Yet Another Demo Thread

[Amimal]

I'm starting this thread in order to list demos I've trouble getting to run properly, peradventure something can be done about them.

Howbeit, because the idea isn't to burden Toni with lazy reports, every demo referenced in here will have undergone a thorough, optimal conditions test, which's explained as follows:

1. All demos will be Break's, latest snapshot, unless otherwise stated.
2. All demos will be tested with default, unmodified settings (as in a first run) + "Best compatibility".
3. All demos will be tested with each of the six A500 Quickstart configurations.
4. All demos will be tested with both the current official WinUAE release as well as with the latest beta.
5. Even though only those demos failing the 12 tests will be reported, logs will be provided for the run of the most common A500 configuration alone.
6. The reports will contain "winuaebootlog.txt" and "winuae.txt" with "Log illegal memory accesses" on (this is kind of overkill for a log, so let me know if I should enable it at all).
7. These text files will be compressed with 7zip and the very archive will be named exactly after the demo file itself.
8. Screenshots may be attached as required.
9. Savestates will be provided if the problem occurs without sufficient immediacy, that time may not be wasted waiting for it to happen (I could easily attach a savestate for every report, but I don't know it that would be wise given the limited amount of space secured for attachments), thus I leave this up to Toni to request it or not.
10. Other tests may be suggested, but I had to keep it within reasonable limits because of time constraints.

As a starter, here's a couple that failed the test:

JoyDivision-DemoDisk01.adf: The second demo won't load at all, outputting the following error: "Unknown command Crusaders".

Ksylitol-Kolmiteho.exe: See the screenshot.

Best regards.

[Toni Wilen]

Logs are not needed in these kinds of tests. (floppy only quickstart config = can be 100% duplicated and it is core emulation related only.)

JoyDivision-DemoDisk01.adf

3.Ram Disk:> xfddecrunch Crusaders
xfdDecrunch 1.9 (09.09.1999) © 1994-99 Georg Hörmann, Dirk Stöcker

Reading "Crusaders" (106696 bytes)...
Decrunching "ByteKiller 1.3" file...
Decrunch failed: corrupted data

-> bad image. (EDIT: image is technically fine, OFS data blocks have checksums and checksums are fine: someone broke the executable and also "fixed" the checksum..)

Ksylitol-Kolmiteho.exe: See the screenshot.

A500 confirmed. Looks like some kind of joke production.

[Amimal]

Thank you Toni. If anything, that may serve as an encouragement to find a working copy.

There's also the fourth demo (F4) from "KORT-UntitledPack.adf" that shows a red garbage line at the bottom of the screen, which's tied to the displaying of the red letters right above it, and because of the extreme proximity, I didn't want to right away discard it as overscan area.

[Amimal]

Morque-Morquality02.adf: Demo works fine (right-click to access), but it refuses to pass the intro test.

Rare-DemoPackLiquid.adf: The fifth demo tends to leave an ugly trace after the scrolling logo/text, and it doesn't really look as it were intended. Very subtle and short-lived, but visible nonetheless. I've occasionally spotted similar behavior for other demos on real A500 hardware, so it could just be a less-than-elegant routine doing it's thing.

[Toni Wilen]

Originally Posted by Amimal

Morque-Morquality02.adf: Demo works fine (right-click to access), but it refuses to pass the intro test.

It has a bootblock virus.

EDIT: It is really your problem to check for viruses and/or scan for disk errors if it is DOS disk
(I prefer fixdisk for disk scanning)

[Toni Wilen]

Originally Posted by Amimal

Rare-DemoPackLiquid.adf: The fifth demo tends to leave an ugly trace after the scrolling logo/text, and it doesn't really look as it were intended. Very subtle and short-lived, but visible nonetheless. I've occasionally spotted similar behavior for other demos on real A500 hardware, so it could just be a less-than-elegant routine doing it's thing.

I didn't bother to check because it does look technically intentional, same keeps happening regularly at the edge of the scroller, whatever its position.

EDIT:

There's also the fourth demo (F4) from "KORT-UntitledPack.adf" that shows a red garbage line at the bottom of the screen, which's tied to the displaying of the red letters right above it, and because of the extreme proximity, I didn't want to right away discard it as overscan area.

Real A500 confirmed. Looks like it is invisible on normal TVs and need 1081/84/8833 with some height adjustments to make it visible.

[Foebane]

Can an Admin please merge this with the "Problematic Demos" thread?

[Toni Wilen]

Don't merge, lets keep this OPs demo thread only. I don't want this to become generic "everyone can post problems here". (Which unfortunately happened with the other one)

[Amimal]

Alrighty, just spotted an horizontal black line that doesn't look like it was intended to be there. It may be nothing, but better safe than sorry.

This happens in the demo "Amiga First" by Panoramic Design. Both the standalone executable and the other one included in Powerlord's "Cream of Norway 01" exhibit the exact same behavior.

Attached is the savestate that will take you straight to the problematic routine.

Protocol followed as with the previous reports.

Best regards and sorry for the inconveniences.

[Toni Wilen]

Oddly enough black line really is supposed to be there. Real ECS A500 and A1000 tested.