mansr Posted February 9, 2019 Share Posted February 9, 2019 2 minutes ago, Superdad said: Probably lots of places since it is widely known that phase-noise/jitter at very low offset frequencies is the most important aspect of clocking performance with regards to audio. I think you mean widely speculated. Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 14 minutes ago, Superdad said: snarky this describes your comment - what is the problem? Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 12 minutes ago, mansr said: I think you mean widely speculated. ok, care to expand on this? Link to comment
Popular Post Superdad Posted February 9, 2019 Popular Post Share Posted February 9, 2019 3 minutes ago, Ralf11 said: this describes your comment - what is the problem? Truth is Ralf, I never know what your point is since 90% of your posts are single-line fragments which are often neither questions or statements. I guess you are just trying to be provocative. mav52, sandyk, 4est and 1 other 4 UpTone Audio LLC Link to comment
jabbr Posted February 9, 2019 Share Posted February 9, 2019 15 minutes ago, Superdad said: Probably lots of places since it is widely known that phase-noise/jitter at very low offset frequencies is the most important aspect of clocking performance with regards to audio. The 100Hz, 10Hz, 1Hz (and even below though rarely published) figures are what audio engineers pay attention to when considering oscillators. Or maybe your question was just rhetorical and snarky. I did not see a question mark... Close-in phase noise has been widely discussed here. The correlation with SQ has been reported but not studied with great precision. That said it’s easy to get a phase noise floor (far out) <120 dBc/Hz so at some point this component is part of the background noise. Close-in phase noise much higher (logarithmically so) in amplitude and hopefully the only component >100, so if clocks are important at all, the $$$ are in close-in phase. Ive posted a ton on this already. I suspect @Ralf11 is referring to that? 0.1, 0.01 offsets can be measured but not by all equipment. Superdad 1 Custom room treatments for headphone users. Link to comment
asdf1000 Posted February 9, 2019 Share Posted February 9, 2019 6 minutes ago, Superdad said: Truth is Ralf, I never know what your point is since 90% of your posts are single-line fragments which are often neither questions or statements. I thought I was the only one 😁 sandyk 1 Link to comment
jabbr Posted February 9, 2019 Share Posted February 9, 2019 1 hour ago, JohnSwenson said: The best ones do have a limit, as the close in phase noise of the reference goes down the phase noise of synthesizer output goes down, to a point. Then lower phase noise of the reference doesn't improve the output. This limit is really good, better than what is in almost all DACs. Thus with this synthesizer and a REALLY REALLY good external clock you probably CAN get an improvement over what probably comes with the DAC. Are you saying that a $3500 external clock probably can get an improvement over the $10 clock in a DAC eg Crystek? That may be — assuming that price point DAC also uses the lowest phase error chips etc ... don’t expect those nice & expensive AD flops there @PeterSt has looked at this IIRC and is probably in a great position to describe the SQ of various clocks holding the rest of the circuitry constant. The way I see it, it’s really hard to be sure unless the circuits are known, as well as measurements ie phase error (or Allen) at the latch with different clocks. Custom room treatments for headphone users. Link to comment
Popular Post mansr Posted February 9, 2019 Popular Post Share Posted February 9, 2019 6 minutes ago, Ralf11 said: ok, care to expand on this? I haven't seen anything resembling a serious study of audibility of clock jitter at various offsets and levels. Really bad jitter is without question audible. I posted some samples here myself a while back. Where I have doubts is regarding phase noise of, say, -100 dB at 1 Hz or whatever. No musical instrument is remotely that stable in the first place. jabbr and Arpiben 2 Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 12 minutes ago, Superdad said: Truth is Ralf, I never know what your point is since 90% of your posts are single-line fragments which are often neither questions or statements. I guess you are just trying to be provocative. try reading in context - look at the post immediately above mine one thing is certain: I am not trying to sell gear Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 1 minute ago, mansr said: I haven't seen anything resembling a serious study of audibility of clock jitter at various offsets and levels. Really bad jitter is without question audible. I posted some samples here myself a while back. Where I have doubts is regarding phase noise of, say, -100 dB at 1 Hz or whatever. No musical instrument is remotely that stable in the first place. I am interested in hearing more from you & Jabbr on this; let's set some boundary conditions to make it easier. e.g. xx ps of jitter; modulated by th esignal or not; etc. ?? Link to comment
Popular Post jabbr Posted February 9, 2019 Popular Post Share Posted February 9, 2019 25 minutes ago, Ralf11 said: I am interested in hearing more from you & Jabbr on this; let's set some boundary conditions to make it easier. e.g. xx ps of jitter; modulated by th esignal or not; etc. ?? I’ve said many times: No question there is a limit in audibility. Don’t know the limit. Is anything <-100 dBc/Hz @ 1 Hz audible? ... but also you can’t merely pluck in a clock spec and think you’ve achieved that at your DAC latch ... Miska and Superdad 1 1 Custom room treatments for headphone users. Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 Yes, and you also said many, many times that close-in phase noise is the/an issue - hence I was a little surprised to see mansr's comment re studies. There is also the question of how much jitter can be heard. I recently saw an audiophile book saying that 10 ps could be heard... Link to comment
Popular Post Miska Posted February 9, 2019 Popular Post Share Posted February 9, 2019 Based on what I've seen on my measurements, I would put more emphasis of at least cleaning up the data and USB communication related interferences out before worrying too much about phase noise. Putting in low phase noise clock component is much simpler than keeping it clean. Superdad, jabbr and Arpiben 3 Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
mansr Posted February 9, 2019 Share Posted February 9, 2019 2 minutes ago, Ralf11 said: There is also the question of how much jitter can be heard. I recently saw an audiophile book saying that 10 ps could be heard... 10 ps with what distribution? barrows 1 Link to comment
Ralf11 Posted February 9, 2019 Share Posted February 9, 2019 It was an "audiophile book"... so of course... ...they did not say. Link to comment
PeterSt Posted February 10, 2019 Share Posted February 10, 2019 6 hours ago, JohnSwenson said: Thus with this synthesizer and a REALLY REALLY good external clock you probably CAN get an improvement over what probably comes with the DAC. I carefully (haha) picked one sentence out of that post, but I could almost have picked any, at random. Without mentioning a concrete example, it tells all nothing. I could also say "I agree with all and thus don't go for any of it". If there is recent improvement on synthesized clocks, jitter cleaners or anything in that realm, it must be very recent and I missed it (which is possible). But I think I can already see the gist: if such a synthesized environment requires lower power supply noise than we can realistically produce, it is all moot. Lush^3-e Lush^2 Blaxius^2.5 Ethernet^3 HDMI^2 XLR^2 XXHighEnd (developer) Phasure NOS1 24/768 Async USB DAC (manufacturer) Phasure Mach III Audio PC with Linear PSU (manufacturer) Orelino & Orelo MKII Speakers (designer/supplier) Link to comment
PeterSt Posted February 10, 2019 Share Posted February 10, 2019 5 hours ago, jabbr said: @PeterSt has looked at this IIRC and is probably in a great position to describe the SQ of various clocks holding the rest of the circuitry constant. Or maybe not. You can't just exchange a (VC)XO with an OCXO and hope that the 1A draw for the latter doesn't influence anywhere, knowing that the whole DAC without oven control maybe uses 400mA otherwise. Just to name something in the rough department. And then not to talk about the invariant current draw this implies, which we would not want. OK, make a shunt supply for it, but ... etc. etc. etc. I am not good at describing the SQ of clock systems. All I can tell is the lower the jitter the better the sound in various aspects and that even going sub fs is still audible but for unexpected reason: acoustical jitter. Ha ! I invented the term on the spot. But if the clock system exhibits so low jitter, it comes under the influences of the acoustical noise in the room, hence the music you produce (or it produces itself, implying oscillation). So there will be a limit, unless that acoustical influence can be within our control (put DAC in basement and such). I recall me tapping on the floor (OK, firmly), that already showing jitter in the ps range (the Crystek being the oscillator). I even made screenshots of that. Lush^3-e Lush^2 Blaxius^2.5 Ethernet^3 HDMI^2 XLR^2 XXHighEnd (developer) Phasure NOS1 24/768 Async USB DAC (manufacturer) Phasure Mach III Audio PC with Linear PSU (manufacturer) Orelino & Orelo MKII Speakers (designer/supplier) Link to comment
jabbr Posted February 10, 2019 Share Posted February 10, 2019 40 minutes ago, PeterSt said: I invented the term on the spot. But if the clock system exhibits so low jitter, it comes under the influences of the acoustical noise in the room, hence the music you produce (or it produces itself, implying oscillation). We have discussed the 1/f pattern of ground vibrations along with the 1/f pattern of close-in phase noise so the idea that crystals are vibration sensitive (as are ceramics etc) has been discussed. I don’t know how much this is an effect ie in dBc/Hz. Close in noise, at least in the Netherlands, with a shout out to Rutgers, does seem to have SQ effects! [ None of this changes the fact that deterministic jitter is typically at least 10x higher than clock jitter —at least- and crosstalk, layout, impedance etc etc etc remain far far more important than the clock itself but there’s no reason not to use a Crystek or similar at that price point. ] Arpiben 1 Custom room treatments for headphone users. Link to comment
PeterSt Posted February 10, 2019 Share Posted February 10, 2019 3 hours ago, jabbr said: [ None of this changes the fact that deterministic jitter is typically at least 10x higher than clock jitter —at least- and crosstalk, layout, impedance etc etc etc remain far far more important than the clock itself but there’s no reason not to use a Crystek or similar at that price point. ] Well, regarding the fact that it is exactly deterministic jitter I could get rid of by means of galvanic isolation (measured at the "nothing changed at the output" behind the clock system), sure. Easy to see by the jitter patterns and the music (level) played. And so indeed, when not first a few things are arranged for upstream, the oscillator itself will matter only when the data jitter etc. is low enough to make it matter. And I believe I got that far with the Phasure NOS1a. The G3 version on top of that could matter again, but then for environmental better power supply. This is to be read as: that again improved a lot which may make it possible to otherwise smeared jitter implications in the analogue stages, to remain (become) audible. Happy now ? IOW, it gets more and more difficult to see through what's involved next when one step has been solved. These iterations are really ad infinitum. Lush^3-e Lush^2 Blaxius^2.5 Ethernet^3 HDMI^2 XLR^2 XXHighEnd (developer) Phasure NOS1 24/768 Async USB DAC (manufacturer) Phasure Mach III Audio PC with Linear PSU (manufacturer) Orelino & Orelo MKII Speakers (designer/supplier) Link to comment
Summit Posted February 10, 2019 Share Posted February 10, 2019 It’s funny that the same persons that constantly reject all kinds of differences between other digital cables seems to think that a very short quality clock BNC cable will have a devastating impact on SQ 😉. Link to comment
Summit Posted February 10, 2019 Share Posted February 10, 2019 12 hours ago, Ralf11 said: try reading in context - look at the post immediately above mine one thing is certain: I am not trying to sell gear Yes, unfortunately that’s not the case for everyone here. Link to comment
Arpiben Posted February 10, 2019 Share Posted February 10, 2019 Those a bit lost with Jitter may find some clues and definitions in this Agilent paper: https://www.keysight.com/upload/cmc_upload/All/Clock_Jitter_Analysis_2008.pdf Those familiar with those aspects in other domains (Telecom ) are still struggling to find any DAC Output measurement showing the eventual benefits. Sorry my ears aren't good enough.😉 Link to comment
mansr Posted February 10, 2019 Share Posted February 10, 2019 1 hour ago, Summit said: It’s funny that the same persons that constantly reject all kinds of differences between other digital cables seems to think that a very short quality clock BNC cable will have a devastating impact on SQ 😉. That's not what anyone is saying. What people are saying is that if you're going to use an external clock, the cabling had better be done right, or any improvement over a decent built-in clock will be lost. 4est 1 Link to comment
Popular Post jabbr Posted February 10, 2019 Popular Post Share Posted February 10, 2019 3 hours ago, Summit said: It’s funny that the same persons that constantly reject all kinds of differences between other digital cables seems to think that a very short quality clock BNC cable will have a devastating impact on SQ 😉. Seems ironic Here is a relatively readable article written by a Dutch clock oscillator expert who is at least as crazy as @PeterSt 😂😂😂 https://www.by-rutgers.nl/PDFiles/Audio Jitter.pdf But seriously he was the first person who explained to me the importance of close-in phase noise in clocks, and I admit that my intellectual interest in this easily outweighs my personal quest for improved SQ. (e.g. the whole 1/f thing) Rutgers, not only discusses clocks, but has pioneered clock oscillator circuits and provides schematics and methods for phase noise measurement devices suitable to be built at home or DIY if one wishes. (look through the site!) At this point we are arguing whether an external clock with a -135 dBc/Hz @ 10 Hz is better than a -100 dBc/Hz @ 10 Hz (offset) and at such low noise levels, small differences e.g. cables and internal circuitry can easily outweigh benefits of small improvements in phase noise at the clock itself (and I'm picking the $3500 external clock as a reference here). The comparison is more RCA vs XLR i.e. is balanced quieter than single ended. Balanced often has advantages but not always. Again you need to know the circuit to predict, and in a similar vein, the internal electronics have to be perfected. So more appropriate to compare a $3500 external clock with a $20,000 DAC, and I suspect that you find that at these price levels, the internal clock in the DAC will equal the external, and all else being equal, the internal clock beats external every time. Superdad, 4est and Ralf11 1 1 1 Custom room treatments for headphone users. Link to comment
jabbr Posted February 10, 2019 Share Posted February 10, 2019 21 hours ago, zephyr24069 said: We can agree to disagree on sonic benefits of an external clock but my standing as a hobbyist-only is absolutely something you can count on and anyone who knows me on the various forums can attest to. I can't easily tell from your profile which stand alone DAC you are using (because this matters). Esoteric seems to be designed for external clocking and it is very reasonable that equipment explicitly designed for external clocking would perform better. For example RME ADI-2 FS does not provide for an external clock. MSB provides internal upgrade: http://www.msbtech.com/support/GalaxyClock_Lit_Press3.pdf ... I like the provision of specs including 0.1 Hz offset -- note that these specs are now achievable at a lower price point given the increased affordability of great clocks -- they also show the base of peak widening effect of close in phase noise. MSB has enough confidence in their onboard clock that they provide a clock output for sync but not input. Again, it all depends on your system Superdad 1 Custom room treatments for headphone users. Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now