STEEM SSE Blitter bug

A forum for anything about the Steem Engine STE emulator, comments, problems, bug reports etc. Steven Seagal regularly provides updated versions of the original STEem code. The current version is v3.9.4.

Moderators: Mug UK, Steem Authors, Moderator Team

WizTom
Atarian
Atarian
Posts: 5
Joined: Tue Jul 05, 2016 10:54 pm
Location: Finland

STEEM SSE Blitter bug

Postby WizTom » Wed Dec 13, 2017 8:44 pm

STEEM SSE doesn't update blitter source and destination address registers ($ff8a24 & $ff8a32) to point to last word accessed when finishing blit.
Real ST Blitter hardware does this (I verified on my old 520STE).
Very useful when doing lots of small blits in fullscreen code as one can just keep setting new XY counts, set busy and blitter will continue right where previous blit ended.

Tested on STEEM SSE 3.9.1.Win32.Debug.
Also tried to test with latest 3.9.4.Debug.Wind32.D3D but could not get that one to work
(TOS 1.62se gave Illegal Instruction at $e00340: move ccr,d0 )

I have vague feeling blitter worked correctly on old STEEM ?

Hatari 2.0.0 seemed to work correctly, but I much prefer STEEM's debugger.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2978
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: STEEM SSE Blitter bug

Postby AtariZoll » Thu Dec 14, 2017 8:32 am

Wellcome to the forum :D
Steven will not like this: why just not using old (v 3.2) Debugger for usual SW ? I'm sure that blitter emulation is OK in it - myself did some blitter code, and all it worked well. Not to mention tons of games and other SW.
I guess that Steven just does lot of changes, even hacks, and it is reason for some unexpected errors.
WizTom: what SW you working on ?
Famous Schrodinger's cat hypothetical experiment says that cat is dead or alive until we open box and see condition of poor animal, which deserved better logic. Cat is always in some certain state - regardless from is observer able or not to see what the state is.

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Thu Dec 14, 2017 8:55 am

No, I'm used to you recommending Steem 3.2 instead of Steem SSE by now.
People use whatever works for them, Steem 3.2, Hatari, Saint...
As far as I'm concerned, there's no Steem SSE bug until I could reproduce/acknowledge it.
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2978
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: STEEM SSE Blitter bug

Postby AtariZoll » Thu Dec 14, 2017 6:12 pm

Steven Seagal wrote:No, I'm used to you recommending Steem 3.2 instead of Steem SSE by now.
People use whatever works for them, Steem 3.2, Hatari, Saint...
As far as I'm concerned, there's no Steem SSE bug until I could reproduce/acknowledge it.

That's not exactly like that. I don't recommend using instead. I say that there are cases when simpler is better - as it was already discussed.
People using whatever works, and whatever is simpler to use. But as usual, people gets offended. Your last sentence confirms that, and really has no sense. You maybe wrote it as sarcasm, because this went not in bug report thread, but it makes not things resolved at all.
Famous Schrodinger's cat hypothetical experiment says that cat is dead or alive until we open box and see condition of poor animal, which deserved better logic. Cat is always in some certain state - regardless from is observer able or not to see what the state is.

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Thu Dec 14, 2017 6:25 pm

An example where Steem SSE goes wrong and not a real STE would be nice.
MOVE CCR,D0 is illegal on the MC68000, or at least I think so.
So I'm not sure those are real bugs due to changes, even hacks, causing unexpected errors, and that's what I meant, not offense.
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2978
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: STEEM SSE Blitter bug

Postby AtariZoll » Thu Dec 14, 2017 6:54 pm

Steven Seagal wrote:An example where Steem SSE goes wrong and not a real STE would be nice.
MOVE CCR,D0 is illegal on the MC68000, or at least I think so.
So I'm not sure those are real bugs due to changes, even hacks, causing unexpected errors, and that's what I meant, not offense.

What about SW for testing which opcodes cause illegal, bus error, or whatever - what I ran on real HW, what made table. Then you run it in emulator, and can see differences - so was made for cases exactly like this ?
I would say that it is not illegal from top of my head. And I'm wrong - it is, and actually has no much sense to use it, since can use move sr,do - with practically same effect.
So, that TOS1.62SE is strange, but maybe not - that could be just CPU test - but causes not bombs, and continues. Only that Steem SSE was set to stop at every exception :D
I think that we should be very nice with people posting here first time. Everyone make mistakes.
Famous Schrodinger's cat hypothetical experiment says that cat is dead or alive until we open box and see condition of poor animal, which deserved better logic. Cat is always in some certain state - regardless from is observer able or not to see what the state is.

WizTom
Atarian
Atarian
Posts: 5
Joined: Tue Jul 05, 2016 10:54 pm
Location: Finland

Re: STEEM SSE Blitter bug

Postby WizTom » Thu Dec 14, 2017 7:55 pm

i was too blind to find some more official bug report thread if there is one. sorry.

doing ste fullscreen demo thing, but that's kind of large and complicated to upload so hacking up little test rout atm.

ijor
Hardware Guru
Hardware Guru
Posts: 3270
Joined: Sat May 29, 2004 7:52 pm
Contact:

Re: STEEM SSE Blitter bug

Postby ijor » Thu Dec 14, 2017 9:43 pm

AtariZoll wrote:
Steven Seagal wrote:MOVE CCR,D0 is illegal on the MC68000, or at least I think so.

I would say that it is not illegal from top of my head.
So, that TOS1.62SE is strange, but maybe not - that could be just CPU test -


Move from CCR is not a valid 68000 instruction. It was added on the 68010 for the purpose of supporting virtual machine together with changing MOVE from SR to become a privileged instruction.

Yes, using MOVE from CCR is a common way to test for the presence of a 68010.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2978
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: STEEM SSE Blitter bug

Postby AtariZoll » Thu Dec 14, 2017 9:56 pm

ijor wrote:
AtariZoll wrote:I would say that it is not illegal from top of my head.
So, that TOS1.62SE is strange, but maybe not - that could be just CPU test -


Move from CCR is not a valid 68000 instruction. It was added on the 68010 for the purpose of supporting virtual machine together with changing MOVE from SR to become a privileged instruction.

Yes, using MOVE from CCR is a common way to test for the presence of a 68010.


Ijor, do you realize how your quoting is inappropriate ? Why you need to quote my wrong assumption, then skipping next sentence after it with proper answer ? At least put there .... for missing part.
Famous Schrodinger's cat hypothetical experiment says that cat is dead or alive until we open box and see condition of poor animal, which deserved better logic. Cat is always in some certain state - regardless from is observer able or not to see what the state is.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2978
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: STEEM SSE Blitter bug

Postby AtariZoll » Thu Dec 14, 2017 10:00 pm

WizTom wrote:...
doing ste fullscreen demo thing, but that's kind of large and complicated to upload so hacking up little test rout atm.

There is plenty of ways and sites where can upload for free even 100MB long files - in purpose of sharing.
Like this: https://www.mediafire.com/
Famous Schrodinger's cat hypothetical experiment says that cat is dead or alive until we open box and see condition of poor animal, which deserved better logic. Cat is always in some certain state - regardless from is observer able or not to see what the state is.

User avatar
Cyprian
Atari God
Atari God
Posts: 1464
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: STEEM SSE Blitter bug

Postby Cyprian » Thu Dec 14, 2017 10:58 pm

WizTom wrote:STEEM SSE doesn't update blitter source and destination address registers ($ff8a24 & $ff8a32) to point to last word accessed when finishing blit.
Real ST Blitter hardware does this (I verified on my old 520STE).
Very useful when doing lots of small blits in fullscreen code as one can just keep setting new XY counts, set busy and blitter will continue right where previous blit ended.

Tested on STEEM SSE 3.9.1.Win32.Debug.
Also tried to test with latest 3.9.4.Debug.Wind32.D3D but could not get that one to work
(TOS 1.62se gave Illegal Instruction at $e00340: move ccr,d0 )

I have vague feeling blitter worked correctly on old STEEM ?

Hatari 2.0.0 seemed to work correctly, but I much prefer STEEM's debugger.



Can you post a small test code?
IMO the BLiTTER emulation in the latest Steem SSE is almost perfect, there are some important (but never documented) improvements added by Steven this year. But if something is still wrong, would be cool to check that.
Last edited by Cyprian on Thu Dec 14, 2017 11:00 pm, edited 1 time in total.
Jaugar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
SDrive / PAK68/3 / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Aranym / Steem / Saint
http://260ste.appspot.com/

WizTom
Atarian
Atarian
Posts: 5
Joined: Tue Jul 05, 2016 10:54 pm
Location: Finland

Re: STEEM SSE Blitter bug

Postby WizTom » Thu Dec 14, 2017 11:15 pm

hmm. posted this once but it seems to have disappeared somewhere.. (really slow validation?)

did small & simple test code and ran that on steem, but that didn't bug like the more complex one..
so reduced my fullscreen demo routine to just the bug inducing part.
ugly haxor code has screenbuffer hardcoded to $100000 so needs 2 megs ste. blitter magic happends on fullscreen scanlines in dbf loop.

attached is zip file with program file and screencaptures on steem 3.9.1.debug, hatari 2.0.0 and pic off monitor on real ste with 2 megs.

i ran steem on windows 10, latest updates, tos 1.62 SE, normal 8mhz 68000, 2 megs etc.

(the previously mentioned "move ccr,d0" thing only happends on steem 3.9.4.debug, not the older 3.9.1 with same configuration,
so i didn't really look into it.)

tomi.
You do not have the required permissions to view the files attached to this post.

ijor
Hardware Guru
Hardware Guru
Posts: 3270
Joined: Sat May 29, 2004 7:52 pm
Contact:

Re: STEEM SSE Blitter bug

Postby ijor » Fri Dec 15, 2017 12:46 am

AtariZoll wrote:Ijor, do you realize how your quoting is inappropriate ?


That's your very personal opinion that is was inappropriate. I don't think it was. I quoted what I believed it was necessary. I didn't have any intention at all to remark your wrong and and ignore you right. As a matter of fact, I think it was pretty obvious from the context. You are free to believe whatever you want.

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Fri Dec 15, 2017 7:09 pm

WizTom wrote:attached is zip file with program file and screencaptures on steem 3.9.1.debug, hatari 2.0.0 and pic off monitor on real ste with 2 megs.


But reproduced & acknowledged :)
On v3.9.4, TOS162UK
Excuse me if I sounded so sceptical but it seemed a big bug that would have been noticed already, and I don't remember having changed those bits.

EDIT:
This was broken in v3.8.1 and it's apparently not a blitter problem, but the scanline starting at the wrong time on "line +20" overscan trick in this case (4 cycles too late?). Thx for the report and the test program. The whole overscan part of Steem will be refactored anyway, but it will take a while.


i ran steem on windows 10, latest updates, tos 1.62 SE, normal 8mhz 68000, 2 megs etc.

(the previously mentioned "move ccr,d0" thing only happends on steem 3.9.4.debug, not the older 3.9.1 with same configuration,
so i didn't really look into it.)

tomi.


I don't have that version of TOS, could you post it too?
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse

WizTom
Atarian
Atarian
Posts: 5
Joined: Tue Jul 05, 2016 10:54 pm
Location: Finland

Re: STEEM SSE Blitter bug

Postby WizTom » Sat Dec 16, 2017 5:51 pm

swedish tos162 and steem.ini file for steem.sse.3.9.4.debug.wind32.d3d i used.

tomi.
You do not have the required permissions to view the files attached to this post.

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Sat Dec 16, 2017 9:27 pm

Thx. In fact I had it, I was confusing Sweden with Spain. :)
About the "crash", it's no STEEM CRASHED alert box, right? Just the boiler stopping on an exception used as a test by TOS.
This can be disabled in options.
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse

WizTom
Atarian
Atarian
Posts: 5
Joined: Tue Jul 05, 2016 10:54 pm
Location: Finland

Re: STEEM SSE Blitter bug

Postby WizTom » Sun Dec 17, 2017 4:41 pm

Yes, boiler room giving popup window about illegal instruction, not steem itself crashing.
was just bit puzzled as previous version of steem sse did not give that popup with same tos image.
found the boiler-room option "Don't notify on exceptions", thank :)
any idea of when the scanline bug might be possibly fixed, like by spring ?

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Sun Dec 17, 2017 8:20 pm

I fixed it Friday when I saw it, and just uploaded dev builds with the fix, you tell us if that's better.
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse

User avatar
charles
10 GOTO 10
10 GOTO 10
Posts: 2217
Joined: Tue Aug 17, 2004 12:11 am
Location: ont. Canada
Contact:

Re: STEEM SSE Blitter bug

Postby charles » Mon Dec 18, 2017 4:48 am

just wondering about bios 6 tickcal , does steem now emulate this aswell?
atari is my lifestyle,not a hobby.
HOLD ON ! ! !,
Im printing unreadable characters ...!

User avatar
Steven Seagal
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2018
Joined: Sun Dec 04, 2005 9:12 am
Location: Undisclosed
Contact:

Re: STEEM SSE Blitter bug

Postby Steven Seagal » Sun Dec 31, 2017 10:49 am

And after video refactoring I get the exact same bug, so maybe it's something else!
Still investigating.
In the CIA we learned that ST ruled
Steem SSE: http://sourceforge.net/projects/steemsse


Social Media

     

Return to “Steem”

Who is online

Users browsing this forum: No registered users and 4 guests