would my processor be= to a 4000+?

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
hi, my 3000+ winnie is chugging along fine at 2.61GHz, and while it has 512k of L2 cache, would the 216MHz more clock make it about even in performance to a socket 939 4000+ (which has a 1MB L2 cache, at 2.4GHz)?
thanks, i want to have something shorter to tell people when they ask me "what processor do you have". i hate saying "oh ive got a socket 939 winchester 64 3000+ (1.8GHz stock) running at 2.61GHz with a 290MHz HTT."
:D
-nick
 

Mik3y

Banned
Mar 2, 2004
7,089
0
0
Originally posted by: zakee00
i hate saying "oh ive got a socket 939 winchester 64 3000+ (1.8GHz stock) running at 2.61GHz with a 290MHz HTT."
:D
-nick

if you tell them you have a 4000+, that'd be lying anyways. it's not a real 4000+, though it operations at the same performance. it's like saying that i have a P4 4.0ghz cpu and i call it a 4000+ just cuz i like the name better. :)
 

slash196

Golden Member
Nov 1, 2004
1,549
0
76
I'd say it's a good deal faster. Cache is not that important anyway. Just tell them you have an FX-55.
 

Mik3y

Banned
Mar 2, 2004
7,089
0
0
Originally posted by: slash196
I'd say it's a good deal faster. Cache is not that important anyway. Just tell them you have an FX-55.

2.6ghz on winchester core is not fx-55 speeds. fx-55 speeds is around 2.7ghz.
 

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
no, 2.6. cache is actually pretty important in cirtain apps...not really games, but in video encoding its very important.
 

Mik3y

Banned
Mar 2, 2004
7,089
0
0
Originally posted by: zakee00
no, 2.6. cache is actually pretty important in cirtain apps...not really games, but in video encoding its very important.

the cache isnt that important to amd because of the shorter pipelines that its core architecture utilizes. it just happens to be that an increase in clockspeed over the extra 512k of cache performs very close with each other. its different for intel since they have much larger piplelines that larger numbers will benefit them more. thats why they have 2mb L2 cache and etc.
 

miketheidiot

Lifer
Sep 3, 2004
11,060
1
0
i would say that that looks about like a "4200" or whatever the next processor is in the 512k cache lineup.
 

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
ohhhh, milky i was thinking of the P4EE that DIFFENATELY benefited from extra cache
now im firm against the FX-55...way overpriced. only good thing is unlocked multi, i can live w/o that
thanks guys :D
 

Zebo

Elite Member
Jul 29, 2001
39,398
19
81
It will beat the 4000 every single test. (see my memory thread which uses a 2610 winch) and compare benches to those at techreport which uses a 4000. PR is a 4100.... 4000's PR is 3969.

1mb lvl 2 can't hope to compete with 210 more mhz. Not to mention overclocked bus...which I even beat FX-55 in some of tests cause of it.


Did'nt we already have this discussion..like a 3400 newcastle owning a 3400 Clawhammer and FX-51?
 

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
thx zebo, nice sys btw ;) 450/1.2 on a 6800gt? thats insane to the membrane lol. how did you do it? is it stable in HL2?
 

VIAN

Diamond Member
Aug 22, 2003
6,575
1
0
Why don't you just use a bench to find out and check out other peoples benches with the similar specs, just CPU benches.
 

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
good idea vian :D
my comp just failed prime95 blend test after 2.5 hours, i need to figure out what the F is wrong. its been fine other then prime95 tho. hmm...i dont think its the ram, or voltage...maybe my cpu just cant handle 290HTT. ill mess with it.
 

jiffylube1024

Diamond Member
Feb 17, 2002
7,430
0
71
Originally posted by: zakee00
good idea vian :D
my comp just failed prime95 blend test after 2.5 hours, i need to figure out what the F is wrong. its been fine other then prime95 tho. hmm...i dont think its the ram, or voltage...maybe my cpu just cant handle 290HTT. ill mess with it.

I'd suspect the 2.61 GHz over the 290HT if it's failing Prime95. Try at a lower speed.
 

mwmorph

Diamond Member
Dec 27, 2004
8,877
1
81
i never liked the fx or Dxxx programs, just tell me the damn performance outright. they might as well have names like the "P4 'if you are a rich bastard and can afford a Lamborgini, buy this chip to piss off all your poor, underpriviliged friends'" cause both ways tell you the same amount of performance info.
 

icejunkie

Platinum Member
Oct 2, 2004
2,326
0
0
Originally posted by: mwmorph
i never liked the fx or Dxxx programs, just tell me the damn performance outright. they might as well have names like the "P4 'if you are a rich bastard and can afford a Lamborgini, buy this chip to piss off all your poor, underpriviliged friends'" cause both ways tell you the same amount of performance info.

HAHA! :p
 

crossphyr

Member
Dec 18, 2004
66
0
0
just tell them you run a 3000+@ fx55 speeds.. than u get the oohhh's and ahhh's and u will be happy.. but if ur not stable which none of us can prove and u dont have a screenshot of prime95 showing so.. for all we know u crash every 5 secs and cant run google without having to cmos ur cpu... oh well goodluck
 

zakee00

Golden Member
Dec 23, 2004
1,949
0
0
Originally posted by: jiffylube1024
Originally posted by: zakee00
good idea vian :D
my comp just failed prime95 blend test after 2.5 hours, i need to figure out what the F is wrong. its been fine other then prime95 tho. hmm...i dont think its the ram, or voltage...maybe my cpu just cant handle 290HTT. ill mess with it.

I'd suspect the 2.61 GHz over the 290HT if it's failing Prime95. Try at a lower speed.

no, its been stable gaming for 4+hrs in css, but it fails prime95 in about 2.5 hours. i just put the speed back down to 2.43GHz, im kinda tired of messing with it. do you guys think that better cooling could enable me to get it up to a stable 2.61? i cant think of what else would be making it fail prime95...im using DDR500 ram, and thats only operating at about 245MHz w/ 2.8V so i know its not the ram...
and im not bragging...i was just wondering what i would be about = to in speed, and vian gave me the idea of just running some bechmarks :p
thanks,
nick