FPGA history

Hi all,

for my PhD thesis I am looking for a citeable publication describing the FPGA technology advances during the recent years (say 10 years - or even from the very beginning), especially what advantages were achieved regarding available logic resources and maximum clock frequencies.

Does someone know about anything like this, e.g. some FPGA "timeline" ?

Regards, S. Heithecker

Reply to
Sven Heithecker
Loading thread data ...

Sorry

for the wrong email address. It must be " snipped-for-privacy@web.de".

Regards, S. Heithecker

Reply to
Sven Heithecker

You could ask that question in de.sci.electronics too.

Xilinx called them Programmable Gate Array or User Programmable ... or Logic Cell Array ( for that they had a trademark ) not FPGAs in

1984/86. But for better or worse they defined complexity in "number of gates" like gate arrays. Their part was "1k gates" and PLDs they defined as "100 gates".

Earliest "field programmable" parts were probably bipolar PROMs. PLDs like John Birkners PALs ( MMI 1975 ) and later GALs ( Lattice

1986 ) were cheaper, less complex, commercially successfull ( not all PLDs could/can claim that ). There were before 1984 complex PLDs around ( Signetics FPLAs ; TI ). Usually based on bipolar PROMs like PALs but often limited to certain applications like state-machines. Expensive and not widely used.

MfG JRD

Reply to
Rafael Deliano

There is an excellent article about the use of fast 4k-bit SRAMs as reconfigurable logic devices. I believe the article was published by fairchild around 1978 an was called "When a memory is not a memory" or similar. Sorry, that I have no better pointer. The article describes how to build multipliers, state machines, etc. from those RAMs.

Also, IBM used a technique in the 60ies where instead of using logic gates directly they used small boards with multiplexers on them. They could be configured using jumpers to implement various logic functions. Nearest neighbour connections where iconfigured by jumpers whereas the remaining interconnect used wire-wrap. The structure looked very similar to mux-based FPGAs (and likely invalidates many FPGA-related patents of the 80ies) Unfortunately I have no pointer to that either.

Kolja Sulimma

Reply to
comp.arch.fpga

There was recently a link to PDFs of old XCELL magazines in this group. Starting from issue 17, second quarter 1995. A nice read, giving perhaps some points for your timeline.

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
 Click to see the full signature
Reply to
Uwe Bonnes

Not FPGAs, but history...

Back in the 70s, it was common to built microcoded systems with the instructions stored in ROMs. The instruction format typically included a next-instruction field with branching done by ORing (or MUXing) a few bottom bits. Those bits often took longer to setup and were the limiting factor on the cycle time.

ROMs were typically implemented as a wide word followed by a mux to select the right chunk out of that word. Sometimes the manufacturers would give you better setup times on the address bits that fed the MUX. Guess which ones were used for branching?

Somebody made a neat part that included the ROM and some programmable branch logic (and registers and whatever). It was UV erasable. I forget who made it, but I think I can find a board and scrape off an old label if anybody is really curious.

--
These are my opinions, not necessarily my employer's.  I hate spam.
Reply to
Hal Murray

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.