680i SLI is too expensive? Asus as the solution : P5N32-E SLI PLUS

[SLINROB]

it has to do with the dividers...

1/2, 2/3, 3/4, 4/5, 5/6, 6/7, 7/8, 8/9, 9/10, 10/11, 11/12, 12/13, 13/14 and a bunch more dividers exist.

so in order to maintain those dividers, the numbers will round down to the nearest divider until you go above the next one.

let's say you set the FSB at 400 and the MEM at 437

then you will be on the 11/12 divider (as will 438, 439)

you will remain on the 11/12 until you pass MEM over 440

then you will be on the 10/11 divider

you will remain on the 10/11 until you pass MEM over 444

then you will be on the 9/10 divider

etc etc etc

just mess around with it... you'll see what i mean...

it pretty much rounds DOWN to the nearest divider

I see you also purchased an open box p5b deluxe wifi off of newegg.com.


I could not believe the low price I paid for that thing and was really shocked when I opened the box to find that all the accessories were in place!

Most importantly the board is running flawlessly! Good riddance 650I plus!

[matthewrsx]

it has to do with the dividers...

That sort of makes sense, but I thought the point of UNlinking the memory was that it didn't go through the dividers.

So what is the point of unlinking the memory if it is going to be linked to a divider anyway?

[shimmishim]

That sort of makes sense, but I thought the point of UNlinking the memory was that it didn't go through the dividers.

So what is the point of unlinking the memory if it is going to be linked to a divider anyway?

it is still unlinked in a sense... seeing as it that with the tradition 965/975 you were limited to 1:1, 2:3, 4:5.

with the 650/680 you now have all these other options.

i think you're right that it's not a TRUE unlinked but it is in the sense that you have like 30+ divider options. i mean what's the difference between 440 and 444?

[matthewrsx]

it is still unlinked in a sense... seeing as it that with the tradition 965/975 you were limited to 1:1, 2:3, 4:5.

with the 650/680 you now have all these other options.

i think you're right that it's not a TRUE unlinked but it is in the sense that you have like 30+ divider options. i mean what's the difference between 440 and 444?

OK, I'll give that a test tonight. You're right in that there is no real difference between 440 and 444, but I'm probably overly meticulous when testing an overclock. I like to only change ONE thing at a time to see what effect it has (ie change the FSB but not the memory speed)...otherwise it's not really a "scientific" test if you know what I mean.

[MrRuckus]

Bah. It seems the 0702 really didn't do all that much for me. I ended up having to back off to something like 3.45 in order to stay 100% stable (which was about the same with the 0602). I have an older Antec 550Watt that I think is mostly to blame. Its an older PWS that requires a 20 to 24 pin and a 4 to 8 pin connector. I ran the board with just the 4 pin for a bit and then with the 8 pin and didn't see a different (saw the question posted a few pages back).

I think once this Antec is replaced with something a little meatier I may be able to get something higher stable. I say this because I can run memtest just fine upto about 3.8Ghz. Speedstepping is turned off in the Bios (heard this can allow memtest to work upto and past 4Ghz). While 3.8 works with just memtest, windows wont accept anything over 3.6 and even 3.6 is very unstable. Most voltages are @ 1.55. I have adequate cooling on everything including a Tuniq Tower 120, 60mm high cfm papst over the NB (good god that gets hot!), a 40MM on each of the Heatpipes and a few 120MM's in the case. Side of the case is off 99% of the time.

3.6 is stable if only at the desktop. But as soon as I try anything past Word or IE, I get reboots. I have a 8800GTS 640MB for video. Once that kicks on the 3D the reboots commence. All is well in windows without 3D or Orthos running @ 3.6.

So, next purchase will probably be a new PWS. I think this Antec needs some rest... But I am pleased with the C2D. Huge leap compared to the P4D I had before it. Even with it @ 4Ghz.

This motherboard is the hottest motherboard I have ever dealt with though. NB gets stupid hot with no active cooling. I dont know how the board survives in cases with no active cooling on it.

[Leeghoofd]

What steppping is your cpu ? I have here an 630F and 744G that both hit max 3400 too or I need wicked voltages and even madder temperatures to get anything higher, tried them on my Striker and P5K both same max speed issues

[The Viper]

Anyone else try this new bios yet? Any other thoughts, a new bios is my only hope for this board, Ive finally decided its simply not worth it to purchase different cooling for the NB.

[SLINROB]

Anyone else try this new bios yet? Any other thoughts, a new bios is my only hope for this board, Ive finally decided its simply not worth it to purchase different cooling for the NB.

At first I thought that new bios had fixed a lot of problems only to have been let down in the end when I discovered that they returned with the addition of a few more issues


Im back to a single 8800gtx and a P5b deluxe board and couldn't be happier. Intel Chipsets from here on out for me.

[hedge]

I don't think this has been asked yet, but I might have missed it in the 30 pages of this thread, but has anyone had a problem with windows not loading when you set your memory speed to less than you actual bus speed. (IE bus speed 450mhz (times it by 4 to get fsb) and ram is set to 400mhz (800mhz ddr)?

As long as I run my memory as the same speeds as the bus then I am fine, but my memory is starting to hit its limit, but I don't think my cpu has yet. Currently, these are my settings:

FSB - Memory Clock Mode: Unlinked
FSB (QDR), Mhz: 1800hz (450mhz)
MEM (DDR), Mhz: 900mhz (450mhz)

[Radical_53]

No. Unlinked memory has worked fine for me so far at various combinations. There are still some holes, but it works out quite nicely.

I'm still not through with this strange issue of instability I encountered. One day the board runs 18h+ Prime, the next day it'll freeze after 3-4 minutes. Without any changes! No temp issues (none that I can measure anywhere), no voltage issues (I don't see any fluctuations, psu can do much more than it has to do right now). I really don't get it.

[SLINROB]

I'm still not through with this strange issue of instability I encountered. One day the board runs 18h+ Prime, the next day it'll freeze after 3-4 minutes.

Thats the exact reason that I got rid of the POS.

[Leeghoofd]

Welcome to the world of Nvidia chipsets... there's nothing more to say , it can deliver but sometimes it craps up , reboots and is fine again, why only the Green team engineers may now...

[Radical_53]

@slinrob: Well, at least I had the chance to see it run stable once So that makes me curious to find a way how to make it work like that all the time

It will do at something low like 3.2GHz stock voltage, but with higher clocks it starts to play the clown...

[SLINROB]

It will do at something low like 3.2GHz stock voltage, but with higher clocks it starts to play the clown...

Yeah thats pretty much my experiences with it. Its not like its a bad board or anything as its actually pretty solid in build and as long as you dont expect a high rock stable OC like you can get with a P965 chipset then you will not be to disappointed with it.


Bottom line if SLI is a must then this is the board for you as im pretty sure its still a better choice than the 680I's in regards to some of the issues that those have.

[hedge]

No. Unlinked memory has worked fine for me so far at various combinations. There are still some holes, but it works out quite nicely.

Okay, it seems I was just running into dividers that didn't want to work at certain clock speeds. I am now going to keep plugging away at this thing and see what more I can get from it.

[Radical_53]

@slinrob: I don't expect clocks like the 965, but I'd expect something at least of the same level as the 975.
And that's where my problem start. For example, if I raise the voltages accordingly to reach FSB 400 "stable" to run some benchmarks, the same voltages are enough to run the same benchmarks at FSB 435 or 440. No problem so far.
But if I try to get my CPU to higher clocks, it starts bugging around after some time. The "after some time" is my problem as I can't seem to find anything temperature related or related to my PSU.

[r0n1n]

I was able to get my board perfectly stable at 475 x 6, for board max OC purposes
Not too shabby, but my E6420 would't follow all the way up in 8 x mode...

500 is just too much asked.
24/7 will be 3.2 though, 8 x 400.
am currently dropping voltages and checking till what point it remains stable: am at 1.37 vcore, 1.50 NB, 1.50 SB, 1.25 1.2 vtt and 2.2 vmem.

then see what we can squeeze out of the mem

cheers

[Leeghoofd]

Radical are you sure it's also not CPU related ? Many peeps run into probs due to a bad batch and apoor clockern too high temp on the cores or too high voltage needed for the CPU to stabilize... don't solely blame the baord as there are many culprits possible....

On my Striker 475 is the max ( CPU doesn't like it, same on my P5K Dlx ), 450 is smooth as a rock....I never reached 500FSB not even with Multi 6x and I thought intially it was the board, but placing the same cpu in the P5K DLx reveals the same stuff so it must be the CPU hitting an FSB wall....

[Wingdings]

To those of you running this board with 4 x 1 GB RAM modules - how is it? Have you encountered any problems, and if so, have you managed to solve them? How is overclocking with 4x1 GB?

[Radical_53]

@leeghoofd: Well, I guess not. I bought the CPU pre-tested and I'm running it with lower clocks and slightly higher voltage (as the board seems to need more voltage to run the CPU stable).
I can run prime some days and some days it bails out after a few minutes. Just yesterday I was able to run it until today, no errors, but then just rebooted after like half an hour of gaming.
I raised board voltages, I lowered voltages, nothing really seems to give me a clear direction where I'm supposed to go, what I'm supposed to do to make it fully stable.
I'm trying to run the stability tests with as little software as possible to keep that out of the equation, still I don't get anywhere close to a perfectly stable system.

[Leeghoofd]

Can you share your settings plz with us to see what you are doing to go from the current FSB to 400 eg with your pretested CPU

AI Tuning: Manual
Nvidia GPU Ex: Disabled
LinkBoost: Disabled

FSB - Memory Clock Mode: Unlinked
x FSB - Memory Raio: N/A
FSB (QDR), Mhz: (means FSB)
Actual FSB (QDR), Mhz:
MEM (DDR), Mhz: (means mhz speeds)
Actual MEM (DDR), Mhz:

Vcore Voltage:
Memory Voltage:
1.2v HT Voltage:
NB Core Voltage:
SB Core Voltage:
CPU VTT Voltage:
DDRII Controller Ref Voltage:
DDRII Channel A Ref Voltage:
DDRII Channel B Ref Voltage:

tCL (CAS Latency):
tRCD:
tRP:
tRAS:
Command Per Clock (CMD): clock (T)

Advance Memory Settings
tRRD: AUTO
tRC: AUTO
tWR: AUTO
tWTR: AUTO
tREF: AUTO
tRD: AUTO
tRFC: AUTO
Async Latency: AUTO

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
MCP PCIE Spread Spectrum: Disabled
SATA Spread Spectrum: Disabled
LDT Spread Spectrum: Disabled

CPU Internal Thermal Control: Disabled
Limit CPUID MaxVal: Disabled
Enhanced C1 (C1E): Disabled
Execute Disable Bit: Disabled
Virtualization Technology: Disabled
Enhanced Intel Speedstep Tech: Disabled
LDT Frequency: 5x

PCIEX16_1 Frequency (Mhz): 100
PCIEX16_2 Frequency (Mhz): 100
PCIEX16_3 Frequency (Mhz): 100
SPP<->MCP Ref Clock, Mhz: 200

[Jultsu]

Ah yes, i got the system running again after a VERY LONG time. I'm benching 9x400 with 0602 bios and 1,472 volts to see if anything's changed. When i get my OCZ PC2-9200 FlexXLC Edition -sticks [ 4 of them ] I'll try with the same settings. Should i go for the 0702 bios?

[Radical_53]

@leeghoofd: I'll write those settings down, no problem.

At least, I think I might have found something during the last days of testing. The Linkboost feature, or at least the raised frequencies it causes, seem to give me some trouble over the time. I disabled that now (had it enabled before as it gave me a slight performance increase).
The other thing is 1.2V HT and CPU VTT. Those are some strange voltages. The system freezes when they are too low, but it'd freeze also when they are set too high. Somewhere in between there's this narrow path that gives stability, and it seems to me that this is different for the varying combinations of FSB, CPU clocks and Vcore.

[Leeghoofd]

Indeed radical 1.4v at 1.2 HT is the most stable for me, some peeps even put 1.5 or more there but my rig just locks up on the boot screen, can't even enter the bios at anything higher then 1.5... CPU VTT I keep as low as possible... till 400FSB 1.45 works fine but if you are looking for higher FSB 1.5 or even 1.55 might be required, EVA2000 aslo foudn this settigns is very critical and dependant on CPU to CPU as it tends to stabilize or destabilize easily your CPU...

Due to this difficult setup many peeps ditch the 680i and surely the ASUS variants as being bad boards, The reference boards from EVGA, BFG, XFX,... and co have done a better job with their biosses, as auto setings in fact allow the user to get great results... we have to do it manually and some peeps think it's only about upping voltages...

[Radical_53]

Exactly, same experiences here. Voltages need to be raised to a certain point to get stability, but too much will make it worse from a certain point.
Definitely not easy to find in the first place
Still, I'm happy now that I found *something* that might do the trick. I'm crawling upwards again slowly, just priming 3.8GHz right now
I'm a little sad that I can't get the Linkboost thingy to work anymore, but that seems to be a smaller improvement anyhow compared to CPU and RAM clock increases.