Restarting the Fizmo Project

[betteroffted]

Dave: I'm really glad you're doing this. I get a lot of use out of my Fizmo and have always enjoyed it's "head-scratching" behaviors. It's pretty fascinating to me that such an oddly complex machine exists without proper documentation or updated editing capabilities. I would have assumed that, even with Ensoniq's buy-out, the Fizmo creators would have still had some independent involvement in its completion.

Anyway, my hat's off to you for taking this on. I'll be excited to try out your editor when the time comes.

[cornutt]

I've done some more over the last couple of weeks; I only have so much time per week to work on it, but I'm getting there. I'm going to be updating the project on my blog, probably once a week or so.

[DLovas]

I use currently to edit these parameters the full version of the (like you stated) VERY outdated full version of Emagic's Sounddiver 3.04 or something. I had a h**l of a time trying to get it to run (if anyone remembers from one of my prior topics on here about it) I admire your dedication and enthusiasm for this project! I'd love to see a new software made specifically for the fiz (preferably with a nicer appearance as well!!! you should tell the guys on the Yahoo group about this - they'd probably all back you up! perhaps you could procure some donations

I'd definently donate to this - however "i aint got no money" at this point so thats a bummer - but keep at it! If you could somehow program it to update things in real time (thats the one major thing i HATE about sounddiver) that would be swell.

Keep with this though! Although i haven't owned it long, I can say that the fizmo is one h**l of a synth (now if only i could get another.....)

lol this is probably a fairly outlandish request - but if you could give the software a way to view the waveform as you change its characteristics!! that'd be pretty crazy...

for me though the most important aspect would be real time updates (or at least have it update itself somehow) and a somewhat improved designed - sounddiver is the stone age to me.

[cornutt]

I'd definently donate to this - however "i aint got no money" at this point so thats a bummer -

Thanks, and the thought is very much appreciated. However, money is not necessary, and I'm not asking for donations. I just need to keep grinding on it. Once I get everything documented, we'll see where it goes from there. I've had the idea suggested to me about doing a cooperative open-source project. I'm thinking about how to put that together.

[DLovas]

I'd definently donate to this - however "i aint got no money" at this point so thats a bummer -

Thanks, and the thought is very much appreciated. However, money is not necessary, and I'm not asking for donations. I just need to keep grinding on it. Once I get everything documented, we'll see where it goes from there. I've had the idea suggested to me about doing a cooperative open-source project. I'm thinking about how to put that together.

Do what you Love, Love what you do

[cornutt]

Progress! I've worked out the locations of most of the sound and layer parameters in the patch dump. If you'll go to my blog post, there's a link to a spreadsheet in ODS format, which you can open with Open Office. You can download Open Office for free from openoffice.org.

So here's the question: who might be interested in working on a patch editor? I can contribute some, but I don't have the time to write the whole thing myself. I was thinking about starting a project on Sourceforge and loading the spreadsheet into there for reference.

[cornutt]

By the way, the dump layout absolutely sucks. It's the biggest mess I've ever seen. I'm really getting the impression that the Fizmo was rushed out the door with the software not really finished. It might be a miracle that it works at all.

[Cybercardinal]

Well i'm not a code writer so I can't help out but I appreciate your effort very much! Maybe later on I can help with the GUI, even though a simple one in purple probably will do

[dnigrin]

Dave, hat's off to you for tackling this project!

I will do my best to find some time to help (though I have many of my own software projects waiting for attention too, so no guarantees). I have experience in reverse-engineering stuff, albeit a different flavor (decoding the audio squelch that the Roland MC-202 and MC-4 sequencers use to store their sequences to cassette - see MC-202 Hack and MC-4 Hack in my sig below). I also have (fading) Java experience, as well as extensive Max/MSP skills.

One thought re: the end product -- the Ctrlr project has really taken off with its latest version: http://ctrlr.org. It now allows for WYSIWYG layout of interface objects, and "no programming" for what those objects send (either CC, NPRN, or SYSEX). You can get even fancier with Lua programming if you want to, but not necessary. It's not very well documented at all yet, but for straightforward work it's not hard to get going. It also offers opportunity for really nice graphics to be used as well (i.e. those made with KnobMan), so even those with no programming skills but who have good graphics skills can get involved. It's cross platform, and can yield either a standalone application, or can be built as a plugin too.

Will take a look at the files you have posted, and be in touch.

Dan

[dnigrin]

Can you post some details of how you used tr and diff on OS X to look for changes between dumps?

[StepLogik]

By the way, the dump layout absolutely sucks. It's the biggest mess I've ever seen. I'm really getting the impression that the Fizmo was rushed out the door with the software not really finished. It might be a miracle that it works at all.

Or the team building the firmware was incompetent.

In my years of consulting for various companies that build hardware devices, I've learned that software is generally treated as an afterthought if the team is composed primary of Electrical Engineers - especially old school guys who still want to do everything in hardware. The code is rarely elegant and is usually very inefficient. To make matters worse, these teams often scoff at the thought of hiring an actual software developer to code the firmware when they can just get an intern or apprentice to hack it out. UX expert to design the LCD pages and button panels? 100% out of the question, "Waste of good money" as one dev manager I worked with put it.

My new Cuisinart Toaster Oven is really nice, but I can tell the EE who designed the firmware was in over his head. Nothing is intuitive, functions are inconsistent, and it takes odd combinations of buttons to accomplish simple tasks. Much like a Yamaha TX-16w

[cornutt]

One thought re: the end product -- the Ctrlr project has really taken off with its latest version: http://ctrlr.org. It now allows for WYSIWYG layout of interface objects, and "no programming" for what those objects send (either CC, NPRN, or SYSEX).

Thanks. I'm going to go look at that and I'll report back.

[cornutt]

In my years of consulting for various companies that build hardware devices, I've learned that software is generally treated as an afterthought if the team is composed primary of Electrical Engineers - especially old school guys who still want to do everything in hardware.

My dad's an old school EE and he would totally agree with you. There's something about studying RFI and grounding that warps you brain. The weird thing, though, is that the guts of the Fizmo are basically the same as the MR/ZR series, but those synths don't seem to have the software bizarreness that the Fizmo has. (I could be wrong about that... someone who's used an MR extensively chime in.) If that's true, I wonder what happened. Did they lose all of the software staff before the Fizmo design was started? Could have been if employees thought that the company was in trouble. I picture a bunch of brave-but-doomed designers pulling all-nighters in basement cubicles trying to crank out the product that will Save The Company. Could make a good movie.

My new Cuisinart Toaster Oven is really nice, but I can tell the EE who designed the firmware was in over his head. Nothing is intuitive, functions are inconsistent, and it takes odd combinations of buttons to accomplish simple tasks. Much like a Yamaha TX-16w

Man, if there's one thing I hate, it's anything that makes you do a bunch of non-obvious button press combinations to get at certain functions. Would it really have added that much cost to put one more damn button on the panel?

[balma]

Yeap, move the transwave selection knob to edit a preset, and then, try to go back to the transwave you were using....how the f**k do you know wich transwaves are being used on each patch?. This synth is so random/matter of luck when creating sounds.... this almost force you to learn by ear each one of the transwaves.

and to make things worse, is custom painted and knobs does not have letters...

when my friends look at it, next to the Vsynth (wich has an impressive "hi-tech" design) they say, balma, what´s that piece of junk?

yeah, that piece of junk is around $1600 right now.... and also they change opinion when they hear its bizarre sound.

[StepLogik]

I wonder if the DSP code/algorithms are still archived somewhere? It would be nice if someone could get their hands on it in order to shore the Fizmo or just build a whole new synth altogether.