🚀 go-pugleaf

In article <9dAz02zs58y201@amdahl.uts.amdahl.com> sbf10@amdahl.uts.amdahl.com (Samuel Fuller) writes:
>In article <688@UALTAVM.BITNET> ECULHAM@UALTAVM.BITNET writes:
>>If we throw away the hardware, and do the decimal arithmetic in software,
>>we get a lot simpler machine to build. But, what do the performance
>>numbers look like?
>> 
>>Hardware ==>   .9 + 1(.1) = 1
>>Software ==>   .9 + 7(.1) = 1.6
 
>>I've clearly exaggerated the performance gain of the decimal hardware.
>>Yet even with that, all that extra hardware cannot even double the
>>performance of the system.
>> 
>>Clearly, decimal arithmetic is one of those high cost, low payback
>>extensions. We should direct our efforts elsewhere.
>
>If a few extra chips can give a machine 60% better performance
>on the workloads that customers run, then it makes no sense to
>not put the chips in.

If a modest amount of hardware makes this much difference, people
would certainly want to put it in.

The REAL question is: how much difference does it really make?
Everybody please remember this was a "what-if" analysis, not an
analysis that claimed that it really cost 1.6X in performance.
At least several vendors who ahve looked at this and did care about
COBOL didn't think that it was this much degradation to overall system
performance, or at least, in designing their RISCs, included a few
simple instructions to help the issue, and chose to stop there.
-- 
-john mashey	DISCLAIMER: <generic disclaimer, I speak for me only, etc>
UUCP: 	{ames,decwrl,prls,pyramid}!mips!mash  OR  mash@mips.com
DDD:  	408-991-0253 or 408-720-1700, x253
USPS: 	MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086