In rec.crafts.metalworking Lawrence Glickman wrote:
On 04 Dec 2004 17:10:21 GMT, Ian Stirling
wrote:

I can't imagine running anything my 2.5 giger can't handle. It gets
the job done.

A 2.5Ghz pentium may only be a couple of times faster than a P100, for
some tasks.

If it requires random access to large (greater than the on-chip cache)
amounts of memory, random access speed of chips hasn't really changed
in the past decade.

Some 'supercomputer' type computers took extreme measures to get round this.

OK, I put in 512MB of RAM and this thing is working as advertised.
Less than 512 is not a good idea, because I can see from my stats that
I only have 170MB of unused physical memory at the moment.

Once you get enough RAM so you don't have to do HD read/writes, you're
on your way to speed. But M$ has shafted everybody, because M$ loves

Not for all tasks.
For a few scientific computing tasks (nuclear simulations are a big one),
the big bottleneck is how long it takes you to read a dozen bytes from a
'random' location in memory.
All the information is in memory - it's just that the memory isn't fast
enough.
The increase in random access speed has not nearly matched the increase
in streaming speed.

IIRC, a 20 year old 30 pin 1Mb SIMM would typically take 150ns (billionths
of a second) for the first byte, then 100ns for the next one.

A modern 1Gb memory DIMM might read out the second (and subsequent) bytes
a hundred times faster than the 1Mb SIMM.
But the initial access to any location in that DIMM is still only around
at best 3-4 times faster.

In some tasks, this can be the limiting factor for calculation.