V0.3.x Firmware Workout: Better late than never

I think the time is ready for a window comparator unit :slight_smile:

1 Like

I’ve noticed that when grain stretch and Looper are sharing the same buffer, Grain Stretch doesn’t visually update it’s window. Though you can hear the changes which means it’s indeed following along nicely, visual feedback in Grain Stretch shows a waveform that is a momentary grab of the buffer.
I have used slice on grid in the Grain Stretch unit, so perhaps it has something to do with this ?

1 Like

Yes, I have noticed this behavior as well when sharing the same buffer across both units. If you “slice on onset” and are feeding the buffer with new material does the 301 keep scanning the incoming audio and readjust slices in realtime? Not sure if this is possible but would be really cool if it could :star_struck:

1 Like

Preamps! Thank you Brian! :love_you_gesture:

0001

7 Likes

36 posts were merged into an existing topic: UI Design Philosophy and Practicalities

This is a known issue that has been the case with shared audio buffers across multiple sample players: ER-301/Tracker - O|D Wiki - would be great to see this fix implemented at Brian’s earliest convenience :slightly_smiling_face:

1 Like

I did find this and, probably inappropriately, opened another thread ER-301 Buffer render bug? - #5 by sixnon , but I must concur with @Julian_Edwardes that it’s specifically noticeable with Grain Stretch, as I’d never come across it in V0.2.x. It did raise questions for me about slicing akin to those of @bc3. Where do the slices reside? Are they fixed at initial slice in the window of the buffer/sample size, or dynamic?

Excellent, no?

Yess! Totally awesome.

Finally got around to updating and I have to say, I am liking the changes.

However, I’m experiencing an issue where the inputs become unresponsive and a power cycle is required to bring them back to life. It happened most recently with a gate input triggering a sample player, but it may have also happened with the audio inputs, as well.

If I can come up with a solid reproducer I’ll post it, but I was just mucking around so I wasn’t in debug mode. This was with 0.3.02 96kHz, by the way.

HELL YO !
I STILL didn’t totally dig into ER301 BUT this huge step will make me doing that.
I’d like to integrate my own custom unit with my own dsp (already done in max and C++)

1 Like

Sweet update! Thanks for the continuous effort.
Unrelated: I would like to record a whole session down to one file. Is there a way to merge the six tracks (well…4 would be enough in my situation) of the recorder to one stereo pair?

  1. Create a stereo Global Chain and call it ‘final mix’ or something like that.
  2. Inside this ‘final mix’ chain put 6 mixers, one for each of the original 6 tracks. Adjust mix levels to taste.
  3. In the 6-track Recorder, clear out everything and link two tracks together to make a stereo track.
  4. Assign the left channel and right channels of ‘final mix’ to the input of the stereo recording track you just made.
  5. Record.
3 Likes

Indeed! I must apologize. Right before replying to your post, I was reading a thread on another forum (that shall not be named) where people were lamenting the lack of modules that react dynamically to triggers of different shapes.

Using .3.2. Had the bit-rate reduction after a while on. Using Clock Delay and Reverb. Didn’t save patch.

v0.3.03 CHANGES

  • FIXED: Legacy (v0.2) presets are loadable again.
  • FIXED: Slice on Onset > Results from median filter of peak heights were being ignored. Slices should be more robust now.
  • FIXED: Slice on Onset > Detection algorithm was not starting from the beginning of the buffer.
  • FIXED: Put back the missing expand all/shrink all function. Focus the chain header and cycle with the ENTER button.
  • FIXED: Unit menu was activated on a button press (instead of a release) which allowed the button release to inadvertently activate a menu item.
  • ENHANCED: Added visual hints as to what will happen if you press a soft button again after already focusing a unit control or header (i.e. focused press). Thank you to @jj.kidder for this idea!
  • ENHANCED: Bump Map now takes 1/3 of the CPU usage as before. Also renamed Bump Map to Bump Scanner.
  • NEW UNIT: Sample Scanner > Same as Bump Scanner except you can also ‘Assign Sample’ to be scanned by the input. Set fade to 0 and width to 1 and height to 1 to use a sample as a lookup table (LUT). Thank you to @brownshoesonly for this idea!
26 Likes

Epic. Thank you Brian!

Amazing… thank you!!!

Super glad I didn’t start re-writing EvilTwin :joy:

3 Likes

So in the sample scanner with width set to 1. “1” would equate to 100% of the sample length? and then i could have a bunch of single cycle waveforms inside the sample to lookup? that doesn’t sound right, so i think no. To do that width would want to be size of the single cycles, right? this way you could float the bump window across any two positions and do a proportion of one cycle and its “width complement” with another cycle? perhaps a poor man interpolation?

I confused myself. however im happy I helped with my video scanner comment

1 Like

wasn’t sure how long i’d last with out the evil lil guy

2 Likes