• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Celeron 2.8Ghz vs Athlon 2400+ (2Ghz)

Caporegime
Joined
24 Dec 2005
Posts
40,065
Location
Autonomy
daz said:
The Celeron will be better at any kind of encoding tasks... but woeful at any kind of task involving repeated accessess to the cache (because it is so small...) such as games.

A Celeron at around 3ghz is equivalent to a normal P4 at 2.3ghz in gaming. :)

Encoding involves repeated access to cache!
The celeron is an inferior awful cpu and to compare it to a p4 is laughable imo. :D
 

daz

daz

Soldato
Joined
18 Oct 2002
Posts
24,078
Location
Bucks
m3csl2004 said:
sorry daz but i find that hard to believe :(

this celeron with a 9800pro cant play css at 1024x768 on low with no aa or af

Your celeron at 2.8 is comparable to a P4 at 2ghz in gaming... which is probably not really enough to play CSS.

easyrider said:
Encoding involves repeated access to cache!
The celeron is an inferior awful cpu and to compare it to a p4 is laughable imo. :D

Of course you can compare a Celeron to a P4... they're the same CPU with a castrated cache!

I'm not sure what encoding benchmarks there are, but I'd encourage anyone to do a test. A Celeron at 3ghz will beat a Athlon XP at encoding a wav to MP3, for example because of the clock rate advantage.
 
Soldato
Joined
24 Dec 2004
Posts
11,355
Location
Knowle, Solihull, UK
daz said:
Of course you can compare a Celeron to a P4... they're the same CPU with a castrated cache!

The later Celerons are based on P4 cores but I believe that there is more to it than just a lack of cache

Take Super PI, for example. This is virtually unaffected by cache (an A64 with 512KB cache performs like an A64 with 1024KB cache at the same clock frequency). The Celeron has to be clocked far higher than the equivalent P4 to get the same result though
 

daz

daz

Soldato
Joined
18 Oct 2002
Posts
24,078
Location
Bucks
ajgoodfellow said:
but I believe that there is more to it than just a lack of cache

Like what? :confused:

You say that SuperPI is unaffected by cache, and use the comparison of 512KB to 1024KB as your example, when really you're comparing 128KB to 512KB+, or even just 256KB+ which is what the Sempron has. The 128KB of cache is just not sufficient to avoid repeated accesses to system memory rather than CPU registers. For a few tasks, 128KB *is* sufficient and the Celeron can make use of it's clock speed.

Seriously... someone do a wav ---> MP3 encode. :o

Either way, a Sempron is better value for money. :)
 
Last edited:
Soldato
Joined
5 Jul 2005
Posts
17,995
Location
Brighton
daz said:
I clearly see the Celeron D 2.8Ghz beating a Athlon 64 Newcastle 2800, Winchester 3000, a Sempron 3100 and a Barton 2500. :confused:


And i clearly see that it is a Celeron D 2.8ghz in that chart, which is faster and STILL being beaten by the 2400+. Which is what is being compared here and not the 2500, 3100 or 3000.
Stop trying make what is clearly a crap cpu look anything but that.

The diferences are a tiny amount of l2 cache and a low FSB. The FSB and cache are the two most important factors with pentiums (bar clockspeed) due to their extremely long pipelines.
 
Permabanned
Joined
24 Jul 2003
Posts
2,584
I had a 2.93ghz celeron d running at 3.6ghz, and then bought a ht p4 3.4ghz, and I didnt see improvement in games...
They are rubbish, but they are still better than semprons.
 

daz

daz

Soldato
Joined
18 Oct 2002
Posts
24,078
Location
Bucks
Raikiri said:
Stop trying make what is clearly a crap cpu look anything but that.

Point is, they're not totally crap. They output little heat and are perfect for running office and web applications.

I'm not entirely sure as to the variables of that particular encode benchmark, but any search on the net will reveal that the Celeron is no slouch when it comes to encoding.

Edit: on this test, you'll see that it's raw clock speed that seems to have the biggest effect. http://images.tomshardware.com/2003/02/17/benchmark_marathon/image030.png
 
Caporegime
Joined
24 Dec 2005
Posts
40,065
Location
Autonomy
lol^^^^^^^^^^^^^^^^^^^^^^^ :D

If I was building an office budget pc I would pick a sempron over the celeron!
And I reckon theres a few on here that would agree with me :p
 
Soldato
Joined
17 Sep 2003
Posts
2,813
Location
Suffolk
daz said:
Point is, they're not totally crap. They output little heat and are perfect for running office and web applications.

I'm not entirely sure as to the variables of that particular encode benchmark, but any search on the net will reveal that the Celeron is no slouch when it comes to encoding.

Edit: on this test, you'll see that it's raw clock speed that seems to have the biggest effect. http://images.tomshardware.com/2003/02/17/benchmark_marathon/image030.png

That was 3 years ago!!!!!

Its a little unfair to compare with todays current processors.
 
Back
Top Bottom