Neither. I use spiflash with a USB dos boot flash drive (I forget what the /Z switch does). Then after flash complete, I power off, turn off power supply, and press the clear cmos button.
Printable View
Neither. I use spiflash with a USB dos boot flash drive (I forget what the /Z switch does). Then after flash complete, I power off, turn off power supply, and press the clear cmos button.
Woke up this morning to my PC rebooting over and over Until resetting the CMOS everything has been running fine until today seems to be a cold boot Issue? any suggestions signature rig with the beta bios F3C JZ
Go back to F2.
Chip needs more vcore would be my guess
yes all C-states are off no sleeping happened from cold boot/ PLL1 PLL setting I did how ever hit power /restart button as soon as it cycled threw the restart, Ya I know it will restart twice after "failed OC" but it was different then that recycle at 1st that was my thinking. Its just odd after 3 days same OC it started acting up this morning I played a epic gaming session of 7 hrs with no problems I did get it to boot after resetting CMOS
which made me to believe it was a bad OC Im going to try the same OC settings over again and see what happens just a little frustrating as I have been w/ AMD&Asus for so many years and its all new and I don't understand Intel like I would like to. Im 110% happy the jump I made to Intel and Gigabyte SB & the black Gigabyte )thank god no ugly blue boards:up: MB made the jump a no brain er
I also thought of that but why would it change after 3 days without problems
4.8ghz@1.38v@ 54c gaming / benching I may have to go back to the F2 bios has anyone had issues with the F3C JZ bios? i know its beta for a reason. just wondering if others had issues with it... Is f2 more stable?
Thanks guys for all your suggestions and help.
*edit* added some pics of the bios settings if you don't mind look threw them and see, anything im doing wrong
http://img155.imageshack.us/g/dscn0129b.jpg/
Why do you have Turboboost technology enabled? (with per core options opened up?) No need for that. Go ahead and set the TDP/TDC (can't hurt anything anyway) to 200/200 for the power limits, then disable that turboboost option since you're using a fixed multiplier for all the cores. Not saying that caused your problem, but no need to have that enabled.
Or did it enable itself after a BIOS reset?
Not happy with your CPU though :( ....your default VID is pretty high...But if it's at 1.385v BIOS at 4.8 ghz, that's probably around 1.345v load, which isn't bad...maybe it can do 5 ghz at 1.450v BIOS -->1.410v load? What is the "load" vcore that the CPU reports at 4.8 ghz at full load? (you can use hwinfo32 to get that while running prime). Can it do 5 ghz?
5294.89 MHz (101.82 * 52) http://valid.canardpc.com/show_oc.php?id=1739610 1.45v LLC2 not rock solid yet maybe more tweaking still unsure of the bios a day or 2 hope it have it stable temps was getting higher then I would like for LinX
Nice.
Try to use LLC1 instead of LLC2, because the board overvolts from the sensors, so that 1.45v might be as much as 0.03v higher than the sensor reading; you may need to use a DMM. It's unclear if the overshoot of LLC2+the overshoot of DMM vs sensor reading was contributing to my and C_N's original chips degrading or if that was irrelevent or not, but since I'm OCD, I'd rather not find out...
I just did a 45 minute prime test (I hate priming) at 4500 mhz and 1.236v load, 1.265v BIOS LLC1 (the same as I did two weeks ago) and the chip hasn't degraded yet, so that's a VERY good sign. Since my first chip degraded at BOTH 5 ghz and 4.5 ghz, though that was on the B2 board with the overshooting LLC1) I assume that if its the same stability at 4.5 ghz, then it should be the same at 5 ghz too. (1.236v load is fine at 4.5 ghz HT on. 1.224v BSOD's after awhile).
So then I went back in BIOS and set some profiles for 4 ghz HT on, 4 ghz HT off, 4.5 ghz HT on, off and 5 ghz HT on/off, which should be fully game stable.
http://img684.imageshack.us/img684/8071/4500mhz.png
I cant seem to get passed 2 seconds worth of 32m pi @ 2133 with hyper ram :( just cuts power out. Any thoughts?
I had the same thing until I lowered the vdimm, my sticks just wouldn’t take the voltage to run tight or properly & the vdimm/vtt needed to be spot on one click higher or lower resulted in a hang. I can get them to complete 32M super slack 2T but what’s the point in that.
gotcha so I need better hyper to be able to run it :)
wont waste my time then I guess they dont run that speed in MIVE or p8p67 deluxe either.
8 8 8 28 2t runs 01 at 2133 but thats it
BBSE till I can find some good hypers i guess :)
EDIT: Whats with all the reboots this thing does. 3-4 cold boot tries, then it will fail settings I was just benching at, re load them and it boots fine again and on and on lol very strange
That boot looping is fixed in B3 with beta bios. Unless you press the reset button...
Got the B3 UD5 today. It doesnt seem to like the Redlines I have using f1 and f2 bios's. I am missing my B2 UD7, hope this gets better with updates.
Edit: either my chip got really good all the sudden or cpuz doesnt read voltages correct? http://valid.canardpc.com/show_oc.php?id=1744563
Did you check the many posts about that?
Use HWinfo32.
I get about half way through 32m before it freezes, 1.80vdimm with 2gb of the hyperx khx2000c8 k3 3x1gb kit from my x58. Looking at the times 2133 c7doesnt seem to help though.
Only place I notice a big difference is in 3dm05, which it has no problem running at those clocks, GT1 and GT2 scores awesome with 2133 c7 7 7. My Dominator gt 2000c8 kit wont even boot at 2133 on this ud5.
1st try using a benching station. Thank you Elano for the nice gift. :up:
http://i51.tinypic.com/24zawiw.jpg
On the new B3 versions are the BIOS's blowing up a lot ?
Not even.
Zero problems so far.
May have a problem if I try to enter sleep mode with PLL override enabled, but this board with the latest beta is stable (as long as you don't do that).
I is :D
+1
This has been my story ever since i got the B2 on release day(with gazillion bios changes), January 9th, up until right now with the new B3 mobo. I'm still using the F1 BIOS tho. I've noticed when you lower the memory speeds it seems slightly more stable. Instead of every cold boot, it does it every second cold boot. If you leave memories at 1333 seems to be booting up fine. Falkentyne keeps saying the new F3 beta fixes all this, but i seriously don't believe him - nothing against you brother, just saying i lost faith completely to even bother changing the bios... :D
ps: That is why im using the multipliers for each core to set up my CPU clock, cause when it fails after a cold boot, this way it only reverts the memory back, but it retains the settings for the CPU multipliers + the CPU vcore. SO now when it says overclocking failed, i just ignore it and it re-boots by itself, and than loads Windows with CPU @ 4.9Ghz and it is rock solid - only the memory is reverted to 1333mhz 1.5v. If I set up a constant multiplier from the top of the Overclocking menu and disable "each core multipliers", than when it fails overclocks it reverts the multiplier back to 34x which is not cool and i have to actually go in and set it up again...
Turbo
Why haven't you flashed the F3C beta bios??
I told you at least 3 times to flash it! You're still mad about the reboots yet you refuse to flash the BIOS that cures it...
it STOPS ALL the reboots! ALL of them!
Only thing it doesn't stop is overclock failed when pressing the reset button and sleep mode with PLL override, which no manufacturer has a fix for yet.
Yet you don't believe me....
1) F3C stops reboots when you leave the power turned off.
2) F3C stops reboots when you unplug the power and remove the CPU for awhile.
3) F3C stops reboot loops when you change settings.
4) F3C stops reboot loops for using a PCI X-Fi.
It will still boot loop, though, if you use an unusuable setting, though, as it's supposed to do (like RAM timings too low for stability or 2133 mhz RAM on 1333 mhz rated memory, etc).
The only power off reboot (this isn't a boot loop) it does with F3C, is, when you make certain changes, for example, turning PLL override off when it's already on, the board will power down after you save the BIOS settings, as it needs to do that.
I'm serious...FLASH it. Then flash the backup BIOS by alt F12 after you're done (Just clear the CMOS after flashing the mains)
There's a good reason I'm using it. It *works*.
You can always go back if you don't like it.
Flashing the backup BIOS with the main helps even more.
I apologize Falkentyne, i didn't mean to sound like rude :banana::banana::banana::banana::banana::banana::b anana:. You've put a lot of effort into the whole thing and even tested a bunch of stuff for me. I promise i will put F3C Tuesday next week when i have time to play. Thanks for your time, really appreciate it. ;)
Just make sure you clear the CMOS after flashing.
I think you'll be pleased with it.
Board feels almost as stable as my old 775 board now, but I only have basic 10600 cas 9 ram and standard hard drives... no idea what will happen if I move up to 12800 or ssd's...
After the first boot, after setting the new settings, reboot and hold down alt f12 to copy the main bios to the backup bios.
You should have a fully functional board.
Anyone been able to hit up the B3 UD5 with a DMM to see how much the board overvolts (compared to Eztune or HWinfo32?). It's already been proven the B3 UD4 overvolts...
anyone have any idea why on cold boot why my pc will restart 3 times like a bad OC reset, and start up and have everything show same as it was the day before and with no adjustments. after the 1st time of the day it starts like it should
That was fixed for me with the latest beta bios f3c. Which one are you using?
If you're on f3c, did you flash the backup with alt-f12?
I had no problems with having the power supply unplugged for 3 hours and then powering on the PC. PLL override was disabled, however. Are you trying this with PLL enabled? I didn't do 8 hours (there shouldn't be a difference between 3 and 8).
F3c Bios, as far as the ALT-F12 Im really lost as to what your saying ,I did save the CMOS to a profile if that's what your saying, How ever I did find for some reason my Turbo boost tech was enabled and did disabled it. can I save the alt-f12 when OCed or do I need to set to default to do it? under the adv cpu core feature internal cpu pll over voltage is enabled thanks again for your help.
What he is saying is at post you hit alt and F12 then flash your backup bios to the same one you are currently using that way both are the same.
AMD freak, do you have a PCI sound card ? Cause mine hung today after it passed the memory test, when it was detecting the IDE channels. After i shut it off and boot up, it loaded windows but my Sound Card was nowhere to be found, its like i never plugged it in my PC. When i restart again, it detected new hardware which was my sound card...
You can do it at any setting., though it's probably safest not to have PLL overvoltage enabled.
And I just turned the power off for 3 hours, with PLL enabled and 5 ghz, and unplugged the power supply, and when I turned it back on, it came back up at 5 ghz. It did take awhile to come up, though. Sounded like I heard the video card fan rev up twice, so maybe it initialized the chipset twice. Didn't power cycle, though.
I don't see the difference between 3 hours and 8 hours, with the power unplugged. I don't know :shrug: Maybe it's a setting you have that's making it lose the power? Or maybe I need to actually leave it off 8 hours?...
@Turbodiv:
I had some issues with the B2 board redetecting the soundcard, but that only happened on early bioses. I also remember when it sometimes didn't detect the NIC for awhile (or I needed to reboot). I haven't seen that on this board, though.
You see this image ?:
http://s4.postimage.org/19l4b5a04/F3_C.jpg
This image is 100% Rock solid for 12 hours PRime95 large FFTS, linX with newest libraries + AVX + 64 bit, also passes Super Pi 32M no problems at all. I just updated my MOBO to F3C bios (main and back up, + i did a few CMOS resets to make sure everything is perfect). Im Shutting my computer down right now and im going to work, so ill be back in 9 hours and im pretty sure it will not boot up with these settings...
ps: Forgot to mention PLL overvoltage is disabled and all the C-states option are also disabled. Also Turbo Boost is disabled, its clocked with the main MULTIPLIER, not the turbo multipliers. TDP limits and the others are AUTO.
[QUOTE=Falkentyne;4803465]Just make sure you clear the CMOS after flashing.
QUOTE]
Ive never heard this. Reboot and load optimized defaults after flashing, but not clear cmos.
Nothing but the GPU / ssd is in the white / dvd - Hds in the black
I backed up the bios didn't see any confirm? but at boot hit the alt+f12 and turned off the PLL and see what happens in the morning after several hrs of gaming tonight, its just odd I can reboot several times in a day no problems and boots in less then 12sec and in the mornings it cycles threw 3 times like bad oc settings but everything stays the same as far as settings from the night before. I can even reboot after w7 loads the 1st time in 12 sec Ive ran the f3c bios from 2nd day. its not really a problem but just very odd only after extended hours being off. its very stable never an issue testing or gaming for several hours at a modest 4.5ghz 1.325v
[QUOTE=OC Nub;4804273]You forgot the /quote
Anyway, this is EXACTLY what I was doing after a flash. Taking the advice here and not bothering to clear cmos...
But I stopped doing that, when I noticed my PCIE running at 8x all of a sudden (just happened to randomly check in the CCC and it said it was at 8x...wth)
And the only way to get it at 16x was to set the BCLK over 100.00.
When I cleared the CMOS, guess what?
It was back to x16 at auto and 99.8 again.
I hadn't noticed this for weeks, because I was usually using 100.03 right away. But when trying to troubleshoot some BSOD's while pushing high overclocks, I set it back...and kept it there later, and that's when I noticed.
(this was on a B2 board).
Summary: always clear the CMOS after a flash.
You never know what lingering thing might be left over from before. Might not be apparent to you, but even the slighest issue could cause something odd down the road.
You have an onboard clear cmos...unlike a jumper, it takes just a few seconds to power off the PSU, open the side cover, and push the button :) One reason I paid more for the UD5 than getting a UD4.
AMD and TurboDIV:
Both of you try clearing the CMOS if you have any issues remaining. You can't blame the board or a buggy BIOS if all you're doing is loading optimized defaults without a proper cmos clear. Loading optimized defaults simply loads preset settings. (you can also go in and set stuff manually too).
More about clearing cmos:
Back in the Abit BH6 days, if you flashed a BIOS, and then restarted the computer without clearing the CMOS, the default CPU voltage would be set to the current CPU voltage you had before clearing CMOS, thus giving you another .3v higher range. This allowed people to bypass the 2.3v limit on 2.0v pentium 2's, by setting the BIOS to 2.3v, flashing the same BIOS on top of itself, and rebooting.
Clearing the CMOS would set the default CPU voltage back to 2.0v.
This is something to keep in mind, if you guys ever have more issues than you feel you should be having....
Just remembered something else, but not precisely; been awhile:
On my P5WDH, when I didn't clear the CMOS after flashing, the board reset the settings, but the saved OC profiles were still there. (but wouldn't work properly). The BIOS version did display the new version, though. When I did clear the CMOS after flashing, the profiles were removed. Also the date/time was reset, so then I knew it was starting fresh.
So if you guys have an issue with the board, that should be fixed after a flash, but still persists, unplug PSU and give the clear cmos button some love.
Thats exactly what i did:
Downloaded the bios, updated @BIOS to ver 2.11 from the twektown forum link. Shut off my antivirus and everything else non essential. Updated my bios through @BIOS than it asked to restart by itself. It restarted, I went into BIOS, set the AHCI and let it load windows. Than i immediately shut the PC off, unplugged the PSU and pressed the CMOS_SW for 10 seconds, than i pumped it for another 10 seconds, than i pumped the RESET_SW, and finally pumped the mobo POWER SW. This all took me like 2 minutes. Than i hooked up the PSU, switched the PSU ON and than i booted the system and immediately went into the BIOS and loaded optimised defaults. After it restarted i immediately pressed ALT+F12 and than the mobo asked me to press ENTER to copy the MAIN BIOS to the BACK UP, and it did that succesfully. Than a message came up "POWER OFF OR RESET", I powered off. Than i pressed CMOS_SW, without unplugging the PSU and wihtout shutting it off, as soon i pressed the CMOS_SW the mobo booted up by itself and than i began clocking my system. I forgot some options on the first round of BIOS changes so i rebooted again and set everyrthing up properly and than i made the screenshot post up there and than i powered off my system. WHen i get back from work i hope it will power up properly since i think i pressed that CMOS_SW plenty enough.
Ok, sounds good to me.
Only difference is that I use spiflash with a dos booting usb flash drive.
Hope everything works out.
Im not blaming the board at all Im 110% happy with it even if it restarts 3 times the 1st time of the day everything is solid, if it happens again Ill try flashing the bios again and clearing the cmos I use Qflash with a USB stick and never have had problems it just the way I learned w/ Asus and it pretty simple.
Im sure they will fix more with future bios.
WOW ! This BIOS really has some magic in it(for now). The PC booted up perfectly fine with the screen shot settings from my earlier post. The bios boot sequence is definitely different with he F3C compared to the F1. Now when it boots up i don't even see the initial window where it tells you the CPU speed and BIOS version, as soon as the screen lights up i see the AHCI sequence which usually comes after the IDE detection with the F1(i disabled the Gigabyte logo). So far im pleased ! :D
Didn't I tell you....?
Now I bet you if AMD freak clears his CMOS, which will flush out all of the old leftover code from previous bioses, he'll be in the ball game too....
BTW I always get the initial screen because I have quick boot disabled. Having quick boot enabled just increases the chances of something happening (like a boot loop)....
I saw someone many pages back who said one of his external devices wasn't detected if quickboot was enabled, but was detected if it was disabled.
Now this may have only been with the B2 revisions, but when I had quickboot enabled (on the B2), I never heard the 30 year old "beep" sound from a successful post...I'm used to hearing that sound when I power on a computer...I even thought maybe the speaker jumper had come out...
Oh ok. So there is a quick boot option than ? I guess i didn't touch it, i just disabled the Gigabyte logo thing.
ps: YEs you did tell me, thanks. I just hope it will boot up properly in the next few days also. :)
looks like everything is good so far with 1st boot this AM. I guess it had something to do with the PLL over voltage being enabled. since I never got the confirm on the backup of the bios. I cleared my CMOS when I changed bios what 2 weeks ago when I got the board. I would like to Thank everyone for helping me sort out my n00bness after the switch to Intel while I learn :up:
Now switch your screen name you freak!:D
Quick boot is disabled form the BIOS but i still don't see the first screen with the BIOS version and CPU speed, Gigabyte logo is also disabled. To be honest i don't care as long as it boots up fine. This morning it also booted up fine with no problems.:)
Glad some people are feeling happy with their board now.
Unfortunately board didn't boot today. Gave the old "Overclcok failed screen" again after 2 success boots. Thats ok, i guess my memories are not compatible or my CPU IMC is bad...
ps: I've just updated the mobo to the F3g bios, only the main bios this time, with the same steps as before to see what happens. Also put the memory to the 18.66 multiplier with its default settings. And this time i clocked the CPU through the Turbo Multipliers and set the TDP and Power limits to 200 each.
new GA-P67A-UD5-B3 - F3g bios anyone tried it yet? anything changed
http://forums.tweaktown.com/gigabyte...beta-bios.html
Is there anyone who can take a DMM to a B3 board and measure the real vcore voltage (compared to HWinfo32) coming from the choke or caps, with LLC1/2/standard, like they're doing in the UD7 and UD4 threads? I don't have one...
DMM is very cheap man, like 10 bucks, you should buy one, they are very useful.
I tried to buy one about 6 years ago and a good one with I think it was called trueRMS or something cost $30+ :( Then I went and looked at those $500 ones with oscilloscopes and wanted one of those :(
lol you gotta know how to use an O-Scope before buying one. They are a huge investment. The difference is that a DMM will give you a fixed reading, while a O-Scope will give you a reading over time, usually much more precise.
A DMM is a great way to start, even a cheap one can do pretty much as good as a $30 one, the truth is the more expensive the more features, and you really only need DC voltage reading, continuity, and resistance(ohm) measurements to do all of the mods/measurements. All the cheap ones have these abilities. its a great tool to have, i have 3. I have been thinking of adding an O-Scope to my arsenal, for reviews really. but $200 bucks is a lot to spend on reviews, especially when its my hobby. but in the future it looks like its going to happen.
I think the issue you are having is memory related also. I have the same G.Skill kit in your sig and I also get the oc failed screen on cold boot, and random bsod's. If I switch out the memory to another kit, with the same timings and voltages it is stable with no issues.
Don’t know how relevant this may be but I'll say it anyway....
On some bios's (from B2 experience) I find the "IO Latency" jumps about between 1 & 2 if using dimm slot 2 & 4, sometimes this is a little random & other times going beyond 100BCLK makes it lock into 1 rather than imo the more stable 2.
I would suggest if you aren’t already to try slot 1 & 3 & leave Performance enhance at std to try to rule out your memory.
@OC Nub, believe it or not but your post up there made me happy ! I would be happy if its just a simple mem issue. The next step for me was to send my kit to someone from here to test on their stable motherboard but since OC Nub tested it already i think its pretty much proven to be a mem issue. Now the Q is which memories to get ? I prefer something 1866 or above.
@C-N im using slots 2 and 4 since it was discussed they were the preffered slots.
PS: Loing time ago (January 9th) when i ordered my parts on release day G.SKILL did not have any P67 certified memories and the only option i had was with these memories. One of their guys in their forums suggested i do an RMA but the exchange policy was only 30 days. Here i am 3 months later with a B3 revision board and my memories are not compatible. I wonder if i explain my pain to G.SKILl would they exchange or im just :banana::banana::banana::banana:ed ?
I think it's time for C_N to get his B3 board. He's going to love the new BIOS if he does...
For UD5-B2 new BIOS F7 is also out.
Btw,he can crossflash with modded B3 BIOS.
Already tested them & went back to F5x...... Didnt like needing more vdimm with the b3 bios
You need more vdimm?
im glad C_N is in this thread :)
Hi C-N,
My PC specs are as follows:
GA P67A-UD4-B3
2600K @ 4.5GHz, @ 1.35V (=1.332V Load), LLC Enabled, energy saving features turned off, Turbo OFF, R-TRCinOS OFF, HT ON.
8GB HyperX Grey @ 1600MHz 1.5V (=1.548V)
2xSapphire 6950 2GB CF
Corsair HX750
I read in one of your posts that for CrossFire PCH core should be at about 1.14V to prevent the graphics driver crashing.
Well, I have had graphics driver crashes (...video driver has stopped responding... then BSoD: 0x116 Video TDR Error) but never under load. I can run 3DMark11, Haven, games etc. Unfortunately, the new Catalyst drivers do cause BSoDs rather often for several people and now I'm not sure if that 3 crashes I have had were brought about by the driver or the low PCH core.
What do you think? Shall I raise the PCH core? I'm a bit worried since I don't know what else would that affect. In fact, I've never read about increasing the PCH core before. :)
GFX pcie lanes are direct to CPU. Why you need to raise PCH?
no it is not, that is actually pretty good.
Good news guys. I wrote to G.SKILL and they agreed to exchange my memory for: F3-17000CL7D-4GBXHD, free of charge. Except maybe i only pay for shipping ? Not sure yet how it will happen. Waiting for my RMA number and instructions. I just hope this solves the cold "bootage" !!!
ps: These guys just won a lifetime customer for sure and this was my first ever G.SKILL experience ! Its all about the customer service in this world today, nothing else ever matters !!!
Sorry guys been out of town for the weekend...
@Falkentyne
My RAM is right on the edge of being able to run 2222 @ 1.64 bios 1.654 ET6 but with the JZ BIOS GB adjusted the voltage so you needed to give 1.66 BIOS to get 1.654 ET6 & although it reads the same just isn’t longtime stable in prime!! :shrug: if I bump to 1.68 ET6 reads 1.68 & its fine but as I cant find any more RAM performance with that BIOS I just as well use what I know F5x.
The B3 BIOS's are pretty much the same for me... no benefit to me, the new one with the 10 load ine levels is also no better for me vcore either droops to far or overshoots more than ever depending on level. In otherwords Its worse for me as the vcore is all over the place.... bear in mind this is not scoped or DMM measured just ET & CPU-Z but its enough for me to make a decision.
@ FL00D
I don’t think there's any literature that says 1 card = x voltage 2 cards = x voltage & I'm fairly sure I didn’t mean to preach to anyone that PCH should be x but I find it helped me to raise it. Bear in mind my rig is @ 5G DDR 2222 & my card is clocked over 900 x 1400 so maybe PCH voltage has more impact on my rig as its running quite high/edgy.
My advice would be to try sticking it to the PCH as it wont hurt especially on B3
got a question, did some testing today to drop temps a little now summer is about here and noticed some cores are warmer then the others as much as 8c
weak core or bad mount, normal? the amb temp is 32c/79f- 4.53ghz@1.275v bios,/QPI/VTT@1.76
Running temps
http://img251.imageshack.us/img251/6...running.th.png
Idle temps
http://img59.imageshack.us/img59/477...ablecut.th.png
The 4 chips I have had have all had some difference, my current 2500K is as much as 10oC on one core under load & I believe its sensor is not reading correctly as at idle in BIOS its temp is lower than ambient with EIST/C1E enabled.
P.S Your temps look great to me considering the ambient... hot hot
New ud5 build from old ud4...here are some pics:)
Kinda bummed out!
Just got my replacement b3 ud5 as my ud4 had cold boot issues with any setting and bios with pll override enabled. Well, exact same thing with ud5 b3, can only avoid cold boot problems if I disbale pll overide and that only gets me 4.5:(
Internal CPU PLL Overvoltage hates me. I can't get above 45x without it, but my computer intermittently cycles power on boot if the setting is enabled--which is more annoying than not having above 45x. I've tried all b3 bios, stock and betas. Doesn't matter what voltages or other settings I tweak, as long as CPU PLL Overvoltage is enabled my PC does the aforementioned power cycling every other cold boot or so.
I have followed the advice on these forums using f3c, making the base bios and back-up bios match, cleared cmos, but STILL the same cold boot issues as on my ud4. Stable as can be in windows at 4.7-4.8, all stress tests are fine, memtest is fine, tried cold boot with memory lower speeds 1066, 1366, still has the same cold boot issues. Board is fine in all other respects, just see no point in buying an unlocked cpu that is stable in windows, but can’t go past 4.5 due to boot problems due to bios issues! Vcore is 1.34 at 4.5 and 1.37-1.38 max to get to 4.7-4.8, which is all the oc I need for what I am using the PC for, which is the annoying game of FSX:)
Any help, this is pretty upsetting as this is my second gig board now with the same issues that have been happening since January! Is there nothing gigabyte can do to resolve the nasty cold boot problems that still exist on stable b3 overclocks?
QPI/VTT@1.76 ?!:eek:
Use HWiNFO32 instead of System Info and install CPU-Z 157.1.
Ok first of all,
Please flash to the NEWEST beta (F3G) for the UD5
This will stop the boot loops when you enable PLL Override.
F1, possibly F2, and all the B2 version betas had weird bugs when PLL is enabled.
Those bootloops are fixed on F3G beta.
Ok, so there is an improvement from f3c? Thanks for this:)
I thought the only difference between f3c (which I have with cold boot issues) and f3g was:
'Fix: remove Floppy Support
Fix: BCLK 100'
Will there be any differnce between f3c and f3g in respect cold re-boots with pll enabled?
Well for what it's worth, I didn't have any boot loops with F3C with pll override enabled. But I didn't do that many boots with it enabled. But at least I was able to SAVE bios settings to enable it. Back on F1, when I was, for example, at 4 ghz and I decided to set it to x50 and enable PLL override and "save and exit", it was like a 50/50 chance if the board would boot up at 5 ghz, or would turn on, turn off after about 10 seconds with no screen (no POST), turn on again, show 34x100, then turn off again, turn on and say "system has been reset due to overclocking or error in voltages", then TURN OFF AGAIN right after that screen, then turn on, and then finally allow me in the BIOS.
I never saw this with F3C.
I did several more more vcore settings with F3G and setting various voltages in BIOS around 1.45v to 1.55v with PLL Override enabled, than I did with f3c, and each time I set them, it accepted the settings, and just rebooted normally. I never saw it boot loop (well, not yet).
The only (three) boot loops I've gotten with f3g were when I pressed the RESET button on the case.
Ok, guys, thanks for the help; testing f3g now. Hope it starts in the morning:) Working well in stress testing and warm boots thus far! Will let you know how it goes.
Maybe it's the CPU?
I've seen 2600K refusing to boot past windows logo with anything above 41x no matter what. Worst CPU ever.
Don't think so. It boots fine and works fine at high mults of x46-48 etc as long as the pc is not left for over 8 hours or so, then it may cold boot loop. Clear Cmos and voila I can boot, just have to keep opening the case etc etc. Can warm boot and run stable all day long!! Most annoying as this is now my second board. Will see how f3g works:)
Guys i think i may have been a little hasty with my summery of the jz bios, i decided to give it another go & did a fresh install. The thing is even though gb have adjusted the vdimm to better represent real vdimm this was not my long run prime stabiliry problem. What i have found partly due to not believing what im seeing combined with falkentyne's constant mention that jz is good & fixed his cold boot problem is jz kind of locks voltages & unless you psu power off after each change it wont register voltages properly in software & is therefore unstable. My board is now running fine on this bios.
Ive ran the jz since the 1st-2nd day and have never unplugged my PSU although I did have some reboots in the beginning and Im leaning towards more of n00bness to Intel and with the help of everyone on the thread... c-n / falkentyne / Stasio ive pretty well and went back to my notes and feel it was my setting the pll enabled was the issue on earlier bios. now with the new f3G im able to again enable it and lower my voltages even more so you might be onto something there. C-N your running the original MB or did you get a B3? also im curious as to the setting on board as to the software how much of a gap in voltages did you read? Is it better to have the PLL enabled or disabled for normal usage?\ barring any problems..
Stasio Ive noticed that the F3G has been removed from the beta thread is there or was there and issue with it or was the G just removed?
GA-P67A-UD5-B3 - F3
Hmm,
C_N how's the vdroop for you on F3g jz?
Just got my RMA number for my B2 UD5. Having a hard time letting go of her......
I have the original B2 board flashed with F3G JZ (B3 Bios)
I noticed CPU-Z & ET didnt always change vcore or vdimm if you used ET6 to adjust voltages although my DMM did register some change & my load temps in LinX went up/down with vcore adjustment. Only way I found to get CPU-Z/ET 6 to register correctly was PSU power down..... V strange but it may have something to do with GB trying to eliminate boot loops & removing the boards ability to set things on a soft power down I guess.
I havent had any problems with int pll with my 2500K I need it to pass x49 & I want x50 so its always been on.
Same as older 3 level BIOS's, no droop set 1.355 BIOS get 1.39~1.4 LinX load with level 1
Thanks for posting links to the B2~B3 mod BIOS's, great work as usual my friend :up:
<------C_N ?
You're using a B3 BIOS (crossflashed), and you're getting RISE at load ??
:shocked:
1.355v BIOS LLC1 for me gives 1.33'ish load... I only get that rise you have if I'm using LLC2.
Did you try a 60 second cmos clear after you flashed? (or if you haven't, you can still do that now, anyway, without reflashing)
(I always flash with flashspi, power off, unplug PSU, wait 60 seconds, press cmos clear button for about 30 seconds, etc...)
This can only mean that the LLC rise in the B2 board is a hardware problem, then....and not BIOS related O_O
Yes, although one BIOS I tested can’t remember which maybe the latest/last B2 or latest official B3 with the 10 load line levels either drooped or overshot... none held a constant voltage on my B2 board.
No...sort of, just flash from DOS or 90% of the time @BIOS, power down, PSU Off, Hold power button down to drain board, Go make a cup of tea!!!, PSU On CMOS clear a few times then power up/boot.
Looks like it & tbh if it could have been fixed with a BIOS it probably would have been by now...