P
PaulCsouls
Is there a general performance enhancement with dual processor boards
or is there only an advantage with certain applications?
Paul
or is there only an advantage with certain applications?
Paul
PaulCsouls said:Is there a general performance enhancement with dual processor boards
or is there only an advantage with certain applications?
JK said:A better question to ask would be whether a dual processor system
will outperform a single processor system of the same cost.
PaulCsouls said:Okay, I'll add some specifics.
Two Intel Xeon 2.4GHz 533MHz FSB 512 Cache for $222.00 each and a $250
Motherboard
VS
One Athlon 64 3700+ 2.4GHz 1MB Cache for $519.00 and a $75.00 Mother
board
Does 64 bits beat 2 processors?
The dual vs. single question has been tested pretty thoroughly over at Tom'sJK said:A better question to ask would be whether a dual processor system
will outperform a single processor system of the same cost. That would
probably mean using a faster processor than those used in the dual
system, a 10,000 rpm hard drive in the single cpu system vs a 7,200
rpm one in the dual processor system, and perhaps a better video card
and more ram in the single processor system. A motherboard for dual
processors is also more expensive than for a single one. If one wants
to use Windows, XP professional needs to be used with dual processors,
while XP home can be used with a single processor.
The dual vs. single question has been tested pretty thoroughly over at Tom's
Hardware and they seem to think that the dual Xeon has some advantages over
the fastest P4. I suspect that, like most benchmarking tests, a lot depends
on which tests you care to run.
http://www.tomshardware.com/motherboard/20040514/index.html
John said:As I said, it's a data point...
^^^^^^^^^^^^ BIG differenceFWIW, the desktop has a 32 MB (dedicated) graphics system, and the laptop
has a 64 MB (shared RAM) system.
Another benchmark (SI Sandra), which separates all the subsystems, confirms
that in EVERY case the dual-CPU desktop is faster. The 2 (arithmetic and
multi-media) CPU benchmarks were each 1.35 times faster on the dual 550,
compared with the single 1.2 GHz; memory bandwidth was 3.7 x; and the file
system (HD) was 2.4 x.
So, even though the 2 CPUs were not able to process data much faster once
the data got to them, they were served up that data much faster from RAM...
David Maynard said:Regardless of which calculation one thinks is the best estimate it's rather
intuitive that two 550s, even summing to 1100, can't possibly be 'twice as
fast' as one 1200 of the same processor class unless something other than
the processors is at play.
JK said:I expect the Athlon 64 3700+ to be much faster in most applications.
The Xeon is essentially a Pentium 4 that can be used in a dual
configuration. You might want to consider an Athlon 64 3800+
if you plan to do gaming or other high bandwith software, as it
has dual on chip memory controllers.
John said:That could well be true for a system running a single app.
However, once
you get into multitasking, a dual-CPU system will be much more effective.
I
run database queries on my systems that hog an entire CPU for 15-60 seconds
at a stretch. On the single-CPU machine, all other processes stop cold.
On
the dual, I work merrily along while the DB ferments in the background.
Also, it is difficult to compare 2 systems -- 1 single-CPU and 1 dual-CPU --
that are otherwise totally equal. Maybe I'll dust off my old 550, pull out
a CPU, and compare scores...
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.