Saturday 3 June 2017

8MB Fast-Ram board in the making.






8MB Fast-Ram board in the making. This will be added onto future STE & STFM boosters. These will be 32MHz speeds to match the 32MHz CPU speeds giving lightning fast access to "alt-ram".  This project will take a while to complete (12 months possibly), again due to lack of my free time to work on such things.

I keep getting asked, "will the boosters work with monster". Well no.  I have tried to get Alan on board with running his alt-ram board faster (along with other things) for like 4 years now and its just time to move on and design my own.

The future project, at least for the STE booster, is to add fast-ram along with 8 banks of flash ROM which can be used in addition to, or in place of the motherboard ROMs. Using my DUAL TOS boards it would give 10 ROM banks.

The design will also have a RTC like the Falcon one. Basically, I needed some NVRAM to store the booster registers in. So the RTC chip is ideal as it has NVRAM and a RTC :)

I know people want IDE, but at the moment I just don't have time to design a IDE, build it , test it.  I really don't want to design yet another IDE interface (actually I did design a new one from the ground up some time ago, but again no time to build it to test).

CF cards IMHO are just not stable and are a bit "old" now.  I actually want to use something like a IDE to SATA bridge and use something more modern. Or a IDE to USB bridge as USB sticks are still common. Though its not high up on my to do list.  There are other great solutions like Ultrasatan and Cosmosex for storage and I don't think we need another storage solution anyway.

One thing to bare in mind is the IDE based devices (which pass data via the CPU) Run a lot slower than DMA devices (hence the name Direct Memory Access). So DMA storage solutions run faster than CPU based once such as internal IDE.

Once this STE booster is designed, I plan to finish the STFM version and have some other ideas to try out. Though I think it will be the end of my booster work then.  For one, I have done so much work on these over the years, I don't want to spend the next 5-10+ years doing the same type of things over and over.

One thing in the mix, is trying faster than 32MHz speeds with the 68000.  This needs the 68SEC000 CPU which wires up much like a 020 CPU (so unfortunately not a simple CPU swap :(  ).  "Word on the street" is they are capable of 50MHz - 80MHz speeds.   So it is something Which is somewhere on my "to do list".

While I had hoped to move into 020 and 030 CPU's, and experiment with 32bit, Realistically, I am just never going to have the time to keep designing such things.  There others such as Rodolphe who are working on a 020 and 030 booster anyway. So there isn't much use in re-inventing stuff which is always being worked on.

More progress is posted on my site as it happens.
http://www.exxoshost.co.uk/atari/last/16mhz/index.htm









1 comment:

  1. Now this is pretty exciting. You could do some really interesting stuff with that kind of power, and instant-access storage, in an ST... or especially an STe...

    As for the mass storage thing, how about a device that foregoes all the IDE or USB bridging nonsense and just directly accesses an SD card? They're super common after all, generally reliable, inexpensive etc, and are designed from the ground up to have a really simple I/O interface. For the sort of IO speeds even a boosted ST can achieve, you probably don't even need to go beyond the most rudimentary MMC mode, intended for very simplistic MP3 players, dataloggers, early semi-smart phones or point-and-shoot digicams...

    ReplyDelete