bruceh@brushwud.sgi.com (Bruce R. Holloway) (01/05/90)
In article <4411@crdgw1.crd.ge.com>, barnett@grymoire.crd.ge.com (Bruce Barnett) writes: > > The Sun 4/110 has the new FPU. > > There are several different FPU units around. Weitek and TI I believe. > I remember something about early SparcStations being shipped with different > FPU's. > > Not all SparcStations are created equal? I thought every SparcStation 1 went out with exactly one Weitek WTL 3170 chip. LSI Logic was running an ad a few months ago with a photo of the processor board which showed it (along with several of their parts). And I've heard it reported that Weitek sold Sun 40,000 of those devices last year. I'm not sure about the Sun 4/110. Is its FPU a board option? Maybe it used to use earlier Weitek chips, but now uses TI's data path.
khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages) (01/05/90)
In article <47516@sgi.sgi.com> bruceh@brushwud.sgi.com (Bruce R. Holloway) writes: In article <4411@crdgw1.crd.ge.com>, barnett@grymoire.crd.ge.com (Bruce Barnett) writes: I'm not sure about the Sun 4/110. Is its FPU a board option? Maybe it used to use earlier Weitek chips, but now uses TI's data path. The 4/110 and 4/260 were shipped long before the 3170 was a gleam in anyone's eye ... FPU1 (or just FPU) was a Weitek 1164(adder)+1165(mult)+sun glue. It was unfortunately an option for the 4/110 (but this sans FPU option was not very popular, for obvious reasons). Early in '89 FPU2 was made available as an upgrade for units already shipped, and made standard for new orders. FPU2 is a TI8847+sunglue. FPU2 is what is employed in the 4/330/70/90 and 4/490 machines. The 3170 is what is customarily employed in the 4/60. It was designed to be a SPARC FPU so there is no glue needed. -- Keith H. Bierman |*My thoughts are my own. !! kbierman@sun.com It's Not My Fault | MTS --Only my work belongs to Sun* I Voted for Bill & | Advanced Languages/Floating Point Group Opus | "When the going gets Weird .. the Weird turn PRO" "There is NO defense against the attack of the KILLER MICROS!" Eugene Brooks
mslater@cup.portal.com (Michael Z Slater) (01/07/90)
>> There are several different FPU units around. Weitek and TI I believe. >> I remember something about early SparcStations being shipped with different >> FPU's. >> >> Not all SparcStations are created equal? > >I thought every SparcStation 1 went out with exactly one Weitek WTL 3170 chip. >LSI Logic was running an ad a few months ago with a photo of the processor >board which showed it (along with several of their parts). And I've heard >it reported that Weitek sold Sun 40,000 of those devices last year. The Weitek 3170 became available just as the SPARCstation 1 was announced. Some early SPARCstations had a piggy-back board in the 3170 socket, which had a TI 8847 FPU and LSI Logic's gate-array FP controller chip. I don't know if Sun shipped these to customers, or used them only in-house. The two FPUs definitely differ in their pipeline structure and performance. There will soon be (or maybe already are?) more FPUs that are pin-compatible with the 3170, but have different performance characteristics. Compiler writers need to pick one to optimize for, and presumably it will be Weitek's 3170, since it's in the SPARCstation 1. Michael Slater, Microprocessor Report mslater@cup.portal.com 550 California Ave., Suite 320, Palo Alto, CA 94306 415/494-2677
pal@murdu.oz (Philip Leverton) (01/11/90)
The Solbourne machines come with either Weitek 1164/65 or something called the Abacus (also from Weitek). (there may be other FPUs used in other models as well). What's an Abacus do for you and where does it fit into the FPU zoo? Does SUN use it at all? (perhaps not - there was a list of SUN machines and their FPU's in a previous article in this thread and it wasn't there). Do any other vendors use the Abacus? (I like the name :-) Philip Leverton Computing Services, University of Melbourne, Australia.
nm@wcfields.Solbourne.COM (Nick Mati) (01/16/90)
The Solbourne Series5 machine uses the Weitek 3171 FPU (Abacus) at 33.33 MHz. The 3171 is similar to the 3170 except that it has a Cypress CY7C601 (IU) interface wrapped around it instead of a Fuji/LSI Logic interface. Running the same code, the performance of the 3171 is basically the same as the TI8847 which is used in the Sun 4/490, the only difference being the TI8847 requires an extra interface chip to talk to the IU. Solbourne's Series4 uses a Weitek 1164/65 chip set. Nick Mati CPU Design Group - Solbourne Computer, Inc.
khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages) (01/16/90)
In article <1990Jan15.223217.6592@Solbourne.COM> nm@wcfields.Solbourne.COM (Nick Mati) writes: >The Solbourne Series5 machine uses the Weitek 3171 FPU (Abacus) at 33.33 MHz. >The 3171 is similar to the 3170 except that it has a Cypress CY7C601 (IU) >interface wrapped around it instead of a Fuji/LSI Logic interface. Running >the same code, the performance of the 3171 is basically the same as >theTI8847 which is used in the Sun 4/490, the only difference being >the TI8847 requires an extra interface chip to talk to the IU. .... This is inconsistent, either the 3171 is like the 3170 or it isn't. 3170 TI8847 fadds|d 5 8 fmults 5 8 fmultd 8 9 fdivs 38(?) 13 fdivd 66(?) 18 fsqrts 15 fsqrtd 22 My slide describing the 3170 has gotten a bit fuzzy (spilt coffee or some such). I'm sure that the data sheets are publically available from Weitek. The TI data is from the '89 Hot Chips Symposium TI handout. My Weitek ref box only has the prelim sheet for the 3170/1 and lacks cycle counts (I suspect that the lit is around here somewhere though... :>) TI seems to have made an algorithmic advance for div/sqrt ... which more than makes up for the 3 cycle lose for fadd,fmults, in most applications that I've studied. Cheers. -- Keith H. Bierman |*My thoughts are my own. !! kbierman@sun.com It's Not My Fault | MTS --Only my work belongs to Sun* I Voted for Bill & | Advanced Languages/Floating Point Group Opus | "When the going gets Weird .. the Weird turn PRO" "There is NO defense against the attack of the KILLER MICROS!" Eugene Brooks