riscy
[Top] [All Lists]

3081+3730 worth it?

To: riscy@pyramid.com (Mips 3000)
Subject: 3081+3730 worth it?
From: Bill Broadley <broadley@neurocog.lrdc.pitt.edu>
Date: Thu, 15 Jul 1993 18:56:56 -0500 (EDT)
Reply-to: riscy@pyramid.com
Sender: owner-riscy@pyramid.com
I was thinking the system might be better without without the 3081+3730.

Would a plain mips 3000 40 Mhz + 3010 be cheaper???

If we ditched the 3730+3081 in favor of a mips 3000+(optional)3010+custom 
logic we could have a bigger cache (say 256k) use 72 pin simms, for max 128 
(256MB?) in 8 slots, save the $130 on the limited supplies of 3730's.  

How hard would it to replace a subset of the functionality of the 3730???
Seems like the 3730 is expensive and does more then we need and limited in 
other areas (simms).  How about moving floppy+IDE+2 serial ports+parallel+game 
to ISA for $25 or with 16550's for $45.  How about if we delete video?(GASP)

I'd think that for X11 stuff an s3-928 ($250) on ISA (5 MB/sec) would be 
faster and take less cpu then the dumb frame buffer ($100+) solution currently 
being considered.  State of the art x-terminals work on 1.25 MB/sec, couldn't 
an accelerated video card do well on ISA at 4 times the speed?  Of course if 
you couldn't afford an S3-928 ($250) you could get a lesser card. 

72 pin simms while more expensive would need 4 times less sockets, and 4 times 
less simms to interleave.  2 8MB simms that allow interleaved access are much 
more attractive then 8 1MB sims, 8 4MB sims, or 4 non-interleaved 4MB x 9's.  
I think once board space and simm sockets are taken into account the price 
increase would be marginal.

Considering that the memory efficient linux+80486 easily needs 16 Mb to run 
X11 while compiling without swapping, 64 MB or greater max would be nice.
I consider 4 MBx9 simms reasonable up to 32 MB's.

The combination of a fast cpu (40 Mhz mips) with the larger risc code size plus
linux+x11 (low locality) seems like the 24k 3081 cache would be a bottle neck 
for the system, I think that 256k or more of cache could make a big difference.
Someone posted a R4000 at 50 Mhz with 16k cache was up to twice as slow as
a 33 Mhz mips 3000 with 64 k cache.

So is this feasible?  

-- 
Bill                                    1st>    Broadley@neurocog.lrdc.pitt.edu
Broadley@schneider3.lrdc.pitt.edu <2nd  3rd>                 Broadley+@pitt.edu
Linux is great.         Bike to live, live to bike.                      PGP-ok

 

<Prev in Thread] Current Thread [Next in Thread>