Hi guys

@DJBase

Yeah the Coldfire and CDTV websites are off-line at the moment.
At Xmas I suddenly had server problems (stupid 1and1) and they came back saying your stressing the server, running three websites and all (it was the SQL database that was slow and the CDTV and coldfire sites didnt use them, go figure) so to prove the point I took them off line... now in a totally unrelated thing my other (business) websites speed picked up. :-/ Mehh
So anyway the sites are going back up but Im redesigning them. (The CDTV website first) but Ive been spending my free time doing things that are more fun so.. ermm.. sorry

Ive just sent off a design to be printed thats a bit of a back to basics board, ive made a few general interface projects and stuff which I will make and test then get the new Coldfire design I did (which is a coldfire stuck to a CPLD stuck to the Amiga bus) printed and try that.
Ive got to e-mail Freescale (or chat on the Coldfire forum) about interrupts and a few other things but thats something to come later.
basically no news to announce at the moment, specially Coldfire based.

Oh looking back at the older questions I didnt answer, the main reason for not editing UAE is because I so dont have the software to do it, I have an old copy of Borlan (think thats what its called) but it needs specific windows sdk's and such..) plus Im not the strongest programmer in the world by any stretch. (lucky for me Freescale include GUI tools and nice examples on how to do all the Coldfire int code )
It would be the obvious way to check how well the Coldfire works on the miggy though, I know that, but I kinda went the other route of making a card that needed a custom kickstart, start off with the Freescale debug code, then clinux and onwards.

@Oli_hd

You should seriously think about going for only CPU, memory interface, glue logic and an expansion bus of some sort in the first version.

Coldfire SDRAM shouldn't be too hard, it's pretty much what the coldfire was designed for

And given a large enough expansion bus it doesn't limit the future possibilities... All new revisions with the same CPU would start off by making an expansion, making it work, then integrating that functionality on the board. New CPU versions would need a new main board, but none of the expansions until it's verified as working with all the plugin cards...

Something like the PC/104 formfactor might work well.

Hmm, this is GPL, but still worth looking at (it doesn't mean you can't make money off of it, it just means you gotta publish your sources. That would actually be worth a bit of money to us users, since it pretty much guarantees the project doesn't die when you loose interest):
http://www.freeio.org/library/toast.htm

Last edited by olegil on 03-Mar-2007 at 12:32 PM.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

While ColdFusion can be resource intensive I think the better solution is a modern dual core CPU. Have a few of those w/ about 4GB RAM and ColdFusion seems to work great!