Mega STE timer speeds up. Bad 68901?

[Xyla]

Hello,

I have a really weird issue with one of my Mega STEs and I'd like to try to fix it. I was wondering if anyone has seen this issue before.

In some games, if I leave the machine running, it will speed up over time. In Airball, it will eventually get so fast you cannot control it. Since obviously the CPU speed can't be slowly going up over time, I assume the game is using a timer, most likely one of the MFP timers. running the game simultaneously on my other Mega STE and a regular Mega ST, only this one machine does this.

I figured the easiest thing to try to fix this is to replace the 68901 since it is a PLCC socketed chip in the Mega STE, but I will have to order one first... but I'm wondering if it is even possible that a bad MFP could cause this, or if there is something else that might be more likely.

Any thoughts or suggestions would be appreciated.

[mikro]

Just an idea, maybe video's vertical synchronisation is slowly drifting away?

[Xyla]

Just an idea, maybe video's vertical synchronisation is slowly drifting away?

No, not at all. I'm using CRTs and the vsync signal is completely normal. It has to be a timer independent of most hardware functions that some software is using to determine delays in loops or something, because overall the machine functions normally.

[frank.lukas]

... may be replace/check Y301 and/or C306/307 ?

mfp_mste.jpg (75.33 KiB) Viewed 1361 times

[Xyla]

... may be replace/check Y301 and/or C306/307 ?

Yeah that's a good idea. I don't have a way to test the crystal but if I'm ordering a 68901 I should get one of those also. I feel like it has to be something in that MFP circuit.

[exxos]

You could try GB6 and see if you get drifting results over time. It uses the 200hz timer tick for the results calculations.

[Xyla]

You could try GB6 and see if you get drifting results over time. It uses the 200hz timer tick for the results calculations.

Oh cool. Ok I'll try that, thanks.

[mfro]

If the 200 Hz system timer ran away, you'd have all kinds of different problems, I suppose.

[Xyla]

Well now I feel like I am losing my mind, because after running a bunch of tests it doesn't seem to be happening anymore.

I wish I had checked it again before I opened it and pulled the 68901, because now I don't know if re-socketing it fixed it or if something else was going on. I had just set this aside 3 years ago and planned to figure out how to fix it, then finally opened it up the other day and pulled the chip without testing again first.

But I have some evidence I'm not insane... I dug up two old videos I recorded when I was trying to diagnose it the first time. Here's Airball after it ran for a while, maybe an hour:



This is how it ran after first booting up, before the speed crept up:



But that's all I have. Letting the machine run for several hours yesterday didn't cause this to happen again. Also Gembench test didn't show any anomalies after looping for a couple hours.

I wish I understood what happened here.