V9958 replacement

Página 3/4
1 | 2 | | 4

Por gdx

Enlighted (6429)

Imagen del gdx

08-11-2021, 01:24

Do the ideas I give above inspire no one? Are they too complicated to do?

sd_snatcher wrote:

HMMC/LMMC blitter commands should activate the /WAIT pin. This would make CPU->VRAM transfers much faster, without the need to to check the blitter status bit, while keeping turbo support. This is the sole reason the VDP access was made so slow on the TR.

It would only be a good idea if the VDP is ultra fast since the Z80 would not be usable during this time.

sd_snatcher wrote:

The /CASX pin was a stupid idea. It should have been A8. This would have easily allowed 256KB and 512KB VRAM configurations using 44246 DRAM chips

/CASX should have been used in all modes from SCREEN 5 and the 64kB only configuration should not have existed.

sd_snatcher wrote:

Sprite priority: low priority = shows up only on transparent-color areas, like on the NES and SMS

Not essential.

sd_snatcher wrote:

There should be a separate palette just for the sprites on 16-color modes (including screen-10).
There should be a bit to enable 15bit palettes on any video mode, since the DAC is 15bit
Screen-8 should have 15bit palette support

Not essential.

sd_snatcher wrote:

Reading the I/O ports 9Ah and 9Bh should read the status registers (R#15)+1 and (R#15)+2, respectively.

I agree, it would be very useful. Or at last, these two ports would have to give the status of the most useful status bits (mirrors of the main bits of status registers).

Por Parn

Paladin (854)

Imagen del Parn

08-11-2021, 12:49

Sprite mirroring, anyone? What about a text mode with full attribute support per on-screen character (and maybe the ability of showing sprites)?

About extra palettes, I don't think it would be especially challenging considering the PC Engine VDP, which was released in 1987, had 16 palettes available to use freely among sprites and background tiles.

I would also love having some kind of tile-based mode without color restrictions, but I suspect this is a pipe dream.

All in all, I think @sd_snatcher's suggestions above are pretty great and, if doable without much effort, useful and cool. I think I'm personally more in tune with what he suggested.

Por Grauw

Ascended (10821)

Imagen del Grauw

08-11-2021, 13:20

What’s the purpose of this thread? I feel like a lot of people are broadcasting their wishes expecting there to be an imaginary person who is supposed to actually implement all this Smile.

And even though I also wish I could rewrite history and address some gripes, if someone would spend all that effort to make a “better” V9958, I’m not sure what good that would do me, because I’m writing software for existing MSX computers. I’d rather see V9958 FPGA reproductions be as faithful and accurate as possible, and an FPGA V9990 clone for more demanding applications (seeing that V9990 chips are becoming harder to find).

Por ivke2006

Expert (90)

Imagen del ivke2006

08-11-2021, 13:47

Well it’s almost Christmas LOL! . Just an replacement would be fine for keeping existing MSX computer alive.
Perhaps same functionality but with higher speed.
Not sure if a ‘turbo-VDP’ would create all kinds of timing issues.

Por gdx

Enlighted (6429)

Imagen del gdx

08-11-2021, 14:24

It would be interesting if there was also a version with the same pinout as the V9938. (or as much as possible for simplified implementation in MSX2s)

Por Parn

Paladin (854)

Imagen del Parn

08-11-2021, 15:30

Grauw wrote:

What’s the purpose of this thread? I feel like a lot of people are broadcasting their wishes expecting there to be an imaginary person who is supposed to actually implement all this Smile.

Why not? Maybe someone with the necessary technical skills could feel inspired.

Por ivke2006

Expert (90)

Imagen del ivke2006

08-11-2021, 17:17

Parn wrote:
Grauw wrote:

What’s the purpose of this thread? I feel like a lot of people are broadcasting their wishes expecting there to be an imaginary person who is supposed to actually implement all this Smile.

Why not? Maybe someone with the necessary technical skills could feel inspired.

How many visitors of MSX.org mastered VHDL or Verilog? Not very likely it will happen but I really hope I’m wrong in that aspect. Smile

Por Grauw

Ascended (10821)

Imagen del Grauw

08-11-2021, 17:50

ivke2006 wrote:

How many visitors of MSX.org mastered VHDL or Verilog? Not very likely it will happen but I really hope I’m wrong in that aspect. Smile

Well KdL, HRA!, Eugeny, Aquijacks, maxis to name a few… It’s certainly a skill that certain people have. V9958 FPGA emulation is being worked on in the context of the 1chipMSX project by the first two gentlemen, and maxis had his Procyon VDP project a few years ago as well.

Por sdsnatcher73

Enlighted (4294)

Imagen del sdsnatcher73

08-11-2021, 21:06

I don’t see this will really lead to anything.it may be my problem of understanding what this could be. What do we expect? If a V9958+ (at least it has a cool name) was released and it would be a drop in replacement (well after desoldering the original V9958) no existing games would benefit. Removing the sprite limitation through a register setting might prove cool but all the other things would mean new games would have to be made for it. And what machines would benefit from the enhancements? Would they work on a Z80 based machine or will only turbo R machines do?

I am not saying it isn’t cool. If this could e.g. be a V9958+ which was V9938 and V9958 pin compatible (switchable somehow with a jumper) then we could get cheap (???) V9958 into MSX2. If the MSX2‘s could benefit from the proposed enhancements as well than maybe software developers could use it. But those are a lot of ifs…

Por PingPong

Enlighted (4155)

Imagen del PingPong

08-11-2021, 21:15

@Grauw: the fidelty is for sure the most important thing. next the improvements, like F18A.
But improvements does not mean anything fancy at all. i think we should stay in what a 8 bit machine is expected to have.
For example i think that v9958 has already very colorful modes, but lacks speed in blitter operations.
So instead of trying to achieve a 19000 color palette (that per se is a good improvement) i judge more interesting a faster blitter and correct most of the vdp flaws that come pratically from the TMS vdp.

Página 3/4
1 | 2 | | 4