Quote from: stallik on November 29, 2022, 04:06:24 PM
So, I've had this board for a year now and only just started populating it  Have a few parts on order but am looking forward to going down this rabbit hole in the new year.

With zero live experience but a little time playing with SpinCad, I have a question. This board is mono using the left output from the FV-1. Would it be possible to use the right output and the relevant circuitry after it and (perhaps using dual pots) create a stereo version usable with a dual amp setup? As I'm not using the FV-1 daughter board, there should be enough room to mount another board above the chip and still keep the same size enclosure.

Or, are patches which use such stereo effects going to be too limited by the processing power of the chip?

You can also put in a feedback loop, perhaps with some clipping diodes to get some crazy effects

Thanks for the replies guys.

Seems to me that if I were to introduce a daughter board for the right channel of the fv1, I wouldn't necessarily have to use it on every patch. So far, I've only played with SpinCAD but some of the stereo patches I've put together sound promising though I've no idea how they would fare in real life.

I think this rabbit hole is going to be fun - it's going to be too cold in the workshop for woodworking till spring

Bumping this thread as I've made some (slow) progress on making my FV-24 work in stereo. It's been a catalogue of errors so far with me not really thinking things through properly, correcting the errors then realizing I've made some more.

Because I'd used a SMD processor and Rob designed his FV-24 board to accept either that or a daughter board, I thought I could recreate most everything on the circuit diagram following the left channel output of the FV-1, shove it on my own daughter board and hang it above the chip.

So, learned KiCad, designed version 1, waited for the PCB's on a slow boat from China then realized that the whole assembly wasn't going to fit into a standard depth enclosure BUT if I redesigned it to have the components under the board, and was careful with positioning, it wouldn't make the thing any taller than the existing board.
So, another board and another slow boat. Version 2 arrived this evening so I populated it and believe it or not, it fits! Well, except for the TL072's. I'd used a DIP-8 (10.16mm) footprint    It was possible to spread the pins to shoehorn them in but it's far from perfect.

Hoping to use dual wafer pots for tone, mix & vol to keep things simple and have created a couple of stereo patches just for test purposes. Wire up everything over the next few days then see if there's any worthwhile musical point to it

Question: what is the DAC's bit depth? AFAIK, the FV-1 wants to run at 32kHz (16kHz bandwidth, which is more than fine for guitar - for anything really). If it's less than 16 bit, do you use a compander (or just go with the grunginess?)?

QuoteThe resolution of the ADC and the DAC is 24 bits

(from Spin Semi architecture page)

So, progress on my stereo board has hit another snag. This time by a mistake I made a couple of years ago. While Deadastro was talking about using a Pickit 2 for programming the chips, I ordered a Pickit 3. New version must be better right? Apparently wrong. Or at least according to the interweb.

I read that the 24lc32a chip can't be programmed using the Pickit 3 (as reported by MPLAB IPE 6.00) unless you tear it apart and remove a couple of resistors and then it might... Other reports suggest just getting the Pickit 2 which does work so - I ordered one but it's on another slow boat from China. I also ordered a small programming board just in case my difficulties are really being caused by a mistake I've made building the main FV-24 board.

Yesterday, I was ready to give up on this, today I have a plan to take it forward but I can't help thinking that something is trying to prevent me going down this rabbit hole.

On a positive note, I've found some stereo SpinCad patches made by users of another FV-1 stereo board (which doesn't seem to be available yet) and they will really help both testing and learning how to construct my own patches.

Yep, pain in the backside about the PICKit3 not progrmming the EEPROM. I mean, who takes features *out* of a product between versions?!? Perhaps there was some good reason I'm not privy to, but it does seem a bit loopy onn the face of it.
Like you, I have both PicKit2 (for that one specific job) and PICKit3 (for everything else).

Seasons greetings everyone.

My pickit 2 arrived this evening and a few minutes later, I'd programmed my first chip. Happy? Well, pleased that I could program a chip but less happy that it doesn't work. My problem is that there is no effect coming through the right channel. Reason for that is that I'm stupid and made some presumptions when I started this. Because Rob designed his board to accept either the smd chip or one on a daughter board, I assumed that all pins were linked between the smd pads and the through holes. Apparently, that's not the case with the right out pin and that's where I'm connecting my daughter board.

I can correct the issue with a simple link but to do that, I'd rather drill a hole in robs board so that I can keep the wire length short. I've had far too much Merlot this evening to perform such a task so will leave it till tomorrow.

It's alive!!!

Yeah, should have waited till tomorrow but you know how it is... Contrary to (my) expectations, this thing sounds awesome in stereo.

Will have to learn how to record it for some sound bites but the rabbit hole is well and truly open and I'm going in

I approve of rabbit holes. 

I approve of Merlot.

You win, Ron.

But. Does the FV-1 make economic sense today? (Honest question).

Quote from: fryingpan on December 28, 2024, 06:53:38 PMBut. Does the FV-1 make economic sense today? (Honest question).

I'm not sure I understand the honest question.

Compared to what exactly? There isn't a lot of direct competition...of course, there are a million other DSP options.
"Economic sense" how? What are you trying to do that th FV-1 might do that you think you could do some other way cheaper?

Quote from: ElectricDruid on December 28, 2024, 07:23:49 PM

Quote from: fryingpan on December 28, 2024, 06:53:38 PMBut. Does the FV-1 make economic sense today? (Honest question).

I'm not sure I understand the honest question.

Compared to what exactly? There isn't a lot of direct competition...of course, there are a million other DSP options.
"Economic sense" how? What are you trying to do that th FV-1 might do that you think you could do some other way cheaper?

Basically, yes. I mean, are there any other chips with decent ADCs/DACs (16 bit is fine already, 12 bit can be fine too) and good processing power that allow better processing to be done? The FV-1 is, dunno, 15 years old now?

I have a question.

I'm using Spincad Designer (version 0.98 build 1039) I create patches 0 to 7, save the bank as hex then open it in PICkit2 programmer then write it to the selected 24LC32A. Each time, patch 0 is ignored and patch 2 ends up in patch 1, patch 3 in patch 2 etc. Finally, patch 0 is inserted into patch 7

I can get round it by imagining the patch order in SpinCad is 7,0,1,2,3,4,5,6 but I'd like to know if I'm doing something stupid or whether it's just a quirk of the FV-24 switching.

The rabbits down here have been no help whatsoever...

Quote from: fryingpan on December 28, 2024, 08:39:04 PMI mean, are there any other chips with decent ADCs/DACs (16 bit is fine already, 12 bit can be fine too) and good processing power that allow better processing to be done? The FV-1 is, dunno, 15 years old now?

Well, the FXCore is the most direct replacement, but it needs an external codec.

http://www.experimentalnoize.com/products_manufacturers.php

If you're ok with having a separate codec, you have a lot more options. You could go with any of the ARM chips. The original Mutable Instruments eurorack modules were all based on STM-somethings, I think, so you can clearly do some decent DSP on these chips, even if they're only general purpose. The dsPIC series would be another option. They're probably not as powerful on paper, but they do have specific DSP instructions and if you program to get the best out of those, you can get them to do about 5 or 6 things on every instruction cycle, which makes up for a lot!

There's still nothing that's as simple and all-contained as the FV-1. Which is a pity, and also probably why it's still around.