PlanetZ forums - scopeusers.com

Somehow, the problem solved itself. Load is now 4000 on both DSPs without the infamous sat connection error.

I'm trying to go from mono- to polyphony. I'm encountering - as expected - a lot of problems (i.e. DSP overload). But something seems to be wrong: I've split my module into 2 modules having 16 partials each. In mono-mode, the load looks normal/balanced: Monophone.png In polymode (2 voices, 1 module ...

And that's how it sounds (tanpura profile)

I looked at a picture of the A16 MKII and noticed it has wordclock. The XITE-1 has wordclock as well. I (embarrassingly) don't know very much about wordclock, but if I understand correctly, it is for syncing, and I think your problem stems from bad syncing. I'd get 2 BNC cables and try that. I'll b...

I had this problem with my XITE-1. Try to connect out AES/EBU from A16 to the input of the XITE-1 and select AES/EBU to sync to. If that doesn't work reverse. Audio will still go through ADAT. Syncing via AES/EBU worked for me. I'd love to try that, but my A16 MK 2 doesn't provide an AES/EBU interf...

Have you tried increasing the latency in ULLI settings? Sounds like an audio latency issue, not a sync issue. BNC works because there's no audio sharing the pipe? Just a wild hunch. Jus' sayin'.... Sync issues would be really jitter and timing but crackles would more likely be latency or IRQ ot IO ...

garyb wrote:yeah, i never saw that behavior.
i think we need to exchange some emails.

If I can be of help, yes!

garyb wrote:bnc should be better than lightpipe, but both should be click-free.
also interesting that it works with other ADAT devices...

And also interesting is: 32K: no crackles on any port. 44.1: crackles on Port A, 48+: crackles everywhere.

wow, that really upsets me(i'm sure it upsets you more!)! i don't know what to say. your XITE handled ADAT exactly as it should here... What upsets me most, is the unruly behavior. Both work fine as is with a 3rd ADAT interface and together differently on the A and B port. And also using BNC wordcl...

Today I got my A16 replacement from ferrofish: same behavior!

Thanks.
And this is the result:

faxinadu wrote:can you explain that? what do you mean use it with a 2:1 switch?

For resynthesis you need either sin or cos signal. The atom has 2 outs (sin and cos), so you need a 2:1 (sin or cos-> next stage). This can be switched dynamically (async).Ecample:

I found out, I can use the complex sine osc with a 2:1 switch.

yes, it's definitely doable, provided the phase-shifter as such is available (not to be confused with a phaser). Even a variable delay or allpass filter can be used. But all these are atomic functions.

If there is a phase shifter, which is voltage controlled (i.e. controlled by the frequency of the osc), that would be fine. I haven't found one.
Though I found one Modular osc, which can produce either sin or cos. I have to dig deeper into that (i.e. if it doesn't consume too much DSP).

Not any one with phase shift option (AFAIK). I need one either sin or cos waveform (dynamically changeable).

Yes, I have to create an osc in dsp. Not possible else (I think). Please correct me, if I'm wrong.
Reason: I need 32 of those, do this on the PC will not work in real time.

As soon as the dongle arrives consider it done!

Do we have an osc, where I can either select sin or cos or do an arbitrarily phase shift? I need this for my resynth project.

It's either a mechanical problem (PCB/compenend getting stressed) or a temperature problem (PCB expands, when getting too hot). Both problems are difficult to fix, if a solder joint or a PCB track is involved.