• Please review our updated Terms and Rules here

Floating Point Systems AP-120B Array Processor

Seems like the ROM table also included a reciprocal table, to seed an implementation of FDIV using the newton-raphson algorithm.
 
Welcome to the forum @myronwhite.

Thanks for filling us in on the history of this processor! It's always great to hear from the original creator of any of the devices that we collect.

- Alex
 
myronwhite said:
I was the primary H/W designer of the FPS-100 (1978) I don't have documentation for it, but I do have a full set of schematics for the AP120B.
The FPS-100 was a clone with somewhat higher logic density.
Click to expand...
would it be possible to get a copy of the schematics to add to the information on bitsavers?
 
I would like to enable that, but I don't know the best way to go about it. As I recall, the sheets are maybe 11x17 (was that 'C' size?), and there are a lot of them (~300 pages, maybe). I guess I could take it to Fedex and have it dis-assembled and scanned, but I would have to see how much that would cost. And the file size would be pretty large.

Can you make recommendations?
 
Last edited:
Upon checking, in addition to a set of schematics for the AP120B, I have a printout of what seems to be source code for the AP120B program ROM.
Also, I have a full set of schematics for the FPS-100, although I am don't think the backplane wiring is included. Can anyone give me contact info for USAGI?
 
I am in Hillsboro, Oregon. I would be happy to ship the documents to you.

I think I mis-spoke above, the program memory in the AP-120B was actually a RAM, not a ROM. Probably it was 512 locations by 64 bits. The FPS-100 schematic has a lot more pages (~2x) with fewer components per page, which was done for readability vs the AP120B schematics. But this makes the scanning task bigger.
 
Howdy! Another FPS alumnus here (Ed Borasky, January 1980 to July 1990). I was in sales and marketing, but did a fair amount of microcode programming as well.
 
Last edited:
myronwhite said:
Upon checking, in addition to a set of schematics for the AP120B, I have a printout of what seems to be source code for the AP120B program ROM.
Also, I have a full set of schematics for the FPS-100, although I am don't think the backplane wiring is included. Can anyone give me contact info for USAGI?
Click to expand...
Hello Usagi here!
I'd be more than happy to give them a scan, but also, you can't go wrong sending them to Al, he does top notch work running bitsavers.

In particular, the FPS-100 stuff would be amazing as we're really kind of up against the wall on building out that machine.
My e-mail is Nakazoto at gmail dot com, you can also hit me up on Discord if you prefer that, username is Usagi Electric!

Cheers,
David
 
myronwhite said:
Regarding the connection to the PDP-11 or DG Nova, each one used a custom PCB dedicated to the "host" (the formatter card, as far as I can recall). The PDP-11 formatter, in particular, had a cable that plugged into the Unibus slot on the PDP-11 expansion chassis.. There were a few other formatter cards, including one that interfaced to a CDC 1700 (??) that was used by Babcox and Wilcox in their work with Nuclear plants (prior to Three Mile Island !).
The reason these were called formatter cards was that the FP data format was specific to each host (in addition to the bus protocal) . The card converted these to the internal FPS FP format (10 bit exp, 28-bit mantissa). This obviously pre-dated the standardization for FP formats.
Click to expand...
Have the FPS systems also eventually been used on PDP-8 systems ?

I am asking bc in the newest usagi video there is an green sticker "fps" on the array processor.

Years ago i got an PDP-8i system which also had this sticker, so i always asked myself for what the sticker stands.
The PDP-8i system once has been used in Heidelberg at the german cancer research center, so there may be chances that they were a customer of FPS.
 
Wildfire said:
Have the FPS systems also eventually been used on PDP-8 systems ?

I am asking bc in the newest usagi video there is an green sticker "fps" on the array processor.

Years ago i got an PDP-8i system which also had this sticker, so i always asked myself for what the sticker stands.
The PDP-8i system once has been used in Heidelberg at the german cancer research center, so there may be chances that they were a customer of FPS.
Click to expand...
As far as I know, the FPS AP120B/FPS-100 array processors were never hosted by PDP-8 mini-computers. The first FPS array processors were produced in 1975, which was after the PDP-11 and Data General minis were being shipped in volumes, and the PDP-8 family was superseded by DEC. Further the PDP-8s did not have 16-bit data formats, which would have complicated things quite a bit for floating point formats.

Prior to 1975, FPS made floating point hardware which was quite a bit simpler, using bit-serial algorithms. These were implemented on 1 or 2 cards that were only ~4x10 inches, and fit in the host chassis (I think). I only ever saw a handful of these, so it is at least possible that there were add-in cards to a PDP-8. The volumes would have been really small, at least relative to the AP120/FPS-100 series, hosted on PDP-11s or DG Novas.

I don't recall seeing green FPS stickers, but it is true that green was the color used on FPS marketing material. (I think taken from the color of the PCB material at the time (FR-11). This is seen on the cover of various FPS documentation which have black rectangles over green backgrounds; these looked like the SSI DIPs on PCB assemblies.)

Do you have a link to the USAGI video that shows the sticker?
 
myronwhite said:
As far as I know, the FPS AP120B/FPS-100 array processors were never hosted by PDP-8 mini-computers. The first FPS array processors were produced in 1975, which was after the PDP-11 and Data General minis were being shipped in volumes, and the PDP-8 family was superseded by DEC. Further the PDP-8s did not have 16-bit data formats, which would have complicated things quite a bit for floating point formats.

Prior to 1975, FPS made floating point hardware which was quite a bit simpler, using bit-serial algorithms. These were implemented on 1 or 2 cards that were only ~4x10 inches, and fit in the host chassis (I think). I only ever saw a handful of these, so it is at least possible that there were add-in cards to a PDP-8. The volumes would have been really small, at least relative to the AP120/FPS-100 series, hosted on PDP-11s or DG Novas.

I don't recall seeing green FPS stickers, but it is true that green was the color used on FPS marketing material. (I think taken from the color of the PCB material at the time (FR-11). This is seen on the cover of various FPS documentation which have black rectangles over green backgrounds; these looked like the SSI DIPs on PCB assemblies.)

Do you have a link to the USAGI video that shows the sticker?
Click to expand...
Thank you very much for your answer !
I will try to take a pic of the sticker at my 8i system later.

Here is the link to the usagi video - look at circa 1.20:
 

Attachments

  • Screenshot_20250724_134210_Firefox.jpg
    Screenshot_20250724_134210_Firefox.jpg
    264.2 KB · Views: 9
You must log in or register to reply here.
Back
Top