TF1230 Compatibility

[Higgy]

What WHDLoad version of the game are they running? v1.3 only released March this Year.
Mine is fine.
1B motherboard, 3.2 + EXIDE, 128mb version

[8 Bit Dreams]

Ok, found Version 1.3 on
https://cdn.gamesnostalgia.com/getwhd.p ... o_v1.3.lha
Works great, so only demos are problem ATM....

Thanks to @Higgy for confirmation of the 3D Demo2 Bug .
Next issue:
AbsenceOFLight - sound only, no picture

[Danoo]

@Danoo

Can you confirm the two issues below are resolved during your tests? They are still present on my test setup.

Cuzco / Oxygene - artifacts on screen during cube / cheese display.

Drool This / Parallax : (can be found in supaduper's archive) With the latest firmware, starts and runs then Gurus the machine about 5 seconds into rendering the "3D FPS room" section near the start of the demo. The same behaviour is shown in the video at 4:12. This one seems to run OK on WinUAE

Just tried Cuzco / Oxygene on the TF1230, confirm artifacts on screen during cube / cheese display.
I also tried Cuzco on a TF1260 and it was even worst with the faster processor. This maybe a issue with the actual demo?

Do not have 'Drool This' on any of my installs, but from what I read previously maybe a WHDload setup issue?

Tried '3DDemo2' and 'AbsenceOfLight' on the TF1230 same issues that @8 Bit Dreams experienced. Both these demos worked fine on the TF1260. :)

TF1230 test rig: 2B, 64MB, 31 custom & ehide

[PaulJ_2.0]

@PaulJ_2.0 Are you still unable to get the card to boot with the new firmware on your 2B?

Tried on Rev 1d.4 both cards boot fine.

I will wait until another FW gets released and see if that fixes it for my 2B, if not I will delve deeper.

[terriblefire]

Sorry folks i'm not at home at the moment. I'm flying down at Goodwood Aerodrome getting my tailwheel endorsement (and also avoiding the COP26 chaos at home). Back Thursday and i can check things.

[terriblefire]

Ok ... dug a bit more into the Oxygene-Cuzco demo.

First thing i tried was disabling the MMU since the TS1230 doesnt have one. Result: no difference.
Second thing i tried was disabling caches with CDIS: Result: no difference.
Third thing i tried was to put the machine into 14Mhz mode with a firmware... Result: significantly worse artefacts on the cube/cheese bit.

I've tried the same demo on the CD32 + TF330 and i get the same issue. So its not specific to the 1200.

[terriblefire]

Bit more digging.. the WHDLoad Readme for the demo has...

Features:
---------
- Full load from HD
- Access fault fixed
- Interrupts fixed
- Bplcon0 color bit fixe
- NewIcon (created by me)
- Source code included
- Quit option that works with NOVBRMOVE (default key is F10)


The patch for this nice demo which was released at Saturne 1994 was
requested by Keito some days ago and as I always liked this very demo
I of course agreed to do the patch. It wasn't much work anyway, there
only was an access fault and some other minor problems to fix.
If you encounter any problems, try "NOCACHE", there is a quite a bit
of SMC in the demo which I didn't feel like patching/recoding.

i'm unsure if the issues are caused by self modifying code or if there is a weird interrupt issue i've not seen anywhere else.

[richx]

With the initial TF1230 sources upload post in the beta thread and the “tf1230r1_2021_10_30_alpha.zip” post only two hours apart, were the demo fixes included in the “initial code upload” upload to github? Any sources are much appreciated.

[terriblefire]

With the initial TF1230 sources upload post in the beta thread and the “tf1230r1_2021_10_30_alpha.zip” post only two hours apart, were the demo fixes included in the “initial code upload” upload to github? Any sources are much appreciated.

Dont understand the question. The initial post hasnt been updated.

[richx]

Dont understand the question. The initial post hasnt been updated.

It appears you started working on the fixes that became the tf1230r1_2021_10_30_alpha.zip release after the post about upload to https://github.com/terriblefire/tf1230. So from the post timestamps it looks like the additional firmware fixes were never committed to github?