IRSS Still Freezing on Resume (1 Viewer)

[Clinton]

Tadr - if all you want is to send commands, then the following should work well enough.
Install IRSS test build 1.4.2.2096.
In input service configuration, uncheck Microsoft MCE receiver. Do not disable Windows Media Center services.
Copy cpp2005's modified Microsoft MCE transceiver files in accordance with his instructions.

Just being able to blast commands and handle events is pretty useful, in my opinion, and I am glad at least that works.
For the time being I have stopped trying on the receiving side. Sooner or later the read thread always aborts.

I will be watching with interest to see what happens with this excellent tool, and hope there will be some developments before too long.

 

[tadr]

Tadr - if all you want is to send commands, then the following should work well enough.
Install IRSS test build 1.4.2.2096.
In input service configuration, uncheck Microsoft MCE receiver. Do not disable Windows Media Center services.
Copy cpp2005's modified Microsoft MCE transceiver files in accordance with his instructions.

Just being able to blast commands and handle events is pretty useful, in my opinion, and I am glad at least that works.
For the time being I have stopped trying on the receiving side. Sooner or later the read thread always aborts.

I will be watching with interest to see what happens with this excellent tool, and hope there will be some developments before too long.

I don't see the modified transceiver files - are you talking about the files posted earlier in this thread? If so, it looks like he removed them.

 

[Clinton]

Tadr - there is a link to the previous version of the modified files in cpp2005's last post above. Although I cannot confirm right now, they should also allow output to blasters without the server restart hassle.

 

[cpp2005]

The suspend/resume fixed version is in the sticky thread, here.
There is another problem of exception in Input Service "The I/O operation has been aborted because of either a thread exit or an application request", which I tried to solve without success in this thread, that's why I removed it.

 

[cafe.racer]

What I don't understand is that it didn't always do it - I'm sure it used to work fine. I'm not sure what changed now, it was either an older version of IRSS, or the fact that I've put the TV Pack on.

It's a shame, at the events still work (so it's turning the TV and Amp on fine, which is the main thing) I just have to get off my @ss to change the source...

 

[Clinton]

I might be on to something. I made a minor change in the source code to handle the Win32 error in the read thread. Testing is continuing, but input now appears to work, and keep working.

While I was at it, I made it so the individual blasters are addressed correctly. Previously selecting blaster 1 only did not work (2 only and 'both' did work). Anyone else notice that?

 

[cafe.racer]

Excellent, marvelous, cheers and lots of other thanks

 

[Clinton]

Cafe.racer - here is a patched version of the MCE Transceiver files that works for me. The patch incorporates cpp2005's suspend/resume solution, and installs in the same way, over IRSS test build 1.4.2.2096.

There is a new StartReceive(port: n) entry in MicrosoftMceTransceiver_DriverVista.log. It should normally occur once as the receive port starts on system startup, and show a change to the learn port and back again before and after each learning operation. Any additional occurrences of this entry will be restarts at the points where the reading 'freeze' used to happen.

The patch also allows blaster port 1 to be addressed correctly.

 

[cafe.racer]

Brilliant - thanks very much.

Stopped the input service, copied the files into plugins, restarted. It seems to be workng great. I will watch it and report back, but THANKS again!

Edit: Still working fine, and I forgot to thank Aaron and cpp2005

 

[Clinton]

Cafe.racer - at least there are now another two happy Vista MCE remote users! As thread starter its your call if you want to formally label this issue 'fixed'.

In the absence of any problem reports with the patch, I will post the tweaked and commented source code in the sticky thread, for posterity.

I second the thanks to Aaron and cpp2005.