is there pci lock on nforce2??

[Shane5578]

is there pci lock on nforce2??

pci lock=no hdd corruption at any fsb

[Royal Oaks]

Yes :)

[Shane5578]

i was reading the sticky in the amd thread, and it says there wasn't one.

oh why do we still get hdd corruption on nforce2, doesn't seem to happen on p4's

[felix88]

there is a PCI and AGP lock.

the hard drive corruption could be caused by a couple of things. some people have found that better RAM fixes it.

[Shane5578]

is the pci lock better on the p4's or something,

i mean when you stress out the ram on p4's, it still doesn't corrupt hdds.

ram just doesn't cause hdd corruptions.

i'm totally confused..

the pci is fixed right, been measured at 33mhz even at very high fsb. but still get corruptions

oh why does epox tech avoid questions about their pci locks on epox nforce2. i mean really you get banned just by asking it.

[felix88]

i'm only stating what i know. some people have replaced the ram on their Nforce2 boards, and the hard drive corruption problems went away. i guess it could also be caused by a lack of voltage to the chipset.

if you want to confirm the PCI lock yourself, buy one of these:

http://www.pcmods.com/details.asp?ProdID=261

otherwise, if you have a problem with the Nforce2 chipset and AMD chips, stick with Intel.

[Shane5578]

so memory and stressed chipset can make OS's become permanently unbootable. mm

[DR. YT]

the nforce boards have an individual lock for the pci and the agp.thus the agp is ajustible without the pci being bothered. some say higher agp speed =better performance some say it has no effect i my self can run it upto 85mhz without a hickup weather it had any positive effects i realy didnt notice:)

[Shane5578]

thats interesting too, no notable performance with higher agp on nforce2. mmm

[DR. YT]

well there could be i for one just didnt notice:)

[Shane5578]

ok, did some research and it does have pci lock (new bios upto 250fsb), and higher agp does increase performance.

[Royal Oaks]

Originally posted by Shane5578
is the pci lock better on the p4's or something,

i mean when you stress out the ram on p4's, it still doesn't corrupt hdds.


HD corruption can be caused by other things. Also, you will notice that a PCI lock is better on AMD based boards, for the reason that 99% of Intel procs are totally locked. When you unlock you run a higher FSB, thus creating a much higher PCI speed if you are running a board with a 1/5 or 1/6 divider. So basically a PCI lock isn't that big of a deal on P4 rigs, just because you dont see many P4's running over 200FSB ;) PCI lock is awesome, cuz you can just run as high a FSB as possible with no worries.

Also, about the AGP thing, I would imagine it would increase performance if ever so slightly. AGP doesn't really play a big role in performance from 4x to 8x, so it would only make sense that you wouldn't see a noticable increase from OC'ing your AGP. I bet all the 3DMARK nutsos are probably big into it though. :D

[lyran]

where is this PCI lock in the A7N8X bios? I can't find it? is it the same as "FSB Spread Spectrum"?

[glock19owner]

Originally posted by lyran
where is this PCI lock in the A7N8X bios? I can't find it? is it the same as "FSB Spread Spectrum"?

No it is not the same as the FSB Spread Spectrum option. The AGP lock is automatic on the NF2 boards.

[KnightElite]

Originally posted by lyran
where is this PCI lock in the A7N8X bios? I can't find it? is it the same as "FSB Spread Spectrum"?

You can't adjust the PCI lock, it is always locked at 33MHz. You can adjust the AGP speed though, at least on some of the boards.

[Shane5578]

i have my p4 on 200fsb with raid 0 (since i have an ES chip), and i had no probs with hdd corruption at all.

so can i have an nforce2 at 200fsb with raid 0 too?

i thought with pci lock, there would be no hdd corruption at all.

[MudWhistle]

Shane loop memtest86 test 5 over and over, I noticed if you get any errors with test 5 it will lead to hdd corruption eventually. Yes even on an Intel i845e with locked pci at 33mhz, it will just take much longer to show up. The problem is more pronounced with dual ddr setups, the corruption you are getting is most likely due to buffered/cached problems between the memory controller / cpu / memory itself. I dont think locking the PCI controller speed matters in this situation. It's more a problem with buffered pages sitting in memory being corrupted then writing to the hdd thus leading to instability.

Great thing about memetst86 is that it exposes weakness with memory controllers, not just the DDR memory itself.

[Greyfire]

Corruption doesn't necessarily come from higher PCI frequencies. It can come from whenever it gets sent the wrong bits. I'd bet that the NF SB is what causes corruption. When it's stressed it gets quite hot.