Gigabyte P35 DS3L not holding overclock overnight

[SerpentRoyal]

Originally posted by: VirtualLarry

Originally posted by: SerpentRoyal
Yup. Abit boards increase the CPU's FSB by 3 to 4MHz from default. If the Gigabyte board cannot maintain this small overclock, then the board is defective.

A dirty trick used to win benchmarks, when reviewers run performance benchmarks on motherboards. That's a negative in my book. If you set it to a certain FSB, it should run that FSB, period.

Any legitimate reviewer would check CPU and RAM speed prior to testing.

 

[SerpentRoyal]

Originally posted by: Billbo
I think it's just luck that any of these "fixes" work. Some people have reported their problems going away by themselves after a few weeks/months. Others have had perfectly good boards for several months before the problems show up. The annoying part for me is not knowing when. For the first week I was so happy to have a board that utilized S3 standby mode properly. Now I can't trust it in case it doesn't wake correctly, as often it will come back with no video signal or it reboots or locks up.

All P35 boards have their little bugs. In general, you should use a BIOS that does not attempt to bypass the double POST. My IP35-E is rock solid with 12 BIOS. The same cannot be said about my IP35 Pro (Guru chip that circumvent the double post). The PRO board is prone to BIOS corruption...fan speed not working and scrambed BIOS.

 

[Billbo]

Yeah sure. And which bios would that be? I've been going to try bios F3 but would lose fixes for PS2 compatibility. I just can't find anything about which bios is the most stable. Only that F3 was meant to be the first worthwhile using.

I have had absolutely no issues today. Not only is it remembering a mild 25% overclock through warm and cold boots, it's doing it with less than default vcore.

 

[mingsoup]

Same EXACT problem after cold cold boots (overnight.) Currently I tried taking out my battery/PSU, reflash to F7, take out my battery/PSU again, then loaded up all my settings, did a save into profile and then a save and quit. I'll report what happens.

I also turned off halt on errors. I will also shutdown and power on before I do the shutdown for the night, to try incorporate all ideas from this thread.

Didn't work. The post restarted and all my settings were lost. Had to put them all back in. UHG! This really is frustrating. I bought this cpu/board to oc it out of the box. Standard speed will not due.

 

[XBoxLPU]

I have had zero problems with my DSL3


The only time I encounter a double post is when I change settings for overclocking but they stick if they are sucessful. To me it seems to be working as designed

 

[mingsoup]

XboxLPU, what bios are you using?

 

[XBoxLPU]

Originally posted by: mingsoup
XboxLPU, what bios are you using?

F7

Rev 2.0

I haven't shutdown the system from Vista yet but I have had various restarts and system crashes (due to unstable overclock settings) but the bios settings have remained in tac

The only time I have seen settings revert back is when it would appear I had enter settings for an unstable overclock.

 

[mingsoup]

This only happens to me when I shutdown for the night and restart. Greater than like 1 hour. I can restart and shutdown power on just fine.

P.S. I'm using 1.4650V on the CPU with all other voltages normal. I'm using 5,5,5,15 timings and a 356 FSB multiplier for a 3.20Ghz OC on an E2160 1.8Ghz. PCI E bus at manual 100mhz. Everything else is default. Again, with this setup...i've primed for hours on end with fine stability/temp.

 

[Billbo]

XBoxLPU, how long have you had your board? Would you like to share what you believe are stable overclock settings please? What voltages are you using etc?

 

[XBoxLPU]

Originally posted by: Billbo
XBoxLPU, how long have you had your board? Would you like to share what you believe are stable overclock settings please? What voltages are you using etc?

Just under a week. Stable settings depend on your hardware. I can't get my E4400 above 3.0Ghz without pusing above 1.4V in the bios

Currently at 300 FSB, 5-5-5-15 timings w/4Gb of PC-6400 Memory at 900. Vcore is at the setting just below 1.4V in the bios and currently testing with Prime to see how low I can drop the vcore and stay stable. Memory is stock voltage. The system is stable at 280 FSB (2.8ghz) with default vcore

 

[Billbo]

mingsoup, since you have the same cpu as me, what other FSB's have you tried? Do you know if there's certain speeds this proc/board combination doesn't like? Even if there were holes, it still doesn't explain the problems because for the first week any settings I tried worked without a hitch. Now those same settings will not stick. Increasing MCH/FSB voltages makes no difference. But once it finally decides to boot it is rock solid in orthos or anything I throw at it, as long as I don't power off or put it in standby.

 

[tigersty1e]

I'm not going through 4 pages of postings, so this may have already been posted/suggested.

I recall reading this bug somewhere else and the cause was the person's PSU unit.

 

[mingsoup]

I've tried 355 and 356 FSB. and thats about it. I've kept all voltages on manual and at default. Timings are hard set at 5,5,5,15. I started at a CPU V of 1.4750, and I've since reduced it two steps from that. Still running perfectly stable in Windows and infinite restarts.

 

[Billbo]

Just an update.

My DS3L has seemingly worked it's way out of the problem. For the past few days it has booted up every morning without a hitch. Think it's only had problems posting 2-3 times and these were when switching hardware and stuff. It's currently running at 2.70GHz and slightly undervolted. A 50% OC is all I wanted to be able to achieve for what I do with the comp. I'll see what it can actually do once I know the problem is definitely gone.

 

[mingsoup]

I've gone all the way down to 2.7Ghz on my E2160. Still won't hold the OC. I'm going to try falling back from F7 to F6 and so on and see if that helps.

 

[Billbo]

I'm using F6 though I don't think there's any difference to F7.

Maybe try what I did and clear cmos then leave everything on default a few days until it's definitely stable. Then increase the FSB slowly. I've also increased FSB, MCH and PCIE voltages by 0.1V. Memory divider is set as needed but timing are on auto though it would depend on your ram what to do there.

I'm wondering now whether something gets screwed up in the northbridge. I say that because when I was having problems, the NB heatsink was VERY hot even at stock. I reckon it runs at least 10C cooler now. Maybe just coincidence but who knows?

 

[Thejbrez]

I've finally decided to RMA my board. I'm sick of this POS losing it's settings every time its turned off. Poor board in my experience.

 

[mingsoup]

F6 didn't help. If anything the board got WAY more finicky about what it likes, not likes. Way more non posts now and still no holding. Any possibility they'd release a new bios to fix this? The longer I have this board oc'ed....its like the voltage requirements are going up.

I can't RMA. I'd prolly get hit with a 15% restock fee. Replacement is just gonna get me the same prob all over again + I'd be computerless.

They'll do something about it....Maybe I will try call Gigabyte just to see.

I'm gonna keep trying different things. Tonight is a auto voltage 2.4Ghz OC on F6. I'll try F5 tomorrow. Oh and the memory is multiplied to exactly 800...which is the memory's native speed.

Its just funny that my post always restarts from cold boot at "Memory Testi"....Does that suggests the memory part of the OC or the memory is to blame? Thanks.

 

[Billbo]

Well, you guys tell Gigabyte about it. I sent them an email today saying that many people are sick and tired of the problems with this board. They deny there's any problem, or deny they've heard about it. Yet I know others have RMA'd their board and received a new one. You can't RMA just because you want a different color, so what's the go there? People must have told them of problems to receive a new board????

What seemed to work for me is stop trying to overclock. Just flash a bios, leave it at stock for a few days and see if it becomes more stable. Test your ram by lowering the cpu multi so it's not in the road. Get memory running properly at the FSB you want to get to. Test the hell out of it with memtest/orthos large fft. Make sure memory is working. Put voltages on manual. Give +0 .1V to everything, except memory you should set to it's rated volts. Set the divider to be at or below what you tested it. Increase cpu speed 5 or 10MHz at a time. Cross fingers, hope for the best.

Honestly, my board was the worst, to the point people were saying it must be a lemon and to RMA it. But by treating it with respect everything came good. It seems people continually try to overclock their board when it's unstable. I don't know what happens with these boards. Gigabyte doesn't either. But I haven't had a problem since I did what I just said. Give it a damn rest and start from scratch. It seems these boards can get confused somehow when pushed too far too quick or something. I could be wrong, but no one, including Gigabyte, have a better explanation.

If we are all suffering the same problem, then you may notice the northbridge heatsink will be much cooler when the board starts working properly. Does this point to the problem? Gigabyte don't want to tell me.

I doubt my luck is going to last for ever. I'm thinking of putting a couple of dollars in a piggybank every time it starts properly. Then by the time it stuffs up again, I'll hopefully have enough money to buy an Abit or something.

Fair dinkum, I'm sick and tired of hearing about the problems with this board and no one, including GB, giving a damn. But I'm happy to not be spending all my waking hours looking for a solution. I really feel for all the others still having probs though.

 

[Idontcare]

For whatever its worth my experience is this: 5 DS3L boards ordered from Newegg right before christmas.

None were DOA.

4 of them were loaded with G0 Q6600, windows installed at stock clocks, then the systems went straight to 9x367 (3.3GHz) on BIOS "auto" voltage control.

They've been pretty near running 100% loaded on all cores for the past week without a single issue.

I had one instance of one of the mobo's losing the overclock settings, a CMOS jumper to short and reset fixed it and hasn't happened since.

1 board is on RMA for replacement. It started out just fine, stock clocks and installing windows. Then game the bad smell, that burnt electronics smell. Could not find the source but I assumed it was my PSU as the fan was running real slow.

Turns out it was a VRM burning out on the mobo, only you could not see the issue from the front of the board. Wasn't until I pulled the board that I could see very subtle shade of brown behind the VRM on the mobo backside. I pulled the board after realizing my Vdroop was some 0.2V!

 

[CDC Mail Guy]

I agree Billbo. I currently have mine set at stock, and for the past several days, I have had absolutely zero problems with it. Where in the past when it was overclocked, about every third time I turned it on I was getting no signal to the monitor. I would have to manually shut down, unplug the monitor cable, plug it back in and then it would work. Was than an overclock problem? I'm not sure, but I am now at almost a week with no abnormalities. Other than CPU-Z showing 6 as multiplier and a core speed of 2000, instead of 2660/2700. All in all I'm happy with the board at stock, and feel no NEED to O.C., I just like to be able to do it, and say that I have an FSB of 1800+

 

[Thejbrez]

I'm not sure why you guys chose this specific motherboard, but I bought it with the intent to overclock. If I can't do it then I might as well return it for a $50 board that will run at stock speeds. I have no idea why it wouldn't hold overclock settings after a cold start but I'll give another board a shot. I'll re-post once I receive the new board and get it tested.

 

[Billbo]

Thejbrez, yes I bought mine to overclock, and am glad I can now. Remember it's only a few of us having these problems. The majority are very stable boards. Then again some people don't have issues until a few months down the track.

Mail Guy, I think you have to disable both C1E and EIST in the bios for the multi to stay put, and maybe even set the power scheme in control panel to "always on". I reckon now your board has been stable several days, try overclocking again. Just do it slowly like 5-10MHz at a time. But get the memory settings right first. And definitely set all voltages to manual.

 

[mingsoup]

My board isn't holding anything. I tried a 2.0Ghz OC on the 1.8Ghz E2160....and it wouldn't even hold that. Voltages were on manual with nothing changed. I'll try the reduce cpu multi...and get memory going at the fsb/multiplier that I would like it to.

 

[Comdrpopnfresh]

So what made it do the start-stop-restart action? I just build a computer for my cousin that does this, an E2180 that I overclocked to 3.0ghz. It can go up to 3.4, but windows wasn't stable, and temps would get bad on load (stock intel cooling). So w/ mushkin 1066 DDR2, I can run a nice 1200mhz RAM, and have a 50% overclock with minimal voltage change- I was happy with it; until I hear that this start-up action is a danger.