**Official Abit Fatal1ty FP-IN9 SLI NF650i Owners Thread**
- Thread starter AFK_Matrix
- Start date
So do we have any more results on the overclocking front guys? I am stuck at the moment deciding between this and the asus one (not to mention the MSI & Gigabyte boards comming soon). At the moment I am swaying towards the Asus one as the bios seems to be pretty good and I should get 3.5+ with an 6600. Only thing I would probably have to do on the asus is the Vdroop mod. The abit isn't giving me very much confidence, specially as there is not even any support on the Abit site!!!
SO at the moment its looking like its the Asus.
SO at the moment its looking like its the Asus.
Caporegime
OC ing is fine but not pushed any mega FSB's. OC'ed to 3.0Ghz but got 2 random freezes within HL2 which i have attributed to memory Vdimm.
There is still VDROOP with this board i.e you set 1.375 in bios and its actually 1.336 in CPUz idle, which drops to 1.328 under load.
it could just be my PSU (400w JEantech), and the fact that im connecting a 4 pin adapter into the specified 8 pin socket in the mobo for CPU voltage...
There is still VDROOP with this board i.e you set 1.375 in bios and its actually 1.336 in CPUz idle, which drops to 1.328 under load.
it could just be my PSU (400w JEantech), and the fact that im connecting a 4 pin adapter into the specified 8 pin socket in the mobo for CPU voltage...
i am new to overclocking but i can tell you i had no problems with this board and took chip upto 3.2 and 3.3MHz but i am concerned about the cooling as i am using the stock cooler.
Wanting to order a new cooler today but am worried about whats going to fit ok.
Wanting to order a new cooler today but am worried about whats going to fit ok.
Caporegime
Chipper39 said:
hi there what NB Voltage did you use? and also did you use LINK and SYNCH modes?
thanks
"what NB Voltage did you use?"
i have everything standard at the mo (3000MHZ) if i remember i upped the NB 1 or 2 steps when i had 3.3 and using unlinked mode
btw this is on an E6600 with GeIL 2GB (2x1GB) PC6400C4 800MHz Ultra Low Latency DDR2 Dual Channel Kit (GX22GB6400UDC) @2.1v
i have everything standard at the mo (3000MHZ) if i remember i upped the NB 1 or 2 steps when i had 3.3 and using unlinked mode
btw this is on an E6600 with GeIL 2GB (2x1GB) PC6400C4 800MHz Ultra Low Latency DDR2 Dual Channel Kit (GX22GB6400UDC) @2.1v
Caporegime
cool, the Arcticfreezer pro 7 works okay, but you will need a knife/similar to get it to fit!
also, using abitEQ what is your NB Temp idle and under load? im sat @ 47~ idle and 51~ load!
also, using abitEQ what is your NB Temp idle and under load? im sat @ 47~ idle and 51~ load!
Was just wondering if this Abit board has Vdroop issues, would the Vdroop pencil mod on the Asus thread work for the Abit board?
Soldato
AFK_Matrix said:
Why not just buy the ASUS as, now there is a Vdroop mod for that I don't see how the Abit is any better?
Some more info having spent some of the weekend overclocking the FP-IN9
CPU Voltage
1.325 1.350 1.375 1.400 1.425 1.450 1.475 1.500 1.525 1.550 1.575 1.600 1.625 1.650 1.675 1.700
Memory voltages
1.80 1.85 1.90 1.95 2.00 2.05 2.10 2.20 2.30 2.40 2.50
DDR2 REF Voltage
-4.0% -2.0% 0.0% +2.0%
Northbridge voltages
1.20 1.24 1.28 1.31 1.35 1.43 1.50 1.58
CPU VTT Voltage
1.20 1.24 1.28 1.31 1.35 1.39 1.43 1.46 1.50 1.58 1.65
Overclocking with E4300
The FSB seems to be the limiting factor. I was pushing more and more volts through the core without any real gain in OCing so I backed down the multiplier to 6x. The same FSB limit seems to be in place.
9 x 1466Mhz - 4 hours orthos stopped manually without errors
9x 1500Mhz - 1 hour orthos, stopped without error reboot / post issues
9 x 1511Mhz -Boot SuperPI 1M @ 17 sec. random reboot / post issues
9 x 1520Mhz - Boot SuperPI 1M @ 17 sec. random reboot / post issues
9 x 1550Mhz - Posted once, restart before XP then wouldn't post - BIOS CLR req
9 x 1600Mhz - Posted a few times, only gets to XP ~ 1in 3. Almost instant crash in XP wouldn't post - BIOS CLR req
6 x 1466Mhz - Boot SuperPI 1M no issues
6 x 1500Mhz - Boot SuperPI 1M reboot / post issues
6 x 1511Mhz - Boot SuperPI 1M reboot / post issues
6 x 1512Mhz - Boot SuperPI 1M reboot / post issues
6 x 1550Mhz - Posted a couple of times restart before XP then wouldn't post - BIOS CLR req
6 x 1600Mhz - Posted a couple of times restart before XP then wouldn't post - BIOS CLR req
6 x 1800Mhz - No Post - BIOS CLR req
6 x 2000Mhz - No Post - BIOS CLR req
Voltage on the core makes little difference. 1.425v in the BIOS (~1.386 cpuz load) gets me to 3.2Ghz. 1.500v in the BIOS (~1.460 cpuz load) gets me to 3.3Ghz.
After that voltage makes little difference, the system us
BIOS issues
There seems to be an issue with how the BIOS settings are applied.
If you apply all the changes at once following a BIOS CLR then save/exit the system will not repost. - no video.
This happens quite often especially following an unsuccessfully post where it recovers some of the settings. It lets you into the BIOS and then will not save, exit and repost.
Pressing the reset key after a failed post often brings you up an older saved set of parameters but then you no matter what changes you apply, no post.
This has been refered to by other OCUK posters wherby a system worked fine one day, wouldn't restart the next and then refused to post at any of the previously working OC's.
The solution that works for me is to change the LDT multiplier between the values of 2x, 3x and 4x. The values do not seem to correlate to any CPU settings and using higher/lower values does not have a specific effect. It's almost as if changing the LDT multi forces a reset in some otherwise problematic system bus. However if you are in this situation, changing the value 90% + allows the next restart to post. Note, you have to have altered the original values that prevented a successful post in order that this will work. It won't force a post where unrealistic values have been applied.
It also seems you need to set voltages, Mode (linked unlinked) etc, then restart before increasing the FSB. This seems the most robust method of getting a good OC without ending up in the posting battle.
I am concerned that there seems to be a wall around 380 FSB with anything much above 366 Mhz being difficult to stabilise even when the resulting CPU clock is relatively low. This is even with max volts on the NB and a cooling fan propped against it. Perhaps this is characteristic of the native 800FSB in the E4300 or it may be something a BIOS update will resolve.
Perhaps this is due to the CPU VTT voltage which I haven't touched as yet?
Has anyone else got > 380FSB with this board, even at the lower multipliers. If you have a 6 series then please report in your results.
Regards
AD
CPU Voltage
1.325 1.350 1.375 1.400 1.425 1.450 1.475 1.500 1.525 1.550 1.575 1.600 1.625 1.650 1.675 1.700
Memory voltages
1.80 1.85 1.90 1.95 2.00 2.05 2.10 2.20 2.30 2.40 2.50
DDR2 REF Voltage
-4.0% -2.0% 0.0% +2.0%
Northbridge voltages
1.20 1.24 1.28 1.31 1.35 1.43 1.50 1.58
CPU VTT Voltage
1.20 1.24 1.28 1.31 1.35 1.39 1.43 1.46 1.50 1.58 1.65
Overclocking with E4300
The FSB seems to be the limiting factor. I was pushing more and more volts through the core without any real gain in OCing so I backed down the multiplier to 6x. The same FSB limit seems to be in place.
9 x 1466Mhz - 4 hours orthos stopped manually without errors
9x 1500Mhz - 1 hour orthos, stopped without error reboot / post issues
9 x 1511Mhz -Boot SuperPI 1M @ 17 sec. random reboot / post issues
9 x 1520Mhz - Boot SuperPI 1M @ 17 sec. random reboot / post issues
9 x 1550Mhz - Posted once, restart before XP then wouldn't post - BIOS CLR req
9 x 1600Mhz - Posted a few times, only gets to XP ~ 1in 3. Almost instant crash in XP wouldn't post - BIOS CLR req
6 x 1466Mhz - Boot SuperPI 1M no issues
6 x 1500Mhz - Boot SuperPI 1M reboot / post issues
6 x 1511Mhz - Boot SuperPI 1M reboot / post issues
6 x 1512Mhz - Boot SuperPI 1M reboot / post issues
6 x 1550Mhz - Posted a couple of times restart before XP then wouldn't post - BIOS CLR req
6 x 1600Mhz - Posted a couple of times restart before XP then wouldn't post - BIOS CLR req
6 x 1800Mhz - No Post - BIOS CLR req
6 x 2000Mhz - No Post - BIOS CLR req
Voltage on the core makes little difference. 1.425v in the BIOS (~1.386 cpuz load) gets me to 3.2Ghz. 1.500v in the BIOS (~1.460 cpuz load) gets me to 3.3Ghz.
After that voltage makes little difference, the system us
BIOS issues
There seems to be an issue with how the BIOS settings are applied.
If you apply all the changes at once following a BIOS CLR then save/exit the system will not repost. - no video.
This happens quite often especially following an unsuccessfully post where it recovers some of the settings. It lets you into the BIOS and then will not save, exit and repost.
Pressing the reset key after a failed post often brings you up an older saved set of parameters but then you no matter what changes you apply, no post.
This has been refered to by other OCUK posters wherby a system worked fine one day, wouldn't restart the next and then refused to post at any of the previously working OC's.
The solution that works for me is to change the LDT multiplier between the values of 2x, 3x and 4x. The values do not seem to correlate to any CPU settings and using higher/lower values does not have a specific effect. It's almost as if changing the LDT multi forces a reset in some otherwise problematic system bus. However if you are in this situation, changing the value 90% + allows the next restart to post. Note, you have to have altered the original values that prevented a successful post in order that this will work. It won't force a post where unrealistic values have been applied.
It also seems you need to set voltages, Mode (linked unlinked) etc, then restart before increasing the FSB. This seems the most robust method of getting a good OC without ending up in the posting battle.
I am concerned that there seems to be a wall around 380 FSB with anything much above 366 Mhz being difficult to stabilise even when the resulting CPU clock is relatively low. This is even with max volts on the NB and a cooling fan propped against it. Perhaps this is characteristic of the native 800FSB in the E4300 or it may be something a BIOS update will resolve.
Perhaps this is due to the CPU VTT voltage which I haven't touched as yet?
Has anyone else got > 380FSB with this board, even at the lower multipliers. If you have a 6 series then please report in your results.
Regards
AD
Soldato
decto said:
I think your problem lies here - the P5N-E SLi has a higher BIOS setting for almost 1.8V on the Northbridge and many of the high FSB clocks have needed all that voltage, which is why so many of us are on water cooling for the Northbridge.
Caporegime
Decto
just got home and had a play.
upped my FSB to 375mhz (x8 multi = 3.0Ghz)
Memtested my ram @ 750Mhz (test 5+6 which previously showed up errors) @ 3-3-3-10.
Booted up fine
and my system froze twice within 20 secs of starting SP2004. (2nd time adjusted LDT to 4x. No Cigar win froze asap under SP2004)
just been back into Bios and upped NB to 1.4 and CPU VTT to 1.39 and it has just passed
Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 1791 Time: 15
CPU: 3000MHz FSB: 374MHz [375MHz x 8.0 est.]
CPU: 3000MHz FSB: 374MHz [375MHz x 8.0 est.]
05/03/2007 18:30 Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
So so far so good @ this mhz and voltage.
edit; just put LDT back up to 5 from 4.
this resulted in a reduction of SuperPi time by 0.4s to 19.1s @ the spec above.
just got home and had a play.
upped my FSB to 375mhz (x8 multi = 3.0Ghz)
Memtested my ram @ 750Mhz (test 5+6 which previously showed up errors) @ 3-3-3-10.
Booted up fine
and my system froze twice within 20 secs of starting SP2004. (2nd time adjusted LDT to 4x. No Cigar win froze asap under SP2004)
just been back into Bios and upped NB to 1.4 and CPU VTT to 1.39 and it has just passed
Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 1791 Time: 15
CPU: 3000MHz FSB: 374MHz [375MHz x 8.0 est.]
CPU: 3000MHz FSB: 374MHz [375MHz x 8.0 est.]
05/03/2007 18:30 Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
So so far so good @ this mhz and voltage.
edit; just put LDT back up to 5 from 4.
this resulted in a reduction of SuperPi time by 0.4s to 19.1s @ the spec above.
Last edited:
Soldato
decto said:
So you don't want to see this then;
