Teensy Qt

stevech - you can scan the thread - for special features. But most stuff is obvious or intuitive and just built to work - just look and click or right click - frustration suits koromix well it seems :)

If you have 1 or 5(+) teensy's online they will show HID or USB. You can see connection details or Monitor serial output. You can Make multiple instances if you have dueling Teensys to monitor. You can UPLOAD any HEX file, or reset a Teensy, the Arduino integration is great if you have more than one active Teensy as it allows you to UPLOAD to a directed Teensy.

Unless he's broken something I haven't found - it has been customized and refined to be a great support tool for the USB debug by .print() nature of the Teensy dev as I use it.
 
Just wanted to add my appreciation to Koromix for this great tool. I like being able to reset the app without going into the bootloader and downloading the code again (or unplugging the USB to cycle power).
 
Koromix: You derive the device serial # using all four low bytes of the MAC it seems - to get (32 bit) -1 as a serial # from 0xffFFffFF. Paul has a post that only the low 24 bits constitute the serial#? That gives room for a mere 16777216 devices. Not that it is critical apparently that top byte is always (normally) zero - but I was reposting code I picked up to make the serial number testing against my 'non-Teensy' "Teensy" and TYQT gave 4294967296 and it puzzled me a bit too long.
 
What I've struggled with is ARM saying that the MCU unique number is x number of bits; it's not a serial number per se. But not which bits are MSB/LSB (endian) and which bits are ascending vs. random.

Not likely a problem here, but in a population of MCUs, taking less than 100% of the bytes may not ensure that the subset is unique as a group. That is, taking 4 bytes, LSBs or not, doesn't ensure that number is unique in the group. Unlikely a problem in a small group (say, 50 or some such).
 
Last edited:
stevech - my 'reposting' link has code that pulls the 128 bit processor ID - that is different and probably not related to the PJRC serial # that Paul crafts and associates with those write once bytes that are the unique MAC/Serial# and USB presented serial#. But indeed which of those 128 ID bits mean what I never saw if documented.
 
Koromix: Trying to use HID_RAW and Serial+Key+Mouse+Joy and the device is going missing? I get this in TYQY log - does it tell you anything? If I push Button I get ''bootloader" - and when I go back to "Serial" it works on the same sketch (the one linked above as 'reposting'):
Permission denied for device 'COM3'
Reset does not seem to work
Permission denied for device '\\.\HID#VID_16C0&PID_0482&MI_00#9&312A0112&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0482&MI_01&COL01#9&1DF9E4D4&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_02#9&C3063E3&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_03&COL01#9&349B11AC&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Reset does not seem to work
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_02#9&C3063E3&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_03&COL01#9&349B11AC&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Reset does not seem to work
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_02#9&C3063E3&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_03&COL01#9&349B11AC&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_02#9&C3063E3&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Permission denied for device '\\.\HID#VID_16C0&PID_0487&MI_03&COL01#9&349B11AC&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}'
Reset does not seem to work
Click to expand...
 
defragster said:
stevech - my 'reposting' link has code that pulls the 128 bit processor ID - that is different and probably not related to the PJRC serial # that Paul crafts and associates with those write once bytes that are the unique MAC/Serial# and USB presented serial#. But indeed which of those 128 ID bits mean what I never saw if documented.
Click to expand...
Ah, yes. Maybe Paul uses his own numbering system.
 
defragster said:
Koromix: Trying to use HID_RAW and Serial+Key+Mouse+Joy and the device is going missing? I get this in TYQY log - does it tell you anything? If I push Button I get ''bootloader" - and when I go back to "Serial" it works on the same sketch (the one linked above as 'reposting'):
Click to expand...

RawHID should work, and indeed it does on my system. Did it occur again? Could be a transient problem, which tends to happen when the Windows driver installer does its thing. Probably need to add a "retry" of some sort to this code, for this kind of thing.

However, Serial + HID does not work on Win32. Interestingly, I cannot find any device interface GUID that lists the Serial node when it is part of an IAD. I looked at the pyserial code, and they found a way around it using the device class GUID instead... but I cannot do that because device class GUIDs are not supported by the RegisterDeviceNotification() API. The windows device tree is such a mess.

Anyway, I have an easy way out but it does require some changes to the Win32 code, so it is for another day.
 
Thx, I was just using a HID sketch and it worked.

I did see my Win10 box pop up some version of:"installing Teensy HID" at some point that I hadn't seen before - not sure what triggered that?

HID does have odd behavior at times as on Raw_HID I have seen the failure to reset and the Teensy disappears - but come back on button press and usually corrects
 
I have been using the command line version for ages now and love it! But recently changed the vid/pid on the teensy for my own. And noticed this tool no longer can detect the teensy :( I assume its using the teensy vid/pid to find it. Can we have some command line arguments added to pass in custom vid/pid so it will still work for people running custom vid/pid?

-vid 0000
-pid 0000

Regards
Russell Pirie
 
TYQT monitor needs a way to specify a fixed pitch font - courier or whatever as some output requires visual alignment. The PJRC audio Tutorial spits this out and the double bars ( || ) should be centered. Sad - but I had to use IDE SerMon - and I'm not sure of a fixed pitch forum font I wanted to show the diff.

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<||>>>>>>>>>>>>>>>>>>>>>>>>> 0.99, 0.86
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<||>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 0.99, 0.98
<<<<<<<<<<<<<<<<<<<||>>>>>>>>>>>>>>>>>>> 0.63, 0.66
<<<<<<<<<<<<<<<||>>>>>>>>>>>>>>>>>> 0.50, 0.63
<<<<<<<<<<<<<<||>>>>>>>> 0.47, 0.29
<<<<<<<<||>>>>>>>>>>> 0.27, 0.38
<<<<<<<<<<<<||>>>>>>>>>>>>> 0.43, 0.45
<<<<<<<<<<<<||>>>>>>>>>>>>>>>> 0.43, 0.55
<<<<<<<<<<<<<<||>>>>>>>>>>>>>>>>>> 0.47, 0.61
<<<<<||>>>>>>>>>>>> 0.19, 0.42
<<<||>>>>>>> 0.12, 0.25
Click to expand...
 
Last edited:
What I'd like is the ability to have \r (CR) in the serial monitor just to return at the beginning of the current line (to overwrite it) and \n (LF) to go to a new line. This would allow to create pseudo static output in the serial monitor. Just a wish...
 
So first, I want to thank Theremingenieur for his gift, which got my MacBook back in working order :) I had not been able to actually test TyQt on OSX for many commits prior to 0.7.5, it seems to work relatively well.

So I'll be quite busy until June 22nd, and I'll be relatively free after that. So I'm keeping a list of requests and I'll work on it at that time.

russdx said:
I have been using the command line version for ages now and love it! But recently changed the vid/pid on the teensy for my own. And noticed this tool no longer can detect the teensy :( I assume its using the teensy vid/pid to find it. Can we have some command line arguments added to pass in custom vid/pid so it will still work for people running custom vid/pid?
Click to expand...

This goes with the customization features Theremingenieur needed... a while ago and it's definitely something I will work on at the end of June. Until then, it just occured to me that you could just tell me the VID/PID/model name you need and I can make a custom build with the changes. TyQt is relatively easy to build... except for the static Qt mess, which you need to do if you want a single executable and avoid the heavy Qt DLLs.

defragster said:
TYQT monitor needs a way to specify a fixed pitch font - courier or whatever as some output requires visual alignment. The PJRC audio Tutorial spits this out and the double bars ( || ) should be centered. Sad - but I had to use IDE SerMon - and I'm not sure of a fixed pitch forum font I wanted to show the diff.
Click to expand...

I'll work on that, probably as a global preference. By the way, I pushed a build that may fix your Serial / HID problem, only on Windows 10 though (for now). I swear the SetupAPI mess will kill me. As I said, quite busy so I didn't spend much time on it.

Theremingenieur said:
What I'd like is the ability to have \r (CR) in the serial monitor just to return at the beginning of the current line (to overwrite it) and \n (LF) to go to a new line. This would allow to create pseudo static output in the serial monitor. Just a wish...
Click to expand...

Definitely a valuable feature, I'll work on it at the end of June.
 
Koromix said:
So first, I want to thank Theremingenieur
Click to expand...

I'll second that - anything that helps TYQT - and a nice gesture!

I'll look to the new build and let you know if anything smoothes out when I get back to Frank's HID based TeensyTransfer.

Indeed, having a global setting to get a fixed font will be handy at times - others the compactness matters.

I did some stuff where I wanted the same code on Two Teensys - took a bit of manipulation with::
>> upload to "A", drop association, upload select "B" - then TYQT will pick the 'first' seen of the two next time - then just have to hit UPLOAD on the 'second' unit.

Picking the 'first' seen possibly a BUG - when two have the same associated upload? - in my case it didn't matter as I did many re-uploads to both, but it re-Associated to the 'first' depending on the 'usb' load order? But I was GLAD it did not purge the item from the 'upload'. In my case I had a cheesy master slave, and if the slave started first I had to reset it after updating the master.

A:>> I have an idea for what might have made it easier to get the ">>upload ..." steps working. When ANY teensy is in 'bootloader' always prompt for "Which Device"? Of course that would need a 'checkbox' to disable that as sometimes it would get in the way - but other times pushing the button would more easily allow uploading to "that' Teensy.

B:>> Optional Feature Creep: When one Teensy is online it gets the upload no questions asked, usually COOL. The exception is that Teensy was running something else and I powered the second Teensy down to swap hardware during the COMPILE/UPLOAD - and I lost the race to get it back online in time so now I have to upload two Teensys. If you consider "A:>>" above - maybe you could also (optionally) pop up the select device dialog when the 'online Teensy' list has changed in the last 'minute'? I'm sure this feature will possibly never be noticed by anyone but me - but handling multiple Teensys I've done it to myself a couple of times now, wiping out a Teensy that had been online for days doing something else, then wondering why the device was no longer running?
 
Koromix - the device drop down that is displayed in Ctrl-Hide/Minimal would be nice to see when the left pane is dragged LEFT and the device panel disappears.

... TYQT still great to have and use. Not back to HID again yet - except incidentally with an OnlyKey I just powered up - but I didn't swap to the new build until I can see an issue to compare.
 
Hello,

I have 21 Teensy's in my project and TyQt is only seeing 15 of them. I am using version 0.7.5-19-g4dd0582 of TyQt. I have 2 7port hubs with 6 Teensy LC's plugged in. If I unplug one set, restart TyQt, then the set that was missing is now shown. It's like it has a hard limit of 15.

Rob
 
Hi raflyer, and sorry for the delay. I just finished my exams and I can come back to life now, and fix stuff.

In theory there is no limitation to how many Teensy you can plug... but right now, the code that tracks which USB port the Teensy is plugged in is quite fragile on Windows. It definitely needs work. Another possibility is that one of your HUB / USB controllers is not well recognized, this happens easily with USB 3.0 controller drivers before Windows 8, as these often do a half-assed job with the USB device tree they expose compared to the Windows drivers.

What is your OS? Are the missing devices from a particular hub, and do they appear if for example you unplug all other Teensies? Could you send me the content of "Tools > Open Log > Full Log" (unplug / plug one of the missing devices once to get more complete log information).
 
Theremingenieur said:
Great to see that you are back. T'as enfin passé le bac? :D
Click to expand...

Tiens je n'avais même pas réalisé que l'ECN médecine finissait en même temps que le BAC. C'était donc ça tout ce bruit à la bibliothèque :)

PaulStoffregen said:
Welcome back... and don't forget to check out the K66 beta thread. ;)
Click to expand...

Thanks, I will do so. I'm very interested, though I'm not sure how much I can contribute to the beta test. I don't do anything fancy with the Teensy itself. I have another Teensy-related project that I hope to post in a few months, but it's very early and I have not touched it in many months. OTOH, now I have four months ahead with a lot less pressure.
 
Hello Koromix! Glad you survived and completed the process.

<edited> to remove beta notes.
 
Last edited:
Just did another thread pointer to TYQT and wondered if anyone tested TYQT 'Integrate to Arduino' (and UNDO integration) on 1.6.9 yet? I had one bad experience - but it may have just been from some private stuff on my machine. And it was fixed with a TeensyDuino 1.29b3 (re)install. ... or maybe the edited files changed on 1.6.8
 
You must log in or register to reply here.
Back
Top