To be honest I myself like digital electronics a bit more so I also started to draw a schematics similar to yours and wanted to test it in Multisim... but you were fasterYes, exactly, I was too lazy to do more math
But again, it would be interesting to see if there could be any improvement/degrading in sound reproduction using tube output stage and some Hi-Fi headphones. As with speakers it would be quite absurd to notice anything
For me it was just a good exercise before making I2S interface to my transport.
You might be right
I found this nice article about: sound perception and later today I will get a free Matlab and try to play with some wavs. I will post my findings then.
Hell it may help headphones have more spatial queues, you never know.
It would be a nice experiment if someone could post two wav files of the same song. The first original and the second with one channel delayed of the dT that the DAC introduce.
I have the AD1865 DAC, so it would be great if the modified file could compensate for the delay. I would be really interested to know your impression on the perception of this delay.
I don't think that the comment on the position of the listener is relevant: I think the problem is that the waves from left a right channel interacts while traveling to the listener, and the phase difference in theory could create strange artifacts.
Now if this effect is audible compared to the room acoustic, this I have no idea.
Best Regards,
Davide
I have the AD1865 DAC, so it would be great if the modified file could compensate for the delay. I would be really interested to know your impression on the perception of this delay.
I don't think that the comment on the position of the listener is relevant: I think the problem is that the waves from left a right channel interacts while traveling to the listener, and the phase difference in theory could create strange artifacts.
Now if this effect is audible compared to the room acoustic, this I have no idea.
Best Regards,
Davide
In my opinion we tend to forget about the most distorting and slow device from the audio chain: the loudspeakers.
I don't have time to google now (at work...) but I would be really curious how much delay can different speakers introduce.
It could be that 11us are well within the delay tolerance of different speakers, be it hi-fi or hi-end.
I don't have time to google now (at work...) but I would be really curious how much delay can different speakers introduce.
It could be that 11us are well within the delay tolerance of different speakers, be it hi-fi or hi-end.
I was lazy to do the delay the hard way (with Matlab) so I found this Stereotool plugin for winamp. I set its gain to 0dB and disabled all its features then used the _Stereo image_ dialog, turned it On and played with the "Channel Delay fake stereo" scroller (everything else on 0 or unchecked), then checked/unchecked the On checkbox without knowing its state. I used some Behringer studio headphones, so far to be high-end.
I played several of my wav/flac files which have one or two instruments/voice in the middle: e.g.: Dave Brubeck - Take Five or A.L.Webber - The Phantom of the opera and found that I could hear a delay of 100uS, although only with low complexity pieces. Unfortunately the software is quite dumb, so the next delay time was 26uS which I could not distinguish, although I tried hard with several pieces. Would be interesting to test with loudspeakers or better headphones.
Zsolt
I get today my AD1865 DAC first time powered on, with delayed right channel - no evidence find that delay is audible. 11 uS is about 90 deg phase shift on 20 kHz, about 4 deg phase shift on 1 kHz. As in one old thread http://www.diyaudio.com/forums/digi...ng-2xpcm1702-feed-i2s-experiment-results.html was written about schematic with delay: "if it does not hurt my purist ear, at least it hurts my purist eye..."
I'm thinking how can this shifter be modified for AD1865 to interface 18bit I2S instead of 18bit LSBJ. We could interface 8414 (FMT 2) plus other I2S devices as well. I would also add logic to switch between 8414 or other I2S.Below I will attach schematic and timing diagram for proper 8414 and 1865 interfacing. This is just my approach, could be done in a few other ways. Couple notes on this one: U2A, U2B, U2D separates data, U2C inverts latch signal. U3B, U3C 2X delays clock, as data is also 2x (or 3x if in out phase) delayed, to keep the sync. With 4XHC164 + HC74 Left data is ~39ns out of sync according to multisim. Everything else is already discussed. Again, I personally would use ALS logic, just to be on the safe side
Zsolt
Will work on that one some time soon, if someone is keen to try and put out his/her ideas (in form of complete schematics), are also welcome to do this before me
The target would be:
Input: switchable in between 2 x true I2S 16 to 24 bit (not LSBJ or MSBJ)
Output: EIAJ 18bit (no delay)
There is group buy onhttp://www.diyaudio.com/forums/group-buys/166354-hiface-usb-i2s-24bit-192khz.html Hiface USB -> I2S 24bit/192Khz going on. Should be perfect source of I2S for 1865 (and any other DAC of course). Good price etc., so if someone else is interested in pure I2S > 1865 interface, hurry up, already half way through (100pcs.) as I'm typing
http://www.diyaudio.com/forums/group-buys/166354-hiface-usb-i2s-24bit-192khz.html
audiodesign,
no panic from my side, as I could not hear any diference (post #184), nor listening directly to 1865, followed by MK132 I/V and just a solid state single fet preamp. I would also like to know if someone can hear any difference, using some more Hi-End realization (tube output and their beloved headphones).
But if things get to the point of interfacing this DAC to I2S, there has to be conversion to EIAJ 18bit, so why not to do things right then ?
More updates on this issue: some how I figured out to look at another DAC outputs. This is output signal on R+ and L+ legs of PCM1793 with DIR9001, both working in native I2S 24bit. Square wave output generated with SpectraLab, 44.1k sampling. And look what I found (see attach.)
Looked again on every legs status (for formats and etc.), everything is in accordance with TI datasheets. It just seems to be the way this dac works, putting each channel at 1 sample (22.6us) separation.And thousand of not suspecting users around the globe are happy with their DVD, blu-ray etc. players
Off course there is slight probability that something is messed up on software side, so if someone could confirm/deny this, are welcome to do so.
no panic from my side, as I could not hear any diference (post #184), nor listening directly to 1865, followed by MK132 I/V and just a solid state single fet preamp. I would also like to know if someone can hear any difference, using some more Hi-End realization (tube output and their beloved headphones).
But if things get to the point of interfacing this DAC to I2S, there has to be conversion to EIAJ 18bit, so why not to do things right then ?
More updates on this issue: some how I figured out to look at another DAC outputs. This is output signal on R+ and L+ legs of PCM1793 with DIR9001, both working in native I2S 24bit. Square wave output generated with SpectraLab, 44.1k sampling. And look what I found (see attach.)
Looked again on every legs status (for formats and etc.), everything is in accordance with TI datasheets. It just seems to be the way this dac works, putting each channel at 1 sample (22.6us) separation.And thousand of not suspecting users around the globe are happy with their DVD, blu-ray etc. players
Off course there is slight probability that something is messed up on software side, so if someone could confirm/deny this, are welcome to do so.
Change your OpAmp
I didn't realize people are using the ancient OpAmp that AN kit put into this board. I first upgraded to the AD744 but the giant leap in SQ and lower noise floor came with the AD825. Astonishing. Its a SMD so you'll need adapter boards from Brown Dog. Its worth the soldering.
Here's the info: http://www.analog.com/static/imported-files/Data_Sheets/AD825.pdf
I didn't realize people are using the ancient OpAmp that AN kit put into this board. I first upgraded to the AD744 but the giant leap in SQ and lower noise floor came with the AD825. Astonishing. Its a SMD so you'll need adapter boards from Brown Dog. Its worth the soldering.
Here's the info: http://www.analog.com/static/imported-files/Data_Sheets/AD825.pdf
.And thousand of not suspecting users around the globe are happy with their DVD, blu-ray etc. players
.
Yea and thousands I mean millions of consumers are happy with the sound of their 4" Bose speakers being fed by these players. Doesn't really prove anything IMO.
Nothing to prove there really, I was just sarcastic on this one...
Bottom line would be that it seems like some big players in the market as BB, are omitting that kinda imperfections as inconsequential. And that is understandable, because targeting market is really not hi-end.
Nothing to prove there really, I was just sarcastic on this one...
Bottom line would be that it seems like some big players in the market as BB, are omitting that kinda imperfections as inconsequential. And that is understandable, because targeting market is really not hi-end.
Yea I'm on the fence on this one as to audibility, but do know that the safest route is always best if not too costly before you have boards made.
If there isn't a simple inexpensive solution to take this issue out of this slight design vulnaribility I have Izotope Ozone and could run some subjective listeing tests with my highend headphone rig .
Would be nice if you could do this
Different ears - different perceptions (even if subjective ones).
My setups were as follows:
Stainberg WaveLab with ESI ASIO > EsiJuli@ analog output > Fet buffer (just a small power follower with 300ma Ids) > HD 650
Stainberg WaveLab with ESI ASIO > EsiJuli@ SPDIF > CS8414 > AD1865 with passive I/V and Zen pre > Fet buffer > HD 650
Was cutting (delaying) any channel in first setup by 10 to 100 microseconds. From ~ 100 us MAYBE something started to change (very subjective). And with 1865 setup, cutting right channel for some 10us gave absolutely nothing.
Would be nice if you could do this
Different ears - different perceptions (even if subjective ones).
My setups were as follows:
Stainberg WaveLab with ESI ASIO > EsiJuli@ analog output > Fet buffer (just a small power follower with 300ma Ids) > HD 650
Stainberg WaveLab with ESI ASIO > EsiJuli@ SPDIF > CS8414 > AD1865 with passive I/V and Zen pre > Fet buffer > HD 650
Was cutting (delaying) any channel in first setup by 10 to 100 microseconds. From ~ 100 us MAYBE something started to change (very subjective). And with 1865 setup, cutting right channel for some 10us gave absolutely nothing.
What material? I'll see if I have the same CD.
2 tracks:
Old and trusty
Gnomus - Vladimir Fedoseev / Moscow Radio Symphony Orchestra
Ripped WAV from this CD.
And just a track that I'm "hooked up" for now. Very nice vocal btw., one of those that makes your hair turn gray
Etro Anime - First Heard Your Name
Ripped WAV from this CD.
Could be anything that you know by heart, if you don't have these.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.