True, curiosity will eventually get the better of me... 😛
If the NAA would run under Botic, that would make it happen much sooner. But to tip my hand a little, I think that cone technology in loudspeakers is the most vulnerable area for 'non-analogue' interference. I have directly addressed 'micro-distortion' in my speaker surfaces, and found that the cone mods erased practically all the sonic advantage of up sampling within the PCM domain. The result of the cone mods were exactly as you guys are describing the sound of the 'no-dac'. So that is why I question whether I am a good candidate to enjoy a 'no-dac'. ...particularly with the PCM-based DSP crossover to which I am now committed.
I'll continue on my current path and read along. The 'leap of faith' may happen in a new portable system that is early in the planning stages.
You guys keep having fun!
F.
If the NAA would run under Botic, that would make it happen much sooner. But to tip my hand a little, I think that cone technology in loudspeakers is the most vulnerable area for 'non-analogue' interference. I have directly addressed 'micro-distortion' in my speaker surfaces, and found that the cone mods erased practically all the sonic advantage of up sampling within the PCM domain. The result of the cone mods were exactly as you guys are describing the sound of the 'no-dac'. So that is why I question whether I am a good candidate to enjoy a 'no-dac'. ...particularly with the PCM-based DSP crossover to which I am now committed.
I'll continue on my current path and read along. The 'leap of faith' may happen in a new portable system that is early in the planning stages.
You guys keep having fun!
F.
Attachments
Last edited:
Dead right it does; when I'm running HQPlayer to upsample 44/16 material to DSD256 the PC is noticeably slower at performing other tasks.
Frank, I don't think I said that. What I said was that it sounds better than the other solutions I mentioned as comparisons. Perhaps HQPlayer is the most accurate DSD upsampling/playback solution?
I have tried to get BBB/Botic running as HQPlayer NAA but I don't have the necessary skills to resolve the issues. I think that either the NAA software has to be backported to Debian Wheezy or Botic needs to be updated to run on Debian Stretch.
BTW, did you note that I also run a BBB/Botic/SO3 based DAC (DAC is Buffalo IIIse/Legato) and using a DSD256 recording (no upsampling) it doesn't sound as good as the direct no-DAC DSD project.
Just do it!
Ray
Eventually I think it will happen...
😀 I did note that. As @stijn001 mentioned, that 'totally analog' sound is pretty sweet. But it only works for me right now if PCM conversions by HQ player also enjoy the same advantages as native DSD. 🙂
I would like to simulate DSD processing through a buffer and LPF. Can anyone post a 1 bit DSD sine signal in WAV format, preferably 1kHz?
v
v
Over the last few weeks I've sent out a few of my prototype direct DSD boards and it would be good to hear back about the results people are getting; I'm sure it would help everyone here.
In the meantime, I've just checked and found I have two of the boards left with one just being spoken for so, if you want the last one PM me. Cost is £2 plus postage. Perhaps the last board is destined for francolargo?!😀
That said, the flip-flop PCBs are on their way and, as long as everything works satisfactorily, I'll offer the spare boards on a similar basis. This is the board;
Plus, to use with my Aikido project, I've also designed and ordered some additional PCBs that just have the mute and DSD indicator circuits (The Broskie Aikido board I'll be using has a second order filter on its input) and again, if they work correctly, I'll make the spares of that board available. This is the board;
Apologies if that sounds like a sales pitch; my intention isn't to do anything other than to make my surplus PCBs available to those who might appreciate them.
Anyway, let's see how things go when I try the boards out.
Ray
In the meantime, I've just checked and found I have two of the boards left with one just being spoken for so, if you want the last one PM me. Cost is £2 plus postage. Perhaps the last board is destined for francolargo?!😀
That said, the flip-flop PCBs are on their way and, as long as everything works satisfactorily, I'll offer the spare boards on a similar basis. This is the board;
Plus, to use with my Aikido project, I've also designed and ordered some additional PCBs that just have the mute and DSD indicator circuits (The Broskie Aikido board I'll be using has a second order filter on its input) and again, if they work correctly, I'll make the spares of that board available. This is the board;
Apologies if that sounds like a sales pitch; my intention isn't to do anything other than to make my surplus PCBs available to those who might appreciate them.
Anyway, let's see how things go when I try the boards out.
Ray
Hi Ray, thanks for your generous efforts.
I'm surprised you've abandoned the flipflop in you Aikido board design. I was under the impression that the common noise rejection as can be done with the balanced output of the flipflop in combination with the Aikido board was the main attraction.
What am I missing?
I'm surprised you've abandoned the flipflop in you Aikido board design. I was under the impression that the common noise rejection as can be done with the balanced output of the flipflop in combination with the Aikido board was the main attraction.
What am I missing?
The Aikido board has a single-ended input, with filter component options on the Aikido PCB, hence not using flip-flop and the board with just the mute and DSD indicator LED. The attraction of the Aikido, apart from its reputation for excellent sound, is the integrated active filter. Also, the aim is to try using the 1.6V DC offset on the output of the USB board to bias the input tube, thus avoiding the use of a DC blocking cap.
The Broskie BCF accepts a balanced input (and outputs single-ended) so that's what I'll be using the flip-flop board with. Obviously balanced operation nullifies the DC offset and offers CMRR and the analogue element of the project is simpler.
Will be interesting to see how the approaches match-up.
Ray
I guess I'm going to have to learn how to use Linux before I get to hear this direct dsd sound, but before I do, just to confirm, no one has been successful in using the JLSounds usb board, windows and JRiver converting pcm to dsd?
Can it be done via DoP ? And what was that stuff about a 4R7 between pins 1&9 on the JL board? Is that windows related?
I've got a Raspberry playing direct I2S, no usb, to a 1541 dac , which sounds excellent , so keen to hear direct dsd, but not keen on going further into linux, I suppose one just has to bite the bullet.
Can it be done via DoP ? And what was that stuff about a 4R7 between pins 1&9 on the JL board? Is that windows related?
I've got a Raspberry playing direct I2S, no usb, to a 1541 dac , which sounds excellent , so keen to hear direct dsd, but not keen on going further into linux, I suppose one just has to bite the bullet.
Mick, you don't need to become a Linux expert, I'm certainly not. The main stumbling block with exploring this direct DSD thing does seem to be the inconsistency around which operating system/driver combos work with which USB boards. At this point in time it seems like Windows and the JLSounds board aren't going to work together for something like native DSD256, but equally, in my experience, nor did AudioLinux (I tried it only a couple of months ago and the kernel wasn't up to date with the JLSounds patches - may be different now?)
My setup, which now works consistently with DSD256, is a powerful Windows 10 PC running HQPlayer and ,currently, Muso, (to provide better library functionality and basic remote control for HQPlayer, but likely to switch to Roon when it interfaces with HQPlayer).
HQPlayer outputs to a Linux NAA device based on a silent Intel Atom motherboard. The NAA is running the Debian Stretch Linux distribution. To be honest the Debian install was pretty simple and just involved running through a list of Linux commands with no great knowledge of Linux required. I also installed Debian on a Beaglebone Black.
I have started to type up some notes on how I did the Debian install and will try to post them here soon.
The stuff about the 4R7 resistor relates to getting the JLSounds board working with the trial version of the Windows driver and isn't relevant in a Linux context.
One additional thing I did do was separate the streaming traffic to the NAA from the general home network traffic; I was experiencing occasional dropouts of the audio when there were peaks on the home network. I achieved the separation by installing a second network card in the HQPlayer PC and setting up a subnet dedicated to a link to the NAA. Again, it wasn't overly difficult and well worth the effort. Perhaps I'll try to produce some notes on that if there is any interest.
Ray
Attached is the first draft of my attempt to document the installation of Debian Stretch and HQPlayer NAA modules.
The document reflects my own experience at setting up an NAA on an Intel Atom based computer. It has worked very well for me and I am able to successfully obtain native DSD256 data from my JLSounds USB board.
Caveats:
Ray
The document reflects my own experience at setting up an NAA on an Intel Atom based computer. It has worked very well for me and I am able to successfully obtain native DSD256 data from my JLSounds USB board.
Caveats:
- I'm not a Linux expert.
- It worked for me, I can't guarantee it will for you.
- If I can I'll help if you have a problem but I cannot guarantee to resolve every issue.
- The document is incomplete and can certainly be improved; I can't offer a definite timeframe for when I'll update it.
Ray
Thanks for the guide Ray 🙂 I can see several cheap Intel nuc type devices which will act as an NAA and I think dsd output from the Waveio board is either ready or imminent, so I'm just trying to work out how powerful a box I need for hqplayer upsampling pcm to dsd. Do you think I need something particularly beefy for this? Could I use another nuc device or old laptop or do I need something powerful?
Assuming you want to upsample to DSD256, you will need a fair amount of processing grunt. I use an HP Z210 Workstation with Xeon E3-1240 3.3GHz Quad Core processor and that does the job nicely. That processor benchmarks at just a tad under 8000 so I would say you should aim for something similar or better. You can find benchmarks for suitable processors here;
https://www.cpubenchmark.net/high_end_cpus.html
BTW, because I wasn't sure what this DSD journey would bring I bought my workstation as 'previously loved' for only £200; worth considering if you're on a shoestring or aren't sure if this is for you.
The DSD filter board I'm sending you is designed to mount on the header pins of the JLSounds USB board so, if you're planning to use a WaveIO you'll have to work out how to hook it up, keeping the data connections nice and short.
Another BTW, if anyone is planning to purchase HQPlayer I recommend contacting Gianluca (Bibo01) who is an authorised reseller and will give you a better price than the HQplayer (signalyst) website.
Ray
All of the original filter/mute boards have now been taken; Frank took the bait!
Ray
I was more talking about degradation of audio/conversion quality but sure, on a low end computer I could see how it would also considerably stress the CPU.
Yes, I know. I was making light of your sweeping, unsupported original statement, which is at odds with what this group is hearing. So, how about relating your experience on the matter. Have you actually heard the results that HQPlayer delivers?