Giulio also uses raspberry pi’s, I’d contact him and ask if he can set-up lv2 builds intructions for raspberry pi/aarch64. In general the build instruction for his projects are a bit underdocumented.
1 Like
Nice move @ReCremisi89, getting in touch with Giulio directly 
.
As with everything in the programming and code compiling world, there is always a good share of inherent professional knowledge, which is not completely documented, and mostly taken for granted among expert users.
1 Like
Pi5 + Hifiberry ADC+ DAC , Vangelis, runs on 64 Samples/48kHz without XRuns.
3 Likes
Good to know, thanks 
Having recently upgraded my Zynthian to Oram-2511, I’ve given virtualjv a try, so despite the hiatus in this thread, I’m assuming this would be a place to post related issues (rather than the upgrade thread).
I’ve noticed one thing with virtualjv and that is that one doesn’t have to go far up on the keyboard for there to be issues with digital artifacts. It sounds a bit like aliasing, but not quite. For most sounds, definitely in the top octave of a five octave keyboard, sometimes lower (I’m assuming depending on the transposition setting in the loaded preset). For many sounds there is a specific key (varies from preset to preset) below which the sound is much cleaner.
The question is if this is the way the JV880 behaves or if it is an issue with the emulator? (I’d like to compare it to my hardware JV880, but typical me I bought it in a non-functioning state, seemed to have a power supply issue, but things came up and I never got around to doing anything but basic checks, so it’s really nice to have the emulator in Zynthian so I can finally get an idea of what is should sound like). Anyone else have any thoughts on this?
2 Likes
Good morning @ricard
Did you consider changing zynthians SR from e. g. 44100 Hz to 48000 Hz or vice versa?
Thorsten
Hi @ricard 
. Your ears are sensitive, and justly catched an intrinsic limitation, by design, of all early 90s manufacturer’s synth-romplers, except the JD-990. Their DACs run at 32 Khz, irrespective of the theoretical resolution of the samples at the core of the synthesiser presets.
This notoriously and unavoidably caused a certain opaqueness in the high keyboard range, and audible aliasing artifacts despite the in-built dedicated filters. It is an inherent part of the instrument’s colour, and is reflected 1-to-1 in the workings of the emulator. Only the JD-990, with a compromise on lower polyphony, was designed to exploit the full 44 kHz spectrum of the samples base, and with better and more transparent DACs (and it shows!).
Thus, the only way to maybe improve things on Zynthian with VirtualJV might be to set the sample rate to 32000.
I hope it helps!
1 Like
Is it possible in some setting parameter?
I think that the resolution of the Zynth audio hardware is more or less freely configurable, in the audio jack parameters through Webconf, in the related tab. Nevertheless, I wouldn’t personally go as low as 32 Khz of sample rate as a system setting, because it would affect every other processor and synth running on the Zynthian. The result would be lackluster FX and sounds, strongly lacking shine in the high frequency range.
A possible solution would be to devote a specific custom Zynth to just running VirtualJV, possibly passing its audio output through the audio input of another Zynthian at full 48 Khz for sound processing.
Not sure if all this trouble would be worth any significant improvement in sonic quality, anyway.
Cheers
1 Like
You could try adding a low-pass filter to the chain which might give some benefit.
1 Like
Hi @riban . I have tried various FX combinations, to add some high-frequency shine to the output of my (beloved) JV-2080, which shares with JV-880 the same limitations in audio resolution (32 Khz) of the whole signal chain, including waveform modulations and multi-effects.
I have had some valuable (but not so dramatic) results with a serial chain of this kind:
- Low-pass filter with 24 or 36 dB/Oct inclination, to shelf the artifacts away.
- Amp simulator to bestow some saturation.
- Shimmer/granulator reverb with mild overtones and short reverb time, to add high range animation.
Anyway, as we know, it is impossible to retrieve something in the spectrum that has already been lost.
2 Likes
I’m not certain whether defining JACK’s SR has effect on HW’s SR.
AFAIK soundcards SR is defined in firmware/config, isn’t it?
Can somebody confirm or contradict?
Hi @Stupps, If I’m not wrong, the audio hardware specification defines the maximum operational frequency of the DACs.
Below that setting, you should be able to apply whatever reduction of duty-cycle is available, in the configuration of the hardware.
Afaik. on the Zynthian environment this is implemented through the Webconf Jackd settings, unlike a desktop computer environment, where there exists some kind of software, to configure the operation of the HW’s audio drivers.
@Aethermind, that is whithout a doubt.
However is it guarantied that DAC will run on JACK’s (via webconf applied) SR or is there a kind of (hidden) SRC between JACK’s output and DAC’s input?
Well just a thought…
Have a nice weekend.
1 Like
The hardware ADC/DAC should run at the jackd samplerate but that is not the issue here. The engine, in this case virtualJV processes chunks of audio at the jackd samplerate. But if that engine is handling samples, it must at some point samplerate convert. It is worth running jackd at 32000fps as an experiment to see how it works but as already discussed here, the original device behaved this way so it is unlikely to have a benefit.
@Aethermind’s suggested chain may proving better result of you want to clean up the sound and it’s what we would have to do (done) with the original hardware.
I don’t have skin in this game. I am happy to use a modern sampler and have no history with these devices to conjour nostalgia.
2 Likes
There’s a bare hardware version based on the MiniDexed code at GitHub - Sterr1/Mini-JV880pi too.
…
I so need to set up a Zynthian with multiple inputs for my diy hardware synths like those 
4 Likes
This just reminds me how disappointing was to get my Proteus 1/XR power supply fixed and listening again how it sounds several years later, compared to what my ears got used to meanwhile…
Of course, one (me!) must find adequate samples to make that modern stuff sound as needed.
Still, I fancy to learn (a lot) with you guys discussing these interesting and relevant technical details. I never knew of this 32 KHz aspect that explains so much of what I’ve been feeling with my gear.
Thank you @riban @Aethermind et al
2 Likes
32000fps seemed like a reasonable approach when comparing with instruments that generally cutoff below 15kHz. Also, analogue broadcast (at least FM in the UK) was band limited to 15kHz so listeners didn’t hear the hf in recordings.
2 Likes
Yes, back in the early 1990s it made absolutely sense to preserve the relatively limited computational power, for features like increased polyphony and on-board FX, reducing the sample rate to what was perceived anyway like a sort-of-hi-fi sound, by the standards and for the habits of the music production, distribution and broadcasting.
By the way, it seems that the previous, really glorious and still very valid today (in my opinion) D-50 had an even lower SR rate of 31400 FPS. It appears that the sonic magic, of a quality that back then stunned people for the cinematic hi-fi colour out of the box, was due to a post-DACs analog LP filter for each of the sixteen voices, which afforded warmth, depth and acoustic covering of the artifacts to the otherwise fully digital signal chain. A similar kind of trick worked even better a few years later, on the mighty JD-990.
Paolo, could you share a snapshot as example to learn better the new starting point?