Once i get to work on one of these i plan to remove all ceramics, replace them with ecpu and probably leave a couple of empty pads for adding electrolytics.
You mean "now my reproduction system is all proper"... maybe you are compensating for a fault in a unit which is not the reason for the error. What you hear is the sum of the chain. It never occurred to you that in the moment you defined your laptop output as being the reference, you started to alter a fine DAC to sound like your laptop...
//
Ha, nice try! There really isn’t much to compensate for in my chain. Hd650 is not known to have an extremely anemic low end. Unfortunately I don’t have ABX results, but I offered what I could for dam1021 users considering the cap mod. I don’t think it was about tweaking a sound against a reference as I didn’t really know what the cap mod would do to the rev4. But you can have your interpretations.
A factual correction: 1541 only has 800uF per rail and 1941 has 400+1000 per rail. Soren probably did listening tests on those so the 1941 configuration may be optimal to his ears, or maybe as good as 1541/1121, maybe even 1021.
I am an audiophile to the extent that I attempted to tell whether dam1021 is better than soundcard, but I’m not in the sense that I trust my listening skills at the moment to finely characterize the sound of two good equipment and choose one over the other when what I’m liking could just be artificial distortion. I don’t think that is likely to be the case with dam1021 since Soren didn’t seem to have tweaked the sound in any way. I’m not a proper DIYer in that I can’t do ABX tests with modded and unmoddd systems, but I do put things together
Last edited:
Sorry I didn't quite understand why you were asking it at first and I'm not checking my messages as much as before. Literally any decent track works, the better mastered the more obvious the change I'd say. I still need to find a time to have my friend come over and do some ABX tests to be sure. That's all I can say really, hope it helps!
@TNT I didn't want to waste time doing more audiophile tests that can't get me much closer to a statement supported by statistics, but looking back you really did quite a bit to keep us sane... It seemed you're interested so I went ahead and did some comparisons on what I was listening to.
Both tracks are dvorak op. 104 1st mvt. I was listening to the Fourier/BPO version 16/44, you can probably get it as a digital download at a very modest price if you don't have it already. At 6:59 the flutes enter and there's a crescendo from there to around 7:15 where you get some nice dynamics. The flutes sound a bit more natural and frankly just quite a bit deeper in the stage on BAL than SE. And as more instruments enter the expansion of soundstage on BAL just went further than SE.
Now, on the Rostropovich/BPO 24/192 version, flutes and cello sound a bit more muted in the same section, so I went straight to the cello solo at 10:40. On SE, the soundstage felt cut off right at my ears, but the BAL naturally extends much further. I went back and forth between the two outputs, and tried to imagine SE with a larger soundstage and BAL with a smaller one, which has worked well in the past to disillusion myself. Only this time there was no way I could convince myself that SE sounds nearly as open as BAL.
Used HD650, -22db with BAL and -16db with SE. They sounded about the same loudness, but I don't quite believe volume changes things in this case. Both are buffered outputs onboard. SE with 10R 2V output and BAL 20R 4V.
This nicely confirmed all of my past experiences with BAL vs SE outputs on my system. The conclusion isn't definite, however, as I would need a proper ABX test to claim that. It could be that my system is noisy and the balanced output helped a lot there? Can't hear any noise at max volume on a silent track though. I'm guessing you're deciding whether to put back your dual-mono... Just do an ABX and if there isn't a difference no point in wasting a second board on the system. I'm a little shocked by the improvement now that I've been listening to SE for a while. The soundstage is almost the same as my laptop output, though other SQ gains match BAL much more closely with a quick listen. The BAL improvement in soundstage is not something you could forget even long after an AB test...
Could it be because I connected SE to the negative buffered outputs of the inverted (left) channels to keep SE and BAL independent? Could the inverted output be inferior somehow? I don't believe it. And my experiences with SE before I moved the SE connection to the inverted channels agree with what I'm hearing now, which is not bad news in my opinion. I've never listened to the BAL on a single board, but if someone claimed that HD650 balanced has a characteristically large soundstage, I would believe him and just attribute the improvement to that. Hope my excursion into audiophilia again satisfies you
Both tracks are dvorak op. 104 1st mvt. I was listening to the Fourier/BPO version 16/44, you can probably get it as a digital download at a very modest price if you don't have it already. At 6:59 the flutes enter and there's a crescendo from there to around 7:15 where you get some nice dynamics. The flutes sound a bit more natural and frankly just quite a bit deeper in the stage on BAL than SE. And as more instruments enter the expansion of soundstage on BAL just went further than SE.
Now, on the Rostropovich/BPO 24/192 version, flutes and cello sound a bit more muted in the same section, so I went straight to the cello solo at 10:40. On SE, the soundstage felt cut off right at my ears, but the BAL naturally extends much further. I went back and forth between the two outputs, and tried to imagine SE with a larger soundstage and BAL with a smaller one, which has worked well in the past to disillusion myself. Only this time there was no way I could convince myself that SE sounds nearly as open as BAL.
Used HD650, -22db with BAL and -16db with SE. They sounded about the same loudness, but I don't quite believe volume changes things in this case. Both are buffered outputs onboard. SE with 10R 2V output and BAL 20R 4V.
This nicely confirmed all of my past experiences with BAL vs SE outputs on my system. The conclusion isn't definite, however, as I would need a proper ABX test to claim that. It could be that my system is noisy and the balanced output helped a lot there? Can't hear any noise at max volume on a silent track though. I'm guessing you're deciding whether to put back your dual-mono... Just do an ABX and if there isn't a difference no point in wasting a second board on the system. I'm a little shocked by the improvement now that I've been listening to SE for a while. The soundstage is almost the same as my laptop output, though other SQ gains match BAL much more closely with a quick listen. The BAL improvement in soundstage is not something you could forget even long after an AB test...
Could it be because I connected SE to the negative buffered outputs of the inverted (left) channels to keep SE and BAL independent? Could the inverted output be inferior somehow? I don't believe it. And my experiences with SE before I moved the SE connection to the inverted channels agree with what I'm hearing now, which is not bad news in my opinion. I've never listened to the BAL on a single board, but if someone claimed that HD650 balanced has a characteristically large soundstage, I would believe him and just attribute the improvement to that. Hope my excursion into audiophilia again satisfies you
Could also just be very bad crosstalk. I used a 20cm triple core (shielded) wire for connecting SE output to the front panel. Tested a simulated crosstalk audio file online and it has a somewhat similar effect.
Tried outputting high levels in one channel. No sound in the other as far as I can tell. But the effect sounds kind of like a -20db crosstalk... anyways I might measure it this weekend.
Tried outputting high levels in one channel. No sound in the other as far as I can tell. But the effect sounds kind of like a -20db crosstalk... anyways I might measure it this weekend.
Last edited:
As for "reference" tracks - I'd like to know what everyone is listening to.
It's generally tricky. At a certain point you realize every recordig is flawed one way or the other, and it becomes a question of preference or even taste. I also have an idea that depending on what was used to record the signal WRT to the AD converter certain filters will work better in a DAC than others.
It's generally tricky. At a certain point you realize every recordig is flawed one way or the other, and it becomes a question of preference or even taste. I also have an idea that depending on what was used to record the signal WRT to the AD converter certain filters will work better in a DAC than others.
Sorry I didn’t realize they had more than one recording/CD release together...
Dvorak: Cello Concerto & Symphony No. 8 - DG: 4784233 - CD | Presto Classical
Could be the same recording just different releases
Dvorak: Cello Concerto - DG: 4794879 - download | Presto Classical
Just out of curiosity, I bought the Rostropovich version again since this is a new release of the PCM studio master (96Khz) and what I had was an earlier release (192Khz) converted from DSD. Did an ABX test in Fb2k and actually was able to get 8/10 right (got impatient on the last one otherwise probably 9/10). I had to cut off the first 2s of silence from the new release to match position, and quickly found that starting at 11:08 (after cut) there's a slight difference in cello higher frequency "harshness" (in the lower and weaker notes) which is slightly more noticeable in 192Khz version. The 96Khz actually has a slightly (0.5db) higher peak level and higher dynamic range but almost the same RMS level, so the extra "harshness" (which could be more real by the way idk) is not from volume mismatch. First ABX test that I passed
Don't waste your time on these things...
Don't waste your time on these things...
Last edited:
I'm glad you like your setup... Though to arrive at a useful conclusion I suggest you replace the 0.01R in the low-res mod with the recommended 0.1R. Paul reported SQ problems with his low-res mod. This would make a better AB test, assuming you did one.
Not that I would switch to 1121 anyway.
Also, are you suggesting your dual-mono sounds worse than SE? Or is the test just between 1021 and 1121?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.