Some explanation to start with. I tried several PC based systems for music playback before, but most (to be honest: all) of them were easily beaten by my not-so-high-end, but not too bad CD player with its internal -Wolfson -DAC. (Onkyo C-7070)
The list includes M-Audio Delta 24/96, Onkyo SE 90 PCI, Auzentech cards and a few others.
Then came ESI Juli@, with its compulsory modifications (op-amps to LM4562, all filter/coupling caps replaced with OS-Cons and Silmic II-s, and the whole thing run on 3 separate analog power supplies. But overall I am using the whole card , not only the "bottom" part of it, as many others do with various additions.
Now the sound is waaay better than the CD, in all aspects: dynamics, resolution, spatial presentation, shortly, much more musical and listenable.
Is it possible that a 15 years old, relatively simple soundcard beats a 6-700 dollar CD player that easy, or really I am fooling myself?
Is it possible, that mechanism induced jitter ruins the sound this much?
Sorry if my questions seems naive, but I was really surprised by the difference. Anyone with similar experience?
The list includes M-Audio Delta 24/96, Onkyo SE 90 PCI, Auzentech cards and a few others.
Then came ESI Juli@, with its compulsory modifications (op-amps to LM4562, all filter/coupling caps replaced with OS-Cons and Silmic II-s, and the whole thing run on 3 separate analog power supplies. But overall I am using the whole card , not only the "bottom" part of it, as many others do with various additions.
Now the sound is waaay better than the CD, in all aspects: dynamics, resolution, spatial presentation, shortly, much more musical and listenable.
Is it possible that a 15 years old, relatively simple soundcard beats a 6-700 dollar CD player that easy, or really I am fooling myself?
Is it possible, that mechanism induced jitter ruins the sound this much?
Sorry if my questions seems naive, but I was really surprised by the difference. Anyone with similar experience?
Last edited:
First off, Juli@ is not 15 years old, but rather more like 12. It wasn't "relatively simple" either, but rather a fairly tricked-out semi-pro card that cost as much as an entire entry-level CD player, with converters only one step down from TOTL, in a time when DACs were pretty much a solved problem already. They were known to be capable of excellent sound, even more so once modified. I still have one floating around somewhere, too bad the Windows drivers really suck (*cough* power management *cough*).
I take it your Juli@ checks out fine measurement wise (e.g. RMAA loopback test)?
If so, I suggest you put in some effort to do accurate level matching between the two, preferably down to 0.1 dB (<0.3 dB in any case, which is 3%-ish). This probably is easiest to do in the computer, just keep >=24 bits of resolution. (You could e.g. recruit Foobar2000's ReplayGain settings or digital volume.) Even a relatively simple multimeter should be adequate as long as it's got a mVAC range and you stick to 50-60 Hz (got a CD with test tones floating around? burn one if need be). Depending on how much deviation between left and right channels you find, individual attenuation values may be needed, which complicates things.
From what you write I very much suspect that Juli@ is the louder of the two, and the CD player in turn is louder than your other cards. Should the CD player actually turn out to be louder, it would have to be fed with properly-processed level-adapted material. Gets a bit more complicated, still doable though.
Once you've got the levels measurably sorted out, try to conduct as close an A/B comparison as possible (with synchronous playback). E.g. have both connected to inputs of the same integrated amp or whatever. Ideally, get someone else to do it so you can do a blind test - but if that's too much hassle for now, do it sighted. I bet the results will turn out to be much, much closer, up to the point where you'll have a hard time telling which is which. That's to be expected anyway.
I take it your Juli@ checks out fine measurement wise (e.g. RMAA loopback test)?
If so, I suggest you put in some effort to do accurate level matching between the two, preferably down to 0.1 dB (<0.3 dB in any case, which is 3%-ish). This probably is easiest to do in the computer, just keep >=24 bits of resolution. (You could e.g. recruit Foobar2000's ReplayGain settings or digital volume.) Even a relatively simple multimeter should be adequate as long as it's got a mVAC range and you stick to 50-60 Hz (got a CD with test tones floating around? burn one if need be). Depending on how much deviation between left and right channels you find, individual attenuation values may be needed, which complicates things.
From what you write I very much suspect that Juli@ is the louder of the two, and the CD player in turn is louder than your other cards. Should the CD player actually turn out to be louder, it would have to be fed with properly-processed level-adapted material. Gets a bit more complicated, still doable though.
Once you've got the levels measurably sorted out, try to conduct as close an A/B comparison as possible (with synchronous playback). E.g. have both connected to inputs of the same integrated amp or whatever. Ideally, get someone else to do it so you can do a blind test - but if that's too much hassle for now, do it sighted. I bet the results will turn out to be much, much closer, up to the point where you'll have a hard time telling which is which. That's to be expected anyway.
Well, the story is a bit more simple than in your description...
A good quality CD is converted to Wav with a Plextor drive -read speed set to minimum- using EAC with AccurateRip plugin, then this file is played with an ASIO-ed CisPlay (RAM player) software, through the Juli@ and the levels are set to be equal (with the CisPlayer), measured with a 1kHz 0dB test record and a two channel scope at the outputs.
Still the ripped CD sounds much better through the PC system, and this is not the fooling myself category. I have a pretty well trained ear -and fortunately listen to live music quite often, so it is beyond the treshold of maybe, or maybe not...
What is the most amusing to me, is that the PC system sounds quieter, meaning the silence and the background beyond the musical signals is totally "black". (checked the CD output with scope, with a -60dB 1kHz test record), nothing else is there only the faint sine wave)
Furthermore -even more surprising- is the layering of the depth plane is again much better this way, sometimes the recording rooms, halls seem to be larger, with more air around the performers.
Well, whatever is the reason, I am very happy with the result, and started to rip my entire CD collection, so finally I can get rid of the cca. 1500 pieces of them occupiyng a full cupboard... If I hurry with it, will be done by next spring 😀
A good quality CD is converted to Wav with a Plextor drive -read speed set to minimum- using EAC with AccurateRip plugin, then this file is played with an ASIO-ed CisPlay (RAM player) software, through the Juli@ and the levels are set to be equal (with the CisPlayer), measured with a 1kHz 0dB test record and a two channel scope at the outputs.
Still the ripped CD sounds much better through the PC system, and this is not the fooling myself category. I have a pretty well trained ear -and fortunately listen to live music quite often, so it is beyond the treshold of maybe, or maybe not...
What is the most amusing to me, is that the PC system sounds quieter, meaning the silence and the background beyond the musical signals is totally "black". (checked the CD output with scope, with a -60dB 1kHz test record), nothing else is there only the faint sine wave)
Furthermore -even more surprising- is the layering of the depth plane is again much better this way, sometimes the recording rooms, halls seem to be larger, with more air around the performers.
Well, whatever is the reason, I am very happy with the result, and started to rip my entire CD collection, so finally I can get rid of the cca. 1500 pieces of them occupiyng a full cupboard... If I hurry with it, will be done by next spring 😀
Last edited:
I think the opamps used have the SSOP8 package. Where do you get them ?
I've blown them on my Juli@ card. Where can I get them ? I wrote to them about it and they said they would charge US$90 to ship two to me ! So I stopped using the card !One channel is OK. Still looking for that package. Got another small package from Singapore but it won't fit . This one is quite narrow.
Edit. Just checked. Mouser has it ! Been a long time since I looked for it. Will be ordering soon.🙂
I've blown them on my Juli@ card. Where can I get them ? I wrote to them about it and they said they would charge US$90 to ship two to me ! So I stopped using the card !One channel is OK. Still looking for that package. Got another small package from Singapore but it won't fit . This one is quite narrow.
Edit. Just checked. Mouser has it ! Been a long time since I looked for it. Will be ordering soon.🙂
Last edited:
I was sure someone will ask this🙂... what I did is the following.
Got the standard small (R-PDSO-G8) package ones, removed the original opamps from the card, and soldered VERY thin copper wires on the PCB (got a few white hairs in the process), reaching up from the board to about 4 mm.
Then soldered the larger packaged ones legs to the wires. Do it only if you are brave enough, have good eyes, and steady hands, and a suitably small soldering iron. I warned you.....
A/B testing is difficult to the brain... it confuses it and the result is jump into conclusions, the end is usually that both are bad OR equaly good😀
Best is to listen it for a while ( a month to 3 or more) and then change the system --> repeat.
IMO it is best approach....
Best is to listen it for a while ( a month to 3 or more) and then change the system --> repeat.
IMO it is best approach....
A decent stereo microscope makes such modifications quite feasible, using just tipped soldering iron. If one can see the magnified image, even a shaking hand gets surprisingly steady.
I got Olympus sz3060 | eBay and it gets used quite often, by me and the children for their needs/interest.
Well, not much more can be said than already written in my first post, but here:
Audio Asylum Thread Printer
you can read about all possible mods in great detail (this page was the base for my mods too....)
I've left my own project aside for now but I've gathered quite a bit of info on the juli@ in there, including links to more modding and info.
http://www.diyaudio.com/forums/digi...uli-s-analog-section-stand-alone-usb-dac.html
http://www.diyaudio.com/forums/digi...uli-s-analog-section-stand-alone-usb-dac.html
- Status
- Not open for further replies.