Intel Accelerates Nehalem Launch

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
I wouldnt call a QX6800 a current processor and if you would have tested say against the Q9650 at 3GHz it would be at best a small increase. The only real improvement will be the 8 core Nehalem versions.
 
[citation][nom]elbert[/nom]I wouldnt call a QX6800 a current processor and if you would have tested say against the Q9650 at 3GHz it would be at best a small increase. The only real improvement will be the 8 core Nehalem versions.[/citation]
I would say that we don't have any real performance numbers until the full test comes out.

Oh, and no, the 6800 isn't the most modern CPU. On the other hand, with both clocked at 2.93 GHz, the Nehalem was ~30% faster. 45nm (I.E. QX9650) does not run even remotely close to 30% faster per clock than 65nm. So, most of the gain has to be architectural.
 
On a more interesting note - WHAT WILL THIS NEW CHIP BE CALLED?

Core 3 Quad? X series?

And as for performance, anandtech found out anything within Cache (L1/L2/L3) this new chip performs the same as the Q9xxx series Intel
 

mothergoose

Distinguished
Jun 26, 2008
109
0
18,680
It will force down the prices of a lot of other today's top end processors regardless, and considering how far ahead the processors are today in the gaming rigs race for ultimate performance, that is definitely a good thing. Although who knows... maybe what it holding games back now is actually the cpu :/ We will find out for sure with this new release!
 

cryogenic

Distinguished
Jul 10, 2006
449
1
18,780
Can't deny that Nehalem is impressive and will bring big improvements over existing architectures and it will shine in certain tasks, but the phrase "We’ll come back with more as soon as we develop a viable suite of benchmarks to show some real-world gaming and productivity performance." doesn't sound quite good to me, I want to know how Nehalem performs in current suite of benchmarks so a meaningful comparison with other architectures can be done, for example how does it compare to Penryn or Barcelona. But that doesn't mean that I'm not interested in future proofing, so having a selection of benchmarks that take advantage of Nehalem's architecture is something that I want to definitely see.

So basically I want complete benchmarking of Nehalem, with current benchmarks and new ones, but not just new benchmarks, because I'm afraid that's a little biased against both older gen Intel CPU's and AMD ones.


 
pitty the cpu is getting more threads and independant cores yet things like HDD's are still lacking - two or more independant access on a single hdd cripples it big time as seek times kill performance
 
Cjl:

I would say that we don't have any real performance numbers until the full test comes out.

Oh, and no, the 6800 isn't the most modern CPU. On the other hand, with both clocked at 2.93 GHz, the Nehalem was ~30% faster. 45nm (I.E. QX9650) does not run even remotely close to 30% faster per clock than 65nm. So, most of the gain has to be architectural.

Were did you get 30%? The latest tomshardware benchmark only showed an 11% advantage per clock between the QX6800 and the 2.93GHz Nehalem. The 11% adbantage is about what a QX9650 has over the QX6800 so I would think my first post is accurate.
elbert :
I wouldnt call a QX6800 a current processor and if you would have tested say against the Q9650 at 3GHz it would be at best a small increase. The only real improvement will be the 8 core Nehalem versions.
The only percentage the Nehalem scored as high as 30% was against the phenom.
 

derek2006

Distinguished
May 19, 2006
751
0
18,990
[citation][nom]elerick[/nom]I've got to be honest I feel both Intel and AMD, I have had a 3.8ghz Air cooled 6300 for over two years. Is it even worth my time to hope that this new chip stock 2.93 beats mine in gaming, etc? I am bored with my old stuff I want something exciting and new is this it?[/citation]

You have been running a near 550mhz FSB for 2 years on air?
 

jaragon13

Distinguished
Jun 30, 2008
396
0
18,780
[citation][nom]elbert[/nom]Were did you get 30%? The latest tomshardware benchmark only showed an 11% advantage per clock between the QX6800 and the 2.93GHz Nehalem. The 11% adbantage is about what a QX9650 has over the QX6800 so I would think my first post is accurate. The only percentage the Nehalem scored as high as 30% was against the phenom.[/citation]
Phenom? No,you mean the Q9450 from Intel.Not a Phenom.
 
elbert :
Were did you get 30%? The latest tomshardware benchmark only showed an 11% advantage per clock between the QX6800 and the 2.93GHz Nehalem. The 11% adbantage is about what a QX9650 has over the QX6800 so I would think my first post is accurate. The only percentage the Nehalem scored as high as 30% was against the phenom.


Phenom? No,you mean the Q9450 from Intel.Not a Phenom.
Well even the phenom 9950BE doesnt fall to 30%. No I was speaking of the phenom 9850BE. The Q9450 at 2.66GHz is faster than a QX6700 which no phenom beats so its around 20~26% slower than Nehalem.
Nehalem is still faster than the fastest Phenom 9950 BE by 28 percent.

Thanks to jumpingjack6 for correcting us on our percentage math. Nehalem is actually 33 percent faster than Phenom 9850 BE with a 17 percent clock difference
 
Anandtech posted all the benchmarks weeks ago showing ~30% over current Intel cpus (clock for clock) - whats this 11%? and we all know overclocking is important to us :p
Click on the link "beat out the competition by fairly significant margins" in the artical. Anandtech done a complete quad only look other than the one benchmark for single core.
Cinebench shows us only a 2% increase in core-to-core performance from Penryn to Nehalem
http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3326&p=7
Tomshardware didn't test with programs using all 4 cores but a mix of what the average pc user would experience.
 
Uhhhhhh elbert - care to include what else was said about that 2% boost?

"Cinebench shows us only a 2% increase in core-to-core performance from Penryn to Nehalem at the same clock speed. For applications that don't go out to main memory much and can stay confined to a single core, Nehalem behaves very much like Penryn. Remember that outside of the memory architecture and HT tweaks to the core, Nehalem's list of improvements are very specific (e.g. faster unaligned cache accesses)."

As we saw with AMD's K7 to K8 overhaul, there are big improvements, and 2% on benchmarks, they wanted to see what the core was like compared to the last - pretty much the same with an IMC and no FSB etc.

What you dont seem to realize is for EVERY home user when there pc there isnt just one heavy app running (and zero benifit etc), there are hundereds - windows alone which can load across all the threads (services etc), antivirus, msn, download apps (torrents), hardware monitoring, drivers etc - the extra threads and performance comes in handy for everone, and its not as if this thing is slower then the last or anything.

Count me in for one of these new beasts, dunno bout you freaks but id like that extra performance and the prices sound about right.
 
and finally:

"One valid concern is with regards to performance in applications that don't scale well beyond two or four cores, what will Nehalem offer us then? Our DivX test doesn't scale well beyond four cores and even then Nehalem's performance was in the 20 - 30% faster range that we've been expecting. The other thing to keep in mind is that none of these tests are really stressing Nehalem's integrated memory controller. When AMD made the move to an IMC, we saw an instant 20% performance boost in most applications. I suspect that the applications that don't benefit from Hyper Threading, will at least benefit from the IMC. We've only scratched the surface of Nehalem here, looking at the benefits of Hyper Threading and its lower latency unaligned cache accesses. We've hinted at what's to come with the extremely well balanced and low latency memory hierarchy of Intel's new baby. Once this thing gets closer to launch, we should be able to fill in the rest of the puzzle." - http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3326&p=9
 
[citation][nom]apache_lives[/nom]Uhhhhhh elbert - care to include what else was said about that 2% boost?"Cinebench shows us only a 2% increase in core-to-core performance from Penryn to Nehalem at the same clock speed. For applications that don't go out to main memory much and can stay confined to a single core, Nehalem behaves very much like Penryn. Remember that outside of the memory architecture and HT tweaks to the core, Nehalem's list of improvements are very specific (e.g. faster unaligned cache accesses)."As we saw with AMD's K7 to K8 overhaul, there are big improvements, and 2% on benchmarks, they wanted to see what the core was like compared to the last - pretty much the same with an IMC and no FSB etc.What you dont seem to realize is for EVERY home user when there pc there isnt just one heavy app running (and zero benifit etc), there are hundereds - windows alone which can load across all the threads (services etc), antivirus, msn, download apps (torrents), hardware monitoring, drivers etc - the extra threads and performance comes in handy for everone, and its not as if this thing is slower then the last or anything.Count me in for one of these new beasts, dunno bout you freaks but id like that extra performance and the prices sound about right.[/citation]
Well apache_lives if you look back at the first benchmarks for phenom the same programs were used to show gains. Using these benchmarks the 9950BE phenom would show some great result. The single core test tho was the problem for phenom. In ending 2% is 2% no matter the suger coating. Reading that second quote I wonder why intel's IMC doesn't get the 20% amd does.
 
Status
Not open for further replies.