EuroDriver Posted October 15, 2015 Share Posted October 15, 2015 This is on Windows 10 with no network audio. The dac is connected directly to my computer using usb and asio driver. I understand what you are saying but this behavior is only with the poly-ext filter engaged. Using other filters does not cause this behavior. I was just looking for Miska to comment if this behavior was unique to that one new filter? I am wondering about this too ! Sound Test, Monaco Consultant to Sound Galleries Monaco, and Taiko Audio Holland e-mail [email protected] Link to comment
ted_b Posted October 15, 2015 Share Posted October 15, 2015 I've noticed something strange regarding the new poly-ext filter and the latest 3.10 version. If I play a 24/96 recording the ext filter causes the DSD sample rate to change automatically from that which is pre-selected. Further, it only allows certain sample rates different from those with 16/44 recordings. I only see this with the ext filter. Perhaps Miska could explain this? As mentioned earlier last week Poly-sinc-ext only does 2x or 4x upsampling, I believe. When I use it I notice the upsampling rate changes from 192 to 176, and my DAC does not do 176 so no sound. You can set upsampling to 88 and it should work. it looks pretty certain that poly sinc ext does integer multiple upsampling. Kinda nice but a real surprise if your dac doesn't like 176, for example (some dacs have blind spots at certain multiples, like dacs not playing 88k and 176k, or same issue with 48k based SDM rates). I'm not sure if ext brings anything else, but feedback seems to be good, right? Edit: I haven't moved to 3.10 yet cuz been busy with other setups (streamers, etc). Will do tonight. "We're all bozos on this bus"....F.T. My JRIver tutorial videos Actual JRIver tutorial MP4 video links My eleven yr old SACD Ripping Guide for PS3 (needs updating but still works) US Technical Advisor, NativeDSD.com Link to comment
Chodi Posted October 15, 2015 Share Posted October 15, 2015 Thanks for the heads up. I guess I missed those comments. In my case I am really enjoying the sound of poly-ext used to upsample to DSD. I am still a bit confused as my dac reads out 5.6 or 11.2 when playing a 96/24 recording although HQP shows 6.1??? I suppose my dac is somehow doing the conversion? It sounds very good so this is just a curiosity nothing more. Link to comment
elcorso Posted October 15, 2015 Share Posted October 15, 2015 This is on Windows 10 with no network audio. The dac is connected directly to my computer using usb and asio driver. I understand what you are saying but this behavior is only with the poly-ext filter engaged. Using other filters does not cause this behavior. I was just looking for Miska to comment if this behavior was unique to that one new filter? Now I understand! Then HQ P version 3.10 and not networkaudiodaemon 3.1.0 But please check Jud comment (the same as mine). To Miska: Regarding networkaudiodaemon 3.1.1 for Windows. What are the changes from 3.1.0 ? Roch Link to comment
Miska Posted October 15, 2015 Share Posted October 15, 2015 This is on Windows 10 with no network audio. The dac is connected directly to my computer using usb and asio driver. I understand what you are saying but this behavior is only with the poly-ext filter engaged. Using other filters does not cause this behavior. I was just looking for Miska to comment if this behavior was unique to that one new filter? "poly-ext", "closed-form" and "minringFIR" are examples that cannot be used for fractional ratios. Whenever you change filter, the set of available sampling rates is updated based on allowed conversion ratios for the currently active material. Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
dean70 Posted October 15, 2015 Share Posted October 15, 2015 "poly-ext", "closed-form" and "minringFIR" are examples that cannot be used for fractional ratios. Whenever you change filter, the set of available sampling rates is updated based on allowed conversion ratios for the currently active material. If that is the case, I found that the closed-form filter does not automatically reset the Output PCM sample rate to an integer multiple like the poly-ext filter does. Alchemy Desktop http://www.origen.net.au/Alchemy/ Link to comment
Miska Posted October 15, 2015 Share Posted October 15, 2015 If that is the case, I found that the closed-form filter does not automatically reset the Output PCM sample rate to an integer multiple like the poly-ext filter does. Yes, there was a bug there that has been fixed for the 3.11 I hope to release tomorrow... It checked divisor, but forgot to make sure modulo is 0... Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
ebrandon Posted October 17, 2015 Share Posted October 17, 2015 I'm really liking the sound of Closed-form, but getting a frequent clicking noise. It happens several times a minute -- little clicks in the music (or in the silence between tracks). My signal path is Mac Mini i7/Yosemite/HQPlayer 3.10 -> wired ethernet -> Mac Mini i5/Yosemite/NAA 3.10 -> Oppo 105D USB DAC My settings are playing 44.1/16 -> Closed Form -> ASDM5 -> 5644800 -> SDM(DSD) If I switch back to poly-sinc-short the clicking disappears. Link to comment
ted_b Posted October 17, 2015 Share Posted October 17, 2015 3.11 up now "We're all bozos on this bus"....F.T. My JRIver tutorial videos Actual JRIver tutorial MP4 video links My eleven yr old SACD Ripping Guide for PS3 (needs updating but still works) US Technical Advisor, NativeDSD.com Link to comment
brightonjel Posted October 17, 2015 Share Posted October 17, 2015 Definitely fixed the low-level popping noise with closed-form filter and no SDM on Mac - thanks, Jussi! Link to comment
dean70 Posted October 17, 2015 Share Posted October 17, 2015 3.11.0 has fixed the closed-form filter output rate issue (and noticed the new closed-form-fast). But is has inadvertently opened up an issue with the hqp-control.exe interface when setting the sample rate. Normally you would have a full list of rates the DAC supports, but when you select poly-ext or closed-form, only a subset are available (44100,88200,176400 and 88200,176400 respectively for 44100 pcm source). The --set-rate option takes the index, which only 0-2 or 0,1 are available, which could change depending on the source sample rate. There is no way of knowing what the valid --set-rate indexes are, and what sample rate they correspond to. Alchemy Desktop http://www.origen.net.au/Alchemy/ Link to comment
Miska Posted October 17, 2015 Share Posted October 17, 2015 But is has inadvertently opened up an issue with the hqp-control.exe interface when setting the sample rate. Normally you would have a full list of rates the DAC supports, but when you select poly-ext or closed-form, only a subset are available (44100,88200,176400 and 88200,176400 respectively for 44100 pcm source). The --set-rate option takes the index, which only 0-2 or 0,1 are available, which could change depending on the source sample rate. There is no way of knowing what the valid --set-rate indexes are, and what sample rate they correspond to. The set of rates has always been intersection of rates supported by the filter and rates supported by the DAC. Only if you use the normal poly-sinc filters which can do conversion from any rate to any other rate the set of output rates is everything that the DAC supports. For this reason, as long as there has been --set-rate in hqp-control there has also been --get-rates to retrieve set of available rates. Set of rates naturally also depends on the output mode (PCM/SDM). So always when using hqp-control, operations should be performed in following order: 1) Set output mode 2) Set filter 3) Set dither/modulator 4) Set output rate (from the available set: --get-rates) Also note that some DACs have bugs in their firmware and/or drivers and claim to support rates that are not really supported, or don't claim rates that they would actually support. - Amanero ASIO driver claims to support every possible rate while it naturally doesn't support everything - Marantz HD-DAC1 doesn't support 48-base rates with DoP while technically there shouldn't be problem - iFi's Thesycon ASIO driver doesn't claim 48-base DSD rates although the hardware actually supports those just fine (Mac&Linux) Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
Miska Posted October 17, 2015 Share Posted October 17, 2015 To Miska: Regarding networkaudiodaemon 3.1.1 for Windows. What are the changes from 3.1.0 ? It has updated ASIO backend, to keep in sync with the one in HQPlayer. Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
Samuel T Cogley Posted October 17, 2015 Share Posted October 17, 2015 I'm really liking the sound of Closed-form, but getting a frequent clicking noise. It happens several times a minute -- little clicks in the music (or in the silence between tracks). My signal path is Mac Mini i7/Yosemite/HQPlayer 3.10 -> wired ethernet -> Mac Mini i5/Yosemite/NAA 3.10 -> Oppo 105D USB DAC My settings are playing 44.1/16 -> Closed Form -> ASDM5 -> 5644800 -> SDM(DSD) If I switch back to poly-sinc-short the clicking disappears. It would seem that closed-form requires a LOT of CPU power. I had similar results with an i7 5930k -> iFI iDSD Micro. After taking the sample rate down from DSD512 to DSD256, the clicking stopped. Looks like you're already on DSD128. You can't go much lower. Have you tried closed-form-fast? Link to comment
John ALSR Posted October 17, 2015 Share Posted October 17, 2015 The Feeble Computer, as previously defined, using V.3.11 of HQP, will NOT upsample base-44KHz to x2 SDM (5.6MHz) to a Mytek DAC via Firewire using the Closed-Form filter. All four 2.5MHz CPU cores at 75-85% giving continuous dropouts. "Closed-form" is more rigorous than Poly-sinc computationally. John Link to comment
Chodi Posted October 18, 2015 Share Posted October 18, 2015 I just wanted to follow up my earlier comment regarding this new closed form filter. This latest version 3.11 seems to have "fixed" the closed form filter and the sound is much better than in the previous version. In fact I am enjoying it and having a difficult time choosing between poly-ext and this revised closed form. It does require considerable computer resources. On my computer the cpu load is about 60% when upsampling to dsd 256. The sound is very open and transparent. Low level detail is excellent. Miska is a genius when it comes to these filters. Where else can you get these options? Nowhere. Link to comment
elcorso Posted October 18, 2015 Share Posted October 18, 2015 I just wanted to follow up my earlier comment regarding this new closed form filter. This latest version 3.11 seems to have "fixed" the closed form filter and the sound is much better than in the previous version. In fact I am enjoying it and having a difficult time choosing between poly-ext and this revised closed form. It does require considerable computer resources. On my computer the cpu load is about 60% when upsampling to dsd 256. The sound is very open and transparent. Low level detail is excellent. Miska is a genius when it comes to these filters. Where else can you get these options? Nowhere. +1 Roch Link to comment
kurb1980 Posted October 18, 2015 Share Posted October 18, 2015 I just auditioned the new filters and poly sinc ext sounds very nice I'm not sold on the closed filters with DSD just yet. I will try these out with PCM to see which is better my instinct says the closed filters will sound better with PCM. I have tried all the filters and I keep coming back to poly sinc and its less computationally heavy brother *2. @ Miska question regarding poly sinc filter why does it tax the processor at 100% for about 8-10sec then drop down to 20% when it starts to play a track? Its the only filter that behaves this way to best of my knowledge? Link to comment
dean70 Posted October 18, 2015 Share Posted October 18, 2015 The set of rates has always been intersection of rates supported by the filter and rates supported by the DAC. Only if you use the normal poly-sinc filters which can do conversion from any rate to any other rate the set of output rates is everything that the DAC supports. For this reason, as long as there has been --set-rate in hqp-control there has also been --get-rates to retrieve set of available rates. Set of rates naturally also depends on the output mode (PCM/SDM). So always when using hqp-control, operations should be performed in following order: 1) Set output mode 2) Set filter 3) Set dither/modulator 4) Set output rate (from the available set: --get-rates) Also note that some DACs have bugs in their firmware and/or drivers and claim to support rates that are not really supported, or don't claim rates that they would actually support. - Amanero ASIO driver claims to support every possible rate while it naturally doesn't support everything - Marantz HD-DAC1 doesn't support 48-base rates with DoP while technically there shouldn't be problem - iFi's Thesycon ASIO driver doesn't claim 48-base DSD rates although the hardware actually supports those just fine (Mac&Linux) Thanks for that info regarding --get-info. I have mapped out each filters behaviour & can now select the correct rate for each filter type and DAC capabilities for my front-end GUI. I can now auto-select closed-form,adsm7 DSD128 for 44100 source & poly-ext,dsd7 DSD128 for dsd64 source Alchemy Desktop http://www.origen.net.au/Alchemy/ Link to comment
ebrandon Posted October 20, 2015 Share Posted October 20, 2015 I'm really liking the sound of Closed-form, but getting a frequent clicking noise. It happens several times a minute -- little clicks in the music (or in the silence between tracks). My signal path is Mac Mini i7/Yosemite/HQPlayer 3.10 -> wired ethernet -> Mac Mini i5/Yosemite/NAA 3.10 -> Oppo 105D USB DAC My settings are playing 44.1/16 -> Closed Form -> ASDM5 -> 5644800 -> SDM(DSD) If I switch back to poly-sinc-short the clicking disappears. HQPlayer 3.1.1 fixes the clicks. I've been comparing my tried-and-true poly-sinc-short to the new poly-ext, closed-form, and closed-form-fast. I go back and forth between poly-sinc-short and closed-form-fast. Closed-form-fast is very smooth and relaxing but not boring. I'm still trying to decide whether the greater "bite" in poly-sinc-short is the "real" sound of the instruments coming through, or undesirable harshness. QUESTION -- what is the difference between closed-form and closed-form-fast? Thanks! Link to comment
ebrandon Posted October 20, 2015 Share Posted October 20, 2015 How come I suddenly have SDM frequency 6144000? I'm amazed my DAC can play this non-standard DSD frequency. Is there any advantage to using this even-multiple of 48/96/192k PCM when using non-integer filters like poly-sinc-short? Link to comment
elcorso Posted October 20, 2015 Share Posted October 20, 2015 HQPlayer 3.1.1 fixes the clicks. I've been comparing my tried-and-true poly-sinc-short to the new poly-ext, closed-form, and closed-form-fast. I go back and forth between poly-sinc-short and closed-form-fast. Closed-form-fast is very smooth and relaxing but not boring. I'm still trying to decide whether the greater "bite" in poly-sinc-short is the "real" sound of the instruments coming through, or undesirable harshness. QUESTION -- what is the difference between closed-form and closed-form-fast? Thanks! From the Signalyst web page: "Adds closed-form-fast interpolation variant for lower CPU load." Roch Link to comment
elcorso Posted October 20, 2015 Share Posted October 20, 2015 How come I suddenly have SDM frequency 6144000? I'm amazed my DAC can play this non-standard DSD frequency. Is there any advantage to using this even-multiple of 48/96/192k PCM when using non-integer filters like poly-sinc-short? All DSD DACs should play "non-standard" DSD frequencies, they are not common mentioned but they are the integer multiple for 48/96/192k PCM. I love the integer and not the floating point conversion. Roch Link to comment
ted_b Posted October 20, 2015 Share Posted October 20, 2015 From the user manual: "closed-form - Closed form interpolation with high number of taps. closed-form-fast - Closed form interpolation with lower CPU load, but also lower precision. Output precision tuned to match about 24-bit PCM" "We're all bozos on this bus"....F.T. My JRIver tutorial videos Actual JRIver tutorial MP4 video links My eleven yr old SACD Ripping Guide for PS3 (needs updating but still works) US Technical Advisor, NativeDSD.com Link to comment
kurb1980 Posted October 20, 2015 Share Posted October 20, 2015 From the user manual:"closed-form - Closed form interpolation with high number of taps. closed-form-fast - Closed form interpolation with lower CPU load, but also lower precision. Output precision tuned to match about 24-bit PCM" Could someone please explain technically what closed form is doing? Please thx? 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