sunflowr / recpu Goto Github PK

Currently when running the sequencer on RE-606/TR-606 you can see the pattern group LED blinking while playing the hi-hats. Looking at the schematic this seem to make sense as the LED signal is also shared with the hi-hat.
But this behaviour is not experienced on the original CPU so I consider this a bug and is investigating what could be causing this, wether it is timing related, a difference in current usage, slew rate or something else.

It's been requested to be able to randomize/mutate patterns.

It would be a nice workflow enhancement if the re-303 could show the currently playing step when running a pattern in normal mode

Tried to flash my original TB-303 but after the sysex dump was successful the device reboots and the missing emulator firmware status message appears again. no luck it keeps the firmware. any ideas? thanks!

It would be a nice workflow enhancement for the re-303 if there was a feature to auto exit pitch mode after all notes have been input, similar to time mode

I've just built an RE-606 and am finding that my patterns aren't saving when I turn the machine off and on again. Repro:

• Turn the machine on
• Wait for a few seconds for the machine to start (if it's relevant, I'm debugging an issue where it doesn't start immediately even after replacing the TIP30)
• Enter pattern write, tap out a pattern, and observe it plays correctly
• Turn off the machine
• Wait ten seconds
• Turn the machine back on

I'm observing inconsistent write behavior, in that:

• Sometimes the pattern is saved correctly but the step length is wrong
• Sometimes the step length is wrong and recent edits (if I go back into a pattern which did save and add notes) aren't saved either

I believe I installed the files from the 1.4.3 release - however, I saw a mention of a similar problem on the RE-303 forum which indicated it was an issue in 1.3 (http://23.235.199.139/~re303c5/forum/topic/1397-pixie-pattern-storage/). Could this be a software bug, or could it be because the unit isn't starting immediately and the rails aren't quite right?

Relevant for RE-303 but could definitely apply to the other RE models: I would love a way to offset the starting octave which triggers notes when external control is on (for example lowest note starts at C4 instead of C1). This would be a great quality of life improvement for when I'm externally sequencing the 303.

Thank you!!! :)

Hi,

great work, thanks. I updated the firmware yesterday, worked well, but I had to overcome a few question marks.

For musicians who bought a pre-built Re-303 and have no IT skills I want to contribute to a more detailed and noob proof manual, with the pitfalls I saw with my fresh noob perspective. How can I extend the source of the .pdf manual ?

3 options:

a) Someone e-mails me the pdf source. If it is word or libre office, I can add to it with "track changes", so the package maintainer can choose what to use and what not.

b) I convert everything to LaTex, check it into the repo, and then everyone can contribute to it in git because the .pdf will be generated from ASCII source code. (I will also add a readme for how to install the LaTex toolchain and I will choose some existing LaTex package for fast setup.)

c) I add the manual fragments I suggest here and someone copy and pastes it into the manual.

And obviously the 4th option d) a contribution from me is not needed at the moment.

Thanks.

Hi,

I want to contribute to your programming. Esp. for isse #4 but also to other issues. Would it be possible that you add the latest version of your code to this github repo ? (Instead of linking to a .zip). Then I can clone it, contribute to your programming and send you a pull-request, that you may or may not pull.

It's not necessary that you carry everything all alone :)

Best regards,
Roelof

I still have the bootloader version on it, which can be accessed via the "Function" and "Power ON" buttons to update the application.
Is dosent work for me.
Does the firmware (bootloader) have to be updated to the latest version? It's not entirely safe!

Steven

Hi,

when Midi In will be implemented (I can contribute to the code if you like), I suggest to keep it like in the X0xb0x:

Velocity > 100 = Accent On
Legato notes trigger slide (portamento on legato)

This way the device will be compatible to Sequencers like Skinnerbox Time&Timbre (Time Sting), who can generate very nice Acid sequences with just one click on a button (much nicer than random by the way). This is great for live playing. Above's Midi interpretation would leave it compatible in: note, swing, accent and glide :) The randomized velocity will generate accents and the here and there occuring legato notes will generate slides. No SysEx needed.

Let me know when I can contribute in C or C++ to this Midi In feature.

Cheers,
R

Looks like there's a bug in 1.3.0 on the RE-303 that erases patterns when the unit is power cycled while in "pattern play" or "track play" modes. I've tested this with both 1.2.0 and 1.3.0, and the issue only exists in 1.3.0.

To reproduce, on the RE-303 with 1.3.0:

1. From "pattern write" mode, program a pattern into any pattern slot (tested with pattern slots 1, 2, & 3, and "pattern group" 1)
2. Switch to pattern play mode
3. Play the pattern, to confirm it's correct
4. Power cycle the unit with the volume/power knob
5. Play the pattern again. It'll play a repeating note, instead of the pattern programmed in step 1.

Another issue that might be related: If power cycled in "pattern write" mode, the first note of a pattern in slot 1 gets overwritten with a 'C'. To test, do the same as above, but use pattern slot 1 and power cycle in "pattern write" mode.

Tested with:

• RE-303 (tested on multiple units)
• 1.3.0 app (reemu.syx) with 1.3.0 bootloader (bootloader.syx) <- bug
• 1.2.0 app (reemu.syx) with 1.3.0 bootloader (bootloader.syx) <- this works fine

Let me know if you need any additional info.

Thanks!

From a user:

Roland Din Triggers are longer than Pulses from Converters like the Doepfer MSY. My attempts prove that short trigger 4ms lead to problems. (Original 303 808 have no problems with short Pulses) I tried with 8ms Pulse length from my midi cv interface and its fine.