CPU Benchmarks at Tech Review

[datamestonic]

So on my quest to figure out if Opterons or x2's are more suitable for gaming I stumbled upon this:

http://techreport.com/reviews/2005q2/athlon64-x2/index.x?pg=1

This review is focused on the x2's but it shows how they rank up against pentium CPUs as well as Opterons. Unless I'm mistaken (which may very well be the case), the 175 that seems to be the Denmark core, falls either directly in front or (in many cases) significantly far behind the 4200+

If this is so, doesn't it suggest that the 4400+, atleast for gaming, should be superior? But I don't understand why, because I was under the impression that Opterons are basically the same except they're higher quality in some aspect, use higher grade silicon, more resilience, etc.

 

[Furen]

I wish people stopped linking to that damned review...

No it does not. That "Opteron 175" is actually a single Opteron 875 or 275 running on a socket 940 motherboard with registered, ECC ram. That WOULD be the equivalent of an Opteron 175 if AMD had released it on socket 940. S939 Opteron 175 performance should be exactly identical to the X2 4400+'s. Socket 940 performs worse not because of the chip itself or the "grade" of the silicon, but because the platform is built for reliability rather than speed, so the memory used has quite harsh (well, not THAT harsh) latency penalties compared to regular (or low-latency) DDR.

I don't think Opterons use any higher grade silicon or anything of the sort, they're just validated for higher reliability (ie. they have better signal-to-noise ratio, lower current draw, higher temperature tolerance, lower heat output at stock clock and/or stuff like that--I dont know exactly what the validation entails, I'm just speculating).

EDIT: grammar, grammar, grammar...

 

[Markbnj]

Good find. Seems very comprehensive. This ought to put to bed some of the debate over X2's and games.

 

[datamestonic]

Originally posted by: Furen
I wish people stopped linking to that damned review...

No it does not. That "Opteron 175" is actually a single Opteron 875 or 275 running on a socket 940 motherboard with registered, ECC ram. That WOULD be the equivalent of an Opteron 175 if AMD had released it on socket 940. S939 Opteron 175 performance should be exactly identical to the X2 4400+'s. Socket 940 performs worse not because of the chip itself or the "grade" of the silicon, but because the platform is built for reliability rather than speed, so the memory used has quite harsh (well, not THAT harsh) latency penalties compared to regular (or low-latency) DDR.

I don't think Opterons use any higher grade silicon or anything of the sort, they're just validated for higher reliability (ie. they have better signal-to-noise ratio, lower current draw, higher temperature tolerance, lower heat output at stock clock and/or stuff like that--I dont know exactly what the validation entails, I'm just speculating).

EDIT: grammar, grammar, grammar...

Ah, I'm sorry as I was saying I wasn't completely sure in myself here, that's why I was posting it in hope to get some feedback.

 

[Hacp]

Originally posted by: Markbnj
Good find. Seems very comprehensive. This ought to put to bed some of the debate over X2's and games.

The X2 4800+ gives up 200MHz to its fastest single-core competitor, the Athlon 64 FX-55, but gains most of the performance back in single-threaded apps thanks to AMD's latest round of core enhancements, included in the X2 chips. The X2 4800+ also matches the Opteron 152 in many cases thanks to Socket 939's faster memory subsystem. Remarkably, our test system consumes the same amount of power under load with an X2 4800+ in its socket as it does with an Athlon 64 FX-55, even though the X2 is running two rendering threads and doing nearly twice the work. Amazing.













So that means the new cores perform better?

 

[Furen]

The FX-55 they use is a 130nm chip, so it does have slightly inferior performance, clock for clock. The 152 is the socket 940 one as well, I think.