Open-source SM?

TomC · 26 Feb 2009

Hehehe 🙂

I often thought about the same thing.

IMO it depends on:

- how much the SM-devs earn each month from SM licenses - how much they want be able to control into what direction SM will go - how good the architecture is right now

Anyway, if this ever should happen, I'm in 😍

Tom

MyCo · 26 Feb 2009

I've been thinking about this too, but I've several problems with open source. There are many projects that died when they switched to open source, because everyone adds bullshit and nobody realy knows about all code parts of the project...

So there has to be a leader or a group of leaders like in the mozilla project, that are responsible for every release.

And I think outsim has to get the money for the work they already did. So that the project is more like a customer contributed closed source...

Andrew J · 26 Feb 2009

MyCo wrote: more like a customer contributed closed source...

I'd certainly be interested in contributing some code! 🙂 However I can easily imagine that it could introduce as many headaches as it relieves for Outsim 🙁 Not saying my code is bad or anything 😉

infuzion · 27 Feb 2009

MyCo wrote: There are many projects that died when they switched to open source, because everyone adds bullshit and nobody realy knows about all code parts of the project.

True, or it is released, & noone does anything to is, like plasq's Wormhole.

Andrew J wrote: However I can easily imagine that it could introduce as many headaches as it relieves for Outsim.

Either open or closed source, both require ALOT of testing. Not to gripe again, but part of my frustration is that SM is finally beginning to feel like it is becoming 1.0, even though it was 1.0+ for a long time. Again, not fault of the devs since they have only so much time, but when you charge

Tom7777 · 27 Feb 2009

I have always like the addon extensions idea, i was thinking if it allowed loading of a certain format people could make stuff like keyboard shorcut menus, helper tools, export to other formats etc. Im not sure a full OS version would happen now that FL has say over things also though (im guessing). I think if it was just a chosen features/options SDK with some sort of extensions folder where you add them and also for 3rd party modules it would be great though.

infuzion · 27 Feb 2009

Tom7777 wrote: I'm not sure a full OS version would happen now that FL has say over things also though....

That would be sad, since it seems about 20% of music computers are Macs.

Thanks for the feedbacks guys!

Tom7777 · 27 Feb 2009

infuzion wrote: Tom7777 wrote: I'm not sure a full OS version would happen now that FL has say over things also though....

That would be sad, since it seems about 20% of music computers are Macs.

Thanks for the feedbacks guys!

By OS in that post i meant open source (just short text) rather than mac OS/OSX, but on the mac subject i wonder whats happening with that. It would be a great feature to export to AU format because as you say theres always a market for mac music software. I think a SDK would be cool even if it's limited to certain features and it also means less work on the devs if theres a good enough 3rd party module/feature. If it ever happened maybe we would be able to have new exporter for AU plugins though as well as other interesting stuff.

infuzion · 27 Feb 2009

Tom7777 wrote: ...but on the mac subject I wonder whats happening with that.

It is not. Seems they need to hire another programmer just for OSX; Malc has more than his hands full. I do not see that really happening either, since I assume most people are happy with Reactor & Max/MSP, even more so since Live-Max will be released in months.

electrogear · 27 Feb 2009

I agree completely with this idea. If SM isn't going to hire more developers then this is the only way of making this software a viable tool for professional developers. SM needs to be improved and I think there's so much work to do it would take one man years to accomplish it on his own. Plug-ins generated with SM are too high in CPU. One shouldn't have to strip stock modules down to their component parts and rebuild them to make them more efficient - having to do all of this time consuming work detracts from the creatability factor 😝 and magnifies the frustration factor! 😠 I still have little problems here and there which I can never seem to track down or repeat which is frustrating. In a big project all of these little problems and inaccuracies can add up to make a wholly unusable plugin. I didn't realise all of this until I got a good way into a large project I was working on, and at that point there was no turning back. I had to put in two or three times the effort I thought I would and the end result is still seemingly useless.

It wouldn't have to be open-source per se, - improvements to a master copy could be moderated - kind of like bug-solving submissions. Instead of reporting a bug we could fix it ourselves, submit the update to Malc, then he could add it to the master copy. Like Myco said - kind of like a custmer-only 'closed source'...

stw · 27 Feb 2009

infuzion wrote: That would be sad, since it seems about 20% of music computers are Macs.

I guess much more in the professional scene... and a multiple of that if you take the part of capable pros willing to buy software - which makes the Mac market IMHO much more valuable to developers than these 20% pretend.

Andrew J · 28 Feb 2009

electrogear wrote: In a big project all of these little problems and inaccuracies can add up to make a wholly unusable plugin. I didn't realise all of this until I got a good way into a large project I was working on, and at that point there was no turning back. I had to put in two or three times the effort I thought I would and the end result is still seemingly useless.

That's what is called a "learning experience" 🙁 Still, the earlier you have this one in life, the better!

infuzion · 28 Feb 2009

Andrew J wrote: electrogear wrote: In a big project all of these little problems and inaccuracies can add up to make a wholly unusable plugin. I didn't realise all of this until I got a good way into a large project I was working on, and at that point there was no turning back. I had to put in two or three times the effort I thought I would and the end result is still seemingly useless.

That's what is called a "learning experience". Still, the earlier you have this one in life, the better!

I've re-noticed for myself, 70%+ of my coding time is actually bug-testing & fixing. So I understand how one programmer can be easily overwhelmed doing all that, while dealing with customers. Malc does a commendable job at programming & customer service, but he is only one person, & IMHO SM needs a few more people hacking away at the code.

Tom7777 · 28 Feb 2009

infuzion wrote: It is not. Seems they need to hire another programmer just for OSX; Malc has more than his hands full. I do not see that really happening either, since I assume most people are happy with Reactor & Max/MSP, even more so since Live-Max will be released in months.

So it's just malc at the moment basically, is jorge no longer working on SM? haven't heard/seen him on the forum for ages so it appears that way.

electrogear wrote: It wouldn't have to be open-source per se, - improvements to a master copy could be moderated - kind of like bug-solving submissions. Instead of reporting a bug we could fix it ourselves, submit the update to Malc, then he could add it to the master copy. Like Myco said - kind of like a custmer-only 'closed source'...

Yeah this would be by far the best thing, also other than that or a selective SDK anything else pushes it towards freeware open source, so i think customer-only source is best. Since theres lots of good coders here this would be very helpful to malc i think, also locked modules as well as SDK however the dll compiler needs to be very hard to extract data from then but i know malc will be working on that.

infuzion wrote: I've re-noticed for myself, 70%+ of my coding time is actually bug-testing & fixing. So I understand how one programmer can be easily overwhelmed doing all that, while dealing with customers. Malc does a commendable job at programming & customer service, but he is only one person, & IMHO SM needs a few more people hacking away at the code.

Yeah, for years i sent malc osm bug reports but it seems sadly even though the app is much more stable in general there is still problems. I have been using it less for various reasons also these days. Malc is a great developer but as you say the work load is so high that he needs a bit of help so i think this would help.

infuzion · 28 Feb 2009

Tom7777 wrote: So it's just malc at the moment basically, is jorge no longer working on SM? haven't heard/seen him on the forum for ages so it appears that way... Yeah, for years i sent malc osm bug reports but it seems sadly even though the app is much more stable in general there is still problems.

Malc still has to approve architecture changes to SM with Jorge, so Jorge is still in the mix somewhere.

I understand your feelings with the bug reports; I think I've reported a few more than you! Yes, SM is still much better, but I'm still finding bugs, & I'm so tired of reporting them.

Tom7777 · 1 Mar 2009

infuzion wrote: I understand your feelings with the bug reports; I think I've reported a few more than you! Yes, SM is still much better, but I'm still finding bugs, & I'm so tired of reporting them.

I didn't know it was a competition 😉 but yeah i used to just send in packs to malc by email quite regularly but over time i sort of gave up on it as it seemed like i was the only one and i would always find more again. Since i have lots of interests SM has recieved less of my time recently although i check the forum often.

There was a time when i would spend all night using SM and 6-8+ hours every day just making things but over time i realized it probably wouldn't be up to the task of the sort of stuff i wanted to make, i have learnt code and asm much better now but still due to reasons like the current unsafe vst compile, buggy busses, slow arrays, loadtimes. Also missing options i asked for over the years from things like copy to last clicked location, sub folders, duplicate, midi player, extra controls on certain modules, hold key drag to make inputs on module and just loads of other things some of which i think i requested via email. Of course a lot of stuff i asked for was also added like wireless modules but it seems the less popular ideas always got left behind.

So due to these and various other things as i don't use SM as much as in the past so far less bug reports, i had a crash today actually also just from disabling DS Out and turning it on again a few times during some tests.

That will have sounded negative i guess but thats how it is for me atm, i think Malc is a great developer, very helpful, SM is a great app just a bit buggy and needs a few changes/updates here and there and hopefully the project will continue and keep getting better. 🙂

RuleT · 2 Mar 2009

It is good idea, guys! But, only in one case - if it will assist SM in being aswim, becoming better and more popular. It should be Firefox-like project, it means that SM is need in good sponsors such as Steinberg etc and more good developers such as Malc.. 🙂 And then Reactor & Max/MSP will be crying.. 😀 Actually it should become world-wide known product because it's "must have" almost for everyone (musicians, producers, DJs, teachers, students, professors, developers, theorists etc) and it's so simple in use! And there can also be much more applications of SM than DSP only. 😉

Tom7777 · 2 Mar 2009

It is world known already though and even more so with the fact it's part of FL Studio now. I don't get what you mean by 'Reactor & Max/MSP will be crying' though because Reaktor already has a huge user base/library and is aimed at the performance side and doesn't make plugins, but the development/building side is very deep and well done. Max/MSP is great if you can use it and can do all sorts of stuff including VJ and external hardware control and both are good apps.

You forgot SynthEdit which does the same concept and did it before SM i think or at least it was more known in the past back then. Similar apps will always come along also over time. It's more of a case of building on whats there already which people like and makes it a great app only with one main developer and constant/regular bugs things can build up request/bug fix wise and things can get left behind easily.

Also a lot of people using the forum have already paid for the app and to suddenly make it free opensource would annoy lots of people im guessing, so it would need to be closed to owners of the app only and also selected parts open to users rather than like firefox where it's free to all. For example if all the code was open, someone could see how the vst was compiled/protections and build a tool to extract people projects with opensource SM. So it has to be restricted/selective parts open to users for reasons like that.

Personally i think a SDK would be the best thing but also to have options to make enhancements to the app itself. I think with this if it happens it will have to be done right. User developments would allow 3rd partys to work on something that was a common request and share it with the community, only doing that can lead to restrictions, rules and also extra costs to the user sometimes so not always perfect but in general great to have available when done well.

exonerate · 2 Mar 2009

I honestly don't think SM would ever be open sourced and I'm sure the thought hasn't ever crossed Malc or Jorges' minds. If they did open source it would be the end of SM I think, it would mean they have had enough, they would no longer be the main developers.

Open source is great in theory, but it doesn't really work so well in practice, unless the project has a lot of funding/donations. A project like synthmaker needs at least 3-4 people working on it full time, which requires money, it's ok saying we'll all chip in with code/donations but I'm certain development would actually be (a lot)slower than it is now.

"Closed customer source" is, to be polite, probably the worst idea I've ever heard, what about new customers? where does their money go? what if I'm submitting more code and doing more work than Malc, shouldn't I get a cut and everyone else who chips in? this would cause a lot of problems.

A better Idea would be an SDK, but an internal one. I'd like an SDK but I am seriously against having to link to lots of external dll's (also imagine the CPU hit calling a dll per sample). SM could ship with it's own C++ compiler, (or possibly link to an existing compiler?) So you write the code in SM and it gets compiled into the schematic. That would be the ideal solution, kind of like the way Lua works in SMX.

Cheers 😉

Exo

hansdump · 5 Mar 2009

Hey,

some kind of new Ideas I like very much:

- a c++ code component would be the killer feature nr.1 in my opinion (or an improved code component with block processing and more math possibilities)

- it s absolutely right, that making of gui is the big advantage of synthmaker

so why not concentrating on stuff, that the other´s don´t have? For now, synthmaker is nothing more than a nice playground with a very good forum. Students can learn so many things... but after that... go away and make it their own with c++. You know why.

Don´t forget, that Reaktor or synthedit improve their products since a long long time. This is their advantage. Sry, but I know developers from native instruments loughing about synthmaker. Indeed... how do you SM developers think you can catch up this disadvantage? Spending day and night for coding?

If you are a professional, you make everything in c++, but the plans for that, you can try out with SM, better Reaktor, PD or Synthedit... so maybe one could embedd the own built synthmaker-gui into the "professional" c++ (vst sdk, linux) workarround somehow? --> this would be the inverse killerfeature.

Interoperability is a good approach for the future in my opinion!

I would pay money for that! Open Source is debatable if you solve the problem of paying developers.

I see here no really new approach with SM, except this gui thing.

Btw you know, what´s coming up next on ableton 8 with max4live?? http://www.ableton.com/home

Sry 4 the hard words...

Hans.

infuzion · 6 Mar 2009

hansdump wrote: So why not concentrating on stuff, that the other´s don´t have? For now, synthmaker is nothing more than a nice playground with a very good forum. Students can learn so many things... but after that... go away and make it their own with c++. You know why...

Btw you know, what´s coming up next on ableton 8 with max4live?? http://www.ableton.com/home

I was hoping with the Open Source, SM can become a hybrid IDE - C++ framework, kindof like if one would use SM's frond end to build & it had an C++ export with JUCE-like libs based on SM.

Then one would have best of both worlds.

I'm very aware of max4live; if I have the money/time, I'm likely to jump ship to that, since Live is my main platform anyway. Maybe I'll also release a mono VSTi with SM & see where that goes, maybe not.

nix · 6 Mar 2009

I don't think NI are laughing at SM. I've found it more accessible than Reaktor so. Cheers

rl · 6 Mar 2009

Right! The GUI part and the GUI design part is where SM shines. Now add improved GUI coding support as the LUA thingy in the (now jurassic) SMX prerelease and an SDK to build own primitives (did I mention that before?)

TheAudiophileDutchman · 6 Mar 2009

rl wrote: Now add improved GUI coding support as the LUA thingy in the (now jurassic) SMX prerelease and an SDK to build own primitives (did I mention that before?)

Votes= Votes + 1 🙂