Problematic Demos.

[Toni Wilen]

All Those Years (Demo) by Phenomena

[A500 1.3 ECS 512+512] In ECS the music slowing down when corruptions appears! (in OCS is fine as expected)

(Ps. I reporting it mostly because of the music is slowing down in ECS so not sure does it happens on a real machine.)

A500 confirmed (glitch uses more DMA time than non-glitch display)

Anzax First (Demo) by The Lost Boys

[A500 1.3 ECS 512+512] If ECS there are glitches on he left side! (in 4400 it is different)

4.9 is 100% correct.

Capone (Import Intro) by Bitstoppers

[A500 1.2/1.3 OCS/ECS 512+512] On the bottom scroller some white points appear! (if CE=disabled seems fine)

(Ps. If CPU=68010 white points does not appear but bottom red stripe glitches?)

Glitches A500 confirmed, A1000 confirmed.


btw, no need to say anything about 68010. It does not help anything.

[amilo3438]

Originally Posted by Toni Wilen

btw, no need to say anything about 68010. It does not help anything.

How it does not help? Aren't you at least a little curious does it happen the same on a real machine too?

From the 4400 release note: "68010 emulation is now cycle-accurate. Including loop mode."

Btw. I usually don't report anything with the 68010 because there was nothing to report until now!

Ps. I don't know how difficult is on your real machine to change between the CPU's... If there is possibility to damage the pins on the CPU then it is probably not worth the risk.

[amilo3438]

Tested on winuae.7z (09.09.2021)
-----------------------------------------------------------------------------------------------------------------------

Demodisk 05 - intro (Intro) by ESA

[A500 1.3 OCS 512] After decompression the music in the demo accelerates! (in 4400 it sounds normal)

-----------------------------------------------------------------------------------------------------------------------

[paraj]

Originally Posted by amilo3438

Tested on winuae.7z (09.09.2021)
-----------------------------------------------------------------------------------------------------------------------

Demodisk 05 - intro (Intro) by ESA

[A500 1.3 OCS 512] After decompression the music in the demo accelerates! (in 4400 it sounds normal)

-----------------------------------------------------------------------------------------------------------------------

Did some quick checks using Quick start A500 + Chipset = Cycle-exact (Full) it seems to work OK with 4.4.0 b20 and b21. b22/b30/b31/b32 exhibits issue.



i.e. change between b21 and b22 seems to be cause. P.S. intro seems to do something funky to display during decrunching in all versions.

[Toni Wilen]

Bozebobs fixed.

Originally Posted by amilo3438

How it does not help? Aren't you at least a little curious does it happen the same on a real machine too?

From the 4400 release note: "68010 emulation is now cycle-accurate. Including loop mode."

Btw. I usually don't report anything with the 68010 because there was nothing to report until now!

Ps. I don't know how difficult is on your real machine to change between the CPU's... If there is possibility to damage the pins on the CPU then it is probably not worth the risk.

I don't care how something timing critical runs on "unsupported" CPU. Practically no one had an 68010. _If_ someone had made timing critical program that requires 68010, that would make a difference

Yes, socket and CPU pins will get damaged sooner or later. It is not worth the risk. (and I do have some other ways to use 68010..)

Originally Posted by amilo3438

Demodisk 05 - intro (Intro) by ESA

This is seriously bad. and probably dangerous to (some) CRT displays. It pokes "random" values to vertical position register which can result in very strange continuously changing refresh rates.

Speed up cause is same side-effect as in Circle scroller / UFO. Not sure if I bother with this too soon, it is simply far too seriously stupid, and not just normal accidentally stupid.

Hex edit the executable, offset 132 ($84). 00 DF F0 2A -> 00 DF F1 80 (to get usual background color flashes during decompression) or F1 FE (if you want to disable decompression "effect")

EDIT: I only got hang and blank screen when I run this on real A500 (connected to OSSC and LCD TV)

[Foebane]

Originally Posted by Toni Wilen

This is seriously bad. and probably dangerous to (some) CRT displays. It pokes "random" values to vertical position register which can result in very strange continuously changing refresh rates.

I had an experience like this with my CRT back in the day with the demo "Boundless Void" by Nemol. It was a plasma effect that looked fine when run from my original A500, but when I ran it on my A1200 years later, the entire screen (vertical and horizontal hold) failed and the screen was rolling as if the TV was broken, but instantly restored to normal when the plasma effect ended and the next effect appeared. My guess is there were some changes to the registers that were poked to achieve the plasma inbetween chipsets.

[amilo3438]

-----------------------------------------------------------------------------------------------------------------------

F7 - Wildfire / Reaktor

[A500 1.3 ECS 512+512] There are after a while (after water balloons) appear glitches (sprite residues) if/when crash to pipe tower!

To faster start the game: After booting go to CLI with Shift+D and type 57g (return)!

Ps. Sorry, seems there is no separate file for this.
-----------------------------------------------------------------------------------------------------------------------

[Zarnal]

Originally Posted by amilo3438

-----------------------------------------------------------------------------------------------------------------------

F7 - Wildfire / Reaktor

[A500 1.3 ECS 512+512] There are after a while (after water balloons) appear glitches (sprite residues) if/when crash to pipe tower!

To faster start the game: After booting go to CLI with Shift+D and type 57g (return)!

Ps. Sorry, seems there is no separate file for this.
-----------------------------------------------------------------------------------------------------------------------

All glitches are confirmed on real hardware ( pic 1 and 2, A500 OCS 512+512, tested by Mutt2828/AF forum ).

[amilo3438]

Originally Posted by Zarnal

All glitches are confirmed on real hardware ( pic 1 and 2, A500 OCS 512+512, tested by Mutt2828/AF forum ).

If there was any doubt to the emulation accuracy, this has now been finally confirmed. Thanks for testing!

[Foebane]

Wonder why so many of these problematic demos are those pre-1990 ones that are so similar to each other, before design and a focus on variety took over? Those coders must've really been copying each others' code when it "hit the metal".

[Zarnal]

Originally Posted by Toni Wilen

Phenomena-Dot (aka Phenomena-SpectreParty) should work now without scroller glitches.

All issues ( scroller glitchs and cli returns in A500/kickstart 1.3/OCS ) are back with b34.

@amilo3438

Can you confirm please ?

[amilo3438]

Originally Posted by Zarnal

All issues ( scroller glitchs and cli returns in A500/kickstart 1.3/OCS ) are back with b34.

@amilo3438

Can you confirm please ?

Yeah, it is back again. (waiting for the second attempt, hope it will be successful)


CEN Demo (Demo) by Rare (or see post #827)

Originally Posted by Toni Wilen

Scroller glitch (left side and line below scroller having garbage) confirmed. Vector objects are correct on real ECS A500. This needs to be investigated more.

Hopefully in the next version! (ECS A500 - needs to be investigated more)

[Zarnal]

Originally Posted by amilo3438

Yeah, it is back again. (waiting for the second attempt, hope it will be successful)


CEN Demo (Demo) by Rare (or see post #827)


Hopefully in the next version! (ECS A500 - needs to be investigated more)

@amilo3438


I've tested again and vector issue happens only with ECS Agnus 512+512. A500/Kickstart 1.3/ ECS Agnus 1Mb is ok.

[amilo3438]

Have tested it again on WinUAE 4.9.0 b34:

QS A500 1.3 OCS 512 = vector object seems fine
QS A500 1.3 ECS 512 = whole screen glitches when vector object appears
QS A500 1.3 ECS 512+512 = only vector object glitches
QS A500 1.3 ECS 1MB = vector object seems fine

(Note: QS = Quickstart configuration!)

Not sure what real config Toni used for confirmation in ECS!?

(JFI - Same happens also in last official WinUAE 4.4.0!)

[amilo3438]

(Ps. Tested in WinUAE 4900b35)


Mystic Places BBS (Demo) by Dual Crew

QS A500 1.3 ECS 512 + 512 = crashes somewhere in the demo! (Note: If KS=3.1 it doesn't crash!)
QS A500 1.3 ECS 1Mb = doesn't crash

Does it crashes on a real A500 1.3 ECS 512 + 512 ?


Full Power (Demo) by Paranoid

QS A500 1.3 ECS 1Mb --> Press the LMB on the "greetings screen with bird" and in the next scene vector objects glitches or sometimes all crashes!

(Note: Press LMB on each screen to move forward faster!)

[Toni Wilen]

Originally Posted by amilo3438

Mystic Places BBS (Demo) by Dual Crew

QS A500 1.3 ECS 512 + 512 = crashes somewhere in the demo! (Note: If KS=3.1 it doesn't crash!)
QS A500 1.3 ECS 1Mb = doesn't crash

Does it crashes on a real A500 1.3 ECS 512 + 512 ?

It does crash, in exact same position. Blitter write corrupts memory. Probably works accidentally (corruption misses important code) in other configs.

Full Power (Demo) by Paranoid

QS A500 1.3 ECS 1Mb --> Press the LMB on the "greetings screen with bird" and in the next scene vector objects glitches or sometimes all crashes!

(Note: Press LMB on each screen to move forward faster!)

A500 confirmed. Tried 3 times, it crashed once identically.

[Zarnal]

Originally Posted by amilo3438

Have tested it again on WinUAE 4.9.0 b34:

QS A500 1.3 OCS 512 = vector object seems fine
QS A500 1.3 ECS 512 = whole screen glitches when vector object appears
QS A500 1.3 ECS 512+512 = only vector object glitches
QS A500 1.3 ECS 1MB = vector object seems fine

(Note: QS = Quickstart configuration!)

Not sure what real config Toni used for confirmation in ECS!?

(JFI - Same happens also in last official WinUAE 4.4.0!)

@Amilo3438


Not confirmed on A500 ECS 512+512 ( tested by Mutt2828 AF forum ). Ship is ok ( pics 1,2,3,4 ).

Edit : oops, she has been tested already.

Originally Posted by Toni Wilen

Scroller glitch (left side and line below scroller having garbage) confirmed. Vector objects are correct on real ECS A500. This needs to be investigated more.

[amilo3438]

Originally Posted by amilo3438

Have tested it again on WinUAE 4.9.0 b34:

QS A500 1.3 ECS 512+512 = only vector object glitches

Seems above problem is fixed now with new winuae.7z (updated from 20.11.2021).

The problem was probably related with topic on this thread: https://eab.abime.net/showpost.php?p...94&postcount=8 (I guess).


EDIT: But in todays WinUAE Beta 41 (RC2) it works fine only if subpixel emulation is enabled! (on above mentioned winuae.7z was working also when subpixel emulation was disabled ?)

[amilo3438]

Tested on winuae.7z (22.11.2021)
-----------------------------------------------------------------------------------------------------------------------

Meltdown (Demo) by Absence & Excellence

[A500 1.3 ECS 512 + 512] At the star part there are some glitches? (same in 4400)

For example on this video it seems there are no glitches (unknown configuration): [ Show youtube player ] (at 1:22)


EDIT: Tested again but in winuae QS A600 or QS A500 ECS 1Mb configuration and it does not glitches!

[Toni Wilen]

Originally Posted by amilo3438

Tested on winuae.7z (22.11.2021)
-----------------------------------------------------------------------------------------------------------------------

Meltdown (Demo) by Absence & Excellence

[A500 1.3 ECS 512 + 512] At the star part there are some glitches? (same in 4400)

For example on this video it seems there are no glitches (unknown configuration): [ Show youtube player ] (at 1:22)


EDIT: Tested again but in winuae QS A600 or QS A500 ECS 1Mb configuration and it does not glitches!

Real ECS A500 test:
- 0.5M chip only: no glitches
- 0.5M+0.5M: glitches
- 1M chip only: no glitches