EDIT: When I later turned my system back on, the envelope magically worked. The LED on the TXo is a bit inaccurate (sometimes the pulse is bright, other times dim), but the CV is very accurate.

It seems like you do need to set the CV for TXo outputs you want to be EGs, similar to how you set CV for the oscillators.

Help no longer needed, but I’m keeping the post below in case anyone wants to play around with the scripts.


Having some trouble with the TXo’s envelope commands (1.4.1). Attempting something along the lines of Teletype Studies 5’s Dolphin and Seagull.

Current script:

M

TO.OSC 1 ADD V 4 N P.NEXT // I couldn't figure out how to bring TO.OSC.N to a "normal" note range
M ADD 50 RSH PARAM 5 // controls speed of metronome
TO.ENV.ATT 2 50
TO.ENV.DEC 2 100
TO.ENV.TRIG 2
I

TO.CV 1 V 10 // amplitude of osc output
TO.CV 2 V 10 // amplitude of EG output
P.N 0
P.L 7
P.WRAP 1
TO.ENV.ACT 2 1

Any ideas?

simple script to create a light show:

L 1 16: TO.TR I MOD ADD I Y 5
Y ADD Y 999```

this is for 4x TXo set up, if you have less just adjust the upper bounds in both loops. try different values for different patterns. i think there is a lot to explore by taking advantage of having 16 outputs in this manner - need to hook it up to 4ms VCA matrix or ER-301.

here is what it looks like:

https://www.instagram.com/p/BUOM_vXg2-f

(should we start a new thread for telex scripts?)
8 Likes

Funny - you edited while I was looking at your patch.

You are right on CV values - think of them as the peak value for your envelope. It sounds like they may not of been set for your earlier tests.

Apologies for not getting back sooner, I’ve been on holiday with my wife and it took a few days before I had a moment to get my lunchbox out of my backpack and give your patch a whirl.

I’m noticing the same with the LED. There is a separate refresh rate for the LEDs in the firmware. I dialed it back to reduce the strain on the CPU, but it looks like with faster envelopes they sometimes miss the peak values due to timing. As you noticed, the goal is to keep the output of the CV values as accurate and smooth as possible.

One of the things I’ve planned a fiddle around with on this trip is a better understanding of the dynamic strain that’s being put on the output expander. Perhaps I can adjust the LED refresh rate when it isn’t working very hard to be faster? I’ll put it on my list of experiments.

Let us know if you have any additional questions!

5 Likes

Any plans to add TO.TR.MUTE to the command set?

very useful for temporarily disabling certain outputs without unpatching.

other than that… wow… blown away by the capabilities of TXO. and the quality of your work in general!

:heart:

Just arrived home form three weeks away to a wonderful package for TXo and TXi waiting for me :slight_smile: Going to get this all setup today and hope to share some results soon.

Hi! Just wondering if I can get on the list for a TXi, assuming there are any extras when the smoke clears. (btw I sent you a message earlier on your instagram)

I’ll add it to my list! I have a few things cooking for the next rev of the firmware.

Thanks for the kind words!!

DO share; hope it goes well!!

I’ll PM you. :wink:

1 Like

Hi, I have just received a teletype in trade and I’m wondering if I can get on the list for a prebuilt i and o. I don’t think I have the proper trust level to pm yet. Thanks.

1 Like

Sent you a PM on the details for leftover units. :slight_smile:

Pt 1. absolutely loving my TXi and TXo!!!

Pt2. a couple small issues I’ve run into on my TXo unit though some seem like they could be issues w/ lookup tables in TT. :

  1. The TO.CV.OFF seems to be adding offsets to my CVs but not the correct offsets.

  2. Sending negative VV values to the TO.CV did not yield expected results while both negative V and negative N values seemed to work fine

  3. TO.CV.N seemed to additionally trip up w/ negative note value,

using my TI.IN ( :slight_smile: ) i measured :

After setting TO.CV 4 0
TO.CV.OFF 4 0 reads at 91
TO.CV.OFF 4 N 12 at 849
TO.CV.OFF 4 N 24 at 1669
(seems like it takes N 24 or V 2 to scale up an octave same w/ negatives)

the VV numbers are a little weirder w/ regards to the negatives
TO.CV 4 VV 100 at 1666
TO.CV 4 VV 0 at 87
TO.CV 4 VV -50 at 12217
TO.CV 4 VV -100 at 16313
TO.CV 4 VV -200 at -16093

i’m guessing this might be the VV table V and N are much closer, tho a little divergent around -1 octave (between the V and N table)
TO.CV 4 N 12 at 1666
TO.CV 4 V 1 at 1669
TO.CV 4 N 0 at 38
TO.CV 4 V 0 at 37
TO.CV 4 N -12 at -1467 (this was a little weird as the others are so close)
TO.CV 4 V -1 at -1523
TO.CV 4 N -24 at -3099
TO.CV 4 V -2 at -3098

The negative TO.CV.N were also just a little strange
TO.CV.N 4 12 at 1666
TO.CV.N 4 0 at 91
TO.CV.N 4 -1 at 16309
TO.CV.N 4 -12 at 14975

I’ll reiterate, I love these things, just curious if anyone else was having this issue, or if perhaps this can be attributed to user error … likely haha. I’m also sure I miss typed something in here, but I can capture some audio later if it would help

1 Like

Hey @tambouri - Thanks for your post.

Totally kicking myself right now. It looks like there was a regression somewhere in the TXo firmware. I have confirmed the following:

  • Things are wacky with negative values sent via VV; positive seem OK
  • CV offsets are not behaving as expected (their value seems halved)
  • I didn’t account for negative note numbers for the TO.*.N functions

These were the first things I had working (save the negative N functions) - but something must have gone amiss when I was optimizing and increasing the device’s internal precision.

I’m on holiday right now, but I’ve got my road kit with me. I’ll shelve some new features and optimizations that I’ve been working on and give this a good look.

Give me a couple of days…

Sincere apologies. Can’t believe this slipped!!

b

5 Likes

Thanks for checking this so quickly :raised_hands:

Enjoy your vacation!

1 Like

@tambouri & @batudal & TELEXo Users:

I was able to find some time to fix the issues that were found with OFFSET commands and negative note numbers for the TELEXo. The update also includes some additional optimizations that you really shouldn’t notice - but are noted below.

I also fixed a bug in the Teletype’s VV behavior for negative values. I submitted a pull request to @sam for that change and hopefully it will make it into the next 2.0 BETA release.


For the TELEXo, you will find an updated binary here:

TELEX Firmware Release 0.14

Release Notes

Updates to the TELEXo code:

  • fixed performance during RECT usage
  • continued to optimize the oscillation loop for performance
  • increased the LED refresh to 50ms from 100ms
  • optimized writes to DAC (only on change of value)
  • BUG: offset was off by a power of two (4 character change - DOH!)
  • BUG: negative values for N (*.QT and *.N) are now handled properly

No changes to the TELEXi.

Apologies for letting this bug slip. I increased internal resolution for the DAC and missed converting the offset value to the proper range. So silly.

Thanks for everyone’s patience and vigilance (especially @tambouri and @batudal)!!!

4 Likes

All good. Thanks a lot for taking care of it promptly. :+1:

I just tested for positive values through power cycles and it’s perfect. Now my quirky Quadnic will stay in tune for the rest of it’s life.

I hope we didn’t ruin your holiday. :stuck_out_tongue:

1 Like

Sorry if I missed this somewhere - are there instructions for how to update the firmware on the TX devices? I’d like to update my TXo with this latest update, but can’t seem to find direction on how to load the new hex file. I assume it’s through usb on the Teensy?

1 Like

There are some details in the manual that came with your expanders. :wink:

First, an all caps moment:

DO NOT POWER YOUR MODULE WHILE CONNECTING THE TEENSY VIA USB TO YOUR COMPUTER - THIS WOULD BE VERY BAD

Ok. With that out of the way, here are some simple steps.

Firmware is loaded on the Teensy processor that piggybacks on the underside of your module. Doing so is pretty simple.

  1. Download the Teensy Loader application for your operating system from https://www.pjrc.com/teensy/loader.html (if you don’t have it already)
  2. Download and unzip the latest TELEX firmware from https://github.com/bpcmusic/telex/releases
  3. (optional) Carefully remove the Teensy from your module
  4. Connect the Teensy to your computer via USB
  5. Use your operating system specific instructions for the Teensy Loader to open the appropriate firmware in the Teensy loader and program it on your Teensy
  6. Disconnect the Teensy from USB when the firmware has been loaded
  7. Reattach the Teensy to your module if you removed it
  8. Party.
1 Like

Thanks! I guess I should have looked at the manual :wink: Don’t you know us musician types are bad a reading things?

1 Like

thanks, that did the trick for offsets and TO.CV.N !!

1 Like

ok, seems i’m doing something wrong with the TO.TR.PULSE.DIV command or maybe is too experimental.
i’m sending a clock pulse to teletype in 8, and trying to divide it by 2 and 4 on TO Outs 3 and 4.
is it ok?

TO.TR.PULSE 3
TO.TR.PULSE.DIV 8 2
TO.TR.PULSE 4
TO.TR.PULSE.DIV 8 4

Pulse dividers cause the TXo to “eat” pulses sent to the unit (or generated by its internal metronomes). Thus, you will need to configure the two TR outputs to their particular division and then send them some pulses.

In the examples below, I’m using the 2.0 Alias for TO.TR.PULSE - TO.TR.P; you can substitute TO.TR.PULSE if you are running the 1.4.1 firmware.

This script will do what you are looking for (assuming M is active):

I
TO.TR.P.DIV 3 2
TO.TR.P.DIV 4 4

M
L 3 4: TO.TR.P I

This script causes every other triggered pulse to emit from output 3 and one out of every four pulses to emit from output 4.

If you want to have them independently triggering (not initiated by a M event), you can use the outputs’ internal metronomes. That would look something like this:

I
TO.TR.P.DIV 3 2
TO.TR.P.DIV 4 4
L 3 4: TO.TR.M.ACT I 1

This make sense?

3 Likes