Uncreative Labs

Guest

I work in a computer store and my boss, the owner seems to believe that the real difference between the Pentium 4 and a Celeron 4 is the chipset. He tells this to customers all the time. I, a computer technician tried to explain to him that this is not the case. So has the other computer tech and our sales associate. He insists that we are wrong when we tell him the difference lies in the difference of L2 cache size. I have even shown him proof from the intel website and many other credible sources and he dismisses it a clever marketing ploy on the part of Intel to get people to dish out a higher amount of money for the Pentium 4 processor. I then rebutted by saying that it costs less for Intel to make a Celeron processor because of the lack of L2 cache and that there is a real noticable difference in performance, in which benchmarking tests have established. He then stated that it costs no more for them to make either processor, that its a lie. He then said that the chipset determines how well the processor will perform. That if you put a Celeron processor in a motherboard with a really good chipset it will perform as good as the pentium 4 because the processors are identical.

I then told my boss that even if what you are saying about performance is true, its still not the chipset that distinguishes a Pentium 4 and Celeron processor. He said I am wrong. He then went on about how was in the computer industry for over 30 years and that he knows what he is talking about and that I should trust his judgement.

I would like to know what you guys think of this. I have read and seen evidence that a chipset can adversely affect system performance if it is of a low end cheapie variety. And I know that the Celeron and Pentium 4 are largely identical. But I also know that more cache is better than little or none. My boss seems to disagree.

Posted: Sun Jun 11, 2006 6:54 am

Depends on app but mostly more cache is better to take a work load but not necessarely faster if u oc a celery they work quite well for a lower cost but a quality MB can also make a difference. It could be little difference in cost of making them but they want to market a cheaper version as well to increase sales.

Posted: Sun Jun 11, 2006 1:36 pm

Get a motherboard that supports both and a benchmarking program. Put the same speed/FSB multiplier chip in for both the P4 and Celeron. Run the benchmark.

Computer Power User magazine likes to use the framerate of a few games as comparison numbers in their hardware reviews.

Posted: Mon Jun 12, 2006 10:56 am

Puckdropper wrote:

Get a motherboard that supports both and a benchmarking program. Put the same speed/FSB multiplier chip in for both the P4 and Celeron. Run the benchmark.

Computer Power User magazine likes to use the framerate of a few games as comparison numbers in their hardware reviews.

In such a set up the higher cache will win but if u oc a celery u get a cheaper
setup with nearly the same performance.

I have setup for my son in an old siemens/nixdorf chassie. 440BX chipset MB
PGA > PGA2 convertor and a Tualatin celeron @ 120mHz / 1.4 gHz with an ATI 9200SE agp graphics card. A cheap setup that works well with most games except for the very latest stuff.

Joined: 19 Apr 2006 Posts: 261 Location: WisConSin

Actually P4's and Celerons depending on models "may" actually cost the same, in some cases the celeron is more than the P4 to manufacture. It really is a marketing ploy is some cases. An example where different models w/ reduced cache or disabled cache has been a cost savings has been with the Recent AMD Duron/Thouroughbred as Thouroughbreds that failed on cache tests could be setup with a simple jumper to run with less cache enabled and pass tests.

I believe intel has internal controls that accomplish similar feats. This is not the case with all models though, die size & non-op rate is what determines cost of manufacture of cpus.
Many celerons and p4s have the same die size hence the same cost but extreme P4s have a higher failure (non-op) rate right after manufacture and a larger die size and hence cost more.

Anyone that says the celeron is NOT a marketing gimick is dreaming (they are) as the celeron is intended to be value and is purposely undermarketed toward price to give intel a value chip when on intels side it really isn't much of a savings to produce. (you could also look at it the other way, the p4's price is too high as the celerons price is about what the p4 should cost)

The problem with either view is that some celerons are disabled p4's and some are produced to be that way so some are a cost savings (and may have better characteristics than another celeron) and others are standard and cost intel about the same to make as a p4. The reason is there is more demand for celerons than there are p4's with specific problems that get rebadged as a celeron.

Cheers

Posted: Tue Jun 13, 2006 12:32 am

Quote:

The reason is there is more demand for celerons than there are p4's with specific problems that get rebadged as a celeron.

No, demand != supply. There's more supply of Sell-a-rons than Pentium 4s therefore the price is lower. (If you believe supply and demand.) Most likely there are two factors that make the contribution to demand for Celerons: Price and name. Pentium's 10+ years old, Celeron's new and exciting yet. (I think it also sounds cooler.)

I don't like Celerons, the same clock speed AMD blows the Celeron out of the water, Perhaps even with the P4. (Celeron D's have on-chip cache which does even the odds a bit.)

Quote:

In such a set up the higher cache will win but if u oc a celery u get a cheaper
setup with nearly the same performance.

That's exactly the point. With everything the same, there's only two changes: Chip series and cache. (You could get a 512k cache and 1 MB cache P4, but the point wasn't more cache wins but the chipset was the same.)