Bug - Follower RMS Setting
- KVRian
- 1094 posts since 23 Sep, 2006
To repro:
- Make a modulator of follower type
- set RMS to 1ms (anything low is fine)
- save modulator preset in the modulator window
- increase RMS setting to max (1000ms)
- play audio, notice graph has a very slow/smooth input signal in the follower due to the large RMS setting
- load the preset you just made, which changes the RMS back to 1ms
- play audio, notice the follower is still very slow/smooth
- manually set RMS setting to the same value it's displaying (1ms)
- play audio, notice the follower is now reacting correctly
Essentially, modulator presets aren't recalling the RMS setting correctly. It's a minor bug, but I've noticed issues with the RMS setting before (think I posted about it a few months back) so it may be a symptom of a bigger issue regarding the RMS setting.
Thanks
- Make a modulator of follower type
- set RMS to 1ms (anything low is fine)
- save modulator preset in the modulator window
- increase RMS setting to max (1000ms)
- play audio, notice graph has a very slow/smooth input signal in the follower due to the large RMS setting
- load the preset you just made, which changes the RMS back to 1ms
- play audio, notice the follower is still very slow/smooth
- manually set RMS setting to the same value it's displaying (1ms)
- play audio, notice the follower is now reacting correctly
Essentially, modulator presets aren't recalling the RMS setting correctly. It's a minor bug, but I've noticed issues with the RMS setting before (think I posted about it a few months back) so it may be a symptom of a bigger issue regarding the RMS setting.
Thanks
-
- KVRAF
- 10375 posts since 2 Sep, 2003 from Surrey, UK
Confirmed here - the RMS parameter value changes back to 1 ms, but the Level graph behaves as if the RMS value is still 1000 ms. This is when I select the preset without stopping playback.
If I then stop / start playback then all is OK. (Or if I load the preset with playback stopped.)
If I then stop / start playback then all is OK. (Or if I load the preset with playback stopped.)
- KVRAF
- 2702 posts since 9 Jul, 2015 from UK
This is actually a larger problem, not connected to RMS, but loading of presets.
If you load the preset (as above) and the change the mod mode to envelope and then back to follower you will see the graph display correctly.
Another repro to try out, which I believe is the same problem.
1. Open MPowerSynth
2. Click on filter 1
3. Lower sustain down to silence
4. Click copy
5. Click on filter 2 (and activate it)
6. Click paste
Now notice how the sustain value correctly displays "silence" however the display is wrong.
7. Click on filter 1
8. Click on filter 2
Now the display has updated to the correct value.
To me this looks like the same problem, and I've noticed it all over the plugins. It seems as though the displays don't update themselves when loading a preset.
If you load the preset (as above) and the change the mod mode to envelope and then back to follower you will see the graph display correctly.
Another repro to try out, which I believe is the same problem.
1. Open MPowerSynth
2. Click on filter 1
3. Lower sustain down to silence
4. Click copy
5. Click on filter 2 (and activate it)
6. Click paste
Now notice how the sustain value correctly displays "silence" however the display is wrong.
7. Click on filter 1
8. Click on filter 2
Now the display has updated to the correct value.
To me this looks like the same problem, and I've noticed it all over the plugins. It seems as though the displays don't update themselves when loading a preset.
Jason @ Melda Production
-
- KVRAF
- 10375 posts since 2 Sep, 2003 from Surrey, UK
Both confirmed here. If I disable / enable Filter 2 then the graph is correct. (Or if I paste the settings when Filter 2 is disabled, then enable it.).
Also, when I drag Filter 1's sustain down to silence I get a blip at ~77ms:
>>> http://i.imgur.com/lq0Kzqz.png
What have I misunderstood?
Also, when I drag Filter 1's sustain down to silence I get a blip at ~77ms:
>>> http://i.imgur.com/lq0Kzqz.pngWhat have I misunderstood?
- KVRAF
- 2702 posts since 9 Jul, 2015 from UK
Oh,that is just to show you a visual of what the release is doing, as dragging the sustain makes the release flat and impossible to read. So it jumps back up so you can visually see whats going on.DarkStar wrote:Also, when I drag Filter 1's sustain down to silence I get a blip at ~77ms:
>>> http://i.imgur.com/lq0Kzqz.png
What have I misunderstood?
Jason @ Melda Production
-
- KVRAF
- 10375 posts since 2 Sep, 2003 from Surrey, UK
- KVRAF
- 2702 posts since 9 Jul, 2015 from UK
No because release is not a level value, it is a time value.
I know it looks like the graph shows a jump up in level but it is actually just showing the shape of the release.
I cant remember now the reason for this, Vojtech explained a while back. But there is a rare but logical reason why you would need to see the release value even if the sustain value is silence.
I know it looks like the graph shows a jump up in level but it is actually just showing the shape of the release.
I cant remember now the reason for this, Vojtech explained a while back. But there is a rare but logical reason why you would need to see the release value even if the sustain value is silence.
Last edited by jmg8 on Wed Jun 29, 2016 10:29 pm, edited 1 time in total.
Jason @ Melda Production
- KVRian
- Topic Starter
- 1094 posts since 23 Sep, 2006
Yeah, jmg8 is correct. The blip at the end is the release time, it has to be shown this way or you may not see it in the case where sustain is -infinity (in your example). If you look at your screenshot you'll notice the blip lasts 50ms (starts at around 75ms, ends at around 125ms), which is the same as the release time you have set.
It took me a while to figure this out originally, there may be a better way of showing the release time, such as colouring the release control and the release portion of the graph the same colour so its clearer to the user. Or, by scaling the release 'height' in the graph to the sustain height (may not be possible/easy because of possible values).
It took me a while to figure this out originally, there may be a better way of showing the release time, such as colouring the release control and the release portion of the graph the same colour so its clearer to the user. Or, by scaling the release 'height' in the graph to the sustain height (may not be possible/easy because of possible values).
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Ok, so it doesn't update RMS properly, I'll check it out!
(or other bugs ? )
(or other bugs ? )
- KVRAF
- 2702 posts since 9 Jul, 2015 from UK
It is not to do with RMS, rather updating the display when loading the presetMeldaProduction wrote:Ok, so it doesn't update RMS properly, I'll check it out!
(or other bugs ? )
Jason @ Melda Production
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
- KVRian
- Topic Starter
- 1094 posts since 23 Sep, 2006
Fixed in 10.01, thanks!
- KVRAF
- 2702 posts since 9 Jul, 2015 from UK
jmg8 wrote:This is actually a larger problem, not connected to RMS, but loading of presets.
If you load the preset (as above) and the change the mod mode to envelope and then back to follower you will see the graph display correctly.
Another repro to try out, which I believe is the same problem.
1. Open MPowerSynth
2. Click on filter 1
3. Lower sustain down to silence
4. Click copy
5. Click on filter 2 (and activate it)
6. Click paste
Now notice how the sustain value correctly displays "silence" however the display is wrong.
7. Click on filter 1
8. Click on filter 2
Now the display has updated to the correct value.
To me this looks like the same problem, and I've noticed it all over the plugins. It seems as though the displays don't update themselves when loading a preset.
Above Is still not fixed for me
Jason @ Melda Production
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Aaaah, that's actually a part of optimization. Let's just say that with potentially infinite number of parameters one needs to update only those who need it, and it's get worse with these compound structures (envelopes, oscillators, adsr), which require considerable CPU time to "reinitialize". Ok so I get the ADSR thing, that needs fixing. If you spot it somewhere else, let me know.
