Hey Everyone!
What might one expect to hear in terms of differences between a CD player with a pair of 18-bit Burr Brown PCM1701s and another with the Sony CXD2552? The two players in question are a Luxman DZ-112 and a Sony CDP-X111ES, respectively. I don't have a strong bias toward R-2R or Delta-Sigma and am not even sure if DAC type contributes much to the overall sound.
I don't want to say too much to bias the results other than that before modifying the DZ-112 I definitely preferred the sound of the modified X111ES. However, after a few modifications, I'm now finding that I prefer the sound of the DZ-112, and that the difference isn't very subtle, IMO. I'm not sure how much might be due to the difference between the two DACs and how much might be for other reasons. Thus far, all listening has been via each player's headphone output with my AKG K702s using two copies of the exact same pressed CD.
Both have 0.01µF polypropylene bypass capacitors on the output electrolytics and ADA4627-1 op amps in the output stage (though I'm currently using OPA827s in the X111ES because I only have two sets of ADA4627-1s that can be installed at the moment). I also changed the DZ-112's M5216 headphone IC to an AD8397 which seems a tad more open and dynamic sounding. The X111ES also has a few PSU electrolytic capacitor upgrades, whilst the DZ-112 has the stock capacitors (except the poly bypass caps). The X111ES also has the original RC4556 headphone amp, which is not as revealing as the AD8397. The X111ES has +/-5V rails for the analog stage whilst the DZ-112 uses +/-12V.
Your thoughts?
What might one expect to hear in terms of differences between a CD player with a pair of 18-bit Burr Brown PCM1701s and another with the Sony CXD2552? The two players in question are a Luxman DZ-112 and a Sony CDP-X111ES, respectively. I don't have a strong bias toward R-2R or Delta-Sigma and am not even sure if DAC type contributes much to the overall sound.
I don't want to say too much to bias the results other than that before modifying the DZ-112 I definitely preferred the sound of the modified X111ES. However, after a few modifications, I'm now finding that I prefer the sound of the DZ-112, and that the difference isn't very subtle, IMO. I'm not sure how much might be due to the difference between the two DACs and how much might be for other reasons. Thus far, all listening has been via each player's headphone output with my AKG K702s using two copies of the exact same pressed CD.
Both have 0.01µF polypropylene bypass capacitors on the output electrolytics and ADA4627-1 op amps in the output stage (though I'm currently using OPA827s in the X111ES because I only have two sets of ADA4627-1s that can be installed at the moment). I also changed the DZ-112's M5216 headphone IC to an AD8397 which seems a tad more open and dynamic sounding. The X111ES also has a few PSU electrolytic capacitor upgrades, whilst the DZ-112 has the stock capacitors (except the poly bypass caps). The X111ES also has the original RC4556 headphone amp, which is not as revealing as the AD8397. The X111ES has +/-5V rails for the analog stage whilst the DZ-112 uses +/-12V.
Your thoughts?
Last edited:
The datasheet for the PCM1701 seems completely inadequate, the CXD2552 seems to be an ancient sigma-delta design with lowish specs. If the 18 bit one really has 16 ENOB or better the two ought to be equivalent, exceeding human perception ability to distinquish, but there's no a great deal of data to go on.
Doing modifications and then hearing a difference can simply be expectation bias - unless you have a way to swap between pre-mod and post-mode in a blind manner you can just be subconciously inventing a preference.
By the way using two copies of a CD is not a safe comparison unless you can check that neither have soft-errors. When ECC fails on audio CD you get interpolated samples due to the interleaved blocks. Most CD players give no indication that this is happening.
You need same CD, same player to guarantee equivalent ECC performance in the absence of soft-error indication.
Doing modifications and then hearing a difference can simply be expectation bias - unless you have a way to swap between pre-mod and post-mode in a blind manner you can just be subconciously inventing a preference.
By the way using two copies of a CD is not a safe comparison unless you can check that neither have soft-errors. When ECC fails on audio CD you get interpolated samples due to the interleaved blocks. Most CD players give no indication that this is happening.
You need same CD, same player to guarantee equivalent ECC performance in the absence of soft-error indication.
Perhaps this topic is a decade or two too late!
Thanks Mark,
I'd rather not focus on a side discussion regarding psychoacoustics and error correction and see if I can get more opinions of the differences in sound between R-2R and DS converters. That said, the pair of identical pressed CDs sound subjectively the same in either player (as expected), making them a controlled variable.
While I'd like to think of a DAC as essentially a "black box" that just converts a data stream into analog audio, I'm hearing quite significant differences between the two players which do not seem to be accounted for variables other than the DACs. This leads me to wonder if there may in fact be a difference in the way the two technologies render sound. Seems like the published data on these two chips is insufficient to account for audible differences.
Thanks Mark,
I'd rather not focus on a side discussion regarding psychoacoustics and error correction and see if I can get more opinions of the differences in sound between R-2R and DS converters. That said, the pair of identical pressed CDs sound subjectively the same in either player (as expected), making them a controlled variable.
While I'd like to think of a DAC as essentially a "black box" that just converts a data stream into analog audio, I'm hearing quite significant differences between the two players which do not seem to be accounted for variables other than the DACs. This leads me to wonder if there may in fact be a difference in the way the two technologies render sound. Seems like the published data on these two chips is insufficient to account for audible differences.
I cannot quite grasp the necessity of using two copies, but if you hate opening cd trays that's ok.
Not sure of the purpose behind this thread either. Do DS chips sound different to multibit? They most certainly do. Countless threads on this and other boards about these differences and about specific favourite chips.
Seems you are ok with opamps sounding different but dramatically different digital technologies sounding different surprises you. Why?
There are two variations of PCM1701(P and KP).I have the KP version in my Denon DCD1560.Assuming that the Luxman digital processing is right!!!! if it's only the different pinout i hope i can help with some ideas to try:
the ADA4627 seem to be a better op-amp for the I/V section due to its lower noise, better phase margin , predictable overshoot behavior from its graphs, more output current vs idle current(will run cooler).
On the other hand , even though it's the lower slew rate and compensated version, it still needs to be tamed a bit and cut the DAC trash output as the op-amp input capacitance is quite low so i'd suggest to try 100pF...200pF on the input of the op-amp to plus input or ground and substracting the coresponding value from the 820pf of your feedback network. Also the de-emphasis is useless today and you might need 2sk246 for your amps , so remove it
You don't have a passive filter like i do in my Denon so modifying it on the Denon recipe need a bit of calculation to get the output level right .I'd try to remove the image filter completely and replace it with a simple first order RC filter , but ada4627 is quite a capable op-amp and should do it right.
the ADA4627 seem to be a better op-amp for the I/V section due to its lower noise, better phase margin , predictable overshoot behavior from its graphs, more output current vs idle current(will run cooler).
On the other hand , even though it's the lower slew rate and compensated version, it still needs to be tamed a bit and cut the DAC trash output as the op-amp input capacitance is quite low so i'd suggest to try 100pF...200pF on the input of the op-amp to plus input or ground and substracting the coresponding value from the 820pf of your feedback network. Also the de-emphasis is useless today and you might need 2sk246 for your amps , so remove it
You don't have a passive filter like i do in my Denon so modifying it on the Denon recipe need a bit of calculation to get the output level right .I'd try to remove the image filter completely and replace it with a simple first order RC filter , but ada4627 is quite a capable op-amp and should do it right.
Last edited:
Sigh...
FWIW, I just started modding CD players recently, whilst others here probably started working on them decades ago. Over the years I've built quite a few pedals (including an A/DA flanger) and modified various amplifiers and recording interfaces. However, this is my first time playing around with a multibit DAC; my first CD player (CDP-65) was probably multibit too, though it was kind of harsh, IMO, and I didn't modify it before it went to recycling.
I'm still trying to formulate my views on the differences in the sound. At the moment, I'm mostly exploring the modding possibilities and listening to the differences as changes are made. After getting the X111ES and making some mods, I had decided that I'd leave the DZ-112 because it just didn't sound that great in comparison (somewhat dull/mellow, which I suspect was due to the AD712 and M5216 op amps and lack of transparency in the output capacitors). However, the DZ-112's new op amps and output bypass caps have left the Sony sounding subjectively worse. That surprised me!
I have two copies of the same CD by accident, but they are useful for making quick A/B comparisons between CD players, which helps because auditory memory is fairly short.
Some objectivists might argue that any two DACs that measure the same (within relevant parameters) must therefore sound the same. Many engineers also claim that op amps impart absolutely nothing to the sound and some have told me that I'm nuts for thinking I can hear obvious differences between them. Some members here seem to be highly objectivist and question every possible variable, whilst others accept the possibility that some people are able to hear differences between op amps and DACs. It's difficult to know where people are coming from sometimes.
I didn't know what to expect from the PCM1701Ps in the DZ-112. Initially I figured that the sound of the DACs in the two players was likely close enough that differences in the analog stages would be more of a factor. But I'm hearing more of a difference than expected. For one, I can turn the DZ-112 up more and it still sounds musical and less fatiguing than the X111ES, just more "natural" somehow. Obviously, such adjectives are insufficient, but that's the best I can do at the moment.
It's possible that the DZ-112 has more distortion, as the player has MSB trimpots which haven't been adjusted since the factory. It's also possible that the low 5V analog section rails in the X111ES are holding it back, perhaps even contributing to some distortion due to lack of sufficient headroom. I greatly prefer the ergonomics and design of the X111ES. The DZ-112 also has trouble with 80min discs.
Links to other pertinent threads would be appreciated.
Hi dreamth,There are two variations of PCM1701(P and KP).I have the KP version in my Denon DCD1560.Assuming that the Luxman digital processing is right!!!! if it's only the different pinout i hope i can help with some ideas to try:
the ADA4627 seem to be a better op-amp for the I/V section due to its lower noise, better phase margin , predictable overshoot behavior from its graphs, more output current vs idle current(will run cooler).
On the other hand , even though it's the lower slew rate and compensated version, it still needs to be tamed a bit and cut the DAC trash output as the op-amp input capacitance is quite low so i'd suggest to try 100pF...200pF on the input of the op-amp to plus input or ground and substracting the coresponding value from the 820pf of your feedback network. Also the de-emphasis is useless today and you might need 2sk246 for your amps , so remove it
You don't have a passive filter like i do in my Denon so modifying it on the Denon recipe need a bit of calculation to get the output level right .I'd try to remove the image filter completely and replace it with a simple first order RC filter , but ada4627 is quite a capable op-amp and should do it right.
Wasn't aware of the PCM1701KP... are they a higher grade or something? How do they sound in your Denon? I'm curious about your subjective impressions. From what I gather, the PCM1701 was designed near the end of the multibit development craze at the end of the 1980s, right before DS took over, and is supposed to be one of the nicer DACs from that era. Sadly, there's very little technical information available on it.
I may make further modifications. But before that I need to get the transport sorted out. It was very susceptible to bumps when I got it and I've spent a few hours playing with the servo trimpots (NOT the laser power) and getting a feel for how they operate. Will probably hook it up to the scope in the next few days for a fine adjustment. However, it is interesting just listening to how the various adjustments affect the sound of the transport, listening for nulls, smoothness, noise, etc....
DO NOT TOUCH the trimpots !
I have no idea of any differences between the PCM1701 versions other than pinout.The Denon was heavily modified in the analog and supply part which achieved lower noise floor and better overall sound, its transport is the best you can get, its digital side is what worries me as part of its datasheet is completely untraceable.
It was my favorite player until i heard some Yamaha player using the Analog Devices equivalent of PCM61... Anyway, I'm thinking of changing the whole digital oversampling and dac with PCM1798 now that my beloved Denon was dethroned by a much cheaper Yamaha... I have a famous NEC using pcm56 and i was thinking of replacing it with pcm61 but now i'm full of doubts as the transport and te digital processing before dac stinks...although i have made the best possible I/V conversion after its DAC and it's not touching the depth of Denon player.
I personally believe that the DAC are all the same to our ears...The digital processing before and the analog processing after them are of much higher importance in my view.Any DAC should sing if delivered the right bits and applied the right analogue treatment after.
You have them both here side by side in a full discussion on changing the I/v convertor of the NEC player. https://www.diyaudio.com/forums/digital-source/338152-nec-cd-810-revival-mods.html#post5811581
!I have no idea of any differences between the PCM1701 versions other than pinout.The Denon was heavily modified in the analog and supply part which achieved lower noise floor and better overall sound, its transport is the best you can get, its digital side is what worries me as part of its datasheet is completely untraceable.
It was my favorite player until i heard some Yamaha player using the Analog Devices equivalent of PCM61... Anyway, I'm thinking of changing the whole digital oversampling and dac with PCM1798 now that my beloved Denon was dethroned by a much cheaper Yamaha... I have a famous NEC using pcm56 and i was thinking of replacing it with pcm61 but now i'm full of doubts as the transport and te digital processing before dac stinks...although i have made the best possible I/V conversion after its DAC and it's not touching the depth of Denon player.
I personally believe that the DAC are all the same to our ears...The digital processing before and the analog processing after them are of much higher importance in my view.Any DAC should sing if delivered the right bits and applied the right analogue treatment after.
You have them both here side by side in a full discussion on changing the I/v convertor of the NEC player. https://www.diyaudio.com/forums/digital-source/338152-nec-cd-810-revival-mods.html#post5811581
Last edited:
One CD is not sufficient for comparison. Try different kind of music and a good recoding selection of acoustic instruments (piano, guitar, flute) solo and orchestral. If you don't have acces to a quality CD's download free "Original CD" samples from the 2l website and burn it to CD: 2L High Resolution Music .:. free TEST BENCH
You may not like such kind of music, but it is the best for comparison. On the other scale try typical recodings of loudness war like Kalvin Harris. The one that gives less fatigue on your favourite material is always better, in the end you will listen to more music...
Once again, don't touch the pots!
You may not like such kind of music, but it is the best for comparison. On the other scale try typical recodings of loudness war like Kalvin Harris. The one that gives less fatigue on your favourite material is always better, in the end you will listen to more music...
Once again, don't touch the pots!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.