FPGA VAX update

Holger Veit holger.veit at ais.fraunhofer.de
Sat Nov 5 11:33:12 CST 2005

Tony Duell wrote:


>>point emphasizes on the "digitalization" issue:
>>in the near future there won't be any analogue AM or FM 
>>transmissions/transmitters any more; even SW band migrates
>THis ia a great pity, and actually, given the number of cheap radios 
>around, I wonder if it will ever happen. I can't believe the public will 
>accept having to replace doxens of sets.
The roadmap for that is already present here in Germany, and I think 
also in the US. This is IMHO driven by the request to close the 
"analogue gap" that still allows to make private copies without 
continuous cash flow. The market forces behind this will force us to 
throw away all our receivers without any shrug of their shoulders - 
besides, if you have to buy new equipment (and maybe even renew 
it/upgrade it every year, it will produce incredible revenues. Your 
maybe expensive B&O receiver set will become obsolete at a simple click 
of someone's fingers. One just has to sell the right trojan horse to the 
unwashed masses: "digital" is cool, is new, is great, and will give you 
much better sound or videos you ever even imagined in your dreams (there 
is some small drawback, but don't take it too serious)....it's 
absolutely great, isn't it?

>>Unfortunately, many of the more interesting TTLs are now no longer 
>>easily available (you find lots of octal
>>drivers and registers,  but almost none of ALUs, multifunction chips, 
>>RAMs - even the classical
>>7490,7491,7492,7493 combo of counters for any purpose is reduced to 
>>7490/7493 - noone
>>needs to divide by 6 ('92) any more - you don't build digital clocks in 
>>TTL) - you have to seek for
>>them in specialized mail-order shops. This is no issue for us old farts 
>>who play with such stuff for
>>long time, but it is another hurdle for starters. A kit with all parts 
>>is much easier to acquire - but then,
>Waht somebody needs to do (and it can't be me for obvious reasons) is to 
>select some CPLDs/FPGAs where the download protocol is documented (that 
>is, you can progam the chip, given the binary file, without a proprietary 
>programmer), and use the tools (that's why it can't be me) to make some 
>useful logic functions -- things like an <n> bit universal shift 
>register, <n> bit ALU, counters, even JK flip-flops, gates, etc. Make the 
>binary files available for free download.
>Anyone wanting to experiment with logic and not wanting to use the 
>horrible CAD tools under an even worse OS, can then program up some chips 
>with the appropriate files and use the resulting devices like the TTL of old.
Do you really believe a newbie will be able in the beginning to select 
among Lattice, Mach, Altera, Xilinx, etc. programmable circuits where 
even the datasheet has to be read twice to understand whether this is a 
CPLD, EEPLD, FPLD, FPGA, with NOR/NAND/Antifuse program cells or 
external boot PROM, with almost uncomparable figures of the transistor 
equivalents, and whatever marketing invents in order to increase market 
share and to distiguish the basically same ideas? And then select the 
one circuit which has a cheap starter-kit, hopefully open-source tools 
for a not worse OS, and finally is not a dead horse that hasn't been 
delisted some time ago? Probably now, after MSEE degree and PhD, I would 
be able to find out after some time, but I won't expect the 13 year old 
boy (who built that TTL scrap computer 30 years ago) to come to a 
decision today.

If I were to start now, I'd be not too different from the horde today - 
have my playstation or the pee-cee and would edonkey or kazaa MP3 files 
as everyone, eventually study for MBA, but surely not EE. It's so bad. :-(


More information about the cctech mailing list