GoldenOne Posted May 21, 2021 Author Share Posted May 21, 2021 2 minutes ago, idiot_savant said: @GoldenOne do you still have the Ares 2 there? your friendly neighbourhood idiot Yep I do https://youtube.com/goldensound Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara Link to comment
numlog Posted May 21, 2021 Share Posted May 21, 2021 26 minutes ago, idiot_savant said: Now I'm guessing that it *is* genuinely NOS at 1.5MHz. Its possible, but also possible the internal rate exceeds 1.5Mhz. On the Soekris R2R dacs the final upsampling rate is 2.8Mhz, exceeding max possible input rate and making ''NOS'' impossible. Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 @PeterSt the stereophile measurements show the frequency response matches a linear interpolate exactly for each sample rate measured your friendly neighbourhood idiot Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 @GoldenOne and do you have HQPlayer? your friendly neighbourhood idiot Link to comment
GoldenOne Posted May 21, 2021 Author Share Posted May 21, 2021 1 minute ago, idiot_savant said: @GoldenOne and do you have HQPlayer? your friendly neighbourhood idiot I do yes https://youtube.com/goldensound Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 Excellent! I'm assuming you can do FFTs and generate test tones? Lets do an experiment: Let us generate a 44.1kHz 24 bit -60dBFS 1kHz sine wave In all cases, NOS mode on the DAC. 1)play that using HQPlayer no oversampling, grab FFT @ 768kHz 2)play that using HQPlayer, 1.411MHz, grab FFT @ 768kHz 3)play that using HQPlayer, 1.411MHz, noise shaped ( I don't know the options - 5th order? ) to 16 bits, grab FFT @ 768kHz 4)play that using HQPlayer, 705.6kHz, noise shaped to 16 bits ( same settings ), grab FFT @ 768kHz I'm honestly guessing at the 16 bit number here, it would be helpful if @The Computer Audiophile shared his settings? If you could, that would be most excellent, your friendly neighbourhood idiot Link to comment
GoldenOne Posted May 21, 2021 Author Share Posted May 21, 2021 3 minutes ago, idiot_savant said: Excellent! I'm assuming you can do FFTs and generate test tones? Lets do an experiment: Let us generate a 44.1kHz 24 bit -60dBFS 1kHz sine wave In all cases, NOS mode on the DAC. 1)play that using HQPlayer no oversampling, grab FFT @ 768kHz 2)play that using HQPlayer, 1.411MHz, grab FFT @ 768kHz 3)play that using HQPlayer, 1.411MHz, noise shaped ( I don't know the options - 5th order? ) to 16 bits, grab FFT @ 768kHz 4)play that using HQPlayer, 705.6kHz, noise shaped to 16 bits ( same settings ), grab FFT @ 768kHz I'm honestly guessing at the 16 bit number here, it would be helpful if @The Computer Audiophile shared his settings? If you could, that would be most excellent, your friendly neighbourhood idiot I'll do this shortly. Though keep in mind my ADC only goes up to 768khz so when running at 1.536mhz on the DAC I may not actually be able to detect all that much, but will see! This is the kind of area where an analyzer would come in very handy. I'm working towards getting an APx555 at the moment, all patreon income is going towards that. That's got a 1Mhz bandwidth so would be great here Additionally though, I'd say that when feeding a DAC 1.536mhz it almost doesn't matter what filter it's running as the information is so high bandwidth anyway. The discussion/OP was about what happens to 44.1khz content. HQP brings a benefit to D/S dacs with oversampling so I'm sure the same would happen here. numlog 1 https://youtube.com/goldensound Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 APx555 is pretty cool, and would make everything much easier What I think might happen is (1) will be classic NOS ( so images everywhere ) (2) will be nicer ( so no images ), but probably a few tones from R2R (3) will be nicer still ( no images, possibly fewer tones, possibly lower noise ) (4) much like (2), maybe a bit noisier All of which should be visible @ 768k rates. your friendly neighbourhood idiot Link to comment
The Computer Audiophile Posted May 21, 2021 Share Posted May 21, 2021 1 hour ago, idiot_savant said: Erm, I've had a bit of a ponder about this, and I think there genuinely could be a problem here, but it's quite complicated so bear with me... So the big problem with R2R DACs is matching the ladder, because the tolerances required are mind-bogglingly tight ( so the MSB has to be exactly 8 million or so times bigger than the LSB ). So the HQPlayer approach suggested by Miska, which is to effectively trade bit depth for sample rate ( much like D/S do ) relies on effectively turning off the bottom bits, and using this noise in a noise shaper - this means that the noise floor changes, so it's actually as good as it was ( potentially better ) in the audio band, but higher in the inaudible region, and the LSBs that effectively only contribute noise and disortion products are removed from the equation. Clever, eh? Now the problems come if the DAC does any further processing, as *if* it does, *any* kind of processing will turn on those pesky bottom bits again - think digital volume control, linear interpolators... Now I'm guessing that it *is* genuinely NOS at 1.5MHz, but we *know* it isn't at say 192k, so when the linear interpolate is active the PC provided noise shaping is effectively ruined. If someone has one with test equipment and HQPlayer, we could do some measurements which would be fun. To be fair, PeterSt saying he didn't think volume control was proper DSP kind of triggered this... As I keep repeating, this *should* be fixable in firmware, and is not particularly obvious... and if anyone from denafrips would like to buy me a drink, that'd be grand. your friendly neighbourhood idiot. Without identifying an issue, you suggest that something is fixable and give yourself credit. I don't follow. 49 minutes ago, idiot_savant said: Excellent! I'm assuming you can do FFTs and generate test tones? Lets do an experiment: Let us generate a 44.1kHz 24 bit -60dBFS 1kHz sine wave In all cases, NOS mode on the DAC. 1)play that using HQPlayer no oversampling, grab FFT @ 768kHz 2)play that using HQPlayer, 1.411MHz, grab FFT @ 768kHz 3)play that using HQPlayer, 1.411MHz, noise shaped ( I don't know the options - 5th order? ) to 16 bits, grab FFT @ 768kHz 4)play that using HQPlayer, 705.6kHz, noise shaped to 16 bits ( same settings ), grab FFT @ 768kHz I'm honestly guessing at the 16 bit number here, it would be helpful if @The Computer Audiophile shared his settings? If you could, that would be most excellent, your friendly neighbourhood idiot Excellent. I'll post the settings I use in HQP. Founder of Audiophile Style | My Audio Systems Link to comment
The Computer Audiophile Posted May 21, 2021 Share Posted May 21, 2021 Here are the settings I use in HQP with the Terminator. Founder of Audiophile Style | My Audio Systems Link to comment
PeterSt Posted May 21, 2021 Share Posted May 21, 2021 16 minutes ago, The Computer Audiophile said: Here are the settings I use in HQP with the Terminator. Chris, DAC Bits 20 would be insufficient for upsampling 32x. Possibly I misinterpret things (I don't know HQP well). 32x in my own book requires 21 bits at least. 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
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 @The Computer Audiophile - thanks for this, denafrips should be thanking us for debugging their product? @GoldenOne Actually, can you make the tone -6dBFS rather than -60? your friendly neighbourhood idiot Link to comment
The Computer Audiophile Posted May 21, 2021 Share Posted May 21, 2021 7 minutes ago, idiot_savant said: @The Computer Audiophile - thanks for this, denafrips should be thanking us for debugging their product? @GoldenOne Actually, can you make the tone -6dBFS rather than -60? your friendly neighbourhood idiot I’d hold off on claiming you’re debugging anything. If it’s working as designed, you’re doing exactly what I feared and wrote about earlier. Founder of Audiophile Style | My Audio Systems Link to comment
The Computer Audiophile Posted May 21, 2021 Share Posted May 21, 2021 9 minutes ago, PeterSt said: Chris, DAC Bits 20 would be insufficient for upsampling 32x. Possibly I misinterpret things (I don't know HQP well). 32x in my own book requires 21 bits at least. I wouldn’t make bold claims without understanding how something works. Questions are always good, but you know people will grab your first statement and run with it, without all the facts. Founder of Audiophile Style | My Audio Systems Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 @The Computer Audiophile if 1,2 and 3 perform as expected, this is great for denafrips and HQPlayer? if 4 responds as I expect this is a minor bug that can be fixed, and if it doesn’t I’m an idiot? @PeterSt - noise shaper… your friendly neighbourhood idiot Link to comment
The Computer Audiophile Posted May 21, 2021 Share Posted May 21, 2021 2 minutes ago, idiot_savant said: @The Computer Audiophile if 1,2 and 3 perform as expected, this is great for denafrips and HQPlayer? if 4 responds as I expect this is a minor bug that can be fixed, and if it doesn’t I’m an idiot? @PeterSt - noise shaper… your friendly neighbourhood idiot I'm not claiming it's great for anyone or you're an idiot. Stop with the nonsense. All I'm saying is that you are edging closer and closer to the camp of folks I mentioned above. You can't conclude much without actually talking to Denafrips. You can't call something a bug, if it's designed to work that way. Perhaps there's a tradeoff involved. The thing is, we don't know until someone talks to Denafrips. The last I heard is that the company didn't want to talk because it didn't want to let others know what it was doing and why. Founder of Audiophile Style | My Audio Systems Link to comment
numlog Posted May 21, 2021 Share Posted May 21, 2021 The point of the discussion is that it isnt NOS, regardless if it was designed that way or a bug. Link to comment
Popular Post feelingears Posted May 21, 2021 Popular Post Share Posted May 21, 2021 4 hours ago, The Computer Audiophile said: There is a movement to take down companies by using measurements as weapons with carefully crafted statements that don’t tell the whole truth. It shoots first and asks questions later. Many mistakes are made that need correction. The losers in all of this are consumers and companies who are slandered due to folks that don’t care about getting all the facts first, don’t care about getting things 100% right, but really care about the appearance of saving people and finding problems with product designs. I’m not saying any of this is going on here, and in fact I know GoldenSound would never do such a thing. I just want to make sure the thread doesn’t accidentally drift in that direction. +1000% I for one appreciate this resistance to the mass mob mentality and I'll just leave it at that. These are "interesting times" and some things are flat out wrong. Do the right thing, as Spike Lee said. Stand firm and call out the issues, @The Computer Audiophile! The Computer Audiophile, Blake and Superdad 1 2 Sum>Frankenstein: JPlay/Audirvana/iTunes, Uptone EtherRegen+LPS-1.2, Rivo Streamer+Uptone JS-2, Schiit Yggdrasil LiM+Shunyata Delta XC, Linn LP12/Hercules II/Ittok/Denon DL-103R, ModWright LS 100, Pass XA25, Tellurium Black II, Monitor Audio Silver 500 on IsoAcoustics Gaias, Shunyata Delta XC, Transparent Audio, P12 power regenerator, and positive room attributes. Link to comment
Popular Post The Computer Audiophile Posted May 21, 2021 Popular Post Share Posted May 21, 2021 28 minutes ago, numlog said: The point of the discussion is that it isnt NOS, regardless if it was designed that way or a bug. Even so, people shouldn’t be claiming to fix anything unless they know something needs fixing. If that is the case, then it’s time to raise a glass. Until then, let’s not create hysteria by assuming things. sonodynesrp205, barrows and Superdad 2 1 Founder of Audiophile Style | My Audio Systems Link to comment
Popular Post idiot_savant Posted May 21, 2021 Popular Post Share Posted May 21, 2021 @The Computer Audiophile I'm certainly not the one trying to take companies down - if you look throughout my comments, I'm saying it's interesting, but in real world use irrelevant if people are using it with your settings. In my *opinion* this is a bug, but a trivial one with limited impact, and I'm interested in how *this* behaviour interacts with other settings. I've also said that the HQPlayer approach to linearising R2R DACs by truncating the LSBs and noise shaping is a good one, and I'm genuinely interested as to how well it works, because all the maths say it should... However, if we have hardcore redbook NOS users, whose entire ideology is that the data should be untouched, that is an issue for *them* with something not doing what they are told. I'm 1000% not in that group btw because I quite like my dog. Now if denafrips were to just turn around and say "after 100's of hours of listening, this mode was preferred by our listeners" then that is as valid a point as any others, but silence just isn't a good look? your friendly neighbourhood idiot semente and The Computer Audiophile 1 1 Link to comment
etane Posted May 21, 2021 Share Posted May 21, 2021 i wouldn't mind it if dac mfgs label that button "LI" instead of "NOS" GoldenOne 1 Link to comment
feelingears Posted May 21, 2021 Share Posted May 21, 2021 (BTW, I am not suggesting that the OP or idiot_savant is trying to take down Denafrips. I too, find this topic interesting in the positive sense of the word.) Sum>Frankenstein: JPlay/Audirvana/iTunes, Uptone EtherRegen+LPS-1.2, Rivo Streamer+Uptone JS-2, Schiit Yggdrasil LiM+Shunyata Delta XC, Linn LP12/Hercules II/Ittok/Denon DL-103R, ModWright LS 100, Pass XA25, Tellurium Black II, Monitor Audio Silver 500 on IsoAcoustics Gaias, Shunyata Delta XC, Transparent Audio, P12 power regenerator, and positive room attributes. Link to comment
etane Posted May 21, 2021 Share Posted May 21, 2021 https://www.kitsunehifi.com/nosvsos/ Link to comment
idiot_savant Posted May 21, 2021 Share Posted May 21, 2021 Wow, I’ve never seen a comparison of OS with putting rats in soup… needless to say, this explanation is a bit lacking… Your friendly neighbourhood idiot semente 1 Link to comment
etane Posted May 21, 2021 Share Posted May 21, 2021 you've heard of the mouse who fell into a bucket of milk? 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