Re: CBM-II Character Set and Colour Expansion

On Tue, Jul 17, 2018 at 1:35 AM, Mia Magnusson <mia@plea.se> wrote:
> Den Mon, 16 Jul 2018 22:31:10 +0200 skrev Francesco Messineo
> <francesco.messineo@gmail.com>:
>> On Mon, Jul 16, 2018 at 9:54 PM, Mia Magnusson <mia@plea.se> wrote:
>>
>> >
>> > A plus with using an extra eprom is that the eprom would be the same
>> > kind that is already needed so there is no risk that anyone building
>> > this thing will be hindered by lack of tools to program  a 16V8.
>> > (But yes, almost all eprom programmers probably can program a 16V8
>> > too, but still).
>>
>> Well, no, not the most vintage ones at least. I have several E(E)PROM
>> programmers
>> (Data I/O, Digelec, Hi-Lo) and none can program PALs. Some can program
>> microcontrollers, with or without adapters.
>> Yeah, I know, I still use '70s and '80s stuff and I can only blame
>> myself for that :)
>
> But then they are really old? I'm not sure but would gess that my
> All-03 and my Data-I/O Unisite would program the PAL's.
>

Yes, the Unisite and the All-03 will probably program PALs, but these
are quite new programmers
compared to what I have :)
The problem is that PAL programming algorithm were never made public,
so in the low cost programmers
they would almost never be included.
PAL manufacturers wouldn't like to be blamed for a marginal programmed
device or a device that would fail
too early, so they licensed the programming algorithms to well known
manufacturers and at a high cost too I believe.
Many people don/t realize how the correct algorithm matters even for
programming a simple EPROM.
Some months ago I helped a guy troubleshooting his fat40 PET and he
was programming 2532 and 2532A EPROMs
with a brand new chinese USB programmer (probably the most famous
current programmer on ebay).
It couldn't program 2532A at all (it's just a reduced VPP 2532) and
the 2532 (non-A) were programmed so badly
that the  PET was not even successfully booting most  of the times.
After he was trying to desolder every other IC on that poor PET, I
decided to send him a complete set of programmed
EPROMs for that board (using my DATA I/O) and they worked fine.
He sent me some of the EPROM he had programmed and they didn't even
verify correctly.
Yes, it could have been his particular unit that was DOA, ot who knows.
And yes, keeping a vintage programmer healthy requires some efforts sometimes.
YMMV
Frank