a 997 Hz signal with levels of 0 dB down to -78 dB in 6 dB-steps, just as I did with fourteen PCM63's reported earlier here.
The 1704's measure as the average PCM63.
I also measured the intermodulation distortion with two signals of - 6dB, 19 and 20 kHz, so that their sum is 0 dB in the peaks.
The intermod seemed to be better, but in fact the intermod of the IV-conv is measured. Because the output current of the 1704 is 1.2 mA in stead of 2.0 mA (with the 63) the intermod should be less, and indeed there is no intermod noticable with my opamp IV-conv. The 63 gives the same pictures with an input - 4dB which correspondents with 1.2 mA.
With recordings that do not exceed the -6 dB, I could not hear remarcable differencies between 1704 and 63.
Attachments
- Status
- Not open for further replies.