27 Jan
2010
27 Jan
'10
7:57 a.m.
I never said it wasn't coming along, or improving.
But it's still a far cry from having specialized chips for various functions.
ok, we've split into say 4 cores, and we have a GPU now,
still, not the same.
the PS3 is about the most advanced and its very hard to code for,
yet that's still multiple cpu's not specialized chips dedicated to tasks.
a general purpose cpu is going to be slower than a chip specialized to a task (for the
same task).
there will be a point were the general cpu speed gets so high it doesn't matter,
however, that's an incredible waste.
this is where early systems excelled, the amiga is the best example I can think of this.
specialized chips for almost every function, offloaded work from the cpu.
yet it still wasn't terribly difficult to code for.
technology improves, but not always for the best. (OS2 was better than windows, scsi was
better than IDE)
the market doesn't always pick the best technology (where is beos today? for example)
the list goes on and on, in the end, resulting in lists like this one.
here we have a huge hobbyist following of various "dead" technologies which will
never be produced again,
yet they continue life, providing things many others can't.
Dan.
Date: Wed, 27 Jan 2010 14:58:42 +0000
From: classiccmp at philpem.me.uk
To:
Subject: Re: General religious wars (was Re: Editor religious wars)
Dan Gahlinger wrote:
_________________________________________________________________
Reinvent how you stay in touch with the new Windows Live Messenger.
http://go.microsoft.com/?linkid=9706116
but still, that sort of trick isn't seen
today.
even the manufacturers of cpus have bought into the "faster mhz is better"
trick,
rather than multiprocessing.
Ultimately that depends on what you're doing. If you're doing something
that can be easily split across multiple CPUs (e.g. matrix maths,
raytracing, some types of image processing), then a multi-core CPU beats
an uber-GHz CPU almost every time (or at least has the potential to).
Best case, you're looking at around (cpu_speed * number_of_cores)
equivalent on a single-core CPU (less a bit for the threading code).
If you're single-threading then your quad-core 3.2GHz uber-CPU is only
going to work as well as a single-core chip of the same speed (although
if the OS has a multithreaded kernel, it might push some of the grunt
work of OS management / HW interfacing onto a spare core).
Take a look at some of the fairly new "GPGPU" concepts -- nVidia's CUDA
for example. A typical GTX200 class GeForce chip has ~192 cores running
at 1.7GHz, and can easily run rings round most modern CPUs (within
certain limitations). It's mainly intended for e.g. running the same (or
a very similar) operation on a ton of input data in one go -- things
like raytracing and games-style (particle engine, visual effects, etc.)
algorithms tend to work well, as do brute-force password-cracking
engines (!), as proven by the likes of Elcomsoft...
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/