newbie building a scratch-built computer
ajp166 at bellatlantic.net
Sun Aug 5 07:22:25 CDT 2007
>Subject: Re: newbie building a scratch-built computer
> From: "Holger Veit" <holger.veit at iais.fraunhofer.de>
> Date: Sun, 05 Aug 2007 12:23:15 +0200 (CEST)
> To: "General Discussion: On-Topic Posts Only" <cctech at classiccmp.org>
>Tony Duell said:
>> IMHO, more important than the EPROM programmer is an EPROM emulator. This
>> is a box of RAM that connectes to the EPROM socket of the target system
>> (the board you'be just made) and also to a host machine (PC, parallel
>> port, serial port, USB?). YOu can quickly download perogams into the
>> emulator, which then appears exactly as the EPROM does ot the target, and
>> the latter can therefroe run said programs.
>> The advantage over using EPROMs is that you can re-write the RAM as many
>> times as you like (EPROMS have a limited number of program cycles) and
>> rewrite it quickly. Unless your programming is a lot better than mine,
>> you will go mad if you have to wait 20 minutes for an EPROM to be erased
>> and reprogammed. each time you want to make a change!
>I haven't so far reached the maximum # of programming cycles for EPROMs in
>any experiment. *) As I recommended you might use EEPROMs instead of
>EPROMs mainly because of their moderate capacity (you won't need PC FLASH
>ROMs with 1MB or more for an 8 bit system) and because they don't need
>erasing in the UV coffin.
Neither have I. I have 2716s, 2732s dating back to early 1980s that
have seen many cycles in many many projects and they are still good.
The thing that seems to kill Eproms is excessive erase times and
>*) Maybe your and my programming strategy differs; now with SW emulators
>for almost any old CPU being available ona PC, I tend to test my code in
>an emulator, often with single stepping first before I make a HEX file to
>burn it; even with EEPROMs - the old turnaround cycle of change a byte,
>program, insert and test how far it works is still too long even with RAM
>boxes. Admittedly, in the old times, without PC emulation, I did a lot of
>paper testing work first before burning the stuff into an EPROM; this
>discipline meanwhile degraded...
More information about the cctalk