Search results

What I found when doing the reverse delay block (Claude assisting with the original code I got most likely from the Spin forum) is that, when using just half the delay RAM, the segments are so short that it becomes difficult to perceive them as backwards. Doing a spectrogram using a sine sweep...

@Cybercow note that in the patch in question, the SKP RUN instructions are used to set up an LFO (SIN or RAMP) using a single instruction for rate and width. Subsequent runtime instructions (e.g. adjusting something via pot control) typically only update rate or width.

Doesn't have to be the same number. In this case it was. e.g. SKP RUN, 1 a ... SKP RUN, 2 b c ... SKP RUN, 3 d e f ... becomes SKP RUN, 6 a b c d e f ----- DL

Hi @Cybercow SKP RUNs are initialization instructions and only don't skip once at the beginning of execution. Meaning that during normal operation they skip over the stated number of lines. Here are the 3 in your program. SKP RUN ,1 WLDS 0,123,23261 ... SKP RUN ,1 WLDR 1, 16384, 2048 ...

I see at least one opportunity for optimizing that I hadn't previously thought about. Aggregate all of your SKP RUNs. LMK if you don't know what I mean. Here's another. before: ;------ Mixer 2:1<br> RDAX ADCL,0.5011872336<br> RDAX ADCR,0.5011872336<br> WRAX REG1,0.0000000000<br> ;------...

You have to change them to SPCDJ files, not .txt files. SpinCAD is looking for a file whose extension is SPCD or SPCDJ. DL

I spent a little time recently to see if I could get Claude to conjure up some oil-can patches directly in SpinASM. Still working on it, to confirm it does what I think (FV-1 is not one of Claude's strong suits). One nice thing about this approach is that it's super easy to ask it to create a...

I'm pretty sure it will not work correctly on an FV-1. SpinCAD's simulator has some wiggle room on resources simply because I decided to make it work that way. So you can go over on instructions etc. to a small amount and still hear it. I'll check it eventually on Mac to see whether it...

It's somewhat odd and it may be a bug in the Oil Can Delay module. I had Claude do the analysis and it seems this is so. Delay Time pin unconnected — the slider directly sets the buffer size: delayLength + modAmplitude + 64 samples. Moving the slider changes memory consumption 1:1 (3341 …...

I got the same result. I'll look into it. Update: Remove the first two lines which say (between the lines): --- JSON: --- The first character in the file should be "{"/ Does it help? Btw I noticed memory allocation is in the red so that will have to be fixed. DL

You may want to nail down exactly which text editor and encoding scheme has been used. DL

Regarding (a), who can say? If you can come up with a repeatable sequence I can look at it. For (b), send it to me and I'll see what's what. In general, if you run the program in debug mode, there may be something useful shown on the console when these things happen. Otherwise they go into...

If you can find any SpinCAD algorithm that does not work in an FV-1, and the instruction/register/memory counts are all within limits, I'm interested to know the details. DL

@BrainGutz one thing I found about the FV-1 is that it may not have enough resources for you to implement everything you have in mind in a single patch. While I can't be sure and I am not going to put it past Cybercow, this did enter my mind as I read through your list. So then the challenge...

Man that is mighty generous of you @Cybercow . At least the guy does know what he wants. As you probably know, I'm in the "teach a man to fish" camp. I went through some painful experiences early on that make me really reluctant to do this sort of thing. I couldn't charge enough to make it...

I've stopped development on SpinCAD for the time being. Who knows what the future will bring but given this is not a money making venture (with all due respect to Cybercow and others who contributed) I have to prioritize my own selfish interests. e.g. I did a recording of "Fear and Loathing in...

AFAIK she does that using a Line6 DL4 with 2 expression pedals.

Yeah my early tests with chatGPT a year ago to write the optimizer code were a waste of time. I started using Claude at work and tried that on the optimizer and it knocked it out of the park. I've used Claude quite a bit since then for a variety of things, and sometimes you get lucky and it is...

Although I haven't tried it, the creator of the Expert Sleepers "Disting NT" Eurorack module came up with a tool "ThreePot" that converts Spin ASM to C code. Seems interesting if for no other reason than to port something you already came up with...

Well ackshewally it was 0.98 until this recent burst, so I gotta let 0.99 simmer for a bit, plus for whatever reason, other than you and a handful of pplz there's not a heck of a lot of enthusiasm any more... either AI is doing it all for everyone or people figured out that even with assistance...