65nm news from Intel

[prep]

We have long had desktop SMP available. Question: what legacy
software runs faster on two cores (whether on one or two chips) than
on one? Answer: none.

Anything run with X*, ie almost everything in existance.

Old tests I did long ago showed 30% increase was the minimum you
should expect.

* `with X' includes xterms and the like.

--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
comp.os.vms,- The Older, Grumpier Slashdot
Raw, Cooked or Well-done, it's all half baked.
EPIC, The Architecture of the future, always has been, always will be.

 

[Eugene Miya]

You may htinf the second is "cheap", but I don't. The second CPU and the
board that dgoes with it are certainly *not* "cheap".


Hmm, TF-1 was what, a decade ot two ago?

Yep, it shows how long it has been cheap.

Ah, back to the /360. ;-)
Yep.

Note that we do have GPUs and DMA masters. SMP doesn't solve all ills.

true.
--

 

[Eugene Miya]

I will use the TF-1 as an example. CPUs weren't boards.
There were multiple CPUs on a chip, so most of the pattern was reusable.

The development costs (board/chipset/BIOS) have to be recaptured across
fewer units sold, so will cost more. Look at the prices of boards with
on-board SCSI, for another example. OTOH, it doesn't cost all *that* much
more to throw another core on a chip.

If you are seeking economies of scale in parallel computing, it doesn't
work well that way. The redundance works toward scale economies.
The problem is that the infrastructure does not last a long.

The desk top is not a likely near term place to find most
multiprocesssors as a final product. Dual processors are marginal for
most uses. In the numeric area, you want to see performance improvements
of factors of 8-16 if not more, not 2-4.

--