RaVeN-05 wrote on 2025-08-06, 14:48:

ok will test, thanks

https://nirvtek.com/downloads/RReady.Alpha.20 … ateIndicator.7z
MD5: b4d4098130f0847d25e0ff8fcf18fa29

- Fixes [ALT+R] and {ALT+X] frame rate indicator not rendering correctly.

[EDIT] Please check the MD5s of the download.

open a command prompt and type:
certutil -hashfile <paste full path to file locally> md5

I just found a bunch of strange folders on my web servers. Not sure whether they were left over from when my brother installed wordpress without telling me. I should hope not. Word press is gone now. [/EDIT]

RaVeN-05 wrote on 2025-08-06, 14:48:

ok will test, thanks

You'll have to use default dosbox-rendition.conf settings:
threadsafe=false
highperformance=true
picdma=true

Got the CGL driver to load my custom build using this:

1    unsigned long error = 0;
2
3    printf("3D Blaster Info 0.1 by Ian Hanschen\r\n");
4    printf("Attempting driver load (fixed)\r\n");
5    printf("Test build\r\n");
6    IFE0(cglLoad("", CGL_BOARD_VERITE_REV01, CGL_CDL_FIXED));
7    //ERR()
8    //IFE(cglLoad("VRITB170.CGL", CGL_BOARD_AUTO, CGL_CDL_FIXED));
9
10    cgl_screen.wReference = CGL_TOP_LEFT;
11    cgl_screen.wMode = CGL_M640x480x60Hz;
12    cgl_screen.wColorFormat = CGL_RGB5551;
13    cgl_screen.wBuffers = CGL_DOUBLE;
14    cgl_screen.wOption = CGL_D16S0;
15    cgl_screen.dwUB_BufferSize = 0;
16    OUT("cglQueryScreen\n");
17    IFE(cglQueryScreen(&cgl_screen));
18    ERR()
19    OUT("cglInitScreen\n");
20    IFE(cglInitScreen(&cgl_screen));

which produced this output:

1V:\DOS\3DBINFO>dos32a 3dbinfo
2DOS/32A -- Protected Mode Run-time Version 7.1
3Copyright (C) Supernar Systems, Ltd. 1996-2002
43D Blaster Info 0.1 by Ian Hanschen
5Attempting driver load (fixed)
6Test build
7cg 1QueryScreen
8Error: 1200 (000004B0)

My modded VRITB*.CGL file redirected the every single call to CGL port 0xC800 with a unique value for each function.

and in the dosbox-renditon log file:

12025/8/7|5:0:11.969 *** CGL input port C800 = value 1612 (bytes: 4)
22025/8/7|5:0:11.969 Registers regsA 00000000, regsB 00000000, EBX: 0017A3E0, ECX:00000001
32025/8/7|5:0:11.969 *** CGL input port C800 = value 1619 (bytes: 4)
42025/8/7|5:0:11.969 Registers regsA 00000500, regsB 00000000, EBX: 00000280, ECX:00000001
52025/8/7|5:0:11.970 *** CGL input port C800 = value 161A (bytes: 4)
62025/8/7|5:0:11.970 Registers regsA 00000500, regsB 00000000, EBX: 00000280, ECX:00000001
72025/8/7|5:0:11.970 *** CGL input port C800 = value 1619 (bytes: 4)
82025/8/7|5:0:11.970 Registers regsA 00000500, regsB 00000000, EBX: 00000500, ECX:00000001
92025/8/7|5:0:11.970 *** CGL input port C800 = value 161A (bytes: 4)
102025/8/7|5:0:11.970 Registers regsA 00000500, regsB 00000000, EBX: 00000500, ECX:00000001

0x280 is 640 and 0x500 is 1280 (2 bytes * 640) I *think*.

Zandvoort, Willow Springs and Watkins Glen are Rendition Ready:
https://youtu.be/6azWiJWRyGw
Zandvoort had to be reduced to 3 mips to load. Watkins Glen 1967 (Nascar 3) managed 4. Willow Springs was already Rendition Ready.

Force feedback support for the dos32a rendition bound version of icr2 is available here:
https://www.icr2.net/forum/showthread.php?tid=1581

(courtest of jxeboy/gplaps).

[EDIT]
GPL conversions in general have issues.
Virignia: https://youtu.be/W1IPto1fQTw
(In addition to the usual GPL polygon transparency issue seams are showing up in the skybox).
Spa-Francorchamps and Stardust Vegas (2000+1967): https://youtu.be/ssQncjPdo0o
Spa's a really difficult track to handle. The number of maximal mip maps retained has to be set to 2 for it to load and this causes a loss of texture on some distant polygons/sprites. Stardust works ok.

Tested/streamed Icr2 in game performance indicator get fixed perfectly, thanks. Visually looks perfect.
Still not tested on laptop, will do.

RaVeN-05 wrote on 2025-08-10, 09:51:

Tested/streamed Icr2 in game performance indicator get fixed perfectly, thanks. Visually looks perfect.
Still not tested on laptop, will do.

Thank you! It's good to know it's fixed. Are there any performance issues?

Yes, obvious fps drops, when too many cars, or complex scenes in custom Raceway, you aware of all of them.

RaVeN-05 wrote on 2025-08-10, 10:34:

Yes, obvious fps drops, when too many cars, or complex scenes in custom Raceway, you aware of all of them.

Ah yes, those. I just watched your stream, you get 30 FPS normally and in demo mode go as high as 190 fps. For some reason even with vsync off I can't go that high. Not entirely sure why.

[EDIT] My bad, it didn't use to go high. It does now with the performance optimisations[EDIT2] and sound disabled.[EDIT3] and carset.

Upcoming dual core update and CPU scaling:
https://youtu.be/v3RdhbDECWM
https://www.patreon.com/posts/upcoming-dual-1 … ntent=join_link

The two systems:
Core 2 Duo/GTX 660/4 GB DDR3
Ryzen 2700X/RTX2060 Super/32 GB DDR4
Demo mode with ALT+X was run without vsync. Unfortunately I forgot to reenable vsync on the C2D system.
Carsets used for demo mode are Cars95 and Reynard '99.
The tracks used for C2D are Laguna Seca (new) and Denver and for the Ryzen
Laguna Seca (new) and Houston. The Ryzen also runs Cars 2020 at Houston.

00:09 Core 2 Duo Demo Mode (Cars95)
00:50 C2D Reynard '99 Demo Mode
01:08 Ryzen Demo Mode (Reynard)
02:04 Ryzen Demo Mode(Cars95)
02:39 C2D Denver (Cars95)
05:14 C2D Denver (Reynard)
07:56 Ryzen: Houston (Reynard) Practice session
08:20 Ryzen: Houston (Cars2020) Practice session
08:52 Ryzen: Houston Race start (Cars20220)
09:21 Ryzen: Houston Race start (Reynard 1999)
09:37 Ryzen: Houston (Cars95) Practice session
09:59 Ryzen: Houston Race start (Cars95)
10:20 Ryzen: Houston Replay (Cars95)

@Raven-05, I'd still like to know if it runs any better on your laptop. Nascar '99 has issues with the last build, stability issues. I'll post an updated build tomorrow.

Tested Monster Track Madness 2 - works perfectly. Thanks)

RaVeN-05 wrote on 2025-08-12, 13:37:

Tested Monster Track Madness 2 - works perfectly. Thanks)

Woohoo and thanks!

[EDIT]
The final set of GPL demos:
https://youtu.be/JmJkwKbZloE

[EDIT2]
Upcoming change to RReady to force V1000 mode (nospecular=1 in rendition.cfg) for Monster Truck Madness 2. By default, RReady ran in V2k mode which meant fog, headlamp beams and reflections and other environmental effects didn't render. Making the change to rendition.cfg works with the current version, but the next release will force it for MTM2 automatically:
https://youtu.be/lRffjPUz9eM

[EDIT3]
Ovals p: Pikes Peak, Phoenix, Phakisa, Ontario and North Wilkesboro
https://youtu.be/t8G0v_gYkAw

N.Wilkes really needs that Nascar transparancey fix.

tested ICR2 on laptop it have 17 fps max in intro, and 1 fps min, often i see something around 9 fps average
vquake 1 min, 4 fps average, 19 fps max first and second demo

retested MTM2 with nospecular=1 and with 0 , everything works perfectly, visually no difference.

RaVeN-05 wrote on 2025-08-16, 14:03:

tested ICR2 on laptop it have 17 fps max in intro, and 1 fps min, often i see something around 9 fps average
vquake 1 min, 4 fps average, 19 fps max first and second demo

Thanks for that. So not great.

RaVeN-05 wrote on 2025-08-16, 14:03:

retested MTM2 with nospecular=1 and with 0 , everything works perfectly, visually no difference.

This is worrying. I wonder whether there's a patch for it that I haven't installed. I have looked and don't seem to be able to find one.

RaVeN-05 wrote on 2025-08-16, 14:03:

tested ICR2 on laptop it have 17 fps max in intro, and 1 fps min, often i see something around 9 fps average
vquake 1 min, 4 fps average, 19 fps max first and second demo

I think you said iCR2 did around 17-30 fps before and vquake didn't run at all.

need re-run to confirm, feels like it was faster before

RaVeN-05 wrote on 2025-08-16, 16:50:

need re-run to confirm, feels like it was faster before

You can test it by setting
[rendition]
Threadsafe=true
Highperformance=false
Picdma=false

It might be faster.

re-tested with old build its works faster even on intro gameplay.

Tested Scorched Planet. Well a lot of things need to be fixed and i think you already aware of them.
in case you have free hour you can re-watch https://www.twitch.tv/videos/2542392799
or 30 mins if you set video speed at 2x =)

P.S. Have another minor not important idea, is to use patches for dosbox which allow savestates and loadstates

RaVeN-05 wrote on Yesterday, 10:53:

re-tested with old build its works faster even on intro gameplay. […]
Show full quote

re-tested with old build its works faster even on intro gameplay.

Tested Scorched Planet. Well a lot of things need to be fixed and i think you already aware of them.
in case you have free hour you can re-watch https://www.twitch.tv/videos/2542392799
or 30 mins if you set video speed at 2x =)

P.S. Have another minor not important idea, is to use patches for dosbox which allow savestates and loadstates

A couple of things. This game must be run in FIFO mode, even on original hardware (Set SPEEDY3D_DMA=0 before running the game).

I *think* the messed up colours is a time of day simulation. I noticed it in yt videos, but I will test it. Anything to fire up the old retro PC.
[EDIT] Not time of day.

RaVeN-05 wrote on Yesterday, 10:53:

re-tested with old build its works faster even on intro gameplay. […]
Show full quote

re-tested with old build its works faster even on intro gameplay.

Tested Scorched Planet. Well a lot of things need to be fixed and i think you already aware of them.
in case you have free hour you can re-watch https://www.twitch.tv/videos/2542392799
or 30 mins if you set video speed at 2x =)

P.S. Have another minor not important idea, is to use patches for dosbox which allow savestates and loadstates

Also frame interpolation is meant to be used for ICR2, GPL and "Die by the Sword."