Rough Idle and Dies If I give any throttle.

I just installed my TBI on my jeep this weekend and I am finally able to get it to start up, however it is not running good at all. When you first start it, the idle is a little slow, after a little time the idle goes up and you think its going to run ok. then the idle drops. This continues until it finally stalls and backfires. I am new to all of this and am trying to get this to a derivable state so I dont have to switch back to the carb for Monday. Please let me know if you have any ideas, Ill post the log below. Thanks in advance.

Here is a video of what it idling. http://www.youtube.com/watch?v=R5LOodVY_Po

I hear a lot of vacuum leak?. Maybe it's the audio though. What happens when you rev it up?

Ok, So I got it idling really well by messing with the timing. reves really nicely, but as soon as you put it in gear it all went to shit, ran like crap for about 30 seconds and then died. Couldnt start it up and eventually realized that the icm had fried. So, I put the distributer back to manually adjusting timing and hooked the tac wire up to the refrence to see how it ran that way.... It runs great when the pcm is not managing the timing. Took it around the block, and it runs really good. However when you give it a bunch of throttle it has a tuff time. Looking at the log I see that the injector pulse width drops during the times that I would have experienced the issue. So would that be the fuel pressure causing that? Adding the log below.I would really rather have the computer controlling the timing, but this is a good failsafe so I can run the tbi during the day, then switch to making things work at night.... Also, if I wanted to leave it this way I have read that I would need a filter for the tac line. I cant seem to find them anywhere, does anyone sell them? how dangerous is it to run without a filter?

Pulse width is commanded, based off a calculation of variables from several different inputs. If the pw is getting smaller while you rev, it's going opposite. Pw commanded has no correlation with fuel pressure (as far as the ecm is concerned) i think. I'm not an expert, but nobody else had responded yet. Hopefully my novice answer will trick a more expert to correct my answer, thereby, answering your original question. Reverse psychology ;)

You need to verify that your base timing and spark advance is rock solid before chasing other problems that may or may not exist. When the timing bypass wire is disconnected, the timing mark should not bounce around AT ALL at idle. If it does, there's your problem to chase. When the bypass wire is connected, the timing should advance to whatever your ECM is set to, and stay there at idle. I have found lots of TBI distributors with cracked reluctor magnets in them. Main symptom is what your engine is doing (rough/erratic idle), plus the vehicle eats ICM's. For a lasting repair, go buy a new or reman distributor. This is assuming the engine doesn't have some sort of big mechanical issue that timing cannot fix.