Post your E4300 Overclock!

Page 15 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
New to the forums and to overclocking, but thanks to all the information from you guys, I think I've been pretty successful.

The parts:
MSI P6N SLI-FI
C2D 4300 @ 3.0 GHz (333.3 MHz x 9)
-VCore +0.025 -> 1.325V
-1:1 ratio with memory
2 GB Wintec AMPX DDR2 800 (at 667 MHz)
-(5-5-5-15-19-1T) - (stock)
EVGA 8800GTS 320MB
2x Seagate Barracuda 7200.10 ST3250820AS 250GB

Power Supply and Case:
Cooler Master RS-600-ASAA 600W
XION Hydraulic XON-566TB

I have stock cooling on all the parts but did add a 120mm intake fan on the front of the case.

I am at the 5 hr mark in Orthos and these are my temps:
Ambient: 21C
NB (load): 40C
C2D Idle: 29C (tCase)
C2D Load: 56C (tCase)

I think that just about covers it, but I have a few questions, if you have the time.

My Idle to Load delta seems to be 27 degrees C, which is more than the recommended 25 max delta that is referred to several places in these forums. Is this something I really need to worry about?

I haven't attempted my memory OC yet, as it seems much more complex/time consuming what with all the timings and I'm not sure if the speed increase will be worth it... thoughts?

And finally, once I have a stable overclock, is there harm in re-enabling C1E and EIST? (or for that matter, Spread Spectrum? What does that do, anyway?) This post: http://forums.techpowerup.com/showthread.php?t=22998 seems to think enabling just C1E could cause instability, but enabling both won't cause any harm and would save me some power/heat/money/wear-and-tear.

what C1E does, is lower the Voltage of the Processor, when at Idle, to save on heat output.
remember, a CPU is meant to only be run at stock.
Intel designs these CPU's to be stable at thier stock speed, with C1E active.

however, C1E for a overclocker, can leave you going nuts if you dot know what it is.
especially if you have a older LGA processor, that doesnt have EIST on it, but does have C1E.

C1E. if used with EIST, is actually a handy feature to have, if you want to keep your CPU temps low while at idle.
but if you choose to disable EIST, I strongly suggest you disable C1E at the same time, otherwise, this feature may drive you bonkers too, just like it did with me.

Thanks in advance. You've already been extremely helpful, even though I haven't posted until just now. 😀
 
333x9 2997mhz with 1.4125 Vcore
FSB + 0.2
MCH + 0.1
Ram at 3-3-3-12 with 1.9 Vdimm 1:1 (DDR2-667)

Idle, 43c (tjunction)
10 Minute TAT, 69c (tjunction)
Passed 12hr Dual Prime 95 Small FTT's.

I don't like having the Vcore up that high but any lower and it fails after a couple of hours and it's still reasonable temp. It fails after around 13hours of Dual Prime95 but it's stable enough for me and I don't really want to vcore any higher. Any thoughts? Suggestions? I only had 3ghz in mind when I made this build as you can tell from the 667. I think I got a pretty ordinary chip to need that much voltage 🙁
 
Hi!
Can anybody tell me which are the best e4300 cpus? I mean which are the best production weeks or model numbers for overclocking.
 
That is fairly typical of E4300's, maybe a hair on the higher vcore side, but not bad at all. Rule of thumb is +/- 10% of stock vcore is safe for long term usage (4 years+). So 1.3 * 1.1 = 1.43vcore :)

Mine is at 350 * 9 @ 1.4375 for daily usage, I maxed out at 375*9 @ 1.55vcore prime stable.

I am hoping to get a Penryn quad core after the first round of price cuts just after christmas...

ANyone need a E4300? I happen to have a spare OEM one, never used. Got it from newegg and ended up not needing it and didn't try to return it in time.

Yes Taco... it does have the IHS on it. :wink:
 
Read from some chinese discussion groups at Hong Kong.

I myself brought the P5LVM-1394 and all overclock is not success so far. Could oc to 233 but then the build-in LAN port would fail.

The P5L-MX seems overclockable for E6300(some successful cases published before) but not for E4300. Similar case on P5LVM-1394.

These two mobo seems unable to lock PCI-E/PCI speed for E4300(FSB800). RAM or CPU seems not the root cause of the failure so most likely the PCI bus is the reason.

I installed a P5L-VM last night and had the same problems, but I was able to fix them. First, I pin-modded the E4300 to 266 FSB. Then I changed the PCI-E speed to 120. Now the motherboard boots at 300+. The lack of a PCI lock crapped out the onboard NIC AND my spare Netgear NIC, but my old 3COM NIC works fine.

266 FSB Pin Mod:
http://forumz.tomshardware.com/hardware/E4300-Pin-mod-ftopict228129.html
 
Gigabyte 965P-S3
2*1 GB DDR-800
E4300 & 3006Mhz (334*9)
EVGA 7100GS-128.
Stock Voltage, Stock HSF.

-----
Coretemp Orthos Load: 67C
Coretemp Idle: 45C (800 rpm CPU FAN)
Coretemp Idle: 35C (2000rpm CPU FAN)
Speedfan Orthos Load: 52C
Speedfan Idle: 30C (800 rpm CPU FAN)
-----

12 hours Orthos stable.

/Chris
 
I am able to tap 3600mhz with the 4300

cpuwo8.png

memhh6.png

better can be done with my e4400 check this out
ss.JPG


:lol:
 
4GHz would suggest a Conroe core not an Allendale core, same for the 3.6GHz. Average E4300's won't make it to 4GHz and most won't make it much past 3.2GHz, there is something fishy about that 4GHz. You posting these numbers prime stable?

Yes it does say E4300, I agree, but the E4400 is replacing the E4300 so it will be interesting to see how they scale compared to the E4300.
 
Hey guys, I'm new to this forum. Sorry for the lengthy post, but I'm having some trouble with my e4300 and was wondering if anyone can share their thoughts...

First, with everything set at stock (1.8GHz), Core 1 runs 7*C hotter than Core 0 at idle (24*C vs. 17*C as per CoreTemp 0.94). I'm not sure there should be such a large difference?

As far as overclocking goes, first off, I am able to drop the multiplier to 6 and got the FSB stable at 1440MHz so the board is fine. My memory is underclocked so that also should not be a problem.

I am able to get 2.7GHz (300x9) with zero voltage adjustments to anything, dual-Orthos stable for over 4 hours. However, in order to go from 1200 to 1333, I need to up the Vcore by +0.15!! (No that is not a typo: 1.45v). Otherwise, it is not stable in Orthos for more than 2 minutes. (Note: when testing with the multiplier at 6, I had to up the North Bridge by 0.05v to run stable at 1333FSB, and I did use the same setting when I turned the multiplier back up to 9).

No matter what overclock I use, the Core 1 temp is 7*C higher at idle than Core 0. Another thing that concerns me is that when I run Super PI, Core 1 assumes 80-90% of load while Core 0 only has 10-20% (as per Task Manager->Performance Tab and SpeedFan). In addition, if I only up the Vcore by 0.05 I can boot windows at 1333FSB, but Orthos will always fail in the first 2 minutes. It ALWAYS fails on Core 0. Is it possible that a processor can have one good core and one bad core? I wonder if NewEgg would allow me to RMA...

I would appreciate any input at all on this issue. And before anyone asks, I am using the stock HSF and yes I verified that is was installed properly.

Thanks in advance
 
This speaks to a heatsink being seated wrong or a concave IHS. The Core temp differential shouldn't be more than 2-3c, so you are right to be slightly concerned.

What was your original vcore? 1.45v for 333*9 isn't terrible. In general, the vcore for the E4300 has to be upped at a higher pace when compared to the E6 series due to manufacturing differences. At 1.4375 I am stable up to about 355*9, but it takes 1.55v plus some uber DFI tweaks to make 375*9 stable.

What heatsink are you using to cool your CPU? What is ambient? Your 17c sounds too low, even my IHSless E4300 doesn't get that cool.
 
Thanks for the reply.

Original Vcore is 1.296 in CPU-Z, 1.30 in SpeedFan (while running at 300x9 with no voltage adjustments). I just wouldn't think I would need to add so much to get 1333 since I can get 1200 with no adjustments. And, many users on another forum get 1333 with the same CPU/Motherboard by only adding 0.05 tops.

Using stock heatsink, ambient is about 17*C actually.

edit: I suppose that if Core 0 can reach ambient temp at idle and Core 1 cannot that would imply an improperly seated heatsink. Why then would Core 1 have the higher load and Core 0 would fail in Orthos? I would think that it would be the other way around if the heatsink was not on properly. I should also point out that at load the temp of the cores even out.
 
Every CPU is unique in it's overclock potential, voltage tolerance, and thermal behavior. If the maximum stable overclock is known at 1.35 Vcore, then each increase of .05 volts will typically allow a stable increase of ~ 100 Mhz, and will result in a corresponding increase in CPU temperatures of ~ 3 to 4c.

This is the general rule of thumb with C2D's.

So 300*9 = 2700 and 333*9 = 2997.

2997-2700 = 300 [yes I am rounding]

300/100 = 3

3 * .05 = .15

1.3+.15 = 1.45v

and that is how we get to your stable vcore.
 
hmm, well I guess that makes sense then. only thing is I hit 60*C quite easily at 1.45 Vcore so i am going to need to get a better heatsink and some AS5.

Thank you superfly...I've read that document, but I guess I was more concerned with the numbers other people have been getting to consider that.
 
Your welcome Zad. Having to up the vcore so quickly can be rather unerving after not having to up it for so long.

Taco-

That's why I'm the man, lol. 8)

The extreme's mount was well configured for use with a IHSless proc, I can't take too much credit. I just want a quad core penryn (yorkfield?), that will be a monster. I am keeping an open mind about K10 though.
 
I'm 22, a grad student, and starting my career. Little different life scenario. I'm lucky to be where I am. 8)

Course I also spend $0 on aestetic crap, which is why my apt looks baren, but I don't care. lol
 
Eh, at least I only lost a CPU and mobo. :?

I hate to lose a GPU... those are more expensive.

I am going to custom build a nice ice cooled water loop next year probably. I think this will involve power tools, a mini fridge, sealent and alot of elbow grease. What initially comes to mind is one inlet and one outlet in a sealed fridge with a metal filter on the outlet so the ice doesn't get into the tubes. Then there is the nice door I am going to seal up with caulk.

Just have to figure out the auto ice refill.
 
If you plan to get a new hsf, I'd pressure mount an ultra 120 extreme and remove your ihs (the extreme's mount only tightens down to the ihs meaning you'll have to pressure mount it yourself if you plan to do what superfly03 did) like superfly, that seemed to work very nicely

hmm....well since I am an OC noob and have never heard of a pressure mounted hsf, nor removing the ihs, i'll research that. sounds risky
 
then I wouldn't mess with it, it's kinda an experienced only thing to do :wink:

well we're all noobs at some point; only way to learn is by doing. i found a video of it being done to a sempron (dont know it its any different that what would need to be done to a C2D, but it looks simple enough).

if it is as easy as the video i saw, i might do it. seems to me that the difficult part will be to "pressure mount" the hsf as there will be a risk of crushing the core(s).

only thing is that hsf costs $60, my e4300 was $114. i dont know if the extra speed im going to get from it will be worth spending an additional 50% of the cost of the CPU in the first place...
 
But what about the e6400? That one wasn't all that much more than the e4300, but I suppose I can't blame you since everyone expected the e4300 to be the next pentium d 805 with the insane mult, while instead the e4300 used a different core and wasn't binned as well so it ended up sucking in oc comparison


Well, that is the way it has turned out. Individual chips may perform better or worse then the average for their group, but (in retrospect) it appears that the E6300 group IS a better/more consistent overclocker than the E4300 group.

My E6300 hit 3Ghz on a DS3 at stock voltage (and only +.1mch and +.1 memory).


Here is what I find interesting..... go back in the forum and read all the people that slammed the E6300 for requiring 430Mhz to reach 3GHz (at stock voltage, of course!). I could name a few *very* vocal highly respected members that pushed the 4300 hard simply because they did not think that memory/motherboards would be up to 400+Mhz speeds, and that the 4300 would be SO much better due to the multiplier. I have both chips. IMHO, the 6300 beats the 4300 hands down at ease-of-overclocking if you have the right motherboard and memory.


So, the "experts" aren't always experts even if they have 1200 postings.
 
hmm, would a celeron D 331 have a similar ihs? could pick one of those up for <$10 on eBay, and could even use my current MB to test if i broke it
 
My Asus p4 P4C800-e deluxe is being retired (cpu died).
I need a replacment. I started to read this thread but it's 12:25AM and I only got through page 3 (of 19) then skipped to 19. Looks like I missed a LOT in between.

I am going to buy a new system. If I can use some of my old hardware, great, if not, oh well.

I need RAID (mirroring), I want the most bang for the buck and would like to overclock the CPU.

I use two 22" monitors with an AGP Geforce 6800Xtreme. I need graphic support for both of them. It would be nice to be able to use my AGP card and also be able to go to PCI-e eventually. But this is not absolutely necessary.

I only have 1GB DDR memory, so if I have to buy new memory, that's ok.

All of my 8 hard drives are ATA (will be happy to move on to SATA)

I need gigabit LAN, and 1394.
I have been very happy with the current Asus motherboard, but I am open to suggestions for boards from other manufactures.

I have been reading all night, but I use my computer for work and need to get it replaced ASAP.

From what I have read so far the e4300 and the P965 chipset may be the way to go?? but on the last page there is talk about the 6600 (?) being a better buy?

I really need your help. I want a fast, reliable, very stable machine.

What do you guys recommend?

Thanks!!