Nubius

Just did my first IHS removal on my opty 165. Was nervous as could be sweatin balls as I was doing it. I wound up taking apart a simple cheap single blade bic razor. The REALLY cheap ones....ya know the orange ones that come 100 in a bag for .50 cents or so lol. Anywho, that blade was REALLY thin and flimsy however it turned out to work extremely well. It sliced through the silicon REALLY easily, but since it was so thin it also cut into my finger and that's not even the sharpened end. Finger isn't bleeding, just lots of loose skin from it diggin in at various places. Successful first top removed. I swear my temperature monitor is off but this is what it read: Before IHS removal: 38C idle - 54/56C Full load After IHS removal: 38C idle(lame) - 44/46C full load So it's approx 10C off at full load, but nothing at idle. I didn't tighten my block down anywhere close to how tight I did it with the IHS on but I don't know how much more I want to tighten it. Temps are definitely better than before according to these readings, but still....this is on water cooling so the temps seem a little whacked. I did see someone else with the CCBWE stepping that was saying theirs was running hot so that could be it as well I suppose. Thank you Angry_Games for that video tutorial as I watched it many many times before attempting. Opty 165 CCBWE 0551WPMW Seems to be a rather mediocre clocker =/ 1.55 yeilded 2.7-2.75 but I haven't tried going further with the IHS off yet since I just completed doing that about 15 mins ago.

Thanks. For some reason I thought you were out of country....didn't realize you were in so cal like myself

Where can I buy a spare BIOS chip for the Lanparty NF4 Ultra-D? I don't immediately need one, but I'd like to have one just incase. I screwed up my old Abit NF7-S v2 way back in the day and was able to find and purchase a BIOS chip and hot flash and got it working and all, and when I went back to that same website, they were no longer around =/

Can I ask why? I simply want to know out of curiosity as I hadnt heard this before.

Little update here: It was definitely the RAM being the issue of all this. I tried each stick seperately in the top slot...upon simply changing the cas latency from 2 to 2.5 all hell broke loose...on one stick...it wouldn't load windows, instead it'd give me a message to the effect of "Windows could not load due to a software error. Could not load DLLs for HAL" ok...that's a new error for me....upon changing it back to Cas latency 2 it boots normally.....when I try the other stick, upon changing it to cas latency 2.5 it simply restarts before it gets to the windows loading screen. Beforehand I couldn't get 2.3GHz stable no matter what combinations of voltages I tried..I put the RAM back to how the board defaults it too, and low and behold I got 2.3GHz stable, 2.4GHz stable, and now have been running prime95 at 2.5GHz for approx 4+ hours now, opening photoshop, checking mail, running sony vegas, etc..etc... so it seems to be trucking along fine...I've got a memory divider at 150 right now so its resulting in like...178MHz speed, which is fine by me, because I've already been in contact with the dude from PDPSys and he will be issueing me an RMA tomorrow, so right now I'm simply testing the CPU. So through tons of meticulous testing, suggestions from Rgone and the fact that he happened to have a similar experience which I still can't believe, I've managed to finally get the CPU running nice and stable, PSU rails don't drop, and I've found the cause of my problems....so everythings going much smoother now, so lets just hope that the sticks they send back to me will be kind to me and overclock like good samsung tccd brainpower pcb modules they should. I wish I hadn't flashed to the 623-3 from the 'magical' 704-2bta now....I did that back before I realized the RAM was to blame, but oh well...everythings running smooth right now, so I guess if it aint broke dont fix it. Thanks again RGone for your insight, I believe from here, assuming I get good modules back, it should be relatively smooth sailing

ok that's good and all, and I understand that, but when it comes to changing the cas latency from 2 to 2.5 or 3 at STOCK settings and not putting it through a stress test then that wouldn't qualify as 'under load' would it? I have a multimeter and I know exactly how I need to test it, I simply haven't gotten around to testing it out yet to see what infact it is drooping down to, but like I said that doesn't explain the inability to change cas latency at stock settings one iota. I've got a friend running a 3200+ venice, 1.5gb of RAM, and the 6800GT AGP that I just sold to him running on a 17a system...hell he's even got the thing overclocked to Ultra specs and the PSU is some unknown brand...and although I've lost the sticker now, the sticker that came on the 24pin power slot of the DFI nf4 board said to the effect of a 3500+ and 2 6800GT's in SLI required 20-22a and considering I'm running only 1 I figured it'd be ample...but I already came to the conclusion I won't truly know until I test it with the multimeter myself.... I really can't afford to swap PSU's at this point anyway as the decent enermax one, and basically any other PSU that has nice strong +12v rails are approx $100....but at the very least my RAM should be able to change cas latency and if Rgones had the same problem before in the past than that shouldn't concern the PSU at all EDIT: Aright I quickly tested it right now using my multimeter. It runs at 12.08 when not under load, and when under load (running prime95 large in place FFTs) it droops down to a whole 12.01...............sooooooo now what?

My drives are on Sata 1&2 however, before I even got this board I looked into the SATA ports and everything I read said that all 4 ports are locked...it's only on the Nf3 chipsets inwhich 1&2 are locked and you've gotta use 3&4 Rgone - This is the response I got back from my manufacturer...it wasn't an RMA request I just told them of my issue: So it seems they will be fully willing to accept it...just gotta do like he says and test each module. Also my RAM HAS to be acting up because I was using 2-3-3-7-1T timings with 2.8v and I was getting BSODs at 220MHz...manufacturer specs it at 2-3-3-6-1T at 2.8 at 233MHz, so somethings definitely up. well at the very least you'll get some fresh sticks Took me all damn day to get to this thread, couldn't access the page....there were a few pages I couldn't access though thanks to a huge power outtage in LA

I care because mine only works at 2 and thats it...I wouldn't care if it'd work at CL3 but it doesn't....and now it's not wanting to even hold 220MHz within manufacturer specs so I definitely gotta get these RMA'd....it seems they are definitely the source of the problem as I'm running 220MHz with the 180 divider right now keeping em at PC3200 and it's running fine while I'm watching shows, sending e-mail, downloading via bittorrent, typing this and running prime95 large in place FFT's Only problem is my tubing is in the way of the RAM slots...primarily the upper one, so it may be an 'interesting' ordeal to say the least trying to get the RAM in and out

hmm that's crazy....was it on this board that you had those problems with as well or did it do that on various boards, because this is the first time I've come across this problem with these sticks. I know people had pointed out my PSU, but it hadn't been acting like PSU related problems yet being that it was certain settings...especially now that I narrowed it down to simply being the cas latency that's causing all this...I suppose with around 2.9v or so I could get it up to 240MHz+ with 2-3-3-7-1T timings but I'd rather have 3-3-3-8-1T timings and hope for 250MHz+ Hmm, man that's a trip though that you have a pair that wont do anything beyond 2CL but another pair work fine....BTW, may I ask what you've gotten the 'good' pair up to? And I'm assuming that you'd suggest I RMA these back to the manufacturer for some fresh sticks? What exactly would I tell them, that the sticks simply won't work with anything but CL2 which doesn't allow me to use them up to the DDR533 which is under manufacturer specs? Would that seem like a valid reason to RMA do ya think? Thanks for your help and the fact that you said you've had a similar problem is a great insight. For a second there I thought I had found the solution because I adjusted it to CL3 after changing the trc from 7 to 12 (I had seen it at 12 on a previous board, and the trfc was 14...and in the DFI Ram timings section it said generally the trc is 2 below the trfc) so after changing that, and the twr to 3 from 2, an DRAM Drive strength to level 1, it booted into windows...got to the desktop (this is at CL3) ...thought I had done it....tried to open CPU-Z and bam! BSOD =/ it's just odd that these symptoms have never showed themselves until I put them in this board, only the CL affecting the system, thats an odd one EDIT: I contacted tech support and relayed my problem to verify that it could be RMA'able but beyond that I also wanted to verify if they are infact still using the TCCD chips and not TCC5s...either way I should be hearing from them in a couple days....soooo now I just need to snag my 512mb of kingston valueRAM from my friend so I will have that to run in the machine if it comes time to RMA em

Ok I think I've somewhat narrowed it down....for WHATEVER reason...I'm not allowed to change my cas latency. I tried each number one by one to see if one specifically caused issues. Put the trp to 3 from 2, booted up just fine, changed the tras to 7 from 5, booted up just fine, trcd to 3 (which is where this board defaults it to) boots just fine....changing the Cas latency to 2.5 leads to automatic reboot before windows loading screen even shows up, changing it to 3 shows the windows loading screen but quickly BSOD's... I'm running 2-3-3-7-1T right now at 210MHz with 2.8v just fine....according to the manufacturer it can run 233MHz at 2-3-3-6-1T, but at 250MHz is where it needs to be 2.5-3-3-7-1T with 2.9v....well....being that for whatever reason this doesn't like me changing the cas latency I don't know whether to consider myself SOL or what. Any clues as to why changing the CL of all things would cause such issues? It absolutely makes no sense to me

Hmm....ok....I followed the 'official DFI cmos clearing procedure' and now this is what I dont understand. I've set everything like you said too from the vcore, to the LTD to the chipset voltage...I even raised the HTT just now to 205 with no problems...but the thing that gets me...is that it seems that the only time that the computer will just randomly reboot right before the windows screen is when I adjust RAM timings =/ I don't understand why it'll reboot simply adjusting timings from what the board defaults it to which is 2-3-2-5-1T, to then changing it to 2.5-3-3-7-1T timings...that alone without raising the HTT or anything will cause it to reboot right before the windows loading screen. However, if I leave those options to 'Auto' in which like I said it'll default to 2-3-2-5-1T it boots into windows no problem =/ Why would loosening the timings cause windows to reboot right before the loading screen, but actually putting MORE strain through the chips at tighter timings it loads just fine...I'm kind of at a loss for that one.

The 6.66 are indeed better...thanks again for that (for whatever weird reason though, after uninstalling previous chipset drivers windows wouldn't boot due to pci.sys missing, but I repaired windows and all is good now)

Oh yeah quick question. As far as the "CPU VID Startup" is concerned....am I supposed to set that relative to how much the Vcore is, or is it ok to leave it at 'Startup' option or what?

Thanks Rgone....I wasn't trying to bash the board in anyway as I KNOW it's basically the leader in terms of stability, overclockability, and the options it has...I'm sure the way I worded some things made it seem like I was pissed with the board, but it was simply because it was late last night and I was frusterated with the things I had tried thus far. Believe me the first conclusion I came to was that I was doing something wrong from me not being knowledgable enough with this board and the options available...such things like I mentioned the LTD and Chipset voltages...I had no clue what were considered safe options for those so I'm glad you pointed me in the right direction for that. Thanks again

Ok thats fine, you don't have to take it as if I'm personally attacking you or the board...I simply said that so I could get an explination as to why it does this which you did manage to provide. I didn't follow the 'OFFICIAL DFI WAY OF CLEARING CMOS' like the way you showed, I followed this: after reset, enter bios -> load optimized defaults->y->save->after POST finishes -> shutdown ->switch off PSU -> unplug -> wait for yellow lights to turn off -> press power a few times-> set clear CMOS jumper -> wait for 10 minutes->set jumper back -> plug in ->switch on -> power up -> enter bios -> load optimised defaults -> now configure your bios. which was provided on the xtremesystems page that has a list of a lot of Ultra-D BIOS's This kind of information is good to know, as by default the chipset is 1.5 and the LTD is set to 1.2v so I didn't know what would be a good range for those to be at, however now I know. As for the testing of the RAM, can I ask is this needed simply to make sure the modules work correctly, or that the modules work correctly in those particular slots? Because I had tested these ram sticks under more strenious conditions, however not on this board (please no need to reittirate 'that wasnt this board' because I understand that) the memory drive and dim strengths are a couple things I have barely read into and don't even fully understand yet so that doesn't help me any (meaning my lack of understanding of those doesn't help me, not you telling me, so dont take it as such) Like I said though, I NEVER once had a boot failure, it simply would either blue screen before even getting to the screen where windows very first starts to load and has the blue scrolling bar, or it'd hanging right where it's supposed to say "Verifying DMI Pool, backing up cmos.....ok!" but before it'd say that it'd hang, then reset. With the nf3 board, I got the ram up to 270MHz 2.5-3-3-7-2T 3v and it ran for quite a while, and then all of a sudden it didn't want to do that anymore....I only ever tried the stock BIOS for the nf3 ultra-D however as all the other BIOS's seemed to add to problems, or fix some but create other problems, so I was leary as to whether or not I wanted to even flash it. Either way I'll take your suggestions and need be post back in a couple days as I figure it'll take that long to do the appropriate amount of testing you've suggested as well as implementing some of the settings you've suggested. Thank you all of you for your help, very much appreciated.