Which is the preferable of the following two processors?


Details:

A few months ago my motherboard died, so I bought a used computer that had a 2.4GHz Celeron. My old system had a 1.7GHz Pentium 4, so now I’m trying to decide which CPU to use.

Obviously a P4 is preferable over a Celeron, but the Celeron is (significantly?) faster than the P4. I’m wondering if the faster Celeron might be better for certain tasks (ie, stronger but dumber is better at some things than smarter but weaker). I tried Googling for some reviews and comparisons for graphs to get a clear depiction of which is better overall, but found nothing that helped. (I did manage to find one page that indicates (apparently by poll, not benchmark) that the Celeron is better.)

So which CPU should I use? Does anyone know of some graphs that I can use to compare the two?

The system is a general-purpose machine for word-processing, Internet, and casual games (not Crysis, but not Solitaire either). It will be running Windows XP. The board is a 478 with 400MHz FSB.

link|improve this question

48% accept rate
You could help us a lot by saying which Pentium 4 and which Celeron you are actually using. You know, there are different P4s and different Celerons. – AndrejaKo Mar 4 '11 at 23:26
I did, 1.7GHz P4 and 2.4GHz Celeron. – Synetech Mar 4 '11 at 23:42
Like most questions like this, your results will vary depending on what you typically are going to do on the system. Gaming? Productivity? Just reading email, watching youtube videos, etc... Also, what OS are you going to use? – deltaray Mar 4 '11 at 23:44
Updated the question for clarity. – Synetech Mar 4 '11 at 23:47
3  
Look up both CPU's on this page, compare the test results...cpubenchmark.net/cpu_list.php – Moab Mar 5 '11 at 0:04
show 6 more comments
feedback

closed as not a real question by KronoS, Jeff Atwood Mar 5 '11 at 6:23

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. See the FAQ for guidance on how to improve it.

3 Answers

up vote 1 down vote accepted

See Tom's Hardware for ton's of benchmarks going back through the years.

link|improve this answer
I managed to find a chart there that had both processors and used the data to make a decision. Thanks for the effort. – Synetech Mar 6 '11 at 20:11
feedback

If you find that the Celeron is faster for the applications you're using, then that's clearly the better choice for you.

If there is some feature that the Pentium 4 has that one of your applications depends on (and that feature isn't available with the Celeron), then you'll have to weigh the trade-off of that feature against the speed of the Celeron.

In nearly all instances the faster option is what usually gets selected.

link|improve this answer
Except I don’t know what applications will be used on this system. It’s a general-purpose system for word-processing, Internet, and casual games (which aren’t exactly Solitaire these days). – Synetech Mar 4 '11 at 23:43
why the downvote? Looks like a valid answer to me... +1 Take Moab's advice and look at those benchmarks. If an app you are using is benchmarked there, pick the faster one! – aCuria Mar 5 '11 at 0:30
Don’t ask me, I didn’t vote and whomever did, didn’t leave a comment. :-| – Synetech Mar 5 '11 at 4:39
feedback

If I had to guess (and apparently I do), I'd take the Hyper-threading and larger cache of the Pentium 4, but really I expect those to perform very close most of the time. Flip a coin. And since they're that close performance-wise I might look to other factors to decide, in which case I'd give the edge to the Celeron on the basis that it's likely to cost you less money on your energy bill. The old P4's really sucked it down.

link|improve this answer
> I'd take the Hyper-threading and larger cache of the Pentium 4 So would I, if I had an HT P4 (in which case I would have said so). > I'd give the edge to the Celeron on the basis that it's likely to cost you less money on your energy bill. The old P4's really sucked it down. That’s a good point. Perhaps I could get one of those P3 Kill-A-Watts from the library and measure the system with each to see if it’s significant enough (but that would mean wasting a ton of time running the systems for nothing since I’d have to have them doing about equal and identical amounts of work…) – Synetech Mar 5 '11 at 5:44
feedback

Not the answer you're looking for? Browse other questions tagged or ask your own question.