So, 45nm sounds hip-cool and all, but what does that mean for performance? Intel's coming forward with some
Penryn numbers -- with the help of Reg Hardware looking over its shoulder to make sure the tests are legit -- and it seems we're in for a decent speed bump, but not necessarily a revolution. Dual core and quad core 3.33GHz Penryns, each with a 1333MHz frontside bus, were pitted against the current desktop-leading 2.93GHz Core 2 Extreme
QX6800 on a 1066MHz bus. Tests were run on top of a well decked test system, running Vista Ultimate. Naturally, the quad core Penryn took the top spot in all the tests, with varying margins depending upon which app was used -- no surprises there. The dual-core Penryn took third behind the spendy QX6800 for the most part, other than the single-threaded Half-Life 2 and the SSE 4-capable DivX test, which capitalized upon the media-friendly instructions Intel is adding to its new processors. Looks like we'll be seeing much more from these processors as more SSE 4 apps hit the market, but for now you shouldn't be feeling too bad about buying a QX6800 -- it's not obsolete by a long shot.
posted April 18th 2007 11:59AM
Hey! That's my pc case...yay! I feel special. 30k or so others who bought it at Newegg do too!
So sure, it's not completely knocking the QX6800 out of the water, but impressive nonetheless. Definite performance gains by Intel are being made practically every month now. I wouldn't be surprised if we see Intel temporarily besting gains predicted by Moore's law, making up for that whole late-era P4 w. HT/Pentium D slouch period.
Could it have just been the bump in clock speed? Anyone could overclock a current processor, but is the new architecture any faster? This doesn't really help answer that question at all.
No it couldn't. There is a 14% clock speed difference but the Quad Penryn performed anywhere between 20% - 35% faster in all test other than the 3D Mark one.Would have been useful to have added C2D E6700 and do some more benchmarks on games and stuff. As it stands though it seems like a decent upgrade with the new architecture to be providing an average of 20% extra performance at the same GHz.
Isn't the more, or equally, significant benefit the heat/power... and therefore noise, savings?
No... I don't think you know what you're saying lol.. 14% OC is nothing on the C2D's and C2Q's
What they should've done is an all stock test.
Penryn
QX6800
E6700
Etc... All the top of the line stock.
Include
E6600, Q6700 QX6800
OC them all, e66 to around 3.4+ and what not... The Quad cores as well to 3.4 or so and test.
"So, 45nm sounds hip-cool and all, but what does that mean for performance?"
45nm has nothing to do with performance, and it's also nothing new. Intel demonstrated their 45nm chips in January of 2006.
oh and in addition to the 14% being nothing, the c2extreme has unlocked multipliers making 14% OC almost a joke to any overclocker.
I never mentioned the overclocking ability of either chips and I definitely never said that you cannot overclock the QX by 14% or more - though it seems you will have to overclock it to almost 4GHz in order to match the Penryn.
What I did say is that at the same clock the Penryn seems 20% faster than the QX. Simple. If the QX is more or less overclockable is of little interest especially given that each and every chip has its own overclocking limits.
Since you did bring it up though I would expect a 45nm chip to be just as overclockable as a 65nm one if not more.
They probably will OC like monsters as well but their price will be quite high...
I say oc your QX6800, and it will match and probably exceed the stock penryn..
I agree with you for the most part :)
Which current generation motherboards support this CPU?
Most motherboards should support it with a simple bios update to allow 1333 FSB.. At the moment though due to them being 45NM I don't think any motherboard does until Intel releases their new chipsets..
Man, this is why I stopped paying attention to the CPU wars about 2 years ago. The word got out that raw clock speed doesn't always mean anything: good. They still tried to distill all the features of a CPU down to a single, now unitless, number: bad. I guess the best thing is that we learned to judge a CPU by how it performs on apps we consider important, instead of a number the manufacturer pinned on to it. The one last step I want to see: instead of the Y-axis of all these benchmark graphs bearing the model number of the processor, just show me the price ;-)
@Paul Miller,
If you go so far as to say "you shouldn't feel bad about buying a QX6800" at least remind people (i'm amazed how many still don't know) to wait until AFTER APRIL 22 due to the planned Intel price drops. Many of the CPU's including the high end ones like the QX6700 and E6700 are going to see 50%+ price drops!
I bought a QX6700 a month ago for $999 and then found out about the price drop on April 22 and was like OH DAMN ... and returned the sucker pronto.. I'm going to buy it again after the drop.
@MichaelG,
"Upclocking a QX6800 to Penryn speed" is wishful thinking and loses out on some benefits. ESPECIALLY when we're talking quadcore Intel CPU's, the speed of the frontside bus is a factor since they must all share it. As it is they're bandwidth starved sticking 4 cores on a 1066 bus. With a 1333 bus things are alleviated somewhat, and you have more headroom to increase that further. Like a MILF that just needs her pipes cleaned, a higher *native* FSB of 1333 opens up and unclogs communication to/from those 4 cores.
45nm will also run cooler, so you won't have to resort to ridiculous extremes with giant car-radiator-sized HSF's or messy water cooling.
Penryn is the part that pro's and enthusiasts are waiting for.
My case too, also from newegg!