I just did a search on "opus111" and "S-D"...
I should have my NOS DAC in the system by next week.
---- We all wanna know all 'bout it! :b
Digital that sounds like analog
- Thread starter opus111
- Start date
---- We all wanna know all 'bout it! :b
Which NOS DAC? Very interested to hear your opinion. In my own limited experience and reading comments, NOS has a purity of tonality that, for those who prize it, can be magic...the trouble seems to be the noise and instrument separation when the implmentation is not so good. The sound becomes a little foggy and dynamic snap. I have found SD that tempted me (Stahl Tek), but none that have made me switch from trusty old NOS DAC.
Sure, that one has an even higher number (132dB) than the 1794. I just gave a couple of examples off the top of my head which beat the PCM1704 on SNR, THD.
As with Lloyd, looking forward to hearing which one you bought and your listening comparisons with your current fave DAC
After my very positive experience with the Vivaldi disc, I decided to invest in this boxed set : http://www.amazon.com/Vivaldi-Concertos-Sonatas-Opp-Complete/dp/B004AB2F8A/
19 CDs, almost 19hrs of Vivaldi I reckon this is a bargain. Only so far listened to the first two discs, but I'm lovin' it Highly recommended.
19 CDs, almost 19hrs of Vivaldi I reckon this is a bargain. Only so far listened to the first two discs, but I'm lovin' it
Highly recommended.
Bruce has mentioned the next two DACs he's getting are Phasure and Lavry. Phasure isn't NOS, its multibit but quite the opposite in that its 16X OS. I reckon this is very bad news as it means the PCM1704s its using are 16 times more glitchy than with NOS. Bruce hasn't said which Lavry yet...
Bruce B
WBF Founding Member, Pro Audio Production Member
The Phasure NOS1 384 is a non-oversampling DAC.
•Frequency response 1 Hz to 192000 KHz (384KHz sample rate). Flat within 0.19dBr in the audio band (DC not sustainable).
•Input of 16 or 32 bits for all Sample Rates (24 bit input is converted to 32 bits).
•Input Sample Rates (Hz) : 22050*, 44100, 48000, 88200, 96000, 176400, 192000, 352800*, 384000*, all auto-select without glitches.
* Under control of XXHighEnd only.
•Output Sample Rates always equal the Input Sample Rates, but notice that in software (PC, XXHighEnd) each native “file” Input Rate can be upsampled to each accepted higher Output Rate (like 96000 to 384000 or 44100 to 176400 etc.) and next is output to the DAC. Output bit depth is 24 - or 16 bits when Input bit depth is 16 bits.
•100% Filterless Non Oversampling, using 8 x PCM1704U-K Burr Brown (Texas Instruments) multibit 24 bit D/A converter chips. Filtering to be applied in software (officially provided by XXHighEnd through Arc Prediction which supports all modes), or not using filtering at all, at own choice.
•Output level for Single Ended Mode (RCA) is 1.5VRMS (-3dBFS relative to normal Full Scale). Output level for Differential Mode (XLR) is 2.7VRMS (+3dBFS relative to normal Full Scale).
•Output Impedance is 33 Ohm. It is the explicit advise to use the NOS1 without pre-amp and without analogue volume control for optimum results (long interlinks can be driven).
•The Clock section consists of replaceable oscillators, separately for the 44100 Hz and 48000 Hz base, both separately fed by shunt regulators. The provided oscillators carry an overall jitter of better than 0.5ps and phase noise of better than 120dBc/Hz at 100Hz and better than 108dBc/Hz at 50Hz.
The Lavry is the DA924
Last edited:
Ah that's my misunderstanding - it can operate up to 8X OS (384kHz) but works in NOS too. Thanks for the clarification Bruce. When I was reading about this DAC over on CA a couple of years ago it was working at 768kHz.
What's missing there is how, if any, the droop correction is applied. Also if there's any filtering of the imaging frequencies, its silent on this point. Without addressing those two issues its going to be sub-optimal.
As regards the DA924, Romy the Cat has some interesting modding here : http://www.goodsoundclub.com/Forums/ShowPost.aspx?PageIndex=1&postID=7038#7038
Does a conventional DAC architecture (not DS) that is oversampled still suffer from noise modulation? I thought that was more a DS issue? Is it in the filter then instead of the modulator loop?
I reckon it does due to glitching. R2R DACs being the worst at glitch (PCM1704 is R2R).
If you think about the colinear architecture adopted by BB, they've moved the major carry glitch from the zero crossing to the 25% and 75% points. So as the signal amplitude reaches -6dB there's suddenly two significant glitches being added in. This is just the major glitch, R2R DACs have gradually decreasing glitch as the 'carries' propagate down towards the LSB.
Another reason why high oversampling ratios have higher noise modulation is settling time. The only trajectory that's allowed that doesn't create noise modulation is pure exponential settling but its highly unlikely any active circuit manages to follow a purely exponential settling tail. The settling time is going to increase with step size, meaning that higher frequencies generate more noise than lower.
If you think about the colinear architecture adopted by BB, they've moved the major carry glitch from the zero crossing to the 25% and 75% points. So as the signal amplitude reaches -6dB there's suddenly two significant glitches being added in. This is just the major glitch, R2R DACs have gradually decreasing glitch as the 'carries' propagate down towards the LSB.
Another reason why high oversampling ratios have higher noise modulation is settling time. The only trajectory that's allowed that doesn't create noise modulation is pure exponential settling but its highly unlikely any active circuit manages to follow a purely exponential settling tail. The settling time is going to increase with step size, meaning that higher frequencies generate more noise than lower.
Agree it is noticable in some of the DS DACs.
Is it possible that other noise-harmonics-etc introduced in other ways are interpreted by some or tools as noise modulation for some conventional DACs?
Cheers
Orb
Here is some thoughts on NOS by peter himself, http://www.computeraudiophile.com/f6-dac-digital-analog-conversion/should-i-nos-dac-14869/. Just focus on peter's posts, the rest is lot of nonsense. His writing is a bit difficult to understand at first (english not his 1st language) but you'll get the hang of it soon. He has some very interesting ideas (Opus should like some of it).
Here is another new development in NOS Dac, DDDAC 1794 NOS by well known german DIY. Whaddaya know, a S-D NOS dac, an oxymoron no? Although I am not sure what advantage can be had with the DDDAC NOS, just explore with an open mind.
The common thing with all the people doing NOS dac is that the digital filters is the main problem, it seems. Removing the filter function results in better SQ hearing wise.
Going back to Peter, he talks about frequency and time domain and feels that not a lot of thoughts/effort has gone into the timing side of things, which ties in with Opus's thing with CM distortion. Lately I have also find that phase and polarity does affect SQ to a large degree but it is a subject that most people dismiss or ignore.
The common thing with all the people doing NOS dac is that the digital filters is the main problem, it seems. Removing the filter function results in better SQ hearing wise.
Going back to Peter, he talks about frequency and time domain and feels that not a lot of thoughts/effort has gone into the timing side of things, which ties in with Opus's thing with CM distortion. Lately I have also find that phase and polarity does affect SQ to a large degree but it is a subject that most people dismiss or ignore.
Glitches, have to cogitate, but that is a good point opus111 and my gut says yes.
There are numerous DS designs that do not oversample the individual modulators (clock rate = 2xNyquist) but use parallel modulators and math or other schemes to get there. All the ones I have seen ultimately do provide the same benefits, and drawbacks, of more conventional DS designs. Any modern DS text describes a few. The idea is popular in beam forming systems for applications like radar, lidar, and sonar (including ultrasonic imaging) as you can implement delays on a single-bit data stream or use an analog mixer with +/-1 input (reduces nonlinearity). I won a contract to research a DS design based upon Hadamard sequences; there were enough practical issues with implementing it that it was not built. Theory said yes, engineering said no, at least for now.
There are numerous DS designs that do not oversample the individual modulators (clock rate = 2xNyquist) but use parallel modulators and math or other schemes to get there. All the ones I have seen ultimately do provide the same benefits, and drawbacks, of more conventional DS designs. Any modern DS text describes a few. The idea is popular in beam forming systems for applications like radar, lidar, and sonar (including ultrasonic imaging) as you can implement delays on a single-bit data stream or use an analog mixer with +/-1 input (reduces nonlinearity). I won a contract to research a DS design based upon Hadamard sequences; there were enough practical issues with implementing it that it was not built. Theory said yes, engineering said no, at least for now.
Bruce, the specs you posted earlier is for the discontinued 1st generation Phasure NOS1. The latest one Phasure NOS 24/768 will take signal up to 32/768 but I think output only 24/768. While it will accept 16/44.1 up to 32/768, this dac is meant to work best at 24/768 with processing done by the computers. Cheers.
Bruce B
WBF Founding Member, Pro Audio Production Member
Seems like an oxymoron to me too
S-D always uses oversampling as the modulator simply cannot run at the input frequency when it has so many fewer bits to play with.<edit> I've digested now what Don's saying about parallel modulators, so I'm inclined to modify with the caveat that 'all the audio S-D modulators I've come across...'
That does indeed seem to be the populist position about the SQ advantages of NOS. I played around with digital filters and found that even with a minimum phase filter (hand crafted on an ARM) the SQ reduction was there at 2X OS. So I decided it wasn't the absence of the digital filter, rather the increased noise modulation (from more glitches, a higher proportion of the time spent settling) caused by running faster.
I have tried in the past, on more than one occasion to understand Peter's writings on CA. Each time I've given up after a fairly short battle - discretion being the better part of valour
I have tried in the past, on more than one occasion to understand Peter's writings on CA. Each time I've given up after a fairly short battle - discretion being the better part of valour
Ha Ha I know. Anyways, if you are up to it, he seems sufficiently succinct on that particular thread.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |