Cat's out of the Bag: Intel Nehalem Benchmarked

jmke · 5th June 2008, 09:24

We had access to a 2.66GHz Nehalem for the longest time, unfortunately the motherboard it was paired with had some serious issues with memory performance. Not only was there no difference between single and triple channel memory configurations, memory latency was high. We know this was a board specific issue since our second Nehalem platform didn't exhibit any issues. Unfortunately we didn't have access to the more mature platform for very long at all, meaning the majority of our tests had to be run on the first setup (never fear, Nehalem is fast enough that it didn't end up mattering).

We've been told to expect a 20 - 30% overall advantage over Penryn and it looks like Intel is on track to delivering just that in Q4. At 2.66GHz, Nehalem is already faster than the fastest 3.2GHz Penryns on the market today. At 3.2GHz, I'd feel comfortable calling it baby Skulltrail in all but the most heavily threaded benchmarks. This thing is fast and this is on a very early platform, keep in mind that Nehalem doesn't launch in Q4 of this year.

http://anandtech.com/cpuchipsets/int...spx?i=3326&p=1

5th June 2008, 20:37

I was right about the performance, but so wrong about the heat... I was so expecting a bit more heat than just a paltry 15 watts clock for clock.

Single-core Cinebench R10 test shows a 25% performance gain over Penryn, and that was with single-channel memory verses dual-channel for Penryn.

Since they tested with DDR3-1067 RAM, will be interesting to test some 1333MHz RAM with 6-6-6-18 timings. Suspect latency will decrease further and might show some small performance gains ontop of everything else.

5th June 2008, 09:24	#1
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 79,019	Cat's out of the Bag: Intel Nehalem Benchmarked We had access to a 2.66GHz Nehalem for the longest time, unfortunately the motherboard it was paired with had some serious issues with memory performance. Not only was there no difference between single and triple channel memory configurations, memory latency was high. We know this was a board specific issue since our second Nehalem platform didn't exhibit any issues. Unfortunately we didn't have access to the more mature platform for very long at all, meaning the majority of our tests had to be run on the first setup (never fear, Nehalem is fast enough that it didn't end up mattering). We've been told to expect a 20 - 30% overall advantage over Penryn and it looks like Intel is on track to delivering just that in Q4. At 2.66GHz, Nehalem is already faster than the fastest 3.2GHz Penryns on the market today. At 3.2GHz, I'd feel comfortable calling it baby Skulltrail in all but the most heavily threaded benchmarks. This thing is fast and this is on a very early platform, keep in mind that Nehalem doesn't launch in Q4 of this year. http://anandtech.com/cpuchipsets/int...spx?i=3326&p=1 __________________ Last edited by jmke : 5th June 2008 at 09:32.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Intel Plans 10W Nehalem CPU	jmke	WebNews	0	5th February 2009 16:17
Intel Nehalem also has a TLB bug	jmke	WebNews	1	1st December 2008 19:29
Power gating and turbo mode: Intel talks Nehalem at IDF	jmke	WebNews	0	20th August 2008 15:04
Intel Nehalem CPU B1 Stepping Photo	jmke	WebNews	0	18th August 2008 16:19
Intel Nehalem 3rd Party Heatsink Pictured	jmke	WebNews	0	18th August 2008 10:45
Intel Nehalem to be released as Core i7 CPUs	jmke	WebNews	0	8th August 2008 15:42
Intel Nehalem Socket 1366 Motherboards Emerge at Computex 2008	jmke	WebNews	0	8th June 2008 10:42
Intel won't let Nvidia make Nehalem chipsets	jmke	WebNews	4	7th June 2008 12:19
Opening the Kimono: Intel Details Nehalem and Tempts with Larrabee	jmke	WebNews	0	18th March 2008 16:14
All Intel Nehalem CPUs not created equally	jmke	WebNews	4	28th November 2007 16:35

5th June 2008, 20:37	#2
Kougar Posts: n/a	I was right about the performance, but so wrong about the heat... I was so expecting a bit more heat than just a paltry 15 watts clock for clock. Single-core Cinebench R10 test shows a 25% performance gain over Penryn, and that was with single-channel memory verses dual-channel for Penryn. Since they tested with DDR3-1067 RAM, will be interesting to test some 1333MHz RAM with 6-6-6-18 timings. Suspect latency will decrease further and might show some small performance gains ontop of everything else.