OCing 2.4C on MSI 875P Neo FIS2R/mini-review

[alcky]

well after playing with my old 2.4B B0 stepping and managing a 162FSB*, I finally got my 2.4C in yesterday

*previous thread

it will POST at 3.54GHz/295FSB but no Windows with Vcore @ 1.7v, POST and get into Windows at 3.42GHz/285FSB on default Vcore but it's unstable. I finally found the 'sweet spot' for now, 3.3GHz/275FSB with Vcore @ 1.6v

I got this mobo firstly because of all the features it has, some initial reviesw showing it as a good overclocker and finally the way it looks After the first week it was available at various retailers and people got it up and running horror stories started cropping up in various forums, anything from not able to OC, all the way to exploding south bridge with pics for proof . I was starting to get worried since my order had gone through, but I was commited to making the best of this as all the other 875 mobos didn't have as many features and were more expensive [a LOT more expensive if you wanted one of the models with comparable features as the Neo].

Likes:
a)full feature set
b)overclocking options
c)bundle pack [software, S-ATA cables and power adapters for all 4 connections, brackets for 1394 and USB 2 ports, etc]
d)looks
e)price

Now some issues:
a) the Vcore fluctuates about .25v +/-, that's why I had to set it to 1.6 in BIOS to ensure stability when in Windows, if left to default setting it would even dip to about 1.5v, now stays in the 1.56-1.61v range.
b) the board does not like CAS 2, very unstable at anything other than DDR400 setting, even though my CorsairXMS is rated for CAS 2 @ DDR434, after much benchmarking and testing out different timings I settled on the ones in the screenshot, even though they're 'horrible' [I'm used to 2-2-2-5], I can't complain when the mem bandwidth still comes in at 5648MB/s, I guess I will wait for some DDR500 [GeiL, where are ya?] so I can at least use a higher bus/mem ratio and get more bandwidth through raw MHz.

By now you're all probably thinking, "is he ever gonna show us the benchmarks?" Well, here you go, I'll let them tell you the rest of the story.

testing system:
MSI Neo 875P FIS2R
P4 2.4C SL6WF/L310
Alpha PAL 8942 with 80mm Tornado
2x512MB CorsairXMS 3500C2
ATI Radeon 9700pro
SB Audigy
WinTV

 

[jdmcnudgent]

looks nice. the week 10 i traded, did 290 on default.

 

[alcky]

looks nice. the week 10 i traded, did 290 on default.

yeah I'm thinking if I didn't have these voltage fluctuations I could maintain the 285FSB on default stable, but gotta live with this MSI or maybe test the PSU, it's an Antec true430 but I have it completely maxed* tried disconnecting 2 drives but no luck... at least I seem to be one of the few lucky ones with virtually 0 problems.

I'm thinking more along the lines of too many people expecting massive OCing right off the bat and toasting their components, it took me about 5 days of playing around [3 with B proc and yesterday and today with C proc] to find the right settings


*various components powered by 430w PSU

DVD drive
CD-RW drive
floppy drive
main HD
backup HD
2 drives for RAID array [disconnected for now]
Radeon 9700pro
4 antec 80mm fans [2LED ones]
2 80mm tornados
TV card
audigy

 

[jdmcnudgent]

you may need a better psu. i have modded both of my enermax psu's and they are now VERY STRONG.: for some reason, my 12 volt line is low, but it reads very high in the bios, something in the mobo.

 

[micamica1217]

you may need a better psu. i have modded both of my enermax psu's and they are now VERY STRONG.: for some reason, my 12 volt line is low, but it reads very high in the bios, something in the mobo.

it's not his psu...I have it and it can be strong, real stong.

what you and alcky are seeing are poor voltage regulators of the mobo...

my asus p4pe was the same way, now all my rails are alot better.

mica

 

[jdmcnudgent]

it's not his psu...I have it and it can be strong, real stong.

what you and alcky are seeing are poor voltage regulators of the mobo...

my asus p4pe was the same way, now all my rails are alot better.

mica

youll know better when you get a c chip and see what spot you hit the wall at with the psu.

 

[micamica1217]

youll know better when you get a c chip and see what spot you hit the wall at with the psu.

and just why do I need a "c" chip?

are you saying that a 2.4c on an asus p4pe would have great voltage rails? it didn't at stock speed with my old 2.4b.

are you saying that you have great rails with both mobos listed above at stock...and what about the asus p4c800?
I know you could tell us about bignuts rig....

mica

 

[jdmcnudgent]

and just why do I need a "c" chip?

are you saying that a 2.4c on an asus p4pe would have great voltage rails? it didn't at stock speed with my old 2.4b.

are you saying that you have great rails with both mobos listed above at stock...and what about the asus p4c800?
I know you could tell us about bignuts rig....

mica

im saying that im curisous how far you would get with a 2.4c on your ic7.

 

[dustybyrd]

i seriously doubt that the antec true 430 can't power his system and a lot more, even overclocked....

i have more power needs than that with my dualie (see sig) and i have a fortron 350...which is probably about as good as an antec 430

 

[theotherphil]

alcky: Try the following mem timings -

CAS Latency - 2.5
Act to precharge time (tras) - 7
DRAM RAS# to CAS# delay - 3
DRAM RAS# precharge - 2

The Corsair should have no trouble running that at your fsb. I am running those timings at 285fsb and get ~6GB/s with the XMS3200LL.

 

[FIZZ3]

and just why do I need a "c" chip?

are you saying that a 2.4c on an asus p4pe would have great voltage rails? it didn't at stock speed with my old 2.4b.

are you saying that you have great rails with both mobos listed above at stock...and what about the asus p4c800?
I know you could tell us about bignuts rig....

mica

I think Mica has a point- the voltage readings many people rely on are those reported by BIOS or programs like MBM5. There is no telling what kind of variance these readings carry.
Aside from that issue, there is also no default cause of any deviant reading; it may be the PS, but it may also be mainboard circuitry that causes it.
Modding a PS may mean the actual voltages are more toasty that you'd like, or it may help nothing at all because the true problem lies elsewhere...

The better way of investigating this is to use your own equipment and measure the lines separately... not many people can or want to do this though.

 

[micamica1217]

im saying that im curisous how far you would get with a 2.4c on your ic7.

no....your original quote was....

"youll know better when you get a c chip and see what spot you hit the wall at with the psu."

your original conversation was on his powersuply...and how he may need a stronger one.
and how your 12v line may be low.

my 12v line is 11.86-11.92, folding right now, and typing.
this is with a non modded true 430w.

my old asus P4PE would show far less at idle...nothing changed inside my case. the same drives, fans, light, and so on.
my other rails sucked too with the P4PE.

yet because on my new abit IC7 I have a better voltage regutator, I now can run at a higer FSB stable.

mica