Wish for new atom (osc)

[soren_jepsen]

Is anyone who is capable of creating new atoms still around?

Cheers
Søren (Ravensound)

[spacef]

in the "third parties", I think you are/could be that person, Soren
(I'm not speaking for SC, just sat. evening thinking).
but yes, anti-aliased oscillators would be great... (we have one or two in Scope Modular)

[valis]

Is anyone who is capable of creating new atoms still around?

Cheers
Søren (Ravensound)

I can’t do that, but how can I help?

[spacef]

I don't know, but I thought Soren could do "C++ to Scope" modules, so I "suppose" he can do an oscillator ? ...

[valis]

I was responding to Soren, but in your case I need to get busy and learn all your tools!

[astroman]

I only remember that back in the days there was a special (very expensive) license to get access to the tools to create custom atoms. Warp69 once complained he was stuck in his development because things couldn‘t be resolved by the existing ones.

[nebelfuerst]

What is the status of thus C++ SDK today ? Still expensive ?

[Spielraum]

i would be very happy if a recognized DSP programmer would come back to us.
unfortunately i don't think so...

[soren_jepsen]

Thank you for the answers

Regarding C++ modules:
Those are definately possible, and not that difficult. If you know C++, creating a module is relatively straight forward. If you look a the code for ScopeSync, and compare it with the old Scope FX pdf that's floating around somewhere, you have what you need. I was thinking about doing a course on it, but I don't know if enough people would be interrested. There are limitations though!!!

- Regarding speed, you are limited to the latency between Scope<>Windows in your ULLI setting. Meaning; the code updating the pads of your module will only update every X millisecond. The lowest possible is around 1 ms. That's relatively fast, but not enough if you want to make a oscillator syncable In the 32bit days, you could make Scope "interrupt" this "wait" at any given time, but I don't think that's possible anymore)
- You can't send a string to Scope from PC. (you can however, send an array of integers, and then in Scope convert each to a character, and then add that together to a string (*sigh*).
- You CAN send a string from PC to Scope, but you can't have any other array pads in that module, and it's probably not stable.
- Generally, Scope still has problems with adressing PC memory. A nice leftover from the 32bit days. This is (to my understanding) why the akai osc doesn't work (besides the "interrupt issue").
- You can also make the GUI with a C++ module, BUT I don't know how to make it follow the z-order of Scope. Meaning; it's either permanently on top of everything Scope......or behind. I think Scope GUI is made with wxWidgets - but that the extend of my knowledge it that direction.

I have several projects in the furnace - but this is sparetime work. Especially with absolutely no help from SC
- A midi-device selection tool/router that lives inside Scope and is presettable (getting those USB keys to Scope as a preset).
- A synth with loadable waveforms, that can do PWM and sync on those
- An advanced ARP/Sequencer
- A dual oscilloscope

With regards to my original question:
I'm looking for a syncable version of the Raw Saw Up (aliasing) osc atom. Anyone got an idea how to do this? I've been missing this on several occasions as a sort of ramp tool.

Cheers!
Søren

[valis]

Do you have Flexor? Assaf got around the pad update issue by making everything with INT math, using cascaded adders. You can generate ramps, and there are NBL oscs too.

Let us hope Holger fixes the 32bit memory addressing issue so that drivers addressing out of that range no longer crash, and for whatever other ways it may help.

wxWidgets was correct in the 32bit era, I presume there's some 64bit version now and the UI code hasn't changed.

[nebelfuerst]

Although I prefer C over C++, I'm able to adopt linuxstuff to my needs.
But just knowing C++ doesn't help without proper documetation of the api of the scope-interfaces and a starting "helloworld.cpp" would be very nice.
Is there some kind of SDK on C++ level ? (I don't think there is anything included in the "normal SDK" )

[Spielraum]

I have several projects in the furnace - but this is sparetime work. Especially with absolutely no help from SC

HUGE THANK you from my side!
i also sacrifice free time and consider it a essential infulsion for our scope universe.

soren, please keep up the good work and continue to surprise us with "excitement, fun and games" in the future.
...currently need the chocolate less ~~

[soren_jepsen]

So....to continue my quest, I whipped up this circuit. In the attached picture I've set it up for 2K Hz. I've calculated the frequency thus:

4294967294/(44100/2000) = 194783097 (rounded)

However I end up with about 1917 Hz, as shown in the second picture. What's wrong with my math? Can someone explain why the frequency is offset?

Cheers

Ramp.PNG (6.54 KiB) Viewed 29814 times

Freq.png (19.9 KiB) Viewed 29814 times

[spacef]

have you tried with a few more values to see whether there is a pattern ? in such case I suppose it would be easier to offset with a constant or additional math function.
otherwise, it might be more difficult (for ex, for CV pitch values, there is no clear curve or pattern, so notes can't be done "mathematically", there must be a table with all required values, as far as I know)

[soren_jepsen]

Thanks for your thoughts

I figured it out though. Actually rather straight forward.

When the threshold for the Compare module is crossed, the Out of the Mix4 module has already seen the max value. Solution is to set the threshold at (Max - frequency value).


Cheers

[Spielraum]

i would welcome it in advance if old dsp programmers would report back to release the old atoms or adapt them to the new dsp's... e.g. the old sq-atoms. that doesn't seem to fall within holger's area of responsibility, otherwise after 14 years of XITE one or the other birth defect would have been corrected, i think. i hope that the errors are not transferred to JOANA...