Slashdot Mirror
Intel Celeron 2.2GHz Reviewed
Detonator 3:16 writes "Black-Ash.net has posted a review of Intels Celeron 2.2GHz Budget CPU; interestingly they have compared it to a common older CPU (PIII-700MHz) to see whether it would be worth using this CPU as an upgrade." Celerons have usually a been a decent processor for the money, and this one looks to continue the trend. It's not the fastest chip ever, but for spending less than $100, it's a good bargain.
I understand that the review is somewhat game centered, I suspect the review site is as well, but this review does nothing for me.
Any frame rate that exceeds the refresh rate of your display is effectively wasted. You just won't see the extra frames. A 23% improvement just means that many more frames you won't see.
In all honesty, since he had to replace both the CPU and the Motherboard, the improvement provided by the combination will touch a few other things that should be presented. Since he chose to use the same video card, how much of the processing load was offloaded to the card? Is there a way to see comparable information wrt the hard drive?
For a closer to purer CPU comparison, I would like to know what kind of improvement to processing Seti@home blocks, or any of the other distributed computing projects.
-Rusty
You never know...
A realistic graphics benchmark would be a program that drew more and more polygons until the frame rate started to drop. That would actually tell you something useful. That's what you care about, after all.
It's amusing how much weight people give to those "refreshes faster than the framerate" benchmarks. NVidia drivers used to spinlock when waiting for vertical sync, instead of blocking. That didn't affect game benchmarks, but that CPU hogging forced OpenGL programs to 100% CPU utilization. I spent some time convincing NVidia's developers that they should block when waiting for vertical sync. The convincing argument was that benchmarkers turned off wait-for-sync, so it wouldn't affect benchmarks. NVidia then fixed it.
Multithreaded game programs speeded up, too.
As for the review, do the grey letters on a black background indicate that it's addressed to an audience that likes "shades of black" games?
Like you can ever get anyone to honor a Pricewatch quote.
I was actually very amused also to see that the Celery 2.2 really didn't give much of an edge over the PIII. Consider this: this is supposed to be a generational leap. Yes, I know about the MHz myth, I have a G3 Blue-and-white at home. You would expect to see a MASSIVE difference, though, between a PIII/700 and a chip that is supposedly 3X as fast. And you don't.
This should silence all the people who think that Pentium 4 must be an improvement over PIII. Pentium-M aka Centrino aka Banias is living proof that the PIII architecture can indeed scale to faster speeds with a little help. And that the P4 architecture, frankly, blows goats.
I would like to see a comparison between the two machines they put head to head, plus an Athlon XP Thoroughbred. I know that when I compare speed between my 733MHz PIII and my 1.4GHz/PR1800+ Athlon systems, there is a DRAMATIC difference in speed. Like the kind of spread that you would expect between the PIII and the Celery if only MHz matters.
Maybe it's the faster RAM. (DDR vs. PC133 Cas=2) But kernel builds are more than twice as fast between the two machines. UT performance is night-and-day between the two, but I blame the aging Rage128 video card and the shitty Linux DRI driver for it for the vast gulf in performance. The Athlon system has a GeForce 4Ti4200 and the nvidia binary driver.
Getting back to the issue at hand, there would certainly have to be a board swap in upgrading a PIII to a P4 Celery. You can get a crappy not-so-Elitegroup mobo for that $54 price, and pray it doesn't blow its caps immediately upon stressing it. Or you can spend a little more than twice the money and get a true Intel board that's built like a tank. You would also have to swap out power supplies, too. The case is not the problem. It is the power supply.
Knowledge is power. Knowledge shared is power multiplied.
While the CPUs are stable, I just can't find a good board that can go more than 6 months without a problem. I've had SiS, nForce, and VIA based boards that have all eventually crapped out.
I have a Duron running right now that has seen very little down time for over 3 years. It just keeps chugging away, overclocked and over-volted all to hell. It's outlived CPU fans, hard drives, video cards, a stick of RAM, even a motherboard.
That's fine, my story goes the other way: Two Duron problems: one a 1.2G and one a 750..er..1000..er..1000..er...1100 -- The heatsink tit on the socket snapped off and cooked the CPU and mobo. Didn't realize it was the mobo too until I put the 1000 in it and fried that. New 1000, new mobo, the bitch ran at 100oC. Far too hot, even with a copper base ThermalTake heatsink rated for the chip. Then I put an 1100 in a new system and I guess my 15+ years of taking care of computers wasn't gentle enough, since I crushed the corner of the die and zapped that one. Now there's a 1200 in there that's working well.
My wife's Duron 1.2 (Athlon 1.2? I don't remember) is flaky under high CPU load. New high-quality 400W (AMD-approved) power supply. RAM is fine. The CPU runs at a constant 55-60oC. High-quality heatsink and fan, installed properly. Ambient is 19oC. The damned thing just runs hot.
I've never had such a run of bad luck with Intel. Hell, I never believed the "Athlon goes up in a ball of flame" stories until that computer did it (it's a rackmount system and hadn't been moved in about a year and a half, don't give me the "you knocked it around" bullshit). -- At least they'll fold back and save themselves instead of practically exploding. And I've _never_ cracked the die on any coppermine.
AMD's nice from a price perspective but I won't buy another one.
My own study of having put together my own systems for the past 14 years. Sorry but IMO the Athlons and the chipsets they are mated to are not in the same league stability wise as my Intel systems. Not to mention it takes a turbo-prop to get these things down to a respectable temperature. Well this time around I went with an Intel CPU/MB combo for my main system, and I couldn't be happier
Also, I give my used systems as hand-me-downs to either my parents or my nephew. So yeah, I would like to get at least get a few years out of the thing.
This is an XP 1700 (1.46GHz), easily but poorly overclocked on a ECS K7S5A (not an O/C-friendly motherboard) to 147 * 11 = 1647 MHz. For reference, BIOS recognizes it as a XP 2000, and I have DDR memory, not SDRAM (this board supports both).
Total computation time : 32.03 seconds
I paid $50 for this CPU several months ago. I don't use the retail fan for my 10% overclock, since I have much higher goals in time. This CPU is extremely cool at stock, I'm sure 10% on a retail sink would be fine.
Here is the method used:
Computation of 4194304 digits of Pi
Method used : Chudnovsky
Size of FFT : 512 K
Physical memory used : ~ 29536 K
Disk memory used : ~ 0.00 Meg
I pressed: 0, 0, 4194304, 512, 1.
The space unintentionally left unblank.
Agreed, I'm hanging onto my SMP Pentium III 800 system for just that reason-- Pentium 4's are SMP'able, and I'm not a big fan of AMD processors (for various reasons, none of which I'll elaborate on because I'll just end up being modded a troll or flamebait, like every other person with a dissenting anti-AMD opinion). Besides, this dual processor P3 800 system doesn't even feel like it's aged much; every P4 based system I play with is just as responsive as this box at everyday tasks, and for gaming this system runs fine except for maybe a 30% drop in frame rates (which, as far as I'm concerned, anything over 50-70 fps is great for me).
And as more games become multi-threaded (since hyperthreading is going to be Intel's next big thing), or at least, SMP aware, I should hopefully see a bump in processing speed if their engines are designed right.
All I know about Bush is I had a good job when Clinton was president.
The fact that desktop users might not need a faster processor is beside the point.. This was a comparison of two processors, ergo, the benchmark used for the comparison should depend heavily on the processing power.
However, the overall speed of desktop applications is much more dependent on processor speed than it is for the overall speed of games, so processor speed is still relevant to them in that sense.
If you want fast games, 90% of the people would probably be better off spending $100 more on their graphics cards.