8088 MPH is REAL finicky with what it will and will not run on. Anything but an IBM 5150 with an original CGA will probably not work so well as intended.

Ampera wrote:

8088 MPH is REAL finicky with what it will and will not run on. Anything but an IBM 5150 with an original CGA will probably not work so well as intended.

That's true but most clones seem to get better results than what Jo22 is seeing. I'm not quite sure what the problem is, but it doesn't seem to be the CPU (since the initial speed test gives correct results) so it could be an incompatibility with the PIC, PIT or some timing-related aspects of the CGA (perhaps the status register). Or it could be something as mundane as a corrupted disk image. I was also going to suggest insufficient conventional RAM, but the effect of that is usually just some effects being skipped. Is that video of the party version or the final version? Whichever it is, you could try the other one and see if that works any better.

Another possibility is that DRAM decay is happening due to the way that 8088 MPH turns off DRAM refresh temporarily. It's supposed to do it in such a way that it preserves RAM contents, but it's possible that this doesn't work correctly on clones.

Thanks for the feedback, people.

I'm not sure which version I tried,
but at least chkdsk /f didn't show any corruption on that 30 years young 20MB drive. 😉

Which one would you recommend ?

And should I run the CGA Compatibility Tester ?
If so, which ones of these tests could give us a clue ?

Edit: The M35 has 640KiB of RAM installed.
About ~500 should be free for DOS, I think (have to double-ckeck).

Jo22 wrote:

Thanks for the feedback, people.

I'm not sure which version I tried,
but at least chkdsk /f didn't show any corruption on that 30 years young 20MB drive. 😉

That would find some sources of corruption but not all.

Jo22 wrote:

Which one would you recommend ?

Normally the final version is more compatible, but it does use some more advanced techniques in the first effect. So in this case it may be interesting to see if the party version works better.

Jo22 wrote:

And should I run the CGA Compatibility Tester ?
If so, which ones of these tests could give us a clue ?

The Vertical and Horizontal retrace detect tests in the M6845 Compatibility section would be the most interesting ones.

Jo22 wrote:

Edit: The M35 has 640KiB of RAM installed.
About ~500 should be free for DOS, I think (have to double-ckeck).

That's not the problem then - that should be plenty for everything except possibly the Kefrens bars effect (which is just skipped if there insufficient free RAM, I think).

Thanks, reenigne. I'll have a look at it in a few days. 😀

Quick update.

It was the party version that didn't work for me.

Oddly enough, another copy of the party version did work.

Even though file sizes were exactly the same, according to the directory listing.
Both came with the same *.nfo file also..

So thanks for the tip with the corruption thing, reenigne. That was a good advice. 😀
For safety reasons, I copied the files manually this time (no extraction using pkunzip on the 8088 PC).

The final version does also work, by the way.
Even better, perhaps. My PAL/NTSC TV set doesn't loose sync anymore.

Videos
Party version
Final version

I also did the CGA test, as promised.
Pictures below.

The more or less complete test can be watched here.

PS: Attached my corrupted version for analysis.

Edit: Links added, title changend (solved).

The files in your broken.zip are identical to the ones from the party version at ftp://ftp.oldskool.org/pub/misc/8088MPH.ZIP . So perhaps the corruption occurred when you unzipped the file on the 8088 machine.

Thank you very much for your reply! 😀
Yes, that's the same link as from pouet.net, where I got both the working versions from (both zip files are fine).

So yes, it likely really just was some form of corruption going on, since DOSBox shows the same glitch with broken.zip.
Perhaps it was just something so subtle, it had no visible effect to the file sizes or ol' Pkunzip "fixed" them.

In either case, I'm glad both the demos themselves and the M35 are fine. ^^