• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

A silly but cool question

in AMD terms, no way. AMD from when?

apparently, at equivalent clock speeds any c2d is twice as efficient / fast as a P4. Thus my E2160 @ 3ghz is 4 times faster (two cores) than a P4 3ghz HT.

i read that from here somewhere, but it seems like a hazarded guess so don't quote me on it.
 
speedy2004 said:
amd64 terms, cos I had a opty [email protected], and thats about 5000+ right, and the 1m super pi is 30secs.

My E8400 @ 3.6 will do 1m superpi in 12secs, so thats over twice the speed of my opty [email protected]. So in amd terms my e8400 @ 3.6 has to be at least 10000+. right or wrong?
Click to expand...

lol. Because superpi is the only benchmark that matters!

The opty was on a slower chipset, with slower ram. The CPU wasn't the only factor in those superpi times ;)

*edit*

Google for some real numbers, ie. how many TFlops each chip can crunch, then you'll know what % speed difference there is between the chips :)
 
the tables turn all the time (although amd's table seems to have been smashed by intel's :p )

ie, xp2400 was about a 1.8ghz chip, but was on par with a 2.4ghz p4

i don't have fanboy views, i just use the best performance/£ setup at the time :)
 
Basically when AMD started their xxxx+ ratings they where better CPU's than the competition ( P4's ) and where running at a lower speed than the P4's but able to perform as well. So rather than confuse people by advertising the rating ( and thus people thinking the P4 was better cos it ran faster ) they chose the equivilency idea.

So a 5000+ is roughly the same as a P4 running at 5K .. ok P4's are hardly used anymore but it's convenient and does mean that AMD can release a faster CPU (with slower clock speed ) again and they just increase the number and everyone knows it's faster.

I know my 5000+ is quicker than my old 3800+ - I don't need to look at benchmarks etc. to find out.

Of thread but I find C2D's numbers just confusing - havn't a clue what's what.

I don't beleive there is a huge difference between C2D and AM2 - at the same clock speed.

So basically a AM2 dual core @ 3.2ghz is 6400+, a 3.0 is 6000+ so logically a 3.6 would be 7200+ - therefore a 8400 at 3.6 logically would be around 7200+ - probably a touch higher but certainly not a 10000+
 
I think the 2.4Ghz Intel is equal to a 2.8Ghz AMD (trying to recall scores I've seen for CPU on 3dmark). Also the new E8400 can do a little bit more work per cycle so I'd say you'd need to be at 4Ghz to 4.4Ghz at the most with that Opteron to draw level but that is just a rough estimation.
 
I just went onto the 3Dmark benchmark thread in the graphics section and pulled out these two scores. It looks as if I've remembered correctly.

The first link is a X2 3800 @ 2.75Ghz.

http://service.futuremark.com/resultComparison.action?compareResultId=3367739&compareResultType=14

The second link is from a stock E6600.

http://service.futuremark.com/resultComparison.action?compareResultId=3667839&compareResultType=14

The CPU scores are about the same but I wanted a 3800 X2 @ 2.8Ghz to compare with. Stupid me forgot about my last system :o. AMD X2 3800 @ 2.8Ghz, 1Gb of DDR400 @ 460 odds 2.5-3-3-7 1T, DFI Ultra-D and 2900Pro. http://img246.imageshack.us/img246/19/28ghz290084799979s1213vki8.jpg

Scores are identical between 2.8Ghz dual core AMD and E6600 @ stock. With the E8x00 series they do more work per cycle so I'll stick by my 4Ghz to 4.4Ghz comparison for now. I think 4.2Ghz sounds about right though. So for your Opteron to be as fast as your E8400 @ 3.6Ghz I think it would have to run @ 4.2Ghz. For your E8400 to be double the power then you'd have to be at about 5Ghz :eek:. Again this is a rough estimation but after seeing the CPU scores I'm a little more confident about this now.
 
Last edited:
Looking at various benchmarks on one well known benchmarking site the 8500 (3.2 ghz right?) seems to average around 30-40% quicker than an [email protected].

Therefore a 8400 @ 3.6 ought to be maybe 50-60% quicker so around a 8000+ - approximatly. (5000 x 1.6 = 8000) probably a little more but I'd still say no-where near a 10000.
 
Super Pi is a poor benchmark for comparison because it returns a much better result for Intel than AMD compared to a wider range of benchmarks. Not to mention outside of scientific applications it's entirely synthetic.
By the same token (though less so) Cinebench performs better on an X2 relative to a Core 2 Duo, 'punching above it's weight' as it were.
Clock for clock on an average of real world benchmarks, a 2.4 (e6600) Core 2 is about equal to a 3.0 (X2 6000) AMD. This assumes simliar ram platform etc. & is approx. Gaming benchmarks can supply a wider range of values but normally only under unrealistic usage conditions.
Id' say C2D clock -0.4 to 0.6 GHZ = AMD X2 clock. Cache /series will make a bit of difference though, EXXX series won't pull ahead so far & the new wolfdales (E8XXX series) & yorkdales (Q9XXX series) will pull ahead a bit further.
AMD Phenoms have better performance per gig than X2's - only a little behind C2Ds - lets say Phenom 2.6 = C2D 2.4ish (not 45nm). Their problem is they aren't clocking high enough & they use more power than they really should.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom