"Asus nFORCE2 (thread retitled)"

[sysfailur]

Originally posted by kanedazan
Sysfailure... You have pluged your athlon, and... multiplier for the cpu were unlocked from the beginning, no problem??

Because as I have a xp2200, but a revA one, I wonder if I buy this board, will it be able to unlock it, because with SUCH an oc friendly board, 200*10 is a must have it seems for a 2200+ for a beginner oc.
Thanks

I have put in my 2600+ and yes all the multipliers I have tried thus far have worked. There is an issue though that I'll have to sort out--some multipliers are showing up as like 7x/14x as in the same selection is the same multiplier... not like I'm gonna use 7, but I'm gonna figure out what works of the two.

[sysfailur]

Originally posted by Henry
Thanks sys

Hope nf2 mobo use one controller for both AGP/PCI clock.

I'm going back down to work on it in a few. I will be sure to check all of the frequencies. I got alot of testing to do!!!

BTW, the ram timings on these things are INSANE (as was stated by someone else in a diff. thread) you can use 1T on any setting!

[DJ_Datte]

hehe, wonder if the mobo is good for it if there was memory that could support it, ie, if the mobo will actually handle it .)

/Damir

[sysfailur]

Originally posted by DJ_Datte
hehe, wonder if the mobo is good for it if there was memory that could support it, ie, if the mobo will actually handle it .)

/Damir

Well, I'm gonna give some 1T timings a try w/ my winbond 5ns. My previous ram (XMS3200) could hit 186 @ Cas 1.5T. So maybe this stuff will do 195 Cas 1T.

[DJ_Datte]

Good Luck

[Wolf]

Hey man can you do a couple of benches with the cpu fsb at 211 and the ram at 166fsb dual channel. I want to see how it holds up bandwidth wise or if there is a performance hit. Can't wait to see what results you get. Fill us in there hasn't been a decent review yet.

[DJ_Datte]

Even if there is say a 35% Scalability loss from not running sync, and from running dual channel (overhead between mem controllers) , its still 333 DDR * 2 = 5328 MB /sec * 0.65 = 3463,2 MB /Sec vs 3376 MB / Sec for Max CPU Bandwith which Should give a almost 100% full mem bandwith ... This mobo will be fun to experiment with .. and I am expecting some crazy "unbuffered" scores, you know when you put SiSoft to run pure bandwith tests, and it usually only reports like 1100 MB /Sec for 166.. I would assume those scores to improve tremndously (SP!)

Am I wrong in my reasoning ?

/Damir

[Wolf]

That's what I'm thinking and if we had the option to run the cpu at 300mhz fsb and run the ram at 60% or 200fsb dual channel that would really open up the pipes. I'm just interested on how bandwidth holds up when the cpu fsb is run asychronously higher than the ram speed. I guess it won't be long now before we find out. I hope epox or someone else allows higher than 211fsb though.

[DJ_Datte]

Abit promised 17 MHz higher then the competition, so I am expecting ATLEAST 228 MHz, which would rock ) I think we are getting somewhere with this platform

I am excited.. I cant wait to get mine !!

/Damir

[vmsrules]

it looks like all of these nforce2 motherboards use the ATXP1 chip from http://www.attansic.com.tw/ to control the cpu voltage. the trick will be finding the way to mod the circuit for this chip.

[QuadDamage]

Sysfailur,

i've tried 1T with my XMS3000 which does 243fsb on unlocked P4 chip, but @ these timings it won't even post, lol. you gotta agree with me that this mobo clocks better than kt400 it's rock solid too.

if i had to list all the features this mobo offers, i would have to spend whole night typing... 5.1 soundstorm sound, LAN nVidia, LAN 3Com, 1394 ports, SATA.........etc etc.

[sysfailur]

Well... sorry for the late reply.... I couldn't get to working on my comp till about 2 hours ago. And since then I've hosed the OS! Hahah! I had it running windows fine and dandy, but when I had it @ 207x12.5=2380mhz it didn't seem to like taht and went buck wild on me. After that it wouldn't boot... BUT I did get a mem bench in @ that and I got ~ 3250/3060 I believe. That was dual-channel, cas 2-5-2-2 (cpu set at "agressive").

I also took some pictures at some chips on the board which may be the voltage chips. I will post thost in a few minutes.. gotta load em in.

I see a great future for this board. I just need more voltage methinks. I am running 1 GB of memory... so I would assume more voltage = better to power everything sufficiently. I also didn't have my heatspreaders on. I suppose I'll put those on and see if it helps.

Overclocking on the 2600+ is mixed so far. I have a feeling this chip is gonna own, b/c I did have it in windows semi-stable @ 2380 @ only 1.85v.... at like 55C. That's my problem, heat and voltage right now. Previously I had my 2400+ at 2.08v @ 52C @ 2339mhz.. that was my best aircool. You can see the difference I'm sure.

Anywho, just wanted to let everyone know, yes it IS working, and it's great so far. I have some learning/testing to do on it (and some modifications) but it's gonna pwn I tell you!

Heat-wise, the northbridge got to around 35C after running it for 2 hours or so @ medium load.

Let's see... what else.. the capacitors are pretty close yes in one spot, and I might cut them off and rewire them a bit for when I put the prommy in, but for right now it's not an issue.

And lastly, although this is off topic--when I get my new hard drive replacements, my case is going to be so jam packed full of crap.. seriously gonna be one of the tightest fit (but everything fits) cases I've ever owned.

[sysfailur]

No need for a seperate thread, I'll just keep all my antics in here. This isn't related to the mobo, but it is related to the cpu ON the mobo.

More voltage on both mem and cpu and this beast would be rock solid.

[sysfailur]

Voltage chips posted here: http://xtreme.routehero.com/forums/s...&threadid=4903

[QuadDamage]

i think these are volt controllers we are looking for. now, we need to know which pins and pots...

Sys,

you mean to say your 2600+ is unlocked on this mobo? without paint job? my 2700+ won't post at anything under 13x.

[DJ_Datte]

Becouse , even if all of thease T-Breads come unlocked, this mobo isnt specially prepped, like som KT400, so only the higher multipliers are unlocked.. you could switch down to lower multipliers (5 - 12.5) instead by doing a L5 job

/Damir

[SKATAN]

keep in mind , it´s an ASUS board , cant wait for the epox and abit nforce2

[Emericana]

Sysfailur i have afew questions. first off for temps is it a insocket probe or a ondie temp reader?

second have you confirmed yet that the AGP lock indeed locks the PCI cards as well? i really hope the answer is yes cause my western digital does not like over 195FSB.

finally, have you had a chance to test the board using just one stick of ram so you can tell the difference between normal and dual channel ddr? i ask cause normaly when you run two sticks of ddr you dont get as good of a overclock. is this true w/ dual channel as well?

[QuadDamage]

i wish i could answer your question. i got only one good stick which is XMS3000, the other one is a wanna-be PC2700 from Twinmos which just can't keep up with corsair...

[sysfailur]

Originally posted by QuadDamage
i think these are volt controllers we are looking for. now, we need to know which pins and pots...

Sys,

you mean to say your 2600+ is unlocked on this mobo? without paint job? my 2700+ won't post at anything under 13x.

Well I have used from like 9.5x to 12.5x and those work. I will try the others out today.

[sysfailur]

Originally posted by Emericana
Sysfailur i have afew questions. first off for temps is it a insocket probe or a ondie temp reader?

second have you confirmed yet that the AGP lock indeed locks the PCI cards as well? i really hope the answer is yes cause my western digital does not like over 195FSB.

finally, have you had a chance to test the board using just one stick of ram so you can tell the difference between normal and dual channel ddr? i ask cause normaly when you run two sticks of ddr you dont get as good of a overclock. is this true w/ dual channel as well?

Well I didn't see a insocket probe.. so.. I would assume it's ondie. The agp/pci lock, I was going to test yesterday but hosed the OS. Reinstalling as we speak! And no I haven't, but I will again today if my OS doesn't explode like yesterday.

[Emericana]

Guys i got some bad news. the AGP lock does NOT affect the pci.

read the last few posts in this thread. the ones i made. http://www.aoaforums.com/forum/showt...0&pagenumber=7

it is about the epox 8rga+ and they say it is a physical limitation to the chipset

[Wolf]

:finger: :finger: :finger:

I'm so tired of motherboard manufacturers :banana::banana::banana::banana::banana: footing arond the specs until they get it on the market especially Epox tech at that site . Man that site is annoying as §§§§. So I guess there is no locked pci on the nforce 2. Then is the highest divider available still 1/5th like the kt333. I knew this was too good to be true. They don't want too much of a speed increase. Small increments so they can milk you for every cent.

[Emericana]

I know exactly how you feel man if there was a PCI lock then the Nforce 2 would be the be all endall Socket A board and we would be able to keep it until the Barton maxed out and we are all forced to upgrade to hammer.

i know that is what i was planning to do. here is what i was going to do...

buy a AMD 1800+ tbred on Monday, watercool it to 2050mhz or more if i can.

use that till may or so until the Bartons go down in price cause Hammer is released. Buy the fastest Barton, and keep it for a while.

i would have been able to do all of this on NF2 cause of the great features but no more. the PCI lock was the best feature of the boar IMHO. i dont wanna upgrade to a Maxtor either, which can support crazy high fsbs. my western digital will only do 195fsb but it is such a high quality drive besides that. 100gig w/ 8mb cache mmmmm

[DJ_Datte]

I really wonder, is the Serial Bridge affected by the pci (I know it is) BUT are the serial channels affected by it too, or would a Pata converter allivate those troubles ?