Sauraen

Updated board (ver. 1b) (as usual, right-click on the fourth image, the "smaller" schematic, and select Open in New Tab to see full resolution)

I met and hung out with Josh Whelchel at MAGClassic, and he agreed to pay me at cost to build two MIDIbox Quad Genesises, one for him and one for me. So the project is officially a GO! The schematics above are not up to date, I should be able to put up better ones in a couple days. The big change I made was replacing the two discrete clock oscillators with a dual oscillator circuit using crystal resonators instead. This was because crystal resonators are available that very closely match the frequencies used in the Genesis, so that when VGM files are played on the synth they will be at the original pitch. I have been working on the Wiki page for MIDIbox Quad Genesis and especially the one for the MBHP_Genesis module. There's a lot of information up, but there will be even more soon! Glad to hear you're on board! It will be extremely helpful to have someone else to test the hardware and software and give me feedback. This is not only MIDIbox Quad Genesis, it's not FM at all. :-P Please start a new topic somewhere else!

Thanks! More info and ideas coming soon. I expect MIDIbox Quad Genesis will take a year. After that I'll keep it in mind.

They're all over eBay for cheap. Don't know if they're genuine of course, but they're cheap enough I can test-order one or two of each from a seller and see if they work. There's a couple marked as "N sold" where N > 50, so it's unlikely that seller is fake.

I've gotten a decent amount of interest so far on the form (for instance, I could sell about 30 module boards if they were $10 and about 50 if they were $5, plus I could sell about 10 front panel boards--which I was not expecting--if they were $70). Please fill it out if you're interested! I have a draft of the MBHP_Genesis module mostly done (values off the top of my head, not checked). The board includes the YM2612 (OPN2), SN76489 or SN76496 (for SN76494, I couldn't find a clock module that was 500kHz in this package, so you'll have to supply an external clock somehow if you want to use that chip), analog mixer/filter with digitally selectable cutoff frequency, and glue logic to access all I/Os from the parallel interface with no additional wires. Four boards may be stacked together, with certain pin headers and sockets, and certain components stuffed or not stuffed on each board--but it does achieve the claim of running four boards without a mess of wires plugging into each of them. It will, however, need a custom bus driver board between the STM32F4 and this module--even if the sound chips could tolerate the +3V inputs (they should be able to), I can't expect the MCU's fanout to be able to drive 12+ chip inputs. Details will come later. Here's the schematic and board, though the forum software resized the schematic, so there's a real full-size version attached to this post. Please forgive the minor graphical issues with the 3D view, I didn't feel like doing custom 3D modeling for the oscillators and the headers, or fixing the TO92.This is the full-size version schematic (click, then click Full Size in the bottom left):

EDIT: For project documentation, see http://www.midibox.org/dokuwiki/doku.php?id=midibox_quad_genesis and the pages it links to, specifically including http://www.midibox.org/dokuwiki/doku.php?id=mbhp_genesis ! This semester I will be building a MIDIbox Quad Genesis synthesizer for Josh Whelchel. The synth will be based on my MIDIbox FM V2.1 (dual OPL3) synth, except it will contain four OPN2 (YM2612) and four PSG (SN76489/SN76494). We are still in the early planning stages of the project, so many of the details are not set yet. However, one thing that's pretty sure is that I will produce a custom PCB for a "MBHP_GENESIS" module, containing one of each sound chip with a parallel interface to the STM32F4 core. These modules will be stackable with pin headers, and their outputs will be correctly mixed with no external circuitry when they are stacked, allowing up to four modules to be connected to the same core. I'm also planning for them to be supplied by the +5V rail (after heavy filtering), with rail-to-rail op-amps, so a separate +/-12V supply would not be necessary. Of course, the more that I have manufactured, the cheaper they will be for everyone. For budgeting purposes, I would like to know how many people here would be interested one or four of these boards. For now I don't need a commitment, just a statement of interest. Depending on what we decide for the project, I also may be producing a large, custom front panel board, which will be VERY expensive unless we find other interested people. The design is barely even started, so of course there would be no commitment; but if you'd consider building a MIDIbox Quad Genesis with a large front panel, let me know that you might be potentially interested. Here's a Google Form for letting me know your interests about this project, please fill this out if you would be interested in a board: http://goo.gl/forms/4Gg5JX6GpX

Yes, I guess I should have put a link. When I replaced the CPU, the project name changed from MIDIbox FM V2.0 to V2.1, and the development thread moved here: http://midibox.org/forums/topic/19073-midibox-fm-v21-on-stm32f4/

So how's the progress on the module drivers going?

Manual scan sniff with velocity to MIDI output is working! (Well, there's some bugs, but you can't hear them here, other than the somewhat uneven velocity response.)

Hey everyone. I am starting work on a new video game music remix. This is an arrangement of some of the emotional pieces from The Legend of Zelda: Skyward Sword, done in the style of Wagner. So I need someone to write lyrics in German, that will fit certain melodies and that will have textual content that refers to the game. Requirements: Must be a Zelda fan :) Must have beaten Skyward Sword Must be familiar with Wagner's music Must be fluent in German I would hope to get the resulting piece published on OC ReMix, but in any case not for profit. The librettist will be credited along with myself and the singers.

Thanks for the suggestion, but that wouldn't be appropriate for my design. I don't need USB or uSD, the fanciest thing going on internally is single MIDI packets! The breakout board I used was $25 and also had the GPIOs in consecutive order; the board I made under that was single-layer and milled by a friend for very low cost. Also, since I wanted to use assembler, I don't know ARM, only AVR (and MIPS ;) ). ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- The Electone saves on memory cards which are a 2kb SRAM with backup battery. I had some 512kb SRAMs lying around, so I decided to make a custom memory unit that would allow the user to select one of 256 "banks" (upper address bits) and show the organ the other address and data pins. The memory is backed up with a supercapacitor instead of a battery, so every time the organ is turned on it recharges. With the exception of the supercapacitor, all the other parts used here were spare parts or salvaged, I didn't buy anything! It was a great idea in theory, but not so much in practice--the bank-selection circuit works, but the organ shows an error when trying to read or write. Guess I'll troubleshoot it later.

I'm pretty confident there's no bugs in opl3.c, due to months of testing, so this is probably relatively accurate. But when you get to the synth engine, mbfm_whatever.c, this is absolutely not true. I have a list of known issues that I never got around to fixing; you'll probably find some of them as you're working with it. So don't be afraid to ask when something looks wrong! I've only learned assembler in the last couple of years, and then the only two I know are AVR and MIPS 1 (from N64 hacking, haha). So I don't recognize those two particular instructions. x86? ARM? I'm probably never going to try to learn a CISC ASM, leave that to the compilers. :) You can see some near-ASM C code in opl3.c for the SendAddrData routine. I tried to write lines of C that would compile to one or two lines of assembler each, since that code is time-critical. I know I tend to make it write-only sometimes, I guess that's not great for collaboration...

My custom processor board is up and running, and connected to the organ's main serial bus. It's just flashing LEDs for now, but hopefully soon it will be converting manual scan data into MIDI messages. As you can see, this is not a MIDIbox; it contains no MIDIbox code. The processor is an Atmel ATXmega128A1 (32MHz)--the top of the 8-bit line. No MIDIbox MCU had enough I/O pins for the manual scan, and reading them with shift registers might not be fast enough.