Hijack v289: no more "button_pacing=n"

Hijack v289 is now available for download.

Note that many of you will get a Config Error message at each startup until you go and nuke the button_pacing=nn line from your config.ini file. This option is no longer relevant, and has been removed from the code.

New in v289:

-- no more button_pacing=nn parameter in config.ini

-- LONG button presses (.L) now work perfectly every time

-- removed some dead code to make Hijack slightly smaller again

-- minor tweaks to the display controller "sendcommand" logic

The major change in the code was to completely eliminate the unsynchronized button queue from empeg_input.c, in favour of the synchronized (timestamps) queue that Hijack provides. This guarantees that the pacing of buttons, short/long presses, etc.. will always be correct, regardless of player load or disk activity.

It works (very well) for me, hope it's useful to others as well.

Cheers

-ml

Thanks yet again for the fast coding!

I think your wife/girlfriend/gay lover/whatever has lost you to the EMPEG! LOL

I believe it is "wife", if I recall correctly.

Back on topic, this changing of the button behavior could be really really cool, and would benefit certain things in the player app too, wouldn't it?

Well, until my voice returns, the wife doesn't mind me coding in silence here (laryngitas).

The new button queuing method should make button macros much more reliable, especially when feeding stuff to the player software. Also, there should be less "lag time" between when a button is pressed and when the player sees it, versus previous versions of Hijack.

So, yes Tony, I think it oughta help all around.

The only downside I can see is that the Hijack button queue is only large enough to hold 24 press/release pairs at a time. This can be changed if it's a problem.

Cheers

Greetings!

Why do I get the feeling that there is a v2.00alpha14 out somewhere???

I've got no idea.

I've upgraded to v289 of Hijack and my Sony Stalk has stopped working!
I tried the same installation configuration with my other empeg (with Hijack v285 installed) and no problems.

The only things that are different on my v289 system as compared to the v285 system is Hijack and I installed charcoalgray99's XSL/XML v1.3 web pages on the v289 hijack system.

I see that Mark indicated that the 'button' queing mechanism is changed in this release - is this responsible somehow for the stalk issues?

My tuner works on both so its not that!

Yup, a typo.. I used "MASK" instead of "MATCH" on one line of the patch.

Fixed for v290.

-ml

Has anyone experienced extreme button lag since loading 289? I threw on a Live album 10 minutes ago and had to wait 5-10 seconds to skip to the next track. After the first button press registered further presses were still very sluggish. I killed the visuals and paused the track for a bit and it came right again.

I know things get sluggish at times, but this was worse than usual. Very wierd.


Regards,
Aragon

I remember that problem happening in beta11 and beta12 occasionally even without Hijack.

FAQ updated as of yesterday. I think i forgot to add a config.ini command somewhere along the way when i was super busy... anyone know which one's missing? I'll look it up otherwise.