@alfkil As annoying as the shadow bug may be on the sam440ep, it does not make the game unplayable, the real problems are the crash game crash when you explode and die and the crash on exit. Have you had any luck working out what causes them and fixing them?
If the crash bugs can be fixed they would make the game quite playable on my sam440. It would also be nice to fix the black box shadow bug you are working on but that seems less important to be, unless the crashes are related of course
Anyway thanks for making Napalm playable and taking a look at wipeout as well
Sam440ep 667mhz 512megs OS4.1 + Minimig, 4MB RAM, ARM add-on board WinUae 2.3.2, OS 3.9, BB2, Catweasel MkIV Amiga 1200, BlizzardPPC 060/200 with SCSI, mediatorSX, Voodoo3, pci lan
I'm only starting to interpret the code, and thus I stumbled uppon the SetChromaTestBla that thellier mentioned. I haven't even located the game loop yet, much less had a chance to get to know what is happening in there. But it is comming all very slowly...
What you did seems to be OK But perhaps your warp3D driver dont implement ChromaTest at all ?!? Wazp3D too wasnt implementing ChromaTest so i have added a simple ChromaTest implementation in v51 = if not black then draw Also as Samo wrote this is not perfect = now on-screen energy-level and texts are well drawn but the ship-shadow is still a black box For the moment i dont know the cause (Wazp3D or WipeOut ??)
Alain Thellier
Note: You can also do (it will avoid any RGBA/ARGB problem) W3D_CHROMATEST_INCLUSIVE (=only texels inside the specified range are drawn, others are rejected) from 1 1 1 1 to 255 255 255 255
However using the new "No coloring" option in WaZp3D-Prefs the shadow under the ship will looks perfectly now altrough we have collateral effect as we miss all graphics details !
So to sum:
With coloring --> All is ok except the shadow under the ship No coloring --> All crap but the shadow under the ship is ok !
In a way or another we need to find a good compromise between the two to be able to make both working fine
So... after having fooled around with the W2097 garbage code for a little while, here is what is at:
- Tried to disable sound, did not fix the issue - Tried to insert lots of little annoying DebugPrintF statements in the main game loop, but it seems that the crash always happens in a different place - Serial output crash logs are useless, telling me an ISI happened right on top of a mr instruction and the like... sigh!
Guys, this is complicated! I am sitting with 210.000+ lines of garbage code, and every time I need to find something in there, it takes me hours and hours... I have to face it, that I might not be able to fix this issue. Sorry...
Sorry for that last outburst, I haven't given up (not completely anyhow), and I have found out new things.
First of all, the bad news is, that with update4, wipeout refuse to start completely. It shows the config window, and the first splash screen, but then it just hangs. One more stupid thing to fix, sigh...
The good news is, that I have found some float accesses, that are not aligned, and if I can fix those, I'm hoping (knock on wood) that they turn out to be the cause of the various crashes.
@alfkil No need to kill yourself trying to do it! But if it gets too much, then perhaps you can take a break & look at it again later (when you feel more motivated)? Don't want to see you get burnt-out, like happened to Olaf Barthel, etc.
First of all, the bad news is, that with update4, wipeout refuse to start completely. It shows the config window, and the first splash screen, but then it just hangs. One more stupid thing to fix, sigh...
First of all, the bad news is, that with update4, wipeout refuse to start completely. It shows the config window, and the first splash screen, but then it just hangs. One more stupid thing to fix, sigh...
I know that i said it before, but i think the only real solution for all of us (if we want to support in end all the warpos based games/apps) its to write our new powerpc.library from scratch. For sure you have good skills for, as well as Frank very-very helpfull in that terms and for sure he can help a lot with it. We can put it on google-pages as opensource, so anyone can do any sort of hacks/workorounds/fixes in any time.
By this way you always will have full access to this, and all the powerpc.library related bugs can be fixed in the library itself. And its library can worry about right alignment and hacks.
All in all it will be interesting in compare with disassembling messy code of games :)
Well, until my sam broke, I was planning to try and replace all the powerpc.library calls with proper aos4 calls. This shouldn't be too hard. If that would fix the issue, then we would know for sure, that the problem is with powerpc.library.
That and the float alignment issues.
Of course now I'm back to crapping my pants for Christmas with no computer to work on...
Of course now I'm back to crapping my pants for Christmas
:))
Maybe indeed a bounty ? Because while you will send it, wait when it will repairs, then wait when it will sends back , that all will take time. But knowing how much you already do, i am sure ppls will collect some euros just to send you new sam motherboard.
While powerpc.library were done at last by Re-Warp project, and Wipeout works with NovaBridge, i still want to make WOS_IRA works for my own use. Can you bring a little help with it plz ?
I just download wos_ira, and tried to disassembly simple "HelloWorld" from warpos example. There are archive (source, binary and stormc makefile):
Checking produce assembler code, and found 2 blocks with "dl.c". Tried in config via WOSCODE both of them, and produced ppc.asm code is real chaotic mess which are far from being readable or sane (just full of addis/cpis and long words). Just like it didn't disassemles properly.
And wos_ira offten crashes, too. But not sure if you still work on it, and if you in interst about crashlogs.
Maybe you can try it yourself and show how to produce correct powerpc output from this example ?
IIRC even the OS4 kernel should have some float alignment exception emulation code for that (it's required for example for the 440 FPU anyway), but if not my powerpc.library does include exception handler code for unaligned float accesses of WarpOS software. Just because of using HUNK instead of for example ELF executables much of FPU the data is aligned wrong in all WarpOS software.
If it helps I could release the exception handler code, or even the complete sources, of my powerpc.library, to someone. However there are several problems with that: - My powerpc.library code is a big mess, I never found the time to properly comment the code. - I don't have the time to help updating, or fixing bugs in it. - I wont release it for free. - And most important: After the very bad experiences with the former OS4 development team no (open) source code release from me any more. I would only release it to a single developer using a NDA, who then could release updated binaries, but not any sources based on it.
Amigans not used your library for a long, long time. She does not work, for example, from x1000 or x5000. A few years ago, a new PowerPC.Library library was created, better because it works on a new hardware. Making for free through a guy who is not looking for a power of problems.
Amigans not used your library for a long, long time. She does not work, for example, from x1000 or x5000.
Of course it can't work on X1000, X5000, A1220 and Sam460, my powerpc.library was implemented a long time before any of those systems were released, and any CPU needs special support, not only for completely different CPUs like the PA6T used in the X1000 but even for very similar CPUs like the 603e, 604, 750, 74xx and 440ep different code was required for working WarpOS emulation.