|
by laudablepuss 04/19/2005, 3:48pm PDT |
|
 |
|
 |
|
Just a guy wrote:
laudablepuss wrote:
Ray of Light wrote:
laudablepuss wrote:
By architecture, are you talking about hyperthreading? I believe only the P4s have that (as opposed to the Celerons), so yeah, if you had an app that took advantage of that you might see an improvement. I don't really know what hyperthreading is supposed to do, so it's hard to say. Hyperthreading . . . maybe the CPU reaches into another dimension to retrieve future instructions which it can subsequently execute in the past. That'd be a pretty cool feature.
"No."
Check this out. (Also here.)
Hyperthreading is apparently not bad. The new dual core Pentium D XXXTREME!@!#!@! or whatever has hyperthreading also, for a total of 4 logical CPUs. Huh.
The question wasn't, "Is hyperthreading a desirable technology," it was whether MM would see a significant improvement by upgrading from a 2.5 Celeron to a 2.5 P4. Ray is right, the answer is No. Besides, I am currently hating all of Intel's pinless CPUs, waiting to see AMD introduce their dual-core chips, hyperthreading or no. The ability to run seti@home and antivirus more harmoniously does not tweak my nobs.
The new dual-cores will cost much and need a new board entirely. Not to mention that there exist no games written to take advantage of multiple-cores, to which you must ask yourself if, for gameing, an AMD64 solution wouldn't offer much more value. Value coming from spending at least $300-$900 for the CPU and $150+ for a top-shelf mainboard. Which I don't believe MM had in mind. The answer to the original question is still. No.
If MM wanted to spend money on a new CPU but keep his current MB, a 3.0+ GHz 478 chip ($180-$300, $900 for the 3.4 EE chip) would indeed demonstrate noticeable speed improvements, but that wasn't the question either. Also, 478 chips are poised to be paperweights sooner than later, spending new money on them hurts my brain.
The answer to the original question is, "No".
The original question (GOD FORBID WE STRAY FROM IT IN THE SLIGHTEST) was, "If I upgrade from my 2.5 Gig Celeron to a 2.5 gig Pentium 4 will I see a significant improvement?" From that it would be reasonable to infer that MM wanted to keep the original motherboard. However, that isn't the only interpretation possible. I jumped into a daring, multi-word answer where I mentioned a feature and asked a question. Pretty ambitious, I know. But yes, it's true that hyperthreading enabled P4s require different motherboards than Celerons based on the P4 core. So MM would need a new motherboard.
Anyway, BACK TO THE ORIGINAL QUESTION AGAIN, knock yourself out. It's tough to compare some of these, but the 2.4 northwood celeron with DDR200 is 169.2 fps in the Open GL Q3 Arena test, while the 2.4 P4 northwood with DDR 266 is at 264.3 and the 2.0 P4 northwood with SDR 133 is at 195.5. So maybe, just maybe, the answer isn't, really, in fact, now that we look at it, "no". I guess it depends on your interpretation of what MM is likely to call "significant". There are several more tests that I didn't look at, but you can feel free.
Incidentally, Tom's HWG suggested that the Pentium Ds would cost in the neighborhood of $300, which isn't very expensive at all. Where are you getting your price information? Is it, as I suspect, nowhere? As for no games taking advantage of the multiple cores . . . correct me if I'm wrong but isn't the OS the thing that handles which core gets which thread? I've never seen a SMP version of Quake3 and yet it appears to run better with multiple processors.
Finally, someone did in fact ask the question: "is hyperthreading a desirable technology" . . . however I was the one who asked it, in a roundabout way. Honest, I totally did! That "huh" at the end of my post was a mental shrug, which you can read as "this is interesting". But enough about me, do tell more about what tweaks your nobs. I can't wait to hear it. |
|
 |
|
 |
|
|
|
How much difference is there between a Celeron and a Pentium 4? by Mischief Maker 04/18/2005, 10:16am PDT 
Re: How much difference is there between a Celeron and a Pentium 4? by Ray of Light 04/18/2005, 10:33am PDT
Re: How much difference is there between a Celeron and a Pentium 4? by laudablepuss 04/18/2005, 1:48pm PDT
Summary by Ray of Light 04/18/2005, 1:52pm PDT
Well . . . by laudablepuss 04/19/2005, 2:20pm PDT
Re: Well . . . by Just a guy 04/19/2005, 2:54pm PDT
Dear just a total genius guy by laudablepuss 04/19/2005, 3:48pm PDT
Minor correction by laudablepuss 04/19/2005, 4:16pm PDT
Whoops nevermind! I thought I was mistaken but I was wrong. by laudablepuss 04/19/2005, 4:18pm PDT
So the answer is still, No, right? NT by Just a guy 04/19/2005, 4:40pm PDT
You should just stick with ragging on me guy. NT by Worm 04/19/2005, 6:29pm PDT
Maybe you aught to figure out what your talking about first, retard. NT by Just a guy 04/19/2005, 6:36pm PDT
No. NT by Worm 04/19/2005, 6:46pm PDT
You're right either way NT by Entropy Stew 04/19/2005, 4:44pm PDT
Wow (the new SB) NT by I need clarification 04/19/2005, 6:54pm PDT
Hi, how's it going? by laudablepuss 04/19/2005, 7:04pm PDT
Quake 3 runs better on any AMD 939 than on dual Xeons, we are talking game box. by Just a guy 04/19/2005, 4:48pm PDT
I THINK YOU ARE STRAYING FROM THE ORIGINAL QUESTION, SIR. NT by laudablepuss 04/19/2005, 4:50pm PDT
|
|
