Please register or login. There are 1 registered and 938 anonymous users currently online. Current bandwidth usage: 326.30 kbit/s September 22 - 02:14pm EDT 
Hardware Analysis
      
Forums Product Prices
  Contents 
 
 

  Latest Topics 
 

More >>
 

    
 
 

  You Are Here: 
 
/ Forums / Overclocking /
 

  Q9300 Overclocking Thread 
 
 Author 
 Date Written 
 Tools 
Continue Reading on Page: 1, 2, 3, 4, 5
john albrich Jan 08, 2011, 03:12pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
Nate Duerksen said:
...Do you mean for my PCI-E video card? I hard-set that to 100MHz once and it didn't seem to make a difference....


Yup. That's the kind of potential interaction I was talking about. If there is no relationship on your system between the PCI subsystem and the settings needed to OC your CPU, then you're in the clear for that specific potential issue.

Want to enjoy less advertisements and more features? Click here to become a Hardware Analysis registered user.
Nate Duerksen Jan 08, 2011, 03:13pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
I just booted to Memtest with everything at default and it froze at 4 seconds. I tried again. Same thing. Bad stick you think?

john albrich Jan 08, 2011, 03:22pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 08, 2011, 03:25pm EST

 
>> Re: Q9300 Overclocking Thread

Isn't there an option in memtest to specify the address range to test? You can isolate testing on a specific stick (or sticks) that way. You can also specify which test to run (e.g. random v. walking v. odd/evens v. etc patterns).

Alternative is to keep close track of each stick and test them physically one at a time. (of course using proper power (e..g unplugging AC mains) and ESD handling procedures to avoid damaging anything)

I'd give the address range test option a try first. I'd run memtest and provide details but my test machine is currently disassembled for some configuration work.

Nate Duerksen Jan 08, 2011, 03:24pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
I didn't see those options. There is a configuration option once it's started, but I don't think I can get in there and get it changed in the 4 seconds before it freezes up. I'm running a windows memory test right now to see what that shows. If that clears, I'll tinker with memtest some more.

john albrich Jan 08, 2011, 03:27pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread

There may be a way to do it using the command line and appropriate options...In that case memtest doesn't actually start running until after you've already set the test parameters.

Nate Duerksen Jan 08, 2011, 03:52pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
I may look into that in a bit. A guy on another OC forum said to raise my vMCH to 1.3 (from the 1.1 default) and see what that does, so I did that and am running Prime95 again. It seems to be running stable, but it's only been 5 minutes so far.

Joe Albergo Jan 08, 2011, 04:22pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
Hey guys, glad to see some people posting about the Q9300.

I wanted to add when I am using Prime95, my 3rd core always fails first and I have no idea why. If it was bad would be I able to run my system as long as I have? I am not OC'ing as hard as I was anymore - I am just at 3.0@400mhz FSB -

I am on liquid and when I go to 3.26 - I play BC2 for about 20 minutes and I get a soft-reset and my BIOS does not report the crash due to overclocked settings.

I guess what I wanted to say all and all is I don't trust Prime95, I've read elsewhere that although its a great program, it might not be the greatest route to test the OC.

Just my opinion, I am no expert in the manner - what do you guys think?

Nate Duerksen Jan 08, 2011, 06:25pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 08, 2011, 06:27pm EST

 
>> Re: Q9300 Overclocking Thread
It crashed after about an hour of Prime95. It didn't show any failures in the text report though, so I too am wondering if maybe it's a Prime95 bug. I'll try the memtest again.

Edit:
Memtest did the same thing again. It froze up at 4 seconds.

Nate Duerksen Jan 08, 2011, 08:33pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
Well, that's interesting. I think I found the problem provided Memtest doesn't have a GB ceiling of 4. I've been switching memory sticks and booting to Memtest for the past hour or so. All of the sticks work fine individually, but whenever I put more than 2 sticks (4GB) in (doesn't seem to matter where), Memtest freezes at 5 seconds. Here's a table of the slots that worked:

slot 1 - 2GB
slot 1, slot 2 - 4GB
slot 1, slot 4 - 4GB

These didn't work:

slot 1, slot 2, slot 3, slot 4 - 8 GB
slot 1, slot 2, slot 4 - 6GB

Nate Duerksen Jan 08, 2011, 11:23pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
4GB also crashed after 45 min on Prime95.

Nate Duerksen Jan 09, 2011, 07:26pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
Well, I think I got a bad MoBo. I set everything to default and reinstalled windows and it still crashes. My copy of windows is good because it's worked perfectly on every other computer I've stuck it on and each individual RAM stick tested good, so that only thing left that I can think of is the MoBo.

john albrich Jan 09, 2011, 11:27pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread

You might try cooling the entire system down with a desktop or floor fan...doing more to get rid of heat build-up on mobo components other than CPU.

For example, I've noticed some people pointing out the heatsinks on the chipsets on their mobos didn't seem very adequate or they didn't seem solidly affixed. People have also mentioned the on-mobo power regulator chips and mosfets for the CPU etc. run hot and some mobos have heatsinks on them and some don't.

I was in that category with a very nice brand new Gigabyte mobo. These chips (19 of them) were running about 10-15C hotter than the CPU, and they did not come with heatsinks...and of course OCing stresses them as well when you up the voltages and power requirements. I've since put heatsinks on these chips. (yet note that these components did have factory installed heatsinks on a cheap oem mobo used in an old eMachines box (and they never even got very warm running at default speeds and the board wasn't designed for OCing)...go figure!).

Nate Duerksen Jan 10, 2011, 07:01am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
I'll see what I can do with that. Supposedly this one is really good at dissipating heat because of its 2oz copper PCB or whatever, but I'll see if that makes any difference.

Erwin Rommel Jan 10, 2011, 09:09am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 10, 2011, 09:19am EST

 
>> Re: Q9300 Overclocking Thread
My Q9300 makes 475X4X7,5 @1,34 v without any problem. Cooler is Armageddon with 2x140mm vent. Mobo is P5Q Deluxe. Memory is 2X2GB AData 1066+ extreme edition (2,2v ; 1:1 @ 950 MHz, 5-5-5-15).
Temps are OK (core is 65-68 C on load), should i leave it for 24/7 or not?

[url]http://www.overclock.net/intel-cpus/815306-overclocking-q9300-...st10598109[/url]
Here are some MoBo sets till 450 FSB.

john albrich Jan 10, 2011, 09:19am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 10, 2011, 09:30am EST

 
>> Re: Q9300 Overclocking Thread
Nate Duerksen said:
I'll see what I can do with that. Supposedly this one is really good at dissipating heat because of its 2oz copper PCB or whatever, but I'll see if that makes any difference.

IMO the use of 2oz copper has an minimal impact on motherboard component heat dissipation. It's simply a marketing claim. They're taking advantage of having to use 2oz copper for reasons that have nothing to do with heat management, and spinning up a new (and questionable) claim for marketing purposes.

Unless they can dissipate the heat from the copper then it will quickly saturate and any added heat dissipation value is quickly lost. In fact, it might even contribute to other heat-sensitive components being conducted more heat than to which they otherwise would be exposed. Think of it this way...it would be like mounting adjacent MOSFETs on the same heatsink as the CPU. Instead of being somewhat thermally isolated from the CPU, the mosfets are now more efficiently coupled to the heat generated by the CPU, and their temps now increase as a result. That could result in the need for adding heatsinks to components that didn't need them before.

Now if they provided additional heat radiators with lots of surface area directly tied-in to these copper layer(s), or even a fluid-cooled layer somehow embedded in the motherboard to remove heat from the copper layer(s) , that would be different. But unless they can get rid of the heat buildup...using 2oz copper instead of say 1.5oz copper is likely ineffective and could cause problems by coupling other components to high-heat sources.

The degree of effectiveness would also critically depend on how much copper is actually left on the various layers...just because they use "2oz copper" does NOT mean a given layer in the mobo actually has much copper left after they remove the copper not needed for electrical signal transmission, signal quality control, and EMI suppression reasons. For example, they might remove 80% of the copper on a given layer and most of what is still left might be mostly extremely narrow width signal lines...which would severely negate any heat dissipation contribution from the copper.

And of course the less copper, the less impact on heat dissipation (as negligible as it may already be). I'd also bet that they're only using the 2oz copper on the power distribution layers in the mobo...and that's primarily to handle the power requirements...not heat...but they make a marketing claim to help boost the sales appeal anyway.

I'd like to see some objective component temperature tests that directly compare temps vs. the use of different motherboard copper layer weights using typical circuit layer designs. Both the temps of high heat-producing components and the temps of adjacent components.

Nate Duerksen Jan 10, 2011, 12:07pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Q9300 Overclocking Thread
You're absolutely right John. There are a couple small~ish heatsinks on the board that I assume the copper goes to, but honestly, I don't see a real difference in 2oz copper vs 1.5oz copper. Copper vs aluminum, yes, but not 2 vs 1.5. I'll throw a fan on it sometime today and see what happens. *crosses fingers* I'm about to give up on Gigabyte if this doesn't work.

vittorio puglisi Mar 14, 2011, 06:51pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Mar 14, 2011, 06:51pm EDT

 
>> Re: Q9300 Overclocking Thread
Just for the record..

I managed to overclock my cpu to 3.63 Ghz / 485 FSB by just changing the Vcore ...

I use a P5Q Deluxe motherboard.

Vcore set to 1.45 V . Any lower would make the system unstable.

Any other changes , like NB,SB,FSB termination,PLL etc ddnt help me at all to increase the fsb by a bit.



Write a Reply >>

Continue Reading on Page: 1, 2, 3, 4, 5

 

    
 
 

  Topic Tools 
 
RSS UpdatesRSS Updates
 

  Related Articles 
 
 

  Newsletter 
 
A weekly newsletter featuring an editorial and a roundup of the latest articles, news and other interesting topics.

Please enter your email address below and click Subscribe.