[ANN] Repro-1

[david.beholder]

I don't think it matters in the long run.

Well it is. It's less rnd, qa and support which is less money/time allocation.

[EvilDragon]

Not really less R&D when u-he has their own framework that is already set for compiling their plugin in all sorts of plugin formats, sooo... It would actually probably take them more R&D to remove the 32-bit plugin compiling code than to just leave it

[Funkybot's Evil Twin]

Not really less R&D when u-he has their own framework that is already set for compiling their plugin in all sorts of plugin formats, sooo... It would actually probably take them more R&D to remove the 32-bit plugin compiling code than to just leave it

At some point though, it's probably better for everyone if we just kill 32-bit plugins as a format. What percentage of folks aren't running 64-bit DAW's and what percentage of those could be if they wanted to? There are fewer and fewer legit reasons for not being 64-bit in this day and age, so any developer that wants to drop 32-bit support to cut down on compiling, testing, support costs, etc. has my full support.

[Funkybot's Evil Twin]

Just to not be completely OT: is there any update on the progress of RePro-1? Are we fast approaching release, or is still looking like it's a few months away? I'm really excited about this synth.

[dark water]

Not really less R&D when u-he has their own framework that is already set for compiling their plugin in all sorts of plugin formats, sooo... It would actually probably take them more R&D to remove the 32-bit plugin compiling code than to just leave it

At some point though, it's probably better for everyone if we just kill 32-bit plugins as a format. What percentage of folks aren't running 64-bit DAW's and what percentage of those could be if they wanted to? There are fewer and fewer legit reasons for not being 64-bit in this day and age, so any developer that wants to drop 32-bit support to cut down on compiling, testing, support costs, etc. has my full support.

There has been too much killing and death.

Save the 32-bit plugin format.

[dark water]

As for the previous page suggestions, this would be rather tasty if anyone ever fancied emulating it in software form:

[david.beholder]

Not really less R&D when u-he has their own framework that is already set for compiling their plugin in all sorts of plugin formats, sooo... It would actually probably take them more R&D to remove the 32-bit plugin compiling code than to just leave it

Since this is c or c++ framework and compilable for both platforms one #define would permanentely turn 32 bit code to legacy. So it's rather showing validity of the idea as they don't need to support 32-bit part of framework code anymore.

What percentage of folks aren't running 64-bit DAW's and what percentage of those could be if they wanted to?

According to Urs himself ~10%. Urs mentioned it after survey.

And it's really on topic because Urs said that asm-optimized version would be only for x64 or something like that.

[Urs]

The one thing that makes 32-bit only interesting is, we could fully drop the remnants of Carbon on MacOS X. Going 64-bit itself still means supporting a convoluted plethora of different CPU features.

It would be rather interesting to drop support for dated CPUs, e.g. anything prior to the AVX instruction set. However, I guess that's not gonna happen anytime soon.

[david.beholder]

It would be rather interesting to drop support for dated CPUs, e.g. anything prior to the AVX instruction set. However, I guess that's not gonna happen anytime soon.

Do you have statistics about percentage of pre AVX users?

[Urs]

Just to not be completely OT: is there any update on the progress of RePro-1? Are we fast approaching release, or is still looking like it's a few months away? I'm really excited about this synth.

The VCOs (and Mod section) are now done. It was a long and stoney path. Self-FM and self-PWM of Osc B are now on a level that will be hard to beat in realtime software.

Yesterday we turned one of our Pro-Ones inside out and had an oscilloscope all over it, measuring the hell out of a few left over things we couldn't fully make sense of. Interestingly, some parts are not what the schematics say. So they probably solder in what they had at times, as long as it was compatible. With those measurements at hand we'll start a final round of sonic tweaks and treatments of the actual audio engine.

The UI is on a good path, but some final trim pots and jumpers need to be decided upon. Also, the sequencer might get a visual editor, which might need to be designed as well.

We've spent a lot of time pondering about some conceptual questions. We dismissed a poly version and we decided to offer a "normal" stereo out synth plug-in with FX and a version with multiple dedicated inputs and outputs. Latter won't have any on-board FX but will be patchable in a modular environment.

FX are a necessity because of the inevitable "hey, this Youtube video of a real Prophet sounds better than your softsynth" from people who are oblivious to otherwise obvious saturation, compression, delay and reverb on that synth demo. I've decided against our fx modules from Diva, Hive, Bazille or Zebra and so I've started implementing a few new ones.

All in all, if we're lucky we can start beta testing in September - but hey, I said it might take till x-mas

[Urs]

It would be rather interesting to drop support for dated CPUs, e.g. anything prior to the AVX instruction set. However, I guess that's not gonna happen anytime soon.

Do you have statistics about percentage of pre AVX users?

nope... but I reckon it'll be a lot.

[Funkybot's Evil Twin]

Sounds like there's been some really good progress, thanks for the update!

Also, the sequencer might get a visual editor, which might need to be designed as well.

I actually thought the three-octave keyboard you posted on Instagram WAS the visual editor for the step sequencer.

If you just added an LED to show the value of the current step, and previous/next button, you could have the sequencing handled entirely in that section of the UI. The downside would be that users could only see one step at a time.

[NerdMcBoon]

If you just added an LED to show the value of the current step, and previous/next button, you could have the sequencing handled entirely in that section of the UI. The downside would be that users could only see one step at a time.

Indeed. A 303-ish sequencer with MIDI input to enter the notes would fit the concept much better than a piano roll like visual editor. However, judging by forum posts about synths without a visual representaion of sequences, I think most people actually prefer some kind of easy visual representation.

[mholloway]

FX are a necessity because of the inevitable "hey, this Youtube video of a real Prophet sounds better than your softsynth" from people who are oblivious to otherwise obvious saturation, compression, delay and reverb on that synth demo.

It makes me sad that developers feel they must tailor / adjust their products to accommodate the dumbest, least-informed members of our society, but I also understand why this choice would be made (and the amount of headache it could save you). But that said, the dumb and uninformed will always find a way to whine and shout...

On board effects don't mean a whole lot to me since I almost always turn them off in favor of my DAW's native effects or various 3rd party effects, so I guess I'm biased as well. But I suppose if you're doing brand new FX that specifically "fit" with the synth (though I have no idea what that would actually be, I guess stuff of the same era?), I suppose this could be potentially interesting. Not really related, but I do love that Reverb in Zebra, one of the best verbs I've ever heard!

-M

[Funkybot's Evil Twin]

It makes me sad that developers feel they must tailor / adjust their products to accommodate the dumbest, least-informed members of our society, but I also understand why this choice would be made (and the amount of headache it could save you). But that said, the dumb and uninformed will always find a way to whine and shout...

I just hope Urs and co are prepared for the chorus of "what do you mean it's not polyphonic, it's software, how hard is it to add polyphony" complaints that will inevitably follow for this release.