31 Jan
2006
31 Jan
'06
1:37 a.m.
http://www.geekpatrol.ca/article/101/geekbench-comparison
Scroll down to "More Integer Performance." Presented without further comment.
--
--------------------------------- personal: http://www.armory.com/~spectre/ ---
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- The only thing to fear is fearlessness -- R. E. M. -------------------------
31 Jan
31 Jan
2:45 a.m.
On 1/30/06, Cameron Kaiser <spectre at floodgap.com> wrote:
http://www.geekpatrol.ca/article/101/geekbench-comparison
Scroll down to "More Integer Performance." Presented without further
comment.
I didn't find "more integer performance" in the article, but rather
"integer
performance".
The test was written with a specific cache size in mind, so would work as
long as
that condition was met, and suck if it was not.
Memory tests are not useful comparing the two systems, as well because of
the
intel need to address memory w/o regard to whether references are word
aligned
or not in the intel architechure. tests can be made to show the Intel
shines with
it's capabilities as well.
I'm not sure what is being shown with a geek benchmark on this page. I
don't see
anything very useful in it.
2:52 a.m.
On Tuesday 31 January 2006 02:45, jim stephens wrote:
On 1/30/06, Cameron Kaiser <spectre at
floodgap.com> wrote:
Then you didn't read very carefully. Copied from the article:
http://www.geekpatrol.ca/article/101/geekbench-comparison
Scroll down to "More Integer Performance." Presented without further
comment.
I didn't find "more integer performance" in the article, but rather
"integer performance". More Integer Performance
For another measure of integer performance, Geekbench executes code for the
MOS Technology 6502 CPU in a virtual CPU. This test is single threaded, and
results are in megahertz.
Pat
--
Purdue University ITAP/RCAC --- http://www.rcac.purdue.edu/
The Computer Refuge --- http://computer-refuge.org
9:57 a.m.
At 01:52 AM 1/31/2006, Patrick Finnegan wrote:
Then you didn't read very carefully. Copied from
the article:
> More Integer Performance
>
> For another measure of integer performance, Geekbench executes code for the
> MOS Technology 6502 CPU in a virtual CPU. This test is single threaded, and
> results are in megahertz.
And the graph shows that the processors in question (a selection of
contemporary Macs plus Intel and AMD for good measure) are able to
emulate the 6502 at virtual speeds of 100 to 400 Mhz.
- John
1:42 p.m.
Then you didn't read very carefully. Copied from the article:
> More Integer Performance
I had copied it from the google mail article into a find box, and then
clicked thru the link. It didnt find it. On my box this label is on the
"fold"
and I did not spot it.
Also I have seen similar performance with a Microdata 1600 emulator
that I have, though I do not have a comparison. It is amazing what the
current generation of processors can do when emulating older
technologies.
Any ideas why the 8/16 bit 6502 runs so much faster on Intel, when
the Hercules 370 (which is 8/16/32) and 390 runs much faster on
power / Mac than on Windows? I am only talking instruction execution
here, not I/O.
Also from his note on the G4 vs G5 comparison, would the exact same
executable run on both processors under whatever Mac OS 10 you have,
so would that maybe explain the G4 being faster than the G5?
Jim
6747
days inactive
6747
days old
4 comments
4 participants
participants (4)
-
james.w.stephens＠gmail.com -
jfoust＠threedee.com -
pat＠computer-refuge.org -
spectre＠floodgap.com