MOS 65245
Ville Muikkula
Is the MOS 65245 IC (used for example in VIC-20) just Commodore's own
version of generic 74LS245? Why did they decide to manufacture this chip
themselves?
Nicolas Welte
>
> Is the MOS 65245 IC (used for example in VIC-20) just Commodore's own
> version of generic 74LS245?
Yes. The datasheet of the device says: The device is fully TTL and CMOS
compatible, and is pin-for-pin compatible with the 74LS245.
> Why did they decide to manufacture this chip themselves?
The datasheet doesn't give an answer to this question :-) Sometimes you can
find more MOS versions of simple TTL chips: the MOS 77xx/87xx series of chips
covers many of the TTLs found in the C64, but you only sometimes see one or
two of them used. Recently I found one in a 1541 as well.
Nicolas
Alex Johnson
>
> The datasheet doesn't give an answer to this question :-) Sometimes you can
> find more MOS versions of simple TTL chips: the MOS 77xx/87xx series of chips
> covers many of the TTLs found in the C64, but you only sometimes see one or
> two of them used. Recently I found one in a 1541 as well.
Maybe it was a supply situation? When they started production they
weren't guaranteed sufficient parts at a reasonable price so decided to
augment their sources with an internal supplier. Once they got going
either their need declined or supply/price improved for external sources
and they used whatever they had excess of or could obtain easiest. That
explains a mix of parts in various units across manufacture dates. It's
a lot like the food industry. You'll see "and/or" between some
ingredients on your snack foods and sodas because they can use
equivalent products to obtain the same effect and they use whatever is
easiest to obtain (usually dictated by surplus produced by really good
prices and over-buying on one of the ingredients).
Alex
Riccardo Rubini
"Ville Muikkula" <vmui...@ratol.fi> ha scritto nel messaggio
news:ai8hts$pn9$1...@pan.oamk.fi...
>Why did they decide to manufacture this chip
> themselves?
Minimizing costs, maximizing profits. This is how the game of running a
company works. I deal with operational research problems very often and I
can tell you that the best decisions on production cycles usually have a
well studied mathematical model behind them. This may be the case, who
knows.
For example, in the C128DCR, the WD1770/WD1772 manufactured by a third party
and the 6526/8521 for burst transfers were substituted by one 5710,
manufactured by MOS/CSG, Commodore semiconductors division, thus cutting off
the necessity to buy silicon from third party.
Probably the reason why not all the TTL logics were manufactured by MOS, but
just few cycles were made for 7707's and 65245, for example, is because
there were more disadvantages than advantages, in terms of production costs
for example, like personnel requirements, scheduling problems, fixed costs
of production etc...After producing some chips maybe they understood that it
was more expensive producing them their own than buying them from a third
party dealer.
Riccardo