Posted by
ryuzaki0
on from the king-of-the-hill dept.
gcshaw2nd writes: "Here it is, the first hands-on review I've seen of Intel's new Northwood chip, running at two gigahertz. It overclocks like a hog, easily to 2.5Ghz."
And that's not all...
by
Glonk
·
· Score: 5, Informative
Aside from the meager "5-10%" performance boost per clock that GamePC reports, the new PC1066 RDRAM and 533MHz FSB coming in a few months offers a "12%" performance boost per clock, when used with the original P4.
Northwood + 533MHz FSB/PC1066 RDRAM should be quite nifty.
According to that chart there, PC1066 RDRAM actually has lower latency than PC133 SDRAM. I don't know how accurate that is, but it says PC1066 RDRAM takes 207 cycles for 128 bytes, and PC133 takes 229 cycles (PC800 took 270)?
Maybe I'm reading that wrong or don't know some specifics about RDRAM architecture, but that sounds nifty...
That's not true, they use RDRAM or SDRAM or DDR SDRAM. In fact there's a brand new artice that benchmarks them all.
Take your pick of RAM (although RDRAM and DDR are the only real choices).
Re:What's the point?
by
zmooc
·
· Score: 3, Informative
You're absolutely right; nobody needs more than 640K RAM too....
The usage of computers changes along with the possibilities and there's still a lot that's not possible. Think about photo-realistic realtime interactive movies (have you seen the latest Chemical Brothers video-clip "Star Guitar"? THAT's what I want to do in realtime and interactively), multi-track samplers that can do a lot of effects without any latency, predicting the weather more exactly without the use of what we call supercomputers nowadays, SETI, simulations of large neural networks etc. etc. That's why we need the Hz's, not for the stuff we we're doing nowadays. As long as I cannot easily create my own Hollywood-production in 16384:1024 with 16-channel sound on my desktop, create the soundtrack for that with a software sampler with professional quality (latency) etc, we're not there yet.
-- 0x or or snor perron?!
Even quicker than 3ghz !!
by
sh0rtie
·
· Score: 5, Informative
now even quicker !! this page claims it has a world record
Re:2 Ghz CPU or marketing BS ?
by
An+Ominous+Coward
·
· Score: 3, Informative
Intel's CEO has an engineering background, and he doesn't let marketing dictate technology. MHz certainly isn't the only issue with regards to performance, but neither is it a non-issue. The P4 is a very intesting chip due both to the technology that enables clockrate scaling and the technology that makes use of clockrate scaling.
Video processing is a CPU hog
by
msobkow
·
· Score: 3, Informative
A number of people I work with have DV video cameras and are buying DVD-R/DVD+RW burners. I don't know of anything that consumes raw cycles like video processing. Even with clean source, it can take 4-5 hours to process a mere 25-30 minutes of video to MPEG2 if you want good video quality (and that's on an Athlon 1.4GHz!)
In the past 3 months, 4 of the 30 people in my work area have picked up DV cameras and looked at DVD burning their home vids. Every one of them has been greatly disappointed to find that they can't do it with their "old" 800MHz PIII boxen without leaving the job running over night.
So I guess the point is that you don't need much more power than currently available for raw compiles and such, but you can expect the upcoming flood of DVD burners and DV cameras to push a significant number of people to upgrade.
-- I do not fail; I succeed at finding out what does not work.
Aside from the meager "5-10%" performance boost per clock that GamePC reports, the new PC1066 RDRAM and 533MHz FSB coming in a few months offers a "12%" performance boost per clock, when used with the original P4.
Northwood + 533MHz FSB/PC1066 RDRAM should be quite nifty.
The PC1066 benchmarks are here.
According to that chart there, PC1066 RDRAM actually has lower latency than PC133 SDRAM. I don't know how accurate that is, but it says PC1066 RDRAM takes 207 cycles for 128 bytes, and PC133 takes 229 cycles (PC800 took 270)?
Maybe I'm reading that wrong or don't know some specifics about RDRAM architecture, but that sounds nifty...
That's not true, they use RDRAM or SDRAM or DDR SDRAM. In fact there's a brand new artice that benchmarks them all.
Take your pick of RAM (although RDRAM and DDR are the only real choices).
The usage of computers changes along with the possibilities and there's still a lot that's not possible. Think about photo-realistic realtime interactive movies (have you seen the latest Chemical Brothers video-clip "Star Guitar"? THAT's what I want to do in realtime and interactively), multi-track samplers that can do a lot of effects without any latency, predicting the weather more exactly without the use of what we call supercomputers nowadays, SETI, simulations of large neural networks etc. etc. That's why we need the Hz's, not for the stuff we we're doing nowadays. As long as I cannot easily create my own Hollywood-production in 16384:1024 with 16-channel sound on my desktop, create the soundtrack for that with a software sampler with professional quality (latency) etc, we're not there yet.
0x or or snor perron?!
now even quicker !! this page claims it has a world record
3023mhz !
Try 2.8 GHz
Or why not 3.0!
-----
Intel's CEO has an engineering background, and he doesn't let marketing dictate technology. MHz certainly isn't the only issue with regards to performance, but neither is it a non-issue. The P4 is a very intesting chip due both to the technology that enables clockrate scaling and the technology that makes use of clockrate scaling.
In the past 3 months, 4 of the 30 people in my work area have picked up DV cameras and looked at DVD burning their home vids. Every one of them has been greatly disappointed to find that they can't do it with their "old" 800MHz PIII boxen without leaving the job running over night.
So I guess the point is that you don't need much more power than currently available for raw compiles and such, but you can expect the upcoming flood of DVD burners and DV cameras to push a significant number of people to upgrade.
I do not fail; I succeed at finding out what does not work.