Who's benching classified test a new bios?

[Shammy]

hey guys, anyone benching on a classified these few days, try out this new bios?

759:
IX58S21X.bin
http://www.filefactory.com/file/ah3eh06/n/IX58S21X_bin


760:
IX58S614.bin
http://www.filefactory.com/file/ah3eh18/n/IX58S614_bin


it tries to improve bootup bclk by going into MRC at 180fsb and upping it to xxx, 230 240 etc, see if it gets same bclk as eleeting?

[chispy]

Tx for sharing Sham. My new 760 Should be here tuesday , i will give it a try.

Angelo.

[sofos1990]

Thanks for sharing:D

[Shammy]

nvm try this newer ones instead:

760:
http://www.filefactory.com/file/ah3818b/n/IX58S618_bin

759:
http://www.evga.com/forums/tm.asp?m=100874952

[DrNip]

HAHA I'll give her a go around this weekend.

[Karmakazi]

Peter, any chance you can get the BIOS gods at evga to add Back to Back Cas Delay settings to next bios? Would be a great help for 32M

[3oh6]

Peter, any chance you can get the BIOS gods at evga to add Back to Back Cas Delay settings to next bios? Would be a great help for 32M
quit being a pansy and just run memory high enough where it doesn't matter :D

[Karmakazi]

quit being a pansy and just run memory high enough where it doesn't matter :D

Hah, if I had 1/10th the uber rams you do :up: Kingstons RMA they sent me seems to be dead already... back to RMAland

Want to sell the 920 gem of yours jody?

[3oh6]

sorry boss, same 220 BCLK bootup limit and a whole lot of issues under cold with this BIOS. i had to go back to 21S and all was fine again. not sure why i wasn't seeing any benefits, there clearly was a new step in the POST codes.

[road-runner]

I tried it also and could not get past 222, which bios works best with W3520/W3570?

[3oh6]

21S has been a dream with me for all chips. been on it since it dropped with 920/975/w3540. i think i briefly tried 21T but got no where so stuck with 21S.

[road-runner]

I dont remember what mine came with and did not save it either, where can I find 21S for the 760?

[3oh6]

you should be able to search the EVGA forum archives. although i am not too sure how far back they keep threads. Peter might be able to help with that as well. also, i am on 759...i keep forgetting about you damn 760 kids. there may be a difference between them and the best BIOS.

[Shammy]

okm tomrooe i upload

[chispy]

Classi 760 Bios S618 have some bugs on it Master Sham , During Subsero bench my pciXpress slots were changing from X16 downclocking to X8 and X4
:( , could not bench 3D at all due to very slow 3D :confused: , One card only in first slot and the speed was droping from x16 to x8 to x4 :shrug: GTX 285 , i tried 2 way sli and tri-way sli with same results very slow 3D ,is it a bios bug or a setting im missing boss ? It does not happen on Air only during subsero bench on LN2 ,Also the board does not work with older cards , i tried FX 5900Ultra , FX 5750 , FX 5550 and none of them worked , it showed only a blank screen with a command prompt like little line :confused: , those cards worked fine on RIIE. those issues as i stated were with latest bios from eVGA website S618 , any work around or fix for those issues Sham , appreciate your help very much , thanks in advance.

Angelo.

[chispy]

I think i got a faulty motherboard :( RMA time as the pciXpress slots do not want to run at 16x anymore , no matter what i do what i try , now it is stuck all pciXpress slots at X8
heres a picture with gpu-z and everest to confirm what i am saying ( i tried and flash thru all the bios and the problem still there ) :/

http://img2.imageshack.us/img2/4361/52533895.jpg
By chispy (http://profile.imageshack.us/user/chispy) at 2009-09-06

[Shammy]

here is the 21S equivalent for 760 - 60Z
http://www.filefactory.com/file/ah629h8/n/IX58S60Z_bin

[Shammy]

chispy i gueess one or more of the small smd caps above the pci express slots got damaged. 32 caps for full 16x bandwidth

[slamms]

That can tell about coldboot? E760, whom I tried, is loaded only at -40C. I can something is wrong I do?

[Nazar]

I'll try to interpret what alamms is asking about :D
He is wondering why his CPU has -40* CBB :D

[3oh6]

because most do on the Classified. there are usually CBB holes throughout the temp range that allow for "power down reboots" like when you change uncore or CPU multi. or change CAS Latency. what you need to do is find the holes where your can make "power down reboots". for example, my one W3540 has a hole it can do this at -87C ~ -92C give or take a degree or two. this same processor under my single stage needs to be colder than -37C in order to do a "power down reboot". any warmer up to about -25C will give me FF (CBB). somewhere around -45C ~ -50C gives me CBB again up to somewhere in the -70C range.

as you can see, i have various holes that eliminate CBB, but each chip seems to have different ones. chances are going colder than -40C you will find many holes that allows for "power down reboots". you just have to test to find them. HTH

[chispy]

Thanks Shamino ,I have RMA the board already.

[slamms]

Nazar thanks!!! :D
3oh6 Thanks for explanations )

[miahallen]

FYI - My CBB changes a bit with uncore clocks. Anyone else confirm this?

[3oh6]

FYI - My CBB changes a bit with uncore clocks. Anyone else confirm this?
haven't looked at it from that angle but my first guess would be higher uncore, more VTT, lower CCB? AKA, the chip is hotter :shrug:

but might have to look into that a little bit if i ever have time. definitely interesting.

[miahallen]

Yeah Jody...I'd be running at ~ 5200 bench stable on the CPU...trying to max uncore for 32M...changing multi around for uncore. I was running 183x29 for 5322 on the CPU and x25 on the uncore for 4575. I was running about 1.625V VTT.

At that point I would CBB at about -90...but as soon as I pushed the uncore multi up to x26, I got CBB and had to bring it up to about -80 to get past FF. Then switching to x27 for my final uncore speed of 4956, my CBB hiked to about -65 or so!!! Once I got it booted, I brought temps back down for a successful 32M run...but it was interesting to see the CBB change with uncore speed like that.

I played with VTT between 1.6V and 1.65V, but didn't seem to change my results much, although I don't feel my limited testing is very conclusive.

[EXPANDER]

Not found link?

[kitfit1]

223 here as well. Gone back to 05/06/09 now, as i couldn't get any of my normal clocks stable. With 05/06/09 my normal "benching" boot is 4.8ghz, but strangely enough, with IX58S618_bin i could boot at 5ghz, go figure.

[MaadDaawg]

Been finding this thread very interesting :up: Being single stage challenged, I haven't given a lot of thought to cold bugs but this conversation has given me pause to think that I maybe should be .. if that made any sense ???

On my phase rig, the ECP always reads FF, even though the BIOS is set to display the cpu temp. On my gamer, which is wc the ECP displays the cpu temp like it's supposed to.

I was thinking this was the effect of the cold on the board but never equated it to a cold bug cause the rig still runs. Now I'm gonna have to play around with the uncore some and see if I can get any changes.

Am I reading this right.. or am I totally off the wall ??

[3oh6]

Am I reading this right.. or am I totally off the wall ??
FF and the system not getting POST is cold bug, FF and a running system is A-OK. the reason you get FF under cold is because the board can't display negative temperatures on the LED so it just shows FF, same as if you turned the feature to show temp on the LED off :up:

[Shammy]

how bout this new one, i tested bclk bootup better
http://www.evga.com/forums/tt.asp?forumid=68

[MaadDaawg]

Thanks 3oh6 .. that was my first thought but then reading the thread thought it might be more.. good to know :) tks again

[hiwa]

http://forum.nexthardware.com/gallery/images/7934/4_screen001.jpg

http://forum.nexthardware.com/gallery/images/7934/2_screen004.jpg


:up: high bclock stable @ air :D Great work sham ;) 235 easy





Frequency Control
CPU Clock Ratio (20X + turbo on)
CPU Host Frequency (Mhz) ( 233 )direct boot
Target CPU Frequency 4183 MHz
Target Memory Frequency 1860
MCH Strap ( 1333 )
CPU Uncore Frequency (Mhz) ( 2 X memory )
CPU Clock Skew ( 0 ps )
Spread Spectrum ( Disabled )
PCIE Frequency (Mhz) ( 101 )
OC Recovery ( Disabled )

Memory Feature
Memory SPD ( Standard )
Memory Control Setting ( Enabled )
Memory Frequency ( 1333 MHz / 2:8 )
Channel Interleave Setting ( 4 Way )
Rank Interleave Setting ( 4 Way )
Memory Low Gap ( Auto )
tCL Setting ( 7 )
tRCD Setting ( 7 )
tRP Setting ( 6 )
tRAS Setting ( 20 )
tRFC Setting ( 60 )
tRRD Setting (4)
RTL'S Settings (53/55/57)
Command Rate ( 1t )


Voltage Control
EVGA VDroop Control ( With VDroop )
CPU VCore ( 1.34 )
CPU VTT Voltage ( +175 )
CPU PLL VCore ( 1.80 )
IOH PLL VCore (1.25)
QPI PLL VCore (1.150)
DIMM Voltage ( 1.65 )
DIMM DQ Vref (-10mV)
IOH VCore ( 1.20 )
IOH/ICH I/O Voltage (auto)
ICH VCore (auto)
**VTT PWM Frequency ( 250 KHZ )
**CPU PWM Frequency ( 940 KHZ )
**CPU Impedance ( Auto)
**QPI Signal Compensation ( Auto)


CPU Feature
Intel SpeedStep ( Disabled )
Turbo Mode Function ( Enabled )
CxE Function ( Enabled )
Execute Disable Bit ( Disabled )
Virtualization Technology ( Disabled )
*****Logical Processor Setting*****
Intel HT Technology (Enabled)
Active Processor Cores (All)
*****QPI Controller Setting*****
QPI Control Settings (Enabled)
QPI Link Fast Mode (Enabled)
QPI Frequency Selection ( 4.800 GT/s )

[miahallen]

I'll try to test this out soon ;)

[Karmakazi]

testsed this bios out last night and it seems pretty stable! 230x23 bclk @ 106mhz pci-e on W3540 real easy.

[Romdominance]

Looking forward to trying this bios on my incoming 760 tomorrow,...I have had great luck with the 21s to this point on the 759 thus far. Thanks all, great info, great thread.:up: