• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

My e6400 overclock

Summers618

n00b
Joined
Dec 26, 2006
Messages
18
Hey guys, I finally put together my rig. Here are the parts:
Intel Core 2 Duo e6400 stepping 6 revision B2 @ 3.60 1.36v
Gigabyte Ds3 F9 BIOS
Sunbeam Tuniq Tower 120
Seagate Barracuda 720.10 250gb perpendicular recording
Ati Radeon X1950XTX @ 702mhz core/1017mhz memory on stock volts
G.Skill PC6400C4 2GBPK @ 900 5-5-5-15 stock volts

Here is my situation. I'm new to the whole scene. This is my first Rig, coming from my family's P4 2.0ghz 756mb ram fx5200 so I'm a newb to the overclocking game. This system was the best I could afford, and i even got several parts at a discount (x1950XTX $275). I cannot afford to do any damage to this system as it is a 24/7 system. I am hesitant to push my system even higher because on the Conroe overclocking database, hardly anyone is above 3.6 ghz without voltages way higher than mine. I am wondering if i just got a really good chip. Coretemp shows me at about 52 degrees during prime95 load. I want to know how to find the limit of my ram and cpu on their own, without frying them with extreme voltages or temps. Being a 24/7 system I have several questions.
-What are the top temps/voltages i should be looking for for a 24/7 system?

-How do I find what is limiting me, ram or Cpu?
-How do i check how many volts are going through my ram? They are supposed to run at 2.1v but when i open easytoon just for fun (I don't use it should I?) it says they are at 1.9v. I am afraid to overvolt them to 2.2 because adding .3 volts if they are already at 2.1 would not be good for 24/7 would it?
-How do I overclock the freaking video card. I use ati tool, and everytime i go above where i am on either setting it crashes, but then resets it to like 500mhz core 500mhz memory and I can't get it back to stock speeds.

Thanks ahead of time, I'm just trying to get the most i can out of this system for 24/7 Right now with this system I'm getting about 13000 3dmarks in 05 which is the lowest of the 3 other x1950xtx out there by about 200 points.
 
your temps are fine, to find out the max you can push your ram lower the multi to 6x then up the FSB and give the ram +0.3v which will make it run at 2.1v (stock is 1.8v) which should give you a bit more headroom in regards to how far you can push it. In regards to a 24/7 stable system just run orthos for at least a good 8 hours to ensure its stability.

I've found using ATI Overdrive works quite well to overclock the x1950xtx you may not get a super high overclock but you will get a nice one.

At this point in time the best advice I can give you is stop worrying your not going to damage anything.
 
Ok, so the ds3 automatically puts ram at 1.8 volts? Ok I'll go tweak it a bit and see what i cant get out of this under 60 degrees and prime stable

*edit Sceenshot if you guys wanted*


*edit x2*
Well it looks like under more testing that my ram is the limiting factor. I can hit 900mhz on them c5 with like 1.9 volts, but increasing voltage to 2.3 and timings to 6-6-6-18 wont boost it up to 950 and be stable, it looks like i'll just stay at 3.6 then, maybe try and tighten the timings and lower cpu voltage
 
Err... my bad, wrong thing... No in 05 i get 13039, the lowest of all the x1950xtx on there
part of it is that even my 6400 clocked higher than other conroes and allendales, still gets lower fps in the cpu section... Is there a reason for this?
 
that's an extremely respectable number from your e6400, given the low low voltage you're feeding it :)

For comparison, my e6400 required around 1.52 volts or so to be stable at 3600 and it put out a ton of heat. Is it box or oem cpu? Any way you could give date? If you have the box, it should be written on there. L6xxxxxxxxxxxxxx etc such as L628Bxx.
 
For a n00b you did get the C4 Corsair, which I, who should know what he's doing, did not (I have C5 - yes cheapness hurts.)

Thus you can get to 3.6 and I'm stuck at 3.36 because of the RAM. It's maxed out.

Just keep temps under 61c on load and you should be good.

[that sound you hear is me slapping myself over the head]
 
I have to run 1.56v to run stable at 3.6
At 1.625 I hit my heat ceiling, I can run 3.7 in orthos -- but for only a good 20mins. With your cpu I could do it indefinately with less voltage even.

I'm jealous. Your cpu cries out to go faster. Poor lil guy.
 
Ok, so the ds3 automatically puts ram at 1.8 volts? Ok I'll go tweak it a bit and see what i cant get out of this under 60 degrees and prime stable

*edit Sceenshot if you guys wanted*


*edit x2*
Well it looks like under more testing that my ram is the limiting factor. I can hit 900mhz on them c5 with like 1.9 volts, but increasing voltage to 2.3 and timings to 6-6-6-18 wont boost it up to 950 and be stable, it looks like i'll just stay at 3.6 then, maybe try and tighten the timings and lower cpu voltage


I only see one instance of Prime 95 running. You need to make a complete copy of that folder and run 2 copies. Or just download Orthos Beta version for dual cores to truly test your stability / heat.
 
you need to run a stress test that runs on BOTH cores not just one. Otherwise you are not stress testing the system.

For example you should be running Orthos that runs on both CPU 0 and CPU 1 and takes them both to 100%. Just running Prime 95 on 1 CPU doesnt do it. All that does is bring your CPU load to 50% not 100%

You also need to run continuously for more than 4 hours on a test that began and ended at the same CPU speed. Not to rain on anybody's parade your screenshot does not indicate that which is why it is better to use Orthos. You should also show both CPU's loaded at 100% for example use the display from speedfan or Windows task manager. That is the only way to prove you were running both cores at 100% for at least 4 hours with the CPU at 3600 the entire time. I thnk when you get to that level you are not going to be running at 1.36V any more.

When you run Orthos it shows the elapsed time since the start of the test and that the entire test was run at the same CPU speed.

When you have one that passes those fair criteria you should reply to the official Conroe overclocking thread and get recognized.
 
you need to run a stress test that runs on BOTH cores not just one. Otherwise you are not stress testing the system.

For example you should be running Orthos that runs on both CPU 0 and CPU 1 and takes them both to 100%. Just running Prime 95 on 1 CPU doesnt do it. All that does is bring your CPU load to 50% not 100%

You also need to run continuously for more than 4 hours on a test that began and ended at the same CPU speed. Not to rain on anybody's parade your screenshot does not indicate that which is why it is better to use Orthos. You should also show both CPU's loaded at 100% for example use the display from speedfan or Windows task manager. That is the only way to prove you were running both cores at 100% for at least 4 hours with the CPU at 3600 the entire time. I thnk when you get to that level you are not going to be running at 1.36V any more.

When you run Orthos it shows the elapsed time since the start of the test and that the entire test was run at the same CPU speed.

When you have one that passes those fair criteria you should reply to the official Conroe overclocking thread and get recognized.

I did the same thing the first time around (my first dual core). I ran 3.6 with prime all night long; but then when someone here pointed my nose to this I tried orthos the following night -- and I was unpleasently surprised I needed to up my voltage.
 
Yeah I know about orthos now... i Have a problem though... I was running 3.7 at about 1.42v and was orthos stable for about 6 hours then i realized the time and date on windows were wrong... I proceeded to change the date and when i came back in to check up on it, i was logged out and it said i needed to activate windows to do further work... But for some strage reason my code wont work... It will be a while for me to get it up and running again
 
haha man, I had something similar happen last night. I'm using w2k3 trial which has 180day timebomb. I had to reset cmos and forgot to set time before booting in. Got locked out. Booted in safemode to backup to my other partition and formatted. guess I got pwned
 
Back
Top