OCing questions on E4300 or E6300

Retsam

2[H]4U
Joined
Sep 2, 2005
Messages
3,905
Ok, not that experienced with C2D chips and ddr2 with ocing.

First. Which chip is better for ocing with using typical memory such as Gskill/OCZ/Corsair 2gb pc2-6400(not micron d9)? E4300 or E6300.

I am trying to figure out what the max overclock would be using 2gb pc2-6400 800mhz memory with one of those chips. I am trying to decide if I need to get good overclocking memory or just buy the $129 Gskill 2gb kit of pc2-6400.

In other words, im not familiar with C2D motherboard bios settings and how ocing ddr2 memory is handled. :p
 
I'm no expert and you don't say what motherboard you are using, but... I would get the E6400 and the PC-6400 memory you want and try your FSB to 400 with a 8 mulitplier for a 3.2GHz overclock. That way you can run the memory at it's rated speed of 800MHz.
 
unless your trying to push 3.6ghz in 1:1 ratio you wont use much more then PC6400 ram.
the e6300 has the nice feature of 1066mhz bus vs the 4300's 800mhz. depending on the motherboard and other hardware, this is a very large speed boost. the 4300 also generaly doesnt see as high of an OC percentage, being a budget chip its much more limited in performance then the 6300.

the 6400 has just about the best value when you factor in cost vs max overclock, its cheaper then the 6600 but it clocks higher then the 6300

the 6600 is second best, its a little more expensive but you see an avg. 200-400mhz max OC gain over the 6400, plus its the cheapest conroe with 4mb L2 cache.

the 6300 is generaly third, you get slightly less top speed then the 6400 but its considerably cheaper.
 
everyone is saying the e4300 seems to overclock less than the e6300

where is the proof for that?
i've seen reviews say that they're identical
 
everyone is saying the e4300 seems to overclock less than the e6300

where is the proof for that?
i've seen reviews say that they're identical

Alot of "user" reports contest to the poor overclockability of the E4300s... The "official" releases show that the E4300's clock similarly.

I don't know who to believe.

I DO know the E4300's use some TIM between the IHS and the core, where their bigger brothers are soldered on... That's gotta hurt temps a bit.
 
The only real difference between the E6300 and E4300 is that the slower E4300 does not include Intel's Virtualization Technology. Ultimately this means almost nothing, as it will not affect the processor's performance or overclocking abilities. The E4300 core is not crippled in any way, so you can expect the same performance out of the E4300 if you were to lower the multiplier to 7x and raise the FSB to 266MHz. Having said that, the E4300 is essentially just an E6300 or even E6400 with VT disabled.
from here this is the first thing that came up in google

well most likely; they are crippled. not on purpose by intel, but just during the mfg. process. when making these chips they dont produce different dies for each of the different core2's they make the x6800s and any dies that have flaws or parts dont work they get reduced in price accordingly and sold as lower clocked/less L2 cache CPUs. typicly the ones that barely work/ need "fine tuning" to make them sell-able, the budget chips, are released several weeks after the premier. so you may have the die of a 6600 in a 4300 chip but one or more part(s) dont work. there is no way of telling the effect of these other flaws on overclock-ability, but it appears that alot of people cant get very much out of them (though a few get lucky).
 
Thanks for info guys. To answer the guy who asked, I dont have a mobo yet. I plan on getting the Gigabyte DS3 though. Its either that or the ASUS P5B-E or whatever.

Yeah its confusing. I know price cuts are coming also. Ill prolly end up getting a E6300. Seems to be the cheapest. I am really only looking for around 3.2ghz anyways.
 
from here this is the first thing that came up in google

well most likely; they are crippled. not on purpose by intel, but just during the mfg. process. when making these chips they dont produce different dies for each of the different core2's they make the x6800s and any dies that have flaws or parts dont work they get reduced in price accordingly and sold as lower clocked/less L2 cache CPUs. typicly the ones that barely work/ need "fine tuning" to make them sell-able, the budget chips, are released several weeks after the premier. so you may have the die of a 6600 in a 4300 chip but one or more part(s) dont work. there is no way of telling the effect of these other flaws on overclock-ability, but it appears that alot of people cant get very much out of them (though a few get lucky).


Im not a computer engineer in fact I dont know much about how a cpu is produced, but what you said here doesnt make too much sense.... If there is a flaw on a die thats made for a 6800, that die is garbage and wont be used in the lower priced cpu.... A flaw is a flaw, if it wont work on a 6800, it wont work for a 4300, to fix that same die to be reused for a 4300 will cost more than the die itself.... Im sure the X6800 gets the primo wafers but I doubt there are much difference between the 6600, 6400 and 6300 in terms of wafer quality.....

As for the E4300 being crippled, its crippled that it doesnt have VT and thats it.... set the bios at 266fsb and lower the multi to 7 and its a 6300, remember that thier stock speed is almost the same so theres no reason why an E6300 would require a better die..... I dont even think Intel bin the lower price cpu cos I would say 99.99% of C2D produced are capable of hitting 2.4gs ( e6600).....

As for best for the buck, nothing beats an E4300, if VT is not a factor, it should be a better buy than a 6400 even at the same price...... Its the same cpu ( Allendale ) and it has a 9x multi, people just dont want to buy them because of Ignorance and the fact that its the lowest ( cheapest ) cpu Intel offers......

Give me a E4300 and Ill show you 3700mhz......
 
From my experience with two E4300 CPUs, it seems to me they have problems working on high FSB frequencies. I haven't been able to run them on 400 MHz no matter what (while I was able to run E6300 on the same type of motherboard at 475+). Granted, it wasn't THE same motherboard, but they were both P5B Deluxe running the same BIOS version. The other E4300 was on Gigabyte DS4 motherboard, but not the latest revision. Most I could achieve was some 355 MHz, maybe more. Anybody else had similar experiences?
 
Just like any other model chip out there, it's the luck of the draw and the equipment used as to how well they will OC, and don't forget the end-users ability to get the most out of the core. I'm running the cheapest setup I could buy and I'm extremely happy with the 100% OC out of my E4300. Enough said, guess I'm lucky.



And a bit more memory tweaking.

 
All C2D overclocking is done with raising the FSB. On Intel boards the FSB affects both the memory and the CPU at the same time. Nvidia boards let you "unlink" the memory speed from the CPU, normally a good thing but not as good as you would first think as adding in wait states and buffers because "everyone is not paddling at the same speed" eats up some benefits.

As mentioned above 3.2 GHz 400FSB x 8 is a nice sweet spot for a good overclock anything more usually requires a lot of attention to detail. a good power supply and superior cooling of everything concerned, chipset chips too.

4300 or 6400 due to their higher multipliers, the 7x mulitplier of the 6300 will get you to 2.8MHz and then you start overclocking your memory. the others will do 8x400 = 3.2GHz and your memory is running at stock speeds. This assumes PC6400(800MHZ - 400FSB x 2) memory.

Look at my sig, 6300 requires extremely high FSB to get to 3.4MHZ and you probally will not make it unless you get very lucky with the board.
 
Im not a computer engineer in fact I dont know much about how a cpu is produced, but what you said here doesnt make too much sense.... If there is a flaw on a die thats made for a 6800, that die is garbage and wont be used in the lower priced cpu.... A flaw is a flaw, if it wont work on a 6800, it wont work for a 4300, to fix that same die to be reused for a 4300 will cost more than the die itself.... Im sure the X6800 gets the primo wafers but I doubt there are much difference between the 6600, 6400 and 6300 in terms of wafer quality.....

As for the E4300 being crippled, its crippled that it doesnt have VT and thats it.... set the bios at 266fsb and lower the multi to 7 and its a 6300, remember that thier stock speed is almost the same so theres no reason why an E6300 would require a better die..... I dont even think Intel bin the lower price cpu cos I would say 99.99% of C2D produced are capable of hitting 2.4gs ( e6600).....

As for best for the buck, nothing beats an E4300, if VT is not a factor, it should be a better buy than a 6400 even at the same price...... Its the same cpu ( Allendale ) and it has a 9x multi, people just dont want to buy them because of Ignorance and the fact that its the lowest ( cheapest ) cpu Intel offers......

Give me a E4300 and Ill show you 3700mhz......



i'll back out of a complete agreement after reading your final fantastic claim, but you're right, VIIV, default FSB and -2MB cache are the only differences. of course not every chip will hit 3 and change, which is where some of the disappointment with this 'budget' processor lies. heh.

as far as any temperature concerns.... ?

i'm quite happy with my e4300 hummin along @ 3.0. SuperPi 1M @ 19s says about the same as my performance in supreme commander.. well, on a system performance basis anyways :p

overall, you get what you pay for, +5 skill magic if you roll 18.
 
If there is a flaw on a die thats made for a 6800, that die is garbage and wont be used in the lower priced cpu.... A flaw is a flaw, if it wont work on a 6800, it wont work for a 4300

well i was simplifying the whole thing, if it wont work for a 6800 doesnt mean it wont work in a 4300.
they are usualy made in batches, hence the code names allendale, conroe, etc. most of the time, all conroes are the same desing, all allendales are the same design, depending on how many options they are making. they take one design, print it on a wafer (several dies per wafer), then test each die. if one fails completely its thrown out, or remelted. if, say, they are making 6600s, and they find a die that has a tiny crack/flaw only in a section of L2 cache making half of the cache fail the test, they can somewhat easily reprogram it to not use that part, leaving 2mb instead, but the rest of the chip is fully functional. this chip would be marketed as a 6400 instead of a 6600. they set some chips with a lower multiplier/fsb if they test it and find its not stable at such and such a speed/voltage.
so technicaly an underclocked 6300 is the same as a 4300 but the 4300 is a budget chip, it almost definately has more flaws, where the flaws are? we have no idea and cannot judge the effects of these on overclock-ability. they passed inspection for running at the rated speed only.
its a little more complicated then that but that is the general idea. i found a great site about this but i cant find it again :(
 
Back
Top