Beta testing for a totally new sampling/triggering alg

[dmitri]

What is your mesh pad?
Can you also please post your best settings for the old alg and the new alg for this pad?

[ignotus]

What is your mesh pad?

It's a 10" pad with a Roland head and a DIY side-mounted trigger. Here's a pic:

trigger.jpg

Here are my settings for the beta firmware:

snare_settings.png

All I change for the old algo is to lower DynLevel/DynTime to 1/4 for the head (I leave the rim as-is because I don't need really accurate fast triggering for it) and raise the rim threshold to 14 (otherwise it auto-triggers).

[dmitri]

With the new alg, can you please take three photos of the 'Samples Graph' from the LCD?
1. For a light hit.
2. For a medium hit.
3. For a strong hit.

[ignotus]

Here you go:

Light:

light_hit.jpg

Medium:

medium_hit.jpg

Hard:

hard_hit.jpg

[dmitri]

This may seem strange, but I can actually set DynLevel/DynTime lower with the old algorithm than with the beta. With the old one, I can set them at 1/4. With the beta, anything lower than 1/12 causes lots of extra unwanted notes. I can also get better flams and rolls with the old algo.

This is really strange. I have completely opposite results.
Judging by the Samples Graph you posted the pad generates very good clean signals and you have a precision rectifier on this input.
I have a mesh pad with similarly good clean signals. With the old alg and DynLevel/DynTime lower than 3/36 I get double triggering for pretty much every hit.
As soon as I enable the new alg double triggering is gone and I cannot see any missed hits in pressrolls. And I can lower DynLevel/DynTime to 1/4 without introducing any double triggering.

With the bad mesh pad, which I was doing most of the testing with, I cannot lower DynLevel/DynTime so much. With the old alg it has to be at least 7/32 but with the new alg I still can lower it to 2/16.

I wonder why we get completely opposite results. I'm totally puzzle:( Do you have other type mesh pads?

[dmitri]

Updated beta firmware in the first post.
No changes to the new sampling/triggering alg but massive speed up for color LCDs handling.

[dmitri]

Updated beta firmware in the first post.
A bit more speed up for color LCDs handling.

[dmitri]

Updated beta firmware in the first post.
Speed increase for Samples Graph for waveform types 0-3. Waveform types for Samples Graph pages are now:
0 - pure levels as registered before Gain adjustment and normalization. For dual piezo/piezo pads signals for both zones will be overlayed with different colors.
1 - same as 0 but split between head (in the top) and rim (in the bottom)
2 - normalized against HighLevel levels after Gain adjustment. For dual piezo/piezo pads signals for both zones will be overlayed with different colors.
3 - same as 2 but split between head (in the top) and rim (in the bottom)
4-7 - same as above but with older slower refresh rate (full clear of the samples graph) just as 0-3 types were in previous versions.
0-3 types may cause samples graph corruption for very fast rolls.

[dmitri]

Updated beta firmware in the first post.
1. 3way Roland type cymbals now should work correctly. Tested with a real 3way Roland type cymbal.
2. Changed Threshold calculations. It now uses Gain setting for calculating raw Threshold so that it now closely matches Threshold setting in the old alg.

As far as I can see I implemented all functionality from the old alg except automatic Edge/Bell switch Threshold detection.

[dmitri]

Updated beta firmware in the first post.
Implemented Edge/Bell switch inputs Thresholds autosetting (<Set All EdgeSw> and <Set All BellSw> as per https://www.megadrum.info/content/misce ... s-settings )