Uncovering the R800

I am just curious... Wasn't this manual (R800UsersManual.zip) already published at File Hunter (https://download.file-hunter.com/Books/EN/) for quite a while?

Don't take me wrong, I understand the importance of the documentation. I simply don't understand how it is "news". :)

Other than that, what would be the real impact for the current emulators (that offer R800) and for the MSX itself as a "community-maintained standard" in terms of what else can be done with the MSX Turbo-R based on those new discoveries?

Besides, in the list of new hardware released in the last few years, is there any new homebrew computer that claims to be compatible with MSX Turbo-R? If so, any impact on those?

Thanks. @msxall

Ok. My bad then...

What about the other considerations in the question about the impact and possibilities?

For emulators it will allow a more accurate memory timing emulation, or at least confirm the accuracy of the current implementation.

For the machines, some have added 1Mb-4Mb to the machines already. It is a much bigger hack than enabling the features we now have documentation for. I can easily see a Turbo-R with 8Mb of RAM and the DMAs/Ints hooked up.

On the other hand, none of those features are really part of the CPU core itself. They are just peripherals. It is not too difficult to add those around a regular Z80 if one feels so inclined.

But DMA could have potential good uses, like faster disk interfaces that didn't depend on the CPU as much, scanners and other stuff... It is too bad that Turbo-R ended up as this half baked effort, it could have been so much more, but I always thinks that MSX was always lagging a generation behind trying to save a few dollars, 2+ is what 2 should have been, and Turbo-R should have been 2+...

Performance wise the R800 performs close to the 68k in early Macs and Amigas. It would have been very "competitive", maybe until middle to late eighties. Just talking raw performance. But we need to keep in mind that the 68K was a 10 year old design by 1987. 386s and 68030's were ruling the world already and risc designs were popping everywhere.

The MSX as a whole was always thought to undercut price in the already competitive market of home computers (even in 1983). By that time the Z80 was almost a decade old already. 5 years latter it was already more than 50 times slower than a high end PC ($$$$$).

Maybe the R800 was conceived in 86-87 era and it looked good at the time. But the vision was not updated as it moved forward.

The DMA in the R800 is very flexible it could be used for video, memory copies, disk IO, etc. I do not believe it was there just for video. No need to have two of it if this was the case. And also the memory mapping and interrupts do not help video as much.

The amount of circuitry needed to enable the new features and use the DMA for disk transfers, for example, is pretty small.

But I hopped that the R800 had a magic way of entering in a 16bit mode that would give it a better way to access memory, even if it was the segmented manner of the 8086. Maybe allow PC, DATA and Stack to be fetched from different 64k windows like the 65816 (6502).