Synth -Sinmad 0.9.9- featured this month (april) in One Synth Challenge, reports thread:

Official support for: nusofting.com
User avatar
KVRian
864 posts since 30 Mar, 2008 from MN, USA

Post Thu Apr 08, 2021 2:50 pm

Sorry for so many messages. There seems to be a bug with the Filter mix control.

When used with the 24db LP filter (non-ladder), with resonance and gain set at 0, the overall volume level drops as you raise the control from dry toward its middle position, and then rises as you continue to move it to 100% wet. This is particularly noticeable when the sub is active. This does not happen on the 12db filter.

On the 18db Ladder filter there is a related issue, but the behavior is different. There, the volume substantially increases across the filter mix range, from dry to wet. There seems to be some kind of interaction going on with gain here. For some reason, increasing the gain will lower the volume of the 18db filter. If I set the gain to about 50%, I get a consistent volume transition using filter mix.

Probably also related: If I use the 24db Ladder, with sub, and set the filter mix to around 55%, keyfollow off, and then play a C2, and then I change the cutoff across its range, around 340 Hz the sub drops out, and then comes back as I continue past 340. This behavior happens on all notes, just at different Hz values for cutoff.

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sat Apr 10, 2021 7:04 am

teilo wrote:
Thu Apr 08, 2021 2:45 pm
Also, is there a way to revert a preset to its saved version short of restarting the plugin?
There should be, but it seems we have designed the [compare] function in way that doesn't work as expected. I will need to change that behavior.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sat Apr 10, 2021 7:07 am

Yes, I guess I should use a custom mix curve for each filter, to reduce the volume variations.
teilo wrote:
Thu Apr 08, 2021 2:50 pm
Sorry for so many messages. There seems to be a bug with the Filter mix control.

When used with the 24db LP filter (non-ladder), with resonance and gain set at 0, the overall volume level drops as you raise the control from dry toward its middle position, and then rises as you continue to move it to 100% wet. This is particularly noticeable when the sub is active. This does not happen on the 12db filter.

On the 18db Ladder filter there is a related issue, but the behavior is different. There, the volume substantially increases across the filter mix range, from dry to wet. There seems to be some kind of interaction going on with gain here. For some reason, increasing the gain will lower the volume of the 18db filter. If I set the gain to about 50%, I get a consistent volume transition using filter mix.

Probably also related: If I use the 24db Ladder, with sub, and set the filter mix to around 55%, keyfollow off, and then play a C2, and then I change the cutoff across its range, around 340 Hz the sub drops out, and then comes back as I continue past 340. This behavior happens on all notes, just at different Hz values for cutoff.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sat Apr 10, 2021 7:09 am

teilo wrote:
Thu Apr 08, 2021 2:35 pm
I have a suggestion for the MW mod. The way it presently works is problematic. Presently, the MW modulates a control across its entire range, regardless of the setting of the control. For example, when modulating filter cutoff, changing the MW to 0 sets the filter to 0 regardless of the cutoff control's position. This has undesirable consequences. When starting from a preset, and then moving the MW, the modulated control jumps downward. This makes the MW almost unusable unless you pre-stage its position before recording or performing.

Instead, it should modulate from the set value of the control, upwards. Ideally, the MW and AT modulators should have their own depth control to set the range of modulation.
I see, you would like to have MW ready on the preset filter spot for each preset. I will see what can I do.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sat Apr 10, 2021 9:27 am

Now I have an update for Windows : Sinmad 0.9.7d (you may ask me a quick DLL update)
- changes for Filter mix control:
1) mix curve keeps the power when slider moves in the middle
2) smoothed this parameter to avoid zipper noise
3) slightly longer slider

Note about the 18dB LP Filter: the resonance "gain" (small knob) behave differently than in the other filters: the higher the frequency the lower the volume. Basically it saturate the filter limiting the gain, but that's not even over the whole frequency range. I use the 18dB LP Filter for big basses.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

User avatar
KVRian
864 posts since 30 Mar, 2008 from MN, USA

Post Sat Apr 10, 2021 9:33 am

liqih wrote:
Sat Apr 10, 2021 7:09 am
teilo wrote:
Thu Apr 08, 2021 2:35 pm
I have a suggestion for the MW mod. The way it presently works is problematic. Presently, the MW modulates a control across its entire range, regardless of the setting of the control. For example, when modulating filter cutoff, changing the MW to 0 sets the filter to 0 regardless of the cutoff control's position. This has undesirable consequences. When starting from a preset, and then moving the MW, the modulated control jumps downward. This makes the MW almost unusable unless you pre-stage its position before recording or performing.

Instead, it should modulate from the set value of the control, upwards. Ideally, the MW and AT modulators should have their own depth control to set the range of modulation.
I see, you would like to have MW ready on the preset filter spot for each preset. I will see what can I do.
Yes. 0 on the MW should correspond to the set position of the filter cutoff or other control it is modulating.

If there were a depth control for the MW, that would give precise control over the range of values that the MW controls. So for the filter, the depth would set the max the MW would be able to raise the cutoff, and moving the MW from 0 to 127 would raise the cutoff from its set position, up to the position defined by depth. This is how most synths function.

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sat Apr 10, 2021 9:53 am

teilo wrote:
Sat Apr 10, 2021 9:33 am
liqih wrote:
Sat Apr 10, 2021 7:09 am
teilo wrote:
Thu Apr 08, 2021 2:35 pm
I have a suggestion for the MW mod. The way it presently works is problematic. Presently, the MW modulates a control across its entire range, regardless of the setting of the control. For example, when modulating filter cutoff, changing the MW to 0 sets the filter to 0 regardless of the cutoff control's position. This has undesirable consequences. When starting from a preset, and then moving the MW, the modulated control jumps downward. This makes the MW almost unusable unless you pre-stage its position before recording or performing.

Instead, it should modulate from the set value of the control, upwards. Ideally, the MW and AT modulators should have their own depth control to set the range of modulation.
I see, you would like to have MW ready on the preset filter spot for each preset. I will see what can I do.
Yes. 0 on the MW should correspond to the set position of the filter cutoff or other control it is modulating.

If there were a depth control for the MW, that would give precise control over the range of values that the MW controls. So for the filter, the depth would set the max the MW would be able to raise the cutoff, and moving the MW from 0 to 127 would raise the cutoff from its set position, up to the position defined by depth. This is how most synths function.
So the filter freq knob would set the min and the MW depth would set the max, right? Yes I could add that (when bugs are fixed). This also means that the MW and AT depths will be always positive/adding to the target parameter value.

BTW I played your mono patch for many minutes but I didn't get the sound off so far.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

User avatar
KVRian
864 posts since 30 Mar, 2008 from MN, USA

Post Sat Apr 10, 2021 11:45 am

liqih wrote:
Sat Apr 10, 2021 9:53 am
So the filter freq knob would set the min and the MW depth would set the max, right? Yes I could add that (when bugs are fixed). This also means that the MW and AT depths will be always positive/adding to the target parameter value.
I think that having it always be positive is better than the current state in which the position of the modulated parameter is essentially ignored. But the way some synths handle this is to have an option to invert (modulate down) or to make the mod bi-directional (in which case 63 or 64 on the modulator = the modulated parameter's set position, and depth determines the max modulation range in either direction).

I should mention that I have a workaround for this. I just turn off MW and/or AT in Sinmad, and use Bitwig's modulators instead. But that's obviously not ideal, as it makes a preset dependent on a DAW project.

FYI - I like to use synths for improv and performance, so these types of tweaks are very useful.

User avatar
KVRist
64 posts since 3 Oct, 2019

Post Sun Apr 11, 2021 8:27 am

When DUO is activated (detuning in cents or intervalls, both) with wave 1 (FM) there is an abrupt jump in the audible overtones around an FM amount of 0.7. Is this the expected behaviour? I would have expected a smoother increase in overtones. I attached a preset, just move the FM amount knob slightly around 0.7...
System:
Win 10, Reaper 6.26, Sinmad v. 0.9.7b (I got the download link just yesterday from Richard)
You do not have the required permissions to view the files attached to this post.

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sun Apr 11, 2021 8:40 am

SeBaer wrote:
Sun Apr 11, 2021 8:27 am
When DUO is activated (detuning in cents or intervalls, both) with wave 1 (FM) there is an abrupt jump in the audible overtones around an FM amount of 0.7. Is this the expected behaviour? I would have expected a smoother increase in overtones. I attached a preset, just move the FM amount knob slightly around 0.7...
System:
Win 10, Reaper 6.26, Sinmad v. 0.9.7b (I got the download link just yesterday from Richard)
You are right! Some more harmonics kick in when FM knob goes from 0.7 to 0.8, when "duo" is on.
I have to see why. Thanks.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Sun Apr 11, 2021 11:22 am

liqih wrote:
Sun Apr 11, 2021 8:40 am
SeBaer wrote:
Sun Apr 11, 2021 8:27 am
When DUO is activated (detuning in cents or intervalls, both) with wave 1 (FM) there is an abrupt jump in the audible overtones around an FM amount of 0.7. Is this the expected behaviour? I would have expected a smoother increase in overtones. I attached a preset, just move the FM amount knob slightly around 0.7...
System:
Win 10, Reaper 6.26, Sinmad v. 0.9.7b (I got the download link just yesterday from Richard)
You are right! Some more harmonics kick in when FM knob goes from 0.7 to 0.8, when "duo" is on.
I have to see why. Thanks.
Now fixed, for next 0.9.8 release.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

User avatar
KVRist
64 posts since 3 Oct, 2019

Post Mon Apr 12, 2021 6:04 am

liqih wrote:
Sun Apr 11, 2021 11:22 am
Now fixed, for next 0.9.8 release.
Will there be a new download link? The link I got from RichardSemper still has the 0.9.7b from april 1.

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Mon Apr 12, 2021 7:09 am

SeBaer wrote:
Mon Apr 12, 2021 6:04 am
liqih wrote:
Sun Apr 11, 2021 11:22 am
Now fixed, for next 0.9.8 release.
Will there be a new download link? The link I got from RichardSemper still has the 0.9.7b from april 1.
Yes, I'm waiting for the mac installer from my part-time partner.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

DASH Guy

Topic Starter

7656 posts since 20 Sep, 2001

Post Wed Apr 14, 2021 8:15 am

Update ready (PC & Mac) at same link or ask Richard. Thanks.

What’s new in 0.9.8

- Fixed uneven wave 1 FM amount when “duo” was on.
- Fixed gain loss when “filter” mix was in the middle.
- Fixed DAW crash when GUI was closed with “save” dialogue still open.
- Fixed Sub that went 2 octaves down when AA was active.
- Redefined “AA” into x4”, since it increases harmonics but it does not
reduce aliasing.
- Reduced master gain of about 2 dB, due to “filter” mix adjustment.
Therefore you need to tweak the volumes in your previously made tracks.
- Fixed FM buffer that could shut sound off.
- Removed unsaved changed when changing preset*.
*This allow to easily (?) restore the original preset since the
functionality of [compare] will be reset when you switch to another
preset. This also means that you will loose your editing if not
promptly saved into a preset file or in the DAW project. We did that
by request of a number of users.

- Reported bug, reproduced but not fixed so far:
Sinmad playing a mono patch in Reaper on macOS can drop the sound
momentarily, when specific conditions of notes timing are met.
http://nusofting.com
NUSofting : Peti Modelonia Sinnah

User avatar
KVRian
864 posts since 30 Mar, 2008 from MN, USA

Post Wed Apr 14, 2021 8:28 am

I'm still getting the spurious pitch glide when first loading a patch. I've noticed that it can happen on multiple voices initially. To eliminate it it is necessary to play a bunch of notes to flush out whatever is causing it, and it is fine from then on.

Return to “NUSofting”