The GHz Race is Back..

...and Intel has leaped ahead with 3.33Ghz pre-production silicon. Not such a small feat considering that this is a working 45nm silicon. If you're asking how significant this is you don't have to look very far and compare AMD's old K8 architecture transition into an 'old' 65nm process technology. Not only were there no performance increase, the clock speed was even slower. If that doesn't sum up the gargantum gap between Intel and AMD in process know how, then I don't know what does.

Yorkfield Performance Advantage vs 2.93Ghz C2D
3DMark '06 V1.1.0 Pro CPU (score) : 21.8%
3DMark '06 V1.1.0 Pro Overall (score) : 7.6%
Mainconcept H.264 Encoder (seconds) : 18.0%
Cinebench R9.5 (CPU test): 24.9%
Cinebench R10 Beta (CPU test): 25.5%
HL2 Lost Coast Build 2707 (fps) : 37.3%
DivX 6.6 Alpha w/ VirtualDub 1.7.1 (seconds): 111%

Queries on Penryn's IPC becomes really insignificant at this point when clearly Intel has a speed demon with a slightly better IPC over Conroe. And with architectural improvements apparently leveling out, raw clock is once again King and by the looks of it Barcelona is just too slow and too late.

From Anandtech:
Obviously we'll reserve final judgments on Penryn for our official review of the CPU, but these initial results look very promising. We would expect to see clock for clock Penryn vs. Conroe improvements to be in the 5 - 10% range at minimum depending on the application. Factor in higher clock speeds and you can expect our CPU performance charts to shift up by about 20% by the end of this year.


Anonymous said...

Where in the hell is 40% boost over Core 2 Duo???????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????

People were right.

Intel was telling lies!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Anonymous said...

Dude, you need Ritalin. Or maybe AMD. Intel is too fast for you.

Anonymous said...

Explain where in the hell is 40% boost!
Intel told us that Penryn will be 40% faster???????????????????????????????????????????????????????????????

Not Penix said...

I know this isn't a very conclusive statement, but if you take the results from the post, the AVERAGE performance increase is 35%

Not quite 40% but i'll take these benches with a grain of salt. I've said it before and i'll say it again, why is it that AMD can say 40% increase and its gospel but when intel does it doesn't mean a thing, at least Intel has released some benchmarks...

Anonymous said...

Penryn was supposed to be a dieshrink yet they promised performance. AMD goes to 65nm and actually manages to under perform........intel goes to 45nm and increases performance by almost 35%+ and stupid AMD fans have the balls to cry about it!!!

K10 will be spanked by a celeron come next year rite before AMD announces bankruptcy!!

Anonymous said...

HAHAHAHHAHA i just read scientia's newest post on his blog. Thank you for the good laugh you have become a joke just like sharikou i had some respect for you but after that post you are just pathetic. Stop making excuses for AMD and post something useful....oh wait you have nothing useful to say about AMD because they HAVE NOTHING TO REPORT!!.

Ahhh well back to excuses you go....

Scientia from AMDZone said...

The gigahertz race is back when they actually release. I've now seen several statements that this won't be out until 2008.

AntiFanboy said...

It's funny how Scientia turns a blind eye to the 3.33GHz Penryn @ IDF.

Penryn is scheduled to be released in Q4 07, at 3GHz+ speeds.

You're just hoping they will push it out to 08 so you don't have to eat crow. ;)

AntiFanboy said...

LOL @ scientias attempt at 'understanding' the 'C2D overclocking myth'.

You are clearly not an overclocker scientia. There is no 'myth' to the overclocking success of C2D. The whole point of overclocking is to run the CPU above spec, and that means higher temperatures too.

I am not sure why you are so anal about the 55C mark, it's an arbitrary figure and it's not like C2Ds start throttling the moment that exceed that temperature.

All Intel has to do to keep a 3GHz+ C2D under that 55C spec is to get a beefier heatsink. The current one is crap, they could take a leaf out of AMD in the heatsink department, the X2 6000+ uses multiple heatpipes and all. Then again, it would have to considering it consumes more power than a QX6800. ;)

Scientia from AMDZone said...


I'm sorry you don't understand my posts but I don't know how to make them any simpler.

Anonymous said...

How about you worry about what AMD is doing instead of worrying about what intel is doing scientia.

It is obvious intel doesnt need anyone to worry about it by the profits they have made in this quarter. None of the crap that you type will change anything. Core 2 duo STILL kicks the crap out of X2 whether at 55C or higher..... and penryn will kick Barcelona's in the nuts when it get released.

I hope for AMD's sake Intel doesnt have readily available penryn before barcelona hits the market that would be another blow to AMD's already shaky reputation.

Anonymous said...

How about you worry about what AMD is doing instead of worrying about what intel is doing scientia.

yomamafor2 said...

scientia from amdzone
The gigahertz race is back when they actually release. I've now seen several statements that this won't be out until 2008.
My guess is that they want to wait and see how K10 fare against Penryn, then clock them accordingly. However if this is true, 3.33Ghz we see now might be the lowest clocked Penryn.

Anonymous said...

4GHz seem unreasonable?

DaSickNinja said...

Very reasonable if you ask me.

Anonymous said...

With Netburst unloaded, Intel is really going to put on the hurt. AMD BK 2008.