Separate names with a comma.
Discussion in 'Motherboards' started by qbackin, Aug 20, 2005.
If you know you have good cooling and the HS feels cool to lukewarm then take 10c off, on average.
We need a new BIOS countdown party.
That made it work LOL... though winflash raped my bios (waiting for a new one now...F*CK!!!!)
Boy do I miss the ASUS UPDATE utility!!!! ->(it never let me down!)
You know I've been using Winflash (Award version) on boards for many years without issue, but posts like the above still make me nervous as hell that one day the same thing will happen to me when I least expect it.
Still, I refuse to install a floppy in my machine for the express purpose of BIOS flashing. Screw that.
And I'm almost as steadfast in my refusal to do any sort of BIOS / firmware flashing in a DOS environment. It's friggin 2006 already and we shouldn't be forced to do this stuff in DOS.
isn't the new bios coming tonight? (sun night) any word on if its coming soon? Its 7pm....
Mario posted an update tonight - check it out: http://www.evga.com/community/messageboard/topic.asp?TOPIC_ID=11391&whichpage=7
Think I'm gonna let others try this out a while before I go ahead and flash it in
Need some help and or advice as just recently I began experiencing BSOD's and reboots
My current rig is as follow:
opteron 165 at 1.5v, cpu 2529 mhz (9x280fsb) under a Mach 1 phase cooler
NF41 EVGA motherboard with the latest bios (evga says new one tommorow as we all know)
2x1gb Mushkin DDR 400 black series 3-4-3-8 1.6v with the ram set for 229.87 mhz (divider)
WD 74gb Raptor hd
EVGA 7800gt video card
PC&C 510asl power supply
Winx XP professional with all updates and SP2
I have installed the hotfix for dual cores
Ok now for the problem. Using the SP2004 program that replaced Prime95 I have been stress testing my computer. When ever I run 2 instances of SP2004 and set one for cpu1 and the other for cpu0 and attempt to run the Large in Place FFTs -stress Ram test for both cpus within a minute either cpu1 or cpu2 will error out.
If however I run just 1 FFTs test on either cpu1 or cpu0 it will run for hours on end no problems. I have ran each cpu independtly for 3 hours in the Large FFTs test mode and had zero problems or error outs. Also if I start one instance and wait a couple minutes and then start another using the other cpu both tests have run succesfully for up to 2 hours (where I ended it)
I have also ran the blend cpu/ram test overnight and in the morning both instances where still green and running strong.
I have used Mem86 in via a floppy boot and let test #5 run for 44 cycles, then let all tests run overnight (9 hours) and received no errors.
So I have narrowed down the problem (I think) to the following scenario. Somehow in windows xp cpu1 and cpu0 both attempt to use some of the mushkin memory at the same time or attempt to start at nearly the same time. In SP2004 this results in the RED screen and halt on which ever cpu failed to gain what it was seeking. In the real enviroment it results in a BSOD or reboot.
Has anyone some feedback, advice or tips to offer? I can not say to myself that the memory is bad since it can survive mem86 and sp2004 independent tests. But why does it error out when one tries to run both cpus to stress the memory? Could the memory be bad in that it can not handle dual requests or is this a motherboard issue?
*UPDATE* stopped using SP2004 and tried using 2 instances of prime 95 but read below*
Ok I finally got fed up and formated the hard drive.
Reinstalled windows xp pro, then all the updates. Got the amd dual core driver but skipped the hot fix
put fear and galactic civ2 back on and gamed for an hour, no problem
then went out on the web and BAM, got a reboot
this occured 3 times so by that point i was fricking pissed
i yanked the mushkin out and put a 512mb stick of twinmos pc3200 onboard and let it run overnight. I have asked my wife to use the computer to surf all over and to let my son play some of his veggie tale's online games to see if they can get it to crash
I was going to test the mushkin in my old epox 8RdA+ board but the case that I stuck that board in wont allow the memory to fit, the heatspreaders hit a metal bracket
When I get home today if the main evga system has not crashed or rebooted then I will pull the epox board out far enough to bench test that mushkin on it
Maybe this EVGA N41 is just unstable, it was semi free but if free means constant headaches no thanks. If its the mushkin then I will rma it newegg and try another batch
Are you running the memory @ 1T OR 2T. You may want to change it back and forth and test it.
Also where is the new bios at?
There was an issue with newegg sending out refurb boards. It's possible you got one of those. If that's the case I would recommend RMA through evga as they are aware of the problem and will ensure you get a brand new board. There is a post somewhere in the 30 pages of this thread where I read that if you want to hunt it down for more info.
You have to e-mail Mario and he'll send it to you. From EVGA forum member reports there are no new RAM voltage options and no new memory tweaking options. Idle Cycle Limit was increased from 16 to 256 (like in DFI boards).
Playing with the new bios now as I too post over there at evga
going to try to get take it back to stock speeds, at 2t and game and surf
if it doesnt crash i will bump upwards
if you want the beta bios let me know, we are spreading it for those guru's to test it hard core style
Is there a changelog as to what this bios has that the previous ones dont?
Check out this topic for good info on the bios beta....
Can you guys post what you are able to get for mhz and whether 1T works or not on UCCC memory. Thanks.
Well it appears that I am in the same boat with others suffering random odd reboots with the EVGA NF-41 motherboard.
I leave my computer running 24/7, and its a sealed phase cooler. I have learned enough to avoid the dangerous "ohhh look water" pain.
I have checked the seal around the cpu and there is no leak. Since the cpu therefore is at sub zero temp's heat is not a factor
I am using a top of the line PC&P 510 asl psu. Bob over there is a TOP NOTCH customer rep and technician who even sent me another unit when I incorrectly suspected that was the culprit. Power supply via my multimeter is rock solid therefore PS is not a factor
I can run memtest86 with my mushkin ddr4000 black series all night long at stock and overclocked to 250mhz and not a single problem is reported. I will go ahead and pull the memory tonight and give it a go on the old epox 8rda+ board for a couple nights and see if it reboots over there. If not, memory is no longer a factor.
Which leave me with either the CPU or the mainboard. If it where the cpu would it not reboot while in DOS under memtest, or just refuse to boot at all? Instead I get reboots while trying to surf the web of all things!
I suspect this beta bios which if you notice on boot up "evaluation bios, not for sale" indicates to me that it indeed is a placebo and not much has changed. I would love to see an honest show of hands how many others out there have experienced random reboots or BSOD's at stock speeds with the EVGA nf-41?
It seems that if I underclock my ram, drop it down to 166 or 133 then the system is happy. But I like others paid for good quality ram that is rated for 250mhz. I have done the time tested metheod of isolating each component to see what is the max each can provide safely. THEN I even backed it down to stock speeds and still got reboots and BSOD's???
One last try for me tonight, I am going to shut her down, double check everything and bench the mushkin on the epox. If it lasts for a couple days then I will be convinced its the motherboard. Since I got it semi-free from newegg along with the 7800gt (which to my knowledge has been error free) there will be no means of rma'ing it short of going through here at EVGA.
I can only hope that evga at least will still send my rebate check to help ease the cost of having to get a new board elsewhere.
Wow, all this for my cheesy little board? I guess I'm one of the lucky ones... I can run my 2x1gb ocz plat at 250MHz, 3-4-4-7 1T almost stable. It works fine when multi'd down at 2-3-2-5. I'm running this with a 3700+, firstname.lastname@example.org, stock cooling for now, but a ninja is in the mail.
Don't take this the wrong way man, but there's no such thing as ALMOST stable. It either is stable or it is not. Sounds like in your case the latter would apply.
LOL - Good point.
Oh btw... I'm running stable Crucial Value at 3-3-3-8 1T 200mhz. Yeah. That's stock. (except for the 1T)
I haven't updated the bios on this thing but I was wondering if this board supports optertons(i think thats how its spelled) On the front of the box it just has athlon 64 logos. Wondering because I am looking into getting an opty 165 x2. Another quick question, is optys as good for gaming as athlon 64s?
Yes, Opterons are supported. See my sig.
I haven't had time to read the whole thread but I was wondering if you guys could help me. I'm probably making some stupid error.
I just got this motherboard and I'm already having problems with it. I can't even get Windows XP to install; on my old motherboard, whenever I wanted to boot from CD I would push any key and it would go, but with this motherboard it automatically boots for me (it doesn't even tell me to push a key). I follow the installation process and Windows wants to reboot and go into the main installation screen. It finishes, and the next thing that happens is it reboots and boots from CD automatically again, right back to where I started from. And yes, CD-ROM is set to boot after the HD's (which happen to be Maxtors). This prevents me from installing Windows at all because I am put in an endless loop of installing.
Is there something I'm missing in the BIOS or what? Thanks for your help.
If Windows finishes installing shouldn't you remove the CD so it doesn't boot from it?
I think I ran into this on my first install, though I am certain that during start-up it asked if I wanted to boot using CD. Once the initial install portion was done and the system restarted, I thought I had to boot from the CD so I selected it. Nope. I'm retarded.
Not sure if you're being asked this, but if you are, don't select it. If you aren't asked and it's doing it anywaay, then remove the CD, let it boot into whatever XP shell it alrerady installed, and then reinsert the CD when needed to finish the install.
It only finishes the pre-install or whatever (blue screen, able to format drives, etc...), and then reboots. I tried taking the CD out to let it boot to what was already installed and then I get a "SYSTEM DISK FAILURE" error and can't do anything but put the CD in and hit enter and it all starts over again. I'm not able to not select Boot from CD. It just goes and boots anyway.
have you contacted EVGA support?
Got a replacement BIOS and now it won't post and just gives a FF code and doesn't cycle I have tried clearing CMOS a lot and removing the battery. Nothing seems to work WTF?
Hmmm, well FF is the post code it gives when the BIOS has booted succesfully
what is your system again?
Everything in my sig except I have the NF-41 mobo instead of the Asus and a EVGA 7800GT CO instead of the 6800GT
Upon power on FF is displayed and nothing happens
With my Epox nForce2 board, there was a saying that FF either meant Fully Functional or Fully F***ed. While FF is normal "running" status, if your board immediately goes to FF as soon as you power on, something is probably wrong.
I can't really say what, but when my Epox did that, it was a power supply issue. The FF thing could be any failure condition I think, but it's an indication the board isn't even getting to the memory self-test, pre-POST stuff.
Did we forget about this thread?
Was eVGA ever going to post the new BIOS one their site to download?
Yeah. What's the deal now that the updated BIOS is out there? RAM Nirvana? I've had one of these laying around here, unused, for months now, waiting for the many issues to be addressed/improved w/a mature BIOS. Are we there yet?
i think if you went on the evga forums and left your e-mail then someone would send you the new BIOS through e-mail. I've got it, but havent had time to flash to it yet.
That was a beta bios wasnt it?
When is that bios going official?
I think they have had enough time to work the bugs from the beta one out.
IOW, still not ready for prime time.
I can't change my vcore off automatic without the computer failing to boot and me having to reset the bios. Is there some bios setting or jumper setting I'm missing? Do I have to turn off overvolt or something?
The beta BIOS didn't accomplish anything apparently (going by the thread @ eVGA post beta "release")..Did nothing for me, but I'm not running one of the more affected IC's, ie: UCCC. I guess it's a good thing that nothing happened, rather than previously stable settings stopped working. Other than the failed attempt @ the 1T issue, there was nothing in it that's been requested - ie: more RAM volt steppings, temp bug, more memory timing options, etc.
There's still been *0* in thread posts by any techs re: updates - only (email) replies (from Mario) to a certain forum member over there. I've all but given up hope that anything will happen (soon), if it happens so be it - but if not it's ok I guess..I have had a pretty good experience overall with my board. I'm more concerned with them at least making an *attempt* at fixing some of the bugs.. Even though they're pretty new to the mobo game, they're marketing it to an enthusiast crowd, and simply ignoring some of them (the bugs/requests) kills the (good) word of mouth that they would otherwise get.
/me goes back to his corner in hopes Mario is actually working on the BIOS, and not occupied totally by a possible GT/GT KO release soon.
/me is also stuck in StepUp queue for a 7900GT CO (154 -> 146 since the 3rd )
RAM nirvana BIOS would be good. I just got my 2 gigs of G.Skill PC4000 (the stuff with the black heatspreader). It's running fine right now @ 260 and advertised timings, but unfortunately in single channel only. I can't get the board to POST if either stick is in either of the two slots closest to the CPU. Which makes dual channel kinda tough to accomplish.
I'm not done playing around with settings yet, but I'm not thrilled either, considering I had memory issues with my older stuff and now again with this new stuff.
Well shit. I think my motherboard has fried, whole computer just shut down in the middle of a game Thursday night and I couldn't power the system up again. Tried all kinds of things, and I don't have enough spare hardware to narrow it down 100%, but it's looking like either the mobo or the BIOS, I either get FF on the LED right away or it flashes to FF for a second and turns off, won't even POST.
I thought it might have been the stupid metal backplate on the Ninja shorting out pins (since the EVGA board doesn't have its pins shaved down and it'd be pretty easy to get them touching that backplate if it was tightened down enough). But I pulled that backplate off and stuck the yellow plastic EVGA one back on and it's doing the same thing.
So I guess my options are:
A) try to get a new BIOS chip from EVGA
B) try to RMA the whole board (though I got it free as part of the combo deal, and I don't know if they'd take it back after having the new backplate on)
C) say screw it and buy a new/better board
I can't get Wake-On-Lan to work for me. I have enabled Wake By PCI/PME in the BIOS and also enabled Wake-On-Lan under Device Properties for the network card. I have used two different methods for testing, sending the packet via MC-WOL and from my router, both of which work for my other PCs. Any ideas?