Skip to main content
Topic: Big news: Massive IR Toy updates (Read 41438 times) previous topic - next topic

Big news: Massive IR Toy updates

In a few days there will be a release candidate for IR Toy firmware v20, and a new version of IR REC&PLAY. This version has huge bug fixes and a new open source USB stack, all thanks to JTR.

The last few release of the IR Toy firmware have been really buggy. Feature bloat has introduced numerous bugs. This release should be a HUGE step towards addressing the numerous REC&PLAY and other issues with firmware v12/v13/v14/v15/etc.

An alpha of the new IR Toy firmware is attached. I've been using it with my TV pc for a while without problems. There are some changes to the IR transmit, so it may not work with the current REC&PLAY apps. If you test it please report how it works.
Got a question? Please ask in the forum for the fastest answers.

Re: Big news: Massive IR Toy updates

Reply #1
Great, I'm really looking forward to checking this out.

Re: Big news: Massive IR Toy updates

Reply #2
keep us updated Ian
gimme a shout when you release the proper version so i can test it out with winlirc :)
not got time to keep testing every upgrade lol

Re: Big news: Massive IR Toy updates

Reply #3
Hi

this is indeed great news. 

A summary of my testing on a V1 IRT so far shows inconsistent and questionable results. These are summarised below.

Terminal Commands & responses:
      t   r   s   
Win XP   V115   OK   S01
Ubuntu   FA02   OK   S01

Notes
   1. V115 was prev FW. IRT was dis/reconnected after FW update.
   2. Response to 't' command was consistent to that above in first round of many tests in both environments.
      Then had one instance in XP - rather surprising result of 3 't' commands was: "FA01V115V115", and the LED was off!
      After flashing a second time, Ubuntu response to 't' was consistently 'V115'.      

LED status at power on:
Win XP   ON, until receive some IR.
Ubuntu   briefly on, then off - first few times.    After second flashing, when response to 't' command was always 'V115', then LED behaviour was then same as XP!

More detail is available if you need it.

Hope this helps progress a great little tool.

Re: Big news: Massive IR Toy updates

Reply #4
I have looked further at the issues and give a short history below:

firmware version history:
    v115
    A02
    v115
    v112
    USBIRTOY-newtransmit
    USBIRtoy-TVBGone
    v115

But after FIVE different firmware being loaded AFTER the A02 - I still get seemingly random version info in respond to 't' command. Here are two latest sequences:
    FA02V115V115V115V115V115V115V115V115V115V115
    FA02FA02V115FA02V115V115V115V115V115FA02V115V115FA02FA02

So if some remnant of A02 has survived all the re-programming/flashing - did I use the correct command?
    'fw_update -e -w -v -m flash -vid 0x04D8 -pid 0xFD0B -ix USBIRToy-v20.hex'

Or is it an issue with using a V1 IRT with firmware or ?????

Re: Big news: Massive IR Toy updates

Reply #5
The self-test isn't perfect, it's a bit of a hack. There's no problem, it's just the best I could do for a quick test prior to the first manufacturing run. A few failures just mean it didn't get triggered right, one success is generally enough to say it works.
Got a question? Please ask in the forum for the fastest answers.

Re: Big news: Massive IR Toy updates

Reply #6
[quote author="ian"]The self-test isn't perfect, it's a bit of a hack. There's no problem, it's just the best I could do for a quick test prior to the first manufacturing run. A few failures just mean it didn't get triggered right, one success is generally enough to say it works.[/quote]

I found the same problems with the self test. I noticed that it depended a lot on the ambient light. I could get it to work correctly 90% of the time if I held the IR TOY under the work bench and away from the fluro light above.

This has not be fixed in anyway by me and don't expect the new firmware to be any different in this regard.

Re: Big news: Massive IR Toy updates

Reply #7
Hi JTR I have just received IR Toy V2 from SeedStudio and I have the same problem as you, the selftest never work and return always FA02 (behind a white paper), without white paper it returns FA22 IIRC.
Anyway it seems to work because i have recorder IR from my Samsung TV remote control and replayed them with success with winlirc-0.9.0.
Just a nota it seems impossible to play any winlirc-0.9.0 with lirc files lircremotes tested with samsung, sony ... and i always have the error winlirc Send failed/not supported.
Best Regards
Benjamin

Re: Big news: Massive IR Toy updates

Reply #8
OK - I feel silly forgetting that the 't' command returned error codes - my poor excuse is that I had never seen any before. Some further checking makes me think the building site next door has some floodlights that might have triggered the test errors, as during the day, I get consistent passes.



I have done some more testing with my V1 IRT on Win XP, and irrecord works fine, but winlirc and terminal commands still give me some issues.


Here is a summary of my queries

   - new firmware - what version does it report? I have only seen 'V115', when using the V2 alpha ( but reverting to other versions gives expected #).



   - Led light behaviour

      - stays on at power on - until IRT receives IR, or leave sample mode.

      - The LED flashing in response to IR is much weaker.

      - using irrecord - at first LED stayed on, then off after receiving IR, but never flashed when receiving IR.



   - After sending 'IR' / 'R' command - IRT seems to hang. Can't get response after, not even after sending string of hex 0s.

The above has been duplicated several times after a downgrade to v12, then reflash with alpha 2. Note that v12 does not give me the above symptoms.

   - First attempts with WinLirc were unsuccessful - may send one command then hang, or hangs on first command. I have a portmon capture if that helps. Later attempts all worked fine - WinLirc/IRT did not hang, although I was unable to verify device response to IR command - as my current location does not have any devices in range.

Re: Big news: Massive IR Toy updates

Reply #9
The led is a known issue. The poetmon dump would be helpful.
Got a question? Please ask in the forum for the fastest answers.

Re: Big news: Massive IR Toy updates

Reply #10
Here's another update, we've been hard at work debugging the IR Toy code and getting it going on the new stack.

Fixes
*Change between modes fixed, that should make it more reliable
*IR sample receive fixes (not TX yet though, but it might be better too)
*LED in RC decoder, power up, sump mode fixes
*A few SUMP mode updates, but errors persist (see below)
*Updated version to v20

Known issues:
*IR transmit and compatibility with any IR TOy REC & PLAY utility unknown
*SUMP logic analyzer mode crashes SUMP client, but works fine in terminal...

The big issues is the SUMP problem. It works fine in a terminal, but with the actual client there are problems:
*At 115K BPS one capture is possible, but at the end the IR TOy disconnects from the PC
*At 900KBPS, pressing the capture button crashes the client completely, even before data is send (just opening the port). The IR Toy still works in a terminal and will return SUMP data just fine.

Windows error report says it's in rxtxserial.dll.

I curious if the stack is doing something unexpected when the client opens the port with certain configurations. The Client log says DTR=LOW.

Code: [Select]
[8/18/11 10:58:36 AM - INFO  -     nl.lxtreme.ols.client.Host]: LogicSniffer v0.
9.4 started ...
[8/18/11 10:58:47 AM - INFO  - logicsniffer.LogicSnifferDevice]: Starting captur
e ...
[8/18/11 10:58:47 AM - INFO  - logicsniffer.LogicSnifferDevice]: Attaching to CO
M2 @ 921,600bps (DTR = low) ...

That is the end of the client log.

Attached:

*Package with firmware and update utility
*Full project archive. Check it out yourself. The USB stack is included so it should compile with one click :)

Wanted:
*Testers of everything
*Tests with OLS client: http://www.lxtreme.nl/ols/
*Developers, always :)

EDIT: test-3 attached
Got a question? Please ask in the forum for the fastest answers.

Re: Big news: Massive IR Toy updates

Reply #11
SUMP mode fixes. Seems to work 100% in my testing.

Oops, uploaded old zip file. Fixed now.

Re: Big news: Massive IR Toy updates

Reply #12
Is there an attachment for that ;)
Got a question? Please ask in the forum for the fastest answers.

Re: Big news: Massive IR Toy updates

Reply #13
What, it took less than a minute for me to change the zip file and you are complaining.... :)

Re: Big news: Massive IR Toy updates

Reply #14
well,
i tested the new firmware with winlirc
good news is receiving still works
bad news is transmitting doesn't appear to work at all. I can't pick up any signal from it.

What have you changed ? I am sure i can fix this without too much trouble.