TMS9118 has also clock out (like V99X8) so you can save more monney with this one
So does TMS9128 and TMS9129. Also they use only two VRAM chips (4-bit chips) instead of eight. Even though they still require the video encoding board they already save a lot of money on design. That's why there's more MSX computers using TMS912x than TMS992x.
Thanks for testing. The reason why ony the main sprite has problems is that it is the sole sprite that is updated on fly on isr.
The glitch occurs only on msx 1 ntsc systems becouse in pal that update falls during vblank.
In ntsc systems insted the border tracing is shorther so that sprite update falls in active area.
On msx 2 i/o timings are less strict so nstc or pal does not matter.
As far as I know, LVA510 is a PAL-Only video encoder so the Fenner must have a TMS9929 as VDP. While the 10 mhz crystal connected to the VDP is the right one for NTSC, the boards with LVA510 always have a built in crystal in them, which is notable in the pictures taken by Werner. If the Fenner is indeed NTSC, I would be interested on more information regarding the encoder board.
GSS = Goldstar
I happen to have a newer Goldstar Z80 chip. An Z8400BPS (Z80B). I tested it with the CPU test program and it detected as standard Z80 so I'd bet same happens with the Z8400APS chip.
Besides the "fugly" repair on the cassette port remote relay, the SPC-800 MSX looks perfect.
Dioniso is working to an interesting test program potentially able to auto detect the difference between TMS and Toshiba chips and tell if your machine has a clone vdp or not.
The test is still in early stage of development and needs testing but looks promising.
Sounds interesting. It won't be able to detect the difference between everything, but it helps a lot (no need to open up machines or have the service manuals for some VDP's).
