One-Synth-Challenge 57: Tunefish - Chat & Gossip

[Mushy Mushy]

I don't really get the logic here. As long as you don't edit whatever audio the synth produces, you can bounce it, right? As for drum consistency, what exactly is the difference between having the synth render the same sound 8 times or taking a sound and multiplying it 8 times? Sorry if I'm missing something obvious here (:

The same sound rendered once and repeated 8 times will always be identical.
The 8 instance loop will have minor variations in the envelope etc (can't explain why but it does). Especially noticable on kicks.

BTW, I have no comment on the appropriateness of the rule. Just stating the reason why it exists (IMO anyway).

[Sonnydeluxe]

Thank you!

Hmmm... Convenience and CPU-saving. So let's say, as an example, I want to use 3 toms with a different pitch. On the one hand, I could spread them across three different tracks with 3 instances of the vst, each instance controlling the pitch. On the other, I could sample them from the three instances (without slicing or modifying them) and could group them into a single drumrack, assigning a different key to each sound. Still, I wouldn't be allowed to use the different options of the sampler (envelope, filter etc.).
Now, the second option I would call "convenient" - I don't change the sound, but I bundle it for the ease of use.
Now, would this be okay?

[crimsonwarlock]

Thank you!

Hmmm... Convenience and CPU-saving. So let's say, as an example, I want to use 3 toms with a different pitch. On the one hand, I could spread them across three different tracks with 3 instances of the vst, each instance controlling the pitch. On the other, I could sample them from the three instances (without slicing or modifying them) and could group them into a single drumrack, assigning a different key to each sound. Still, I wouldn't be allowed to use the different options of the sampler (envelope, filter etc.).
Now, the second option I would call "convenient" - I don't change the sound, but I bundle it for the ease of use.
Now, would this be okay?

I think it's already made clear that repeating a sampled sound in any way, is not allowed.

[Sonnydeluxe]

ok, thanks.

[V'ger]

Yes the rule is that you can bounce and also sample sounds, but not do it for 'creative' purposes.

For example, some synths don't have synced oscillators and f.i. a bassdrum will therefore sound different each time, and you can't sample one nice one in order to bypass the synth's limitations as that would be simple cheating.

To avoid this issue altogether, it's best to bounce a larger part or even whole tracks.

Basically any sampling or bouncing must only be done when the EXACT same result could be produced without sampling or bouncing. And with EXACT I mean EXACT..

[bjporter]

Yes the rule is that you can bounce and also sample sounds, but not do it for 'creative' purposes.

For example, some synths don't have synced oscillators and f.i. a bassdrum will therefore sound different each time, and you can't sample one nice one in order to bypass the synth's limitations as that would be simple cheating.

To avoid this issue altogether, it's best to bounce a larger part or even whole tracks.

Basically any sampling or bouncing must only be done when the EXACT same result could be produced without sampling or bouncing. And with EXACT I mean EXACT..

And of course this is part of the challenge... some Synths are very consistent while others are all over the place

[Kondarivan]

The same sound rendered once and repeated 8 times will always be identical.
The 8 instance loop will have minor variations in the envelope etc (can't explain why but it does). Especially noticable on kicks.

I see. Thanks for clarifying. Never noticed this, sorry for being clueless...
But hey, it can sound more realistic this way

-------

Back to Tunefish, there's some odd behavior that I'm experiencing here:
- If inside the modulation matrix everything is set to <none>, the spline oscillator is triggered by common MIDI note on and note off signals, while the additive synth keeps playing even after the note has ended. Not sure what the expected behavior should be (as both make sense in a certain context, but I suppose they should be consistent (if you're wondering why your additive synth doesn't stop playing, it's probably because of this). I incline to think that the never-ending additive synth sound is not intended, as setting the volume to zero and back to a positive value stops the sound.
- If a modulation source of Adsr1 is present inside the modulation matrix and provided that nothing else in the matrix has Volume as a modulation target, the note is released about 500 ms later than expected (applies to the spline oscillator; the additive synth simply keeps playing).
- If a modulation source of Adsr2 is present, in the same conditions as above, the note is never released.
Also, I'm just nitpicking here, but why does mouse-wheel scrolling up meets boundaries at 6, 13, 24, 27, 30, 49, 55 and 61 for every knob?

Can anyone confirm these? I hope I'm not "hearing things"

[loft_electronics]

what exactly is the difference between having the synth render the same sound 8 times or taking a sound and multiplying it 8 times? Sorry if I'm missing something obvious here (:

The difference is that the synth produces the sound with (often) free running oscillators, so each note played is slightly different to the previous one. That's one of the ways in which synths are different from instruments that just replay samples. Many synths offer the option to restart a waveform in just the same place each time a key is pressed (for example, on Synth1 there's a 'phase' knob to do this) which makes it possible to get more consistency. There are many other things in synths which are dynamic and ever-changing, so it would be kind of cheating to just sample a synth when it's doing something good and then replaying that bit.
UPDATE - sorry, didn't realise this had already been answered.

[crimsonwarlock]

the spline oscillator is triggered by common MIDI note on and note off signals, while the additive synth keeps playing even after the note has ended. Not sure what the expected behavior should be (as both make sense in a certain context, but I suppose they should be consistent (if you're wondering why your additive synth doesn't stop playing, it's probably because of this). I incline to think that the never-ending additive synth sound is not intended, as setting the volume to zero and back to a positive value stops the sound.

I have reported this problem yesterday and it's a confirmed bug; the dev is working on solving it. The workaround is to link an ADSR to volume which should stop this behavior.

[Kondarivan]

I have reported this problem yesterday and it's a confirmed bug; the dev is working on solving it. The workaround is to link an ADSR to volume which should stop this behavior.

Thanks for letting me know.

And many thanks to payne_bc for changing the knobs to linear.

[payne_bc]

Oh, I found a crash with the updated x64 version of Tunefish running in Reaper 4.55 (obviously also x64):
- Set the "Unisono" control (from the spline osc) to something other than 1
- Set the "Frequency" (from global) to 0
- Start playing a note
- Move the "Frequency" knob for a bit.

Can anyone confirm?

Unfortunatelly I cannot reproduce that here. Seems fine here when I do that.
However the hanging addsynth when no ADSR is set is fixed. Here you go:

http://www.tunefish-synth.com/beta3/vst24_tf3_x64.zip
http://www.tunefish-synth.com/beta3/vst24_tf3_x86.zip

[crimsonwarlock]

However the hanging addsynth when no ADSR is set is fixed. Here you go:

Thanks for the update, will test tomorrow.

[psmacmur]

Testing in Studio One 64-bit. Since switching my system to 44.1 kHz, the only issues I've encountered are the occasional hanging additive note when playing via the QWERTY keyboard input, and the odd behavior of the dials in arc mode. Arc mode was clever actually, just with that edge case at -pi/2, alas. Linear input works well, very stable. Kind of sad, though; if its linear, it should be a slider, in my UX opinion. However, space is tight, so don't change that!

The UI fits just perfectly on my 720p laptop, good use of space. Props for the clean look and eschewing the skeuomorphism that clutters so many VSTs these days.

Also, the other day I was walking past a condo under construction and heard a cool sound of metal flapping in the breeze. Managed to synthesize something sufficiently evocative thereof tonight in Tunefish, with no external effects. Between that and the more vanilla patches I was exploring last night, I'm quite pleased with the capabilities of the synth. Nice work, @payne_bc! Thanks for the quick fixes, and props for the 64-bit build.

Finally, I do have one question, what are "Slop" and "Slp"?

Cheers!

[H-man]

Testing in Studio One 64-bit. Since switching my system to 44.1 kHz, the only issues I've encountered are the occasional hanging additive note when playing via the QWERTY keyboard input, and the odd behavior of the dials in arc mode. Arc mode was clever actually, just with that edge case at -pi/2, alas. Linear input works well, very stable. Kind of sad, though; if its linear, it should be a slider, in my UX opinion. However, space is tight, so don't change that!

The UI fits just perfectly on my 720p laptop, good use of space. Props for the clean look and eschewing the skeuomorphism that clutters so many VSTs these days.

Also, the other day I was walking past a condo under construction and heard a cool sound of metal flapping in the breeze. Managed to synthesize something sufficiently evocative thereof tonight in Tunefish, with no external effects. Between that and the more vanilla patches I was exploring last night, I'm quite pleased with the capabilities of the synth. Nice work, @payne_bc! Thanks for the quick fixes, and props for the 64-bit build.

Finally, I do have one question, what are "Slop" and "Slp"?

Cheers!

Slope??

[H-man]

This sure is one unruly beast. Count me in

Works reasonably well in Podium at 41khz (and above?) but I'll be saving often.