1) Stability/No lockups - ok
2) Corruption - after execution , no corruption
3) Red Square appearing on screen, no, corruption here i think
4) 60 Hz on exit, yes capture card can capture it.

odd thing windows can't start it tells problem with videocard it restarts to safe mode, than i can restart to normal.

https://fex.net/s/czkobnz

on fex 3 videos, of vga capture and one from phone

Thanks dude.

And now this:

https://1drv.ms/u/s!As-dKk-N73dSmDI0-mvdo0VcFFCZ?e=ltnX2t
MD5: a33f3f76be8a2d8bfba440662290ebd4

https://fex.net/s/t0cp8pp

video

Thanks.

One more:
https://1drv.ms/u/s!As-dKk-N73dSmDOuYbmvEkulMKTj?e=STDJ9Y
MD5: 4869ac0c70584dc406ef4cfcd2d978e2

This has the mystery mouse call INT 33h that Nascar fails on. It also has tweaks to the rendering. You won't see anything on screen, because the output display buffer is almost certainly wrong.

https://fex.net/s/mlpayy4

last execution, no corruption, i connected to monitor and dos ok.

Thanks!

Something funky happens to dosbox's mouse handler after running SPeedy calls.

That Int 33h mouse call isn't know on your system. It just returns the same parameters that it started it.

But with Nascar it calls it with the duff values and the mouse handler fails. I think I need to a Speedy3D Int 33h call which redirects mouse commands from the RReady window to dosbox, scaling the coordinates to fit Speedy3D 's(640x480).

Wanted to say this thread is going at the speed of a bullet 💀

So much to do, so little time.

SPlanet also fails with a mouse error. It's safe to say it's Rendition related. @ripsaw8080 suggest the interrupt vector's corrupted.

Does anyone know what the two parameters passed to RenUtil are?

1RenUtil a b

produces:

1INT 10h with AX = 0x158B, BX = a
2INT 10h with AX = 0x158C, BX = a
3
4INT 10h with AX = 0x1585, BX = b
5INT 10h with AX = 0x1586, BX = b

Re: RReady (Rendition Verite wrapper) Test Build (Alpha 1)
renutil.txt here

Thanks.

Yep, I discovered the txt file for it earlier. My post about it is here (same thing RaVeN-05 linked) and what I observed with it in the post below.
Re: Rendition Verite Thread

@Agent of the BSOD:
1) Can I have a copy of your V1000 BIOS, I just realised I don't have an image of it. Your (Rendition) ROM BIOS offset is 0x9FC, but I no longer appear to have a copy of it. The reason I ask is I'm trying to avoid loading a BIOS file into RAM and maintaining just a minimal image, but I don't have your BIOS.

2) Is RenUtil actually a TSR or does it just flip some switches in the BIOS?

Here's a dump of mine. Also, renutil is not a TSR.

Ta.

V1000/V2000 Microcode disassembler (Incomplete,):
https://1drv.ms/u/s!As-dKk-N73dSmDXYLf6oCXyrcZzL?e=cJNqWR
MD5: edc9bbaae092c1ce4c1c47e8f1678c52

The endianness might be wrong for the instructions, I can't really tell. Also some instruction opcode names are dumped without correctly listing params.

I'm not sure I'll even need this. I'm also not sure the ELF header's being dealt with correctly. Some files, like VOH, have duff ELF headers. Not sure whether they're even ELF files.

To decode:
RenDecode <filenam>

Source:
https://1drv.ms/u/s!As-dKk-N73dSmDa-3h_bPHrH_xYX?e=Bsn4Bm
MD5: 2dcd2c888aadc7255e5d0fa5dba13a6e

The instruction endianness might have to be swapped, not sure. To enable that decomment this line:

1Line 465 in RenditionUCode:
2		//if(bigEndian)
3			//instruction = _byteswap_ulong(instruction);

Capping the load of the Rendition BIOS to 32KB fixed the mouse bug with Nascar. Now it just goes into a loop waiting for a Sound Blaster interrupt, without issuing any Rendition commands.