i think i know what happened here, ill try to simplify my notions with this handy example:the pentium d went from 1fps to 2 fps....massive percent increase, while the conroe went from 50 fps 60fps...not as great as a percent increase lol
Printable View
i think i know what happened here, ill try to simplify my notions with this handy example:the pentium d went from 1fps to 2 fps....massive percent increase, while the conroe went from 50 fps 60fps...not as great as a percent increase lol
:rofl:Quote:
Originally Posted by DoubleZero
But 50fps - 60 fps would be a massive absolute value...Quote:
Originally Posted by grimREEFER
why not get a computer that goes from 0 fps to 1 fps. then the infinite increase in FPS is completely unbeatable :woot:Quote:
Originally Posted by riptide
lolQuote:
Originally Posted by nn_step
Complete and utter Hogwash...
I am running out of time to analyse some of the more interesting/puzzling results and won't come to conclusions before the 4th of July weekend.
I'm throwing some raw data at you, numbers are CPU time used (this is single-thread):
This is compiling/building a Common Lisp system. (BTW, Netburst completely sucks here, I'm glad that episode is over.)Code:Lisp build:
fast safe (fast/safe = target policy)
2.4 939-DC-Opt 505 826 (memory 218 MHz)
2.6 939-DC-Opt 469 731 (memory 216 MHz)
2.4 940-SC-Opt 537 846
2.4 Conroe 328 439
3.66 Conroe 222 296
You can build the program two ways: safe-mode and fast-mode. The safe-mode build will insert a lot of extra checks like type checks, integer overflow checking, function parameter counting etc.
As you can see, the Conroe completely spanks the AMD64s in the safe-mode build, more so than in the safe-mode build. That is puzzling, I have never in my 15 years of Lisp programming seen anything like this. Suddenly you get the insertion of the extra checks almost for free, whereas so far you had to weight getting the checks against twiddling your thumbs for a while duing the build.
Here is a relative index, the first rows giving you:
- performance/MHz fast-mode
- (same, relative to 2.4 Opteron)
- performance/MHz safe-mode
- (same, relative to 2.4 Opteron)
That means Conroe reaches a 50% advantage over AMD64 per clockspeed for the fast-mode build but goes up to 90% advantage for the save-mode build.Code:
fast (rel) safe (rel) fast safe
------------------------------------------------------------
1212.0 (1.0) 1982.4 (1.0) 2.4 939-DC-Opt 505 826
1219.4 (1.0) 1900.6 (1.0) 2.6 939-DC-Opt 469 731
1288.8 (0.9) 2030.4 (1.0) 2.4 940-SC-Opt 537 846
787.2 (1.5) 1053.6 (1.9) 2.4 Conroe 328 439
812.5 (1.5) 1083.4 (1.8) 3.66 Conroe 222 296
Possible explantions include:
- It's the larger cache. Since this compilation is single-thread the AMD64 has 1 MB, the Conroe has between 2 and 4 depending on how it is associated with the one core running the compiler
- Some parallel execution unit in the Conroe picks up the extra instructions to insert the safety checks
- We get the extra instructions cheaper than on AMD64 due to some clever rescheduling in Core2
- The extra instructions to insert the safety check in the compiler normally blow memory prefetch in AMD64 and Netburst but Core2's prefetch is clever enough to find the stuff in advance
- I am doing something wrong here (not likely, I re-ran it a few times)
- All of the above
I need to run a few programs that report cache hits and the like but I don't think any of the performance counter programs that exist are running on Conroe yet.
It was some months ago, so it might not be true. I found this article first,Quote:
Originally Posted by duploxxx
http://www.smarthouse.com.au/Computi...forms/R7G5G6U4 and than I found this
http://www.pcpro.co.uk/news/85506/vi...forrester.html
You really do got me curious here...Quote:
Originally Posted by uOpt
The first thing that cameup to my mind was...***m
thats not for core thats for ViiVQuote:
Originally Posted by gOJDO
Or someone could just use Win 2003 server and 2003 64 Bit server, and use 32/64 Bit Cinebench....Quote:
Originally Posted by nn_step
Is this really this difficult?
I don't understand how comparing results of these two processors in 32bit, and then comparing them again in 64bit would be a revealing test? Wouldn't you be getting rid of more variables if you did a comparision of the differences of A64 machine (32bit vs. 64bit benchmarks) and also a Conroe machine (also 32bit vs 64bit benchmarks) and then compare the deltas of these two machines? Also, are the same benchmarks being used across the two OS's? I understand that an advantage over A64 is trying to be illustrated, but the advantage Conroe holds over A64 is already apparent. It is more interesting to see if Conroe in 32bit gains an advantage over itself in 64bit.Quote:
Originally Posted by uOpt
I was strickly comparing whether Conroes speed advantage in 64 bit code over AMD64 is the same as Conroes speed advantage in 32 bit code over AMD64. It is.Quote:
Originally Posted by phi|os
I did not compare 64 bit performance to 32 bit performance. I'd like to do that but this is more complicated than it sounds. Compilation benchmarks have to target one common platform (which makes one a crosscompiler and the other not unless you make both crosscompiler). Most multimedia programs have hand-coded MMX or SSE assembly routines thrown in that might or might not trigger based on whether you are on x86_86 or i386. Most other benchmarks are either useless artificial benchmarks and/or are one-trick ponies.
source here
http://www.xbitlabs.com/images/cpu/c...t/sandra-1.png
http://www.xbitlabs.com/images/cpu/c...t/sandra-2.png
http://www.xbitlabs.com/images/cpu/c...t/sandra-3.png
http://www.xbitlabs.com/images/cpu/c...t/sandra-4.png
http://www.xbitlabs.com/images/cpu/c...encemark-1.png
http://www.xbitlabs.com/images/cpu/c...encemark-2.png
http://www.xbitlabs.com/images/cpu/c...bit/7zip-1.png
http://www.xbitlabs.com/images/cpu/c...bit/7zip-2.png
http://www.xbitlabs.com/images/cpu/c...-64bit/wme.png
http://www.xbitlabs.com/images/cpu/c...t/pdnbench.png
http://www.xbitlabs.com/images/cpu/c...athematica.png
http://www.xbitlabs.com/images/cpu/c...inebench-1.png
http://www.xbitlabs.com/images/cpu/c...inebench-2.png
http://www.xbitlabs.com/images/cpu/c...inebench-3.png
http://www.xbitlabs.com/images/cpu/c...inebench-4.png
http://www.xbitlabs.com/images/cpu/c...bit/povray.png
http://www.xbitlabs.com/images/cpu/c...bit/ut2004.png
I just read that article too StyM. They said overall Conroe gets ~10% boost in Windows XP 64 over 32 bits. Not bad, AMD gets about ~15%...
Vista isn't coming out for a while longer. Might be even more delayed than planned. Plenty of time for both company's to fix what they need to fix.
C2D still beats FX-62in 64 bit appplications. In some cases out performing FX-64 in 32 bit applications. There is no other AMD process to bench to imply or
suggest what kind of performance gains C2D should be making. There are no AMD processors that even beat C2D in 64 bit applications. Therefore, the gains that FX chip make are mute, period. FX would need to make on average 40% gains in 64 bit applications to be competive with C2D and in most cases beat C2D.
What I don't understand is why the fanboism that FX is better then C2D in 64 bit applications when the gains themselves are so meager. Is this the best you can do? AMD still loses out on every 64 bit benchmark out there. Therefore, it is not better then the C2D in gains from 32 bit applications. What that is called is diminishing returns not gains.
Now is that really what this is about ? This shows how the GAINS are in 64bit, not if conroe is faster then the FX.Quote:
Originally Posted by Eastcoasthandle
Sorry man, but you are paranoiac :para: No one is saying here that C2D is slower than FX. Jesus.. :rolleyes: Thats what I call fanboism. :slapass:Quote:
Originally Posted by Eastcoasthandle
umm Dude who said that about current AMD models?Quote:
Originally Posted by Eastcoasthandle