How does that 20 mhz z80 turbo works? It was said that the z80 in the card becomes the main z80, and the MSX z80 becomes an "I/O z80".... Does that mean that all existing software needs to be patched to make this work? Or maybe there is an updated version of the BIOS to use with this card?
Для того, чтобы оставить комментарий, необходимо регистрация или !login
last night I commented on it in this thread, although I should have opened a new one link
the truth is that it is amazing what sharksym has achieved and I am looking forward to knowing more information.
P.S. it seems that he plans to use the z280 to be an R800
Ademir Carchano did 20 years ago a cartridge with a Z180 at 18 or 20 mhz, don't remember exactly. IIRC, you had to write new software to make use of it, and since he never released it, most people never knew about it.
If this one can use existing software, it will be something really remarkable.
Ademir Carchano did 20 years ago a cartridge with a Z180 at 18 or 20 mhz, don't remember exactly. IIRC, you had to write new software to make use of it, and since he never released it, most people never knew about it.
If this one can use existing software, it will be something really remarkable.
the z380 also came out at the time.
What is remarkable about this project is that it has already put soft without retouching and that it promises to be transparent to the user.
Cool stuff, really intriguing how he got it to work. I am wondering how he switches from the internal CPU to the external one. Z80 /busreq and /busack signals are not available on the MSX cartridge slot.
Cool stuff, really intriguing how he got it to work. I am wondering how he switches from the internal CPU to the external one. Z80 /busreq and /busack signals are not available on the MSX cartridge slot.
I wonder if there's not some FPGA magic happening here.....
If you look at this video, you will notice that I/O is a bottleneck. It looks like the Z80 is used as a proxy for performing I/O. So the raw processing power only applies to code which runs solely on the CPU. As such the for loop test speeds up a lot, but the movie player slows down because it frequently accesses the VDP and there is extra overhead (latency) introduced by the proxy running on the Z80.
If you look at this video, you will notice that I/O is a bottleneck. It looks like the Z80 is used as a proxy for performing I/O. So the raw processing power only applies to code which runs solely on the CPU. As such the for loop test speeds up a lot, but the movie player slows down because it frequently accesses the VDP and there is extra overhead (latency) introduced by the proxy running on the Z80.
Yes, I understand the VDP / PSG issue. Despite that, I believe that a huge gain can be obtained if you can process code with a faster z80 and a fast RAM to exchange data. It would be better of course if there was not the VDP to slow things down, but nothing is perfect.
I was wondering, the VDP output comes from the MSX or from the cartridge?
If you look at this video, you will notice that I/O is a bottleneck. It looks like the Z80 is used as a proxy for performing I/O. So the raw processing power only applies to code which runs solely on the CPU. As such the for loop test speeds up a lot, but the movie player slows down because it frequently accesses the VDP and there is extra overhead (latency) introduced by the proxy running on the Z80.
Yes, I understand the VDP / PSG issue. Despite that, I believe that a huge gain can be obtained if you can process code with a faster z80 and a fast RAM to exchange data. It would be better of course if there was not the VDP to slow things down, but nothing is perfect.
I was wondering, the VDP output comes from the MSX or from the cartridge?
If you look at this video, you will notice that I/O is a bottleneck. It looks like the Z80 is used as a proxy for performing I/O. So the raw processing power only applies to code which runs solely on the CPU. As such the for loop test speeds up a lot, but the movie player slows down because it frequently accesses the VDP and there is extra overhead (latency) introduced by the proxy running on the Z80.
Yes, I understand the VDP / PSG issue. Despite that, I believe that a huge gain can be obtained if you can process code with a faster z80 and a fast RAM to exchange data. It would be better of course if there was not the VDP to slow things down, but nothing is perfect.
I was wondering, the VDP output comes from the MSX or from the cartridge?
Internal VDP (MSX)
Cool! Is the main ram shared by both Z80s or each one has its own RAM?
