Help Overclocking a Phenom II x6 1055t on a asus m4a785td-v evo

[Clay_Fury]

hows everyone doing?

In the last couple of days I have acquired my first real custom PC as before I was only ever upgrading factory built computers. Even then I wouldnt call it upgrading. I was so stupid then that I went out and bought a £300 nvidia graphics card spending every last penny I had and whacking it inside of a PCworld machine expecting an alienware dream machine to be born. Instead I got a highly unstable comp because of cpu bottlenecking and undersized PSU, not to mention the damn graphics card only just wedged in there. so I could only imagine the temps I was getting. So obviously now skint I sold the comp to cash converters for £200 and till now havent touched a pc that wasnt bought from a shop with a warranty lol. Anyhoo I was selling my laptop on Ebay as I wanted to have another go however some guy was selling this pc I have now and amazingly he was after a laptop so I put some money towards and I drove 2 hours to his and we swapped. So I get back very excited and turn it on then it freezes, so reset started to install fresh os and froze again. In fact it froze every 10 mins and I was fuming. I rang him up and he said I just have to tighten the cpu cooler as he said it gets loose. so ok I did and it carries on freezing on me for hours also the cpu idle temp was about 55. Eventualy I give up and couldnt care less if I damaged it anymore and removed the cooler and cpu. After doing my research I removed the poorly applied standard heatsink and drove down to pc world and got some artic mx-2. I applied this and refit and its been running like a dream however I have had a freeze once but thats it. Now the CPU idles at 23c. I know idle is not important when overclocking however I was impressed at the difference in temp just by changing the heatsink. I was worried that it may of been a ram fault causing the freezing or a faulty motherboard. Also the ventalation is restricted as he has squeezed the lot into a £25 tower case. Anyway sorry about all the waffling but what I am having trouble with is the mobo and cpu setup I have for overclocking. See ive already fuly updated the bios and read any OC guides i could get on the phenom however this mobo has some auto features in the bios that go with amd processors aswell as amd's overdrive software. Now im trying to research and people are either saying OC in the Bios its safer but give no guide on how to do it or to use overdrive however you have to turn off some features in the bios. So my OC knowledge is about as good as its defenition so when people start mentioning voltage adjustments and Front side bus OC and nearly everyones answer is different I didnt want to mess it up. So after all that what I wanted to know is if somebody could give me some tips and values I could run by and what im looking for in the results. Once I have experemented a few times and got some feedback of you guys im sure il be good to go then I can take responsibility of answering noob questions like this. cheers Clay Fury.

P.S if you have any feedback on why my comp kept freezing before id love to know. if it helps I have a geforce gtx 465 1024mb gddr5, 4gb ddr3 ram and a 765w psu. aftermarket cpu cooler. Il show a screen shot of overdrive if it helps?

 

[trents]

Okay, the first thing to do is to analyze what the settings are in bios for voltages and frequencies and some info about your hardware. Fortunately there is some free software that will facilitate this and present the info in pictures. So download the following two programs:

CPU-z

HWMonitor

Run CPU-z and report back to us with attached pictures of these tabs from CPU-z:

"CPU", "Memory", "SPD" and "Motherboard"

Then run HWMonitor and attach a pic of its user interface. It will show us some important voltages and temps.

Now, how to attach pics with your posts:

1. If you are running Windows Vista or Windows 7 there is a handy little program in Windows Accessories called "Snipping Tool". Find it and place it in your task bar for handy access. If you are running XP you can download a freeware "screen capture" tool from the net. Do screen captures of the above requested program user interfaces, i.e., CPU-z tabs and HWMonitor. Crop the pics so that you capture only the program window, not the entire Windows desktop. Save the file to desktop or wherever.
2. Find the "Go Advanced" button at the bottom of the new post window and click on it.
3. Click on the "paper clip" icon and click on it. That will open up a file upload tool. The rest should be obvious.

 

[trents]

Admin, this might be more appropriate for General Hardware.

 

[Clay_Fury]

thanks for the quick reply and for putting up with me. Ive taken the screen shots and attatched them with the paperclip function as you said. Sorry about before. Im starting to get the freezing problem but its only happening when running bad company 2 and shogun total war. However the games run beautifully with no lag and flawless looking graphics. Also no component was overheated. All im guessing is insufficient RAM maybe however on old comps of mine when the RAM got full I got a little yellow triangle warning and an advisement that my comp was eating into the virtual memory. Im worried I have been sold faulty hardware I was just so excited about getting the PC I rushed. ty again for giving me your time Clay Fury

 

[trents]

Look at your 12v+ rail in HWMonitor. It shows 12.61. That's high and may be out of spec. Also, why are you running your memory in single channel mode? How many sticks are you using?

 

[trents]

Pic of "SPD" tab from CPU-z also, please.

 

[Clay_Fury]

I dont know its how it came. So how do I adjust all this? The attachments are also the only two slots being used. I tried moving the ram before and I couldnt even get a display. So are those settongs bad?

 

[Clay_Fury]

just in case this is the mem in overdrive Im still such a noob to all this but it facsinates me. Hsve I been sold a bad comp?

 

[trents]

You can't adjust the 12v+ rail voltage. That's what the PSU is putting out. It may indicate something going out on the PSU or just a cheap PSU that doesn't have very good voltage regulation. What make and model is it and how long has it been in service?

If you can't run your memory in the slots that would give you dual channel mode then that's a concern. It could mean something is wrong with the motherboard. Before we conclude that, however I would suggest that you reset the CMOS with the jumper on the motherboard, just to clear all the current settings and reset everything to stock values. When you move the jumper, though, make sure you have your PSU unplugged first. Move the jumper, leave it there for 15 sec. then move it back to the default position. Then plug the PSU back in. Move your two memory sticks into same color slots (dual channel mode usually is reflected by same color slots but there are exceptions) and try to reboot. The CMOS jumper is usually near the little flat, round battery in the motherboard.

 

[Apht3rThawt]

First off, don't worry about the psu +12v rail. 12.6 is considered tolerance so 12.61 is nothing, but, your -12v output is not even close. You should watch hwmonitor to see if the -12v output comes up(or goes down) to -12v. I've seen mine drop but not -6v. Your ram should be in 2 different colored slots. 4 gigs of ram is plenty, you won't get errors from too little.

 

[trents]

Man, I just looked at HWMonitor on my system an it tells me the 12+ rail is only putting out 2.4. That can't be right! My system is completely stable. Maybe the program is not to be trusted for PSU voltage monitoring. And it doesn't give me a line for 12v- at all.

 

[trents]

. . . Your ram should be in 2 different colored slots . . . View attachment 96564

Yeah, but it looks like he's got it in two different colored slots now and its showing single channel in CPU-z. Now, I'm taking it that slot 2 and slot 4 (CPUz) are equivalent to DIMM_B2 and DIMM_B1 in your board diagram. Are you interpreting the diagram correctly? The idea is not to populate the same channel, either A (A2,A1: different colors) or B (B2,B1: different colors) but to populate two channels (hence, "dual") such as A2,B2 (same color) or A1,B1 (same color) or A2,B2 and A1,B1 if using 4 sticks.

 

[Apht3rThawt]

Man, I just looked at HWMonitor on my system an it tells me the 12+ rail is only putting out 2.4. That can't be right! My system is completely stable. Maybe the program is not to be trusted for PSU voltage monitoring. And it doesn't give me a line for 12v- at all.

I find that different mobos give different ouputs on hwmonitor. But radical voltages should be looked into.

And then to ram. Right, dual channel, same colored slots, hah. Hard to remember sometimes mobo to mobo and color to color. My 2 jiggy 890gx and 890fx are different; 890gx has same colored slots next to each other, 890fx has slots staggered, both mean put ram in same colored slots. Dual channel, same color, duh...

 

[Catscratch]

Don't pay attention to software's voltages. The best is seen in bios. My +12 shows 9v on monitors but 12.4v in bios. Even bios may not show the real value either. Only way is to measure with a volt meter.

PS: Put your system specs in your signature so we don't have to search in your first post everytime

 

[Clay_Fury]

hey guys been offline for a couple of days anyhow I have reset the BIOS and installed so many different bios optional updates but I have changed a few things. I Have rehoused my PC into a bigger and better tower bought a further 4gig ram bought a samsung 500g f3 SATA and I have this problem which I had before come back again. When I rang the guy up before complaining that the comp would not boot up he said retighten the cpu cooler and jiggle it and it worked however I couldnt see that particular component causing the problem so I re-applied the heatsink compound. Anyhow happened in new housing so i reapplied heatsink again and it failed to boot a few more times. Anyway just before I went to try fixing it with my foot and the wall I jiggled it and pressed FIRMLY and it booted. What do you think the problem is? any ideas

P.S HWmonitor has stopped giving me temp readings of the CPU, and this has always been the case. They come back on a fresh install of HWmonitor but then dissapear again

 

[Apht3rThawt]

It's not the cpu cooler, it's the cpu/socket. You can run a pc without a cpu cooler, it will just get too hot and fail, but it will boot. If it boots with pressure on the cpu cooler then the cpu is loose or the socket is not holding the cpu correctly. Or you have a short circuit or bad ground on the mobo. Mobo may be cracked and stress will cause it to reconnect.