• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

2400XP + Nforce2 vs. P4 2.4B + Granite Bay vs. P4 2.4B + i845PE

mryoso

Member
Really confused on what to get here.

Price difference doesnt matter, which would be the fastest setup, stock and overclocked .

The Granite Bay board would be the MSI GNB Max-L ( only one available here ), yup overclocking options would be severely limited but i was under the impression that even a highly overclocked 845PE would be no match for it.

The Nforce2 would prolly be an Asus A7N8X Deluxe and the 845PE an Epox board.

additionally the GB board wont have an ide raid controller. would a Promise SMA Raid card be slower than an integrated one since its limited by the PCI bus speed? ( the asus A7n8x's raid controller ).

Will be paired by at least 2x512 of DDR333 or 4x256 DDR400 ( Corsair )
Thank You!

 
The 2400+ setup would be comparible to the GB setup. But the P4 could probably OC farther. The P4 at an equal speed is probably even with an AthlonXP, unless at high speeds then the scales tend to tip in the P4's favor. But the P4 can almost always OC farther.

 
Now, this is pure guessing, but i'd guess that the GB board would give it an advantage (in most applciations) equal to 150MHZ more clockspeed.

Thus, a P4 2.4 on a GB might get whupped pretty bad by a P4 2.66.

The GB starts to flex it's muscles when you get up to 2.8GHZ+
 
Originally posted by: FishTankX
The 2400+ setup would be comparible to the GB setup. But the P4 could probably OC farther. The P4 at an equal speed is probably even with an AthlonXP, unless at high speeds then the scales tend to tip in the P4's favor. But the P4 can almost always OC farther.
This isn't true. The P4 will overclock farther in raw MHz, but the AXP will get more performance from an overclock. For example, you can get a 1.6a to 2.4GHz fairly easily, and using the same cooling, you can get a 1700+ to the speed of a 2800+ (2.25GHz) fairly easily. The 2100+s are OCing pretty well, but if you're overclocking, unless you just want to waste money, then don't get an AXP that's over a 2100+ because they do hit a ceiling at about 2.4GHz (about equivalent to what a 3000+ or 3100+ would be).
 
uh... its a little bit more confusing this time! lol!

so uh...
which would be faster?

1) An Athlon XP2400+ ( need to get this to make sure it will be a Tbred B ) on a DC Nforce2 platform on a typical overclock . just guessing here but maybe around 2.2Ghz???

2) A P4 2.4B ( or 2.66 if need be ) on a i845PE? again guessing.. but maybe i can run around 160FSB for 3.2Ghz.

3) A non-to-slightly overclocked P4 2.4B/2.66 using DDR266 on a GB platform ( its hard to use DDR333 on the GB right? even if it isnt, the only available board i can grab would be the MSI GNB, eliminating all overclocking hopes 🙁 )

this would have been really easy if someone had a review of the GB, i845PE and the 850E against the Barton+Nforce2...

just about ready to buy any one of these three...

thanks

 
I'm working on benching my setup now for you w/ a XP 2100+ oc'ed to 2.27ghz. More details will follow 🙂
 
Aight, here's a sampling of my Week 50 2k2 2100+ on an 8RDA+ w/ 512mb of PC3500 (2x256mb sticks)

Temp at full load running Prime95 for 6+ hours:
55*c

Vcore: 1.775v

Ram: 175mhz 2-2-2-5 1T

Sandra Scores:

CPU Arithmetic:
Dhrystone: 8236 MIPS
Whetstone: 3428 MFLOPS

Interger: 12609
Floating-Point: 13335

Memory
Int: 2597
Float: 2449

3dMark2001:
13811 3D Marks on a Geforce4 Ti4400 @ 294/679

Ut2k3 Benchmarking Utility 2.0 Provided by HardOCP.com:

CPU Test
Resolution dm-antalus
640x480 77.373421
800x600 77.423943
1024x768 77.217575
1280x1024 76.741631

Guess I'm seeing a GPU limitation there, should know when I receive my 9700 Pro in the next 2 weeks.

High Quality Test:
Resolution dm-antalus dm-asbestos ctf-citadel
640x480 135.829773 181.997757 101.257301
800x600 134.468185 164.050705 90.284340
1024x768 102.499413 132.880447 70.057487
1280x1024 69.380630 100.987450 50.037205

Low Quality Test:
Resolution dm-antalus dm-asbestos ctf-citadel
640x480 191.982666 260.892456 182.451477
800x600 187.138611 260.756866 172.121887
1024x768 143.428558 257.875549 135.067368
1280x1024 96.047165 250.533905 93.714577

When I have time to remove the Northbridge heatsink and slap some AS3 on there, I should be able to get a higher FSB.
 
Sandra Scores: ( Mine 2.26b@160FSB i845g 1x512MB@400Mhz )

CPU Arithmetic:
Dhrystone: 8236 MIPS ( 7270 )
Whetstone: 3428 MFLOPS ( 1520/3553 )

Interger: 12609 (10714)
Floating-Point: 13335 (13008 )

Memory
Int: 2597 ( 2938 )
Float: 2449 ( 2957 )

3DMark2001SE ( GF4ti 4200@315/560 Max Driver would give me 🙂 )
12851

--

cant come up with anything here...still a bit lost on what i should be getting... lol.. but its nice to see some comparisons with my current rig thanks!. is it that far off?

anyway, what would be best upgrade among the three?

1) An Athlon XP2400+ ( need to get this to make sure it will be a Tbred B ) on a DC Nforce2 platform on a typical overclock . just guessing here but maybe around 2.2Ghz???

2) A P4 2.4B ( or 2.66 if need be ) on a i845PE? again guessing.. but maybe i can run around 160FSB for 3.2Ghz.

3) A non-to-slightly overclocked P4 2.4B/2.66 using DDR266 on a GB platform ( its hard to use DDR333 on the GB right? even if it isnt, the only available board i can grab would be the MSI GNB, eliminating all overclocking hopes )

this would have been really easy if someone had a review of the GB, i845PE and the 850E against the Barton+Nforce2...

thanks
 
I don't value synthetic benchies all that much, do you have the UT2k3 demo installed? If so, download the HardOCP benching util and see what kinda results you get from that.
 
Don't necessarily buy into all that duall ddr hype and high bandwidth numbers...if the rambus reviews showed us anything it is that not all application need that type of bandwidth and thus real performance can be rather minor even with a large lead....


That being said I want to look at the question of a highly overclocked p4pe versus a MSI GB mobo...

Example: My cpu...listed below...

MSI GB mobo....Let say I can only get stable 150fsb for a total of 2.4ghz and 300mhz ddr for 3700-3800mb/s of bandwith in sandra...

Now...

A P4pe mobo with say a 4:5 memory ratio I can likley get to 171fsb I do now for 2.736ghz and run my memory at 428mhz ddr and I score mid 3200's in sandra...

I can tell you from extensive testing I did 4-6 months ago that that much raw cpu speed can overcome that lead in bandwidth....336mhz of cpu can offset closer to 800-1000mb/s of bandwidth lead so unless you can get that cpu up to say 2.53ghz (159fsb) and 320mhz ddr it is unlikley that will be anything then a push....

I tested divx, video encoding, cadd, etc...no sythetic BS like Sandra...
 
wow, so i guess the GNB isnt worth it if it doesnt overclock well...

would you know how the NForce2 would stand agaisnt an overclocked p4 on a non-rambus platform?
btw, ill be doing the exact same thing as you did, encode videos!


thanks!
 
A non oc'd GB mobo will act like a pc1066 rdram system which is used in most cpu reviews of late...that being said a 2800+barton with nforce2 will likely beat a any p4 2.8ghz with ddr (non-oc'd or pc2700 only level), but be a wash against a GB (non-oc'd setup)....

The GNB is basically a i850e platform with rambus when it is not oc'd....Ocing will help get bandwidth scores up but then how much will they help in apps varies...gaming could help quite a bit, but otherwise may do little or nothing or be so minute the difference in the real world is not felt.

An athlon does quite well in video encoding, with the p4 having the edge in some apps that especially encode to the sse2 features....

example of bandwidth in divx encodings....

My system:

1.6@2.64 w/ 440mhz ddr cas 2.5= same time encoding same movie with all settings as a 1.6@2.72 w/ 342mhz ddr cas 2....

basically 100mhz in memory almost equated to 100mhz in cpu gain...cas had something to do with it as well...

I have no real test wth athlon xps...I did an athlon 1600xp stock and it roughly did a movie a little bit faster as my 1.6a did at stock
 
thank you.

it looks as though i might have to go the nforce2 route for video encoding.

hope you dont lose patience here.

but will a xp2400+ ( when overclocked ) beat a p4 2.66 on a 845pe platform?

with both chips on average expected overclock... 2400 tbred-b ( 2.2GHz?) and the p4 2.66 ( ~3Ghz?)

just about ready to buy ... really.. im serious.. lol. been up for nights straight looking for an answer... 🙂
 
no barton here in our place 🙁 best i could get would be a xp2600+, if it would mean more overclock compared to the xp2400+ never mind the price difference! im desperate! loL!

thanks for that link MTDEW, gonna be printing it and spend time reading it! 😀
 
Originally posted by: mryoso
no barton here in our place 🙁 best i could get would be a xp2600+, if it would mean more overclock compared to the xp2400+ never mind the price difference! im desperate! loL!

thanks for that link MTDEW, gonna be printing it and spend time reading it! 😀

Seems the 2400+ are overclocking just as high if not better then the 2600+, so just stick with the 2400+ then.
 
Np, just came across it while trying to answer the same question you asked 4 myself.

Seeing how I already have an Epox 8RDA+ Nforce2 board with 2000xp palimino here, I believe I may go the 2400Tbred route also.
Heck I might even change waterblocks and stick my Nforce2 in my watercooled setup along with my 9700pro.
 
i think if someone did an Nforce2 + Athlon XP vs various P4 chipsets ( ie not only the i850E, since DDR users far outnumber RDRAM users) reviews, then it would be whole lot easier! of course this will take a whole lot of time :Q

 
Back
Top