E8400 & EP45-UD3R Overclocking question

Jensenfc

Member
Jul 27, 2003
67
0
0
I ordered a e8400 to upgrade from my e6400 CPU. Anyways while I wait for the new CPU to arrive I thought I would try to figure out the best way to overclock the e8400 to 4Ghz.

My system:
Motherboard: Gigabyte GA-EP45-UD3R
CPU: Intel E8400
RAM: 2x2GB OCZ Reaper PC2-6400
Power Supply: Corsair HX520w

You have to excuse me but I am a noob at overclocking and need some advice. From what I was reading from the reviews on newegg, my ram should be able to run at 1066 @ 2.1v with 5-5-5-15 or 18 timings. Some reviews on newegg also say that the e8400 should be able to do 7.5x mutiplier with 533Mhz fsb (4ghz) at stock voltage. This would make the cpu ram 1:1. Does anyone know if this should work? From what I have read if I keep the multiplier at 9x I would have to increase the Vcore to around 1.4v. Is it better to do that since I wouldn't have to crank up for ram so much?

Anyone have a better way for me to get this cpu to around 4ghz?

Thanks!
 

fffblackmage

Platinum Member
Dec 28, 2007
2,548
0
76
I think you may have to increase voltages either way. I personally think it's better not to stress the ram too much. You can get to 3.6GHz with the FSB at 400MHz, which means stock ram speeds for 1:1. You could then overclock ram and fsb up to around 445MHz. That'll get you 4.0GHz and you wouldn't have to OC the ram so much.

Of course, having the ram and fsb at 1066MHz/533MHz would mean higher memory bandwidth, but I don't think there's a significant gain doing that. Well, someone can correct me if I'm wrong.
 

polarbear6

Golden Member
Jul 14, 2008
1,161
1
0
Originally posted by: fffblackmage
I think you may have to increase voltages either way. I personally think it's better not to stress the ram too much. You can get to 3.6GHz with the FSB at 400MHz, which means stock ram speeds for 1:1. You could then overclock ram and fsb up to around 445MHz. That'll get you 4.0GHz and you wouldn't have to OC the ram so much.

Of course, having the ram and fsb at 1066MHz/533MHz would mean higher memory bandwidth, but I don't think there's a significant gain doing that. Well, someone can correct me if I'm wrong.

Well think about it, Your stressing the North bridge instead of your processor. If your ram can hit that high its ok.I'm also planning on making my processor a 4.0 ghz one.
I wish i could know whether my ram can handle that much of fsb
 

fffblackmage

Platinum Member
Dec 28, 2007
2,548
0
76
Originally posted by: ganesh1
Originally posted by: fffblackmage
I think you may have to increase voltages either way. I personally think it's better not to stress the ram too much. You can get to 3.6GHz with the FSB at 400MHz, which means stock ram speeds for 1:1. You could then overclock ram and fsb up to around 445MHz. That'll get you 4.0GHz and you wouldn't have to OC the ram so much.

Of course, having the ram and fsb at 1066MHz/533MHz would mean higher memory bandwidth, but I don't think there's a significant gain doing that. Well, someone can correct me if I'm wrong.

Well think about it, Your stressing the North bridge instead of your processor. If your ram can hit that high its ok.I'm also planning on making my processor a 4.0 ghz one.
I wish i could know whether my ram can handle that much of fsb

How does setting the FSB at 450-ish MHz be stressing the north bridge more? I'm pretty sure the FSB at 533MHz is much more stressful for the NB, which is why I would recommend leaving the multiplier at 9x rather than decrease it to 7.5x. My reasoning is that higher memory bandwidth past 400MHz fsb/DDR2-800 in dual channel offer insignificant gains and could possibly shorten the life expectancy of the mobo significantly. (Or did I misunderstand something?)

Either way, the processor will be running at 4GHz, but stress on the north bridge/ram would be different.
 

polarbear6

Golden Member
Jul 14, 2008
1,161
1
0
Originally posted by: fffblackmage
Originally posted by: ganesh1
Originally posted by: fffblackmage
I think you may have to increase voltages either way. I personally think it's better not to stress the ram too much. You can get to 3.6GHz with the FSB at 400MHz, which means stock ram speeds for 1:1. You could then overclock ram and fsb up to around 445MHz. That'll get you 4.0GHz and you wouldn't have to OC the ram so much.

Of course, having the ram and fsb at 1066MHz/533MHz would mean higher memory bandwidth, but I don't think there's a significant gain doing that. Well, someone can correct me if I'm wrong.

Well think about it, Your stressing the North bridge instead of your processor. If your ram can hit that high its ok.I'm also planning on making my processor a 4.0 ghz one.
I wish i could know whether my ram can handle that much of fsb

How does setting the FSB at 450-ish MHz be stressing the north bridge more? I'm pretty sure the FSB at 533MHz is much more stressful for the NB, which is why I would recommend leaving the multiplier at 9x rather than decrease it to 7.5x. My reasoning is that higher memory bandwidth past 400MHz fsb/DDR2-800 in dual channel offer insignificant gains and could possibly shorten the life expectancy of the mobo significantly. (Or did I misunderstand something?)

Either way, the processor will be running at 4GHz, but stress on the north bridge/ram would be different.

When did i say setting the FSB at 450 will be more stressing. I just said 533 fsb is a bit of too much for the NB as quad pumping it would give us 2132 FSB. As far as i know P45 works at 1600 FSB, So its like 532 Mhz of over clocking the board.

Which could be disastrous for the board.
 

Jensenfc

Member
Jul 27, 2003
67
0
0
Wow, didn't mean to start a fight lol. So what I got out of this is that running it at 9x 450 would be slightly better for my ram but if my ram can take the 1066 speed then that would be fine too? From what I have read is that most ddr2-1066 ram is just ddr2-800 ram with looser timings (atleast this OCZ reaper stuff). I currently have my ram running at 4-4-4-12 so if i overclock them and loosen the timings it should be fine based on the reviews I read on newegg. Anyone who has overclocked at e8400 to 4ghz know how much to voltage needs to be increased with a 9x multiplier? How about 500mhz x 8? That way the ram isn't pushed so hard and maybe the vcore wouldn't need to be pushed so high.
 

polarbear6

Golden Member
Jul 14, 2008
1,161
1
0
Originally posted by: Jensenfc
Wow, didn't mean to start a fight lol. So what I got out of this is that running it at 9x 450 would be slightly better for my ram but if my ram can take the 1066 speed then that would be fine too? From what I have read is that most ddr2-1066 ram is just ddr2-800 ram with looser timings (atleast this OCZ reaper stuff). I currently have my ram running at 4-4-4-12 so if i overclock them and loosen the timings it should be fine based on the reviews I read on newegg. Anyone who has overclocked at e8400 to 4ghz know how much to voltage needs to be increased with a 9x multiplier? How about 500mhz x 8? That way the ram isn't pushed so hard and maybe the vcore wouldn't need to be pushed so high.

Well my processor is not even satisfied with 1.418v of vcore. So I'm running at 1.36v(thats the voltage where it at least makes through the post and boots into vista). Although i randomly keep getting BSOD. But i don't care for them and reset the settings.

 

fffblackmage

Platinum Member
Dec 28, 2007
2,548
0
76
Originally posted by: Jensenfc
Wow, didn't mean to start a fight lol. So what I got out of this is that running it at 9x 450 would be slightly better for my ram but if my ram can take the 1066 speed then that would be fine too? From what I have read is that most ddr2-1066 ram is just ddr2-800 ram with looser timings (atleast this OCZ reaper stuff). I currently have my ram running at 4-4-4-12 so if i overclock them and loosen the timings it should be fine based on the reviews I read on newegg. Anyone who has overclocked at e8400 to 4ghz know how much to voltage needs to be increased with a 9x multiplier? How about 500mhz x 8? That way the ram isn't pushed so hard and maybe the vcore wouldn't need to be pushed so high.

Oh no, it was just my fault for misunderstanding.

Jensenfc and ganesh1, sorry. orz

There's one thing I'm rather confused about though. You want to use a lower multiplier because you want to lower the required voltage, but either way, you want 4.0GHz. I don't understand why that might be so. I'm under the impression you'll need a certain amount of voltage at a frequency of 4.0GHz regardless of what the multiplier is at.
 

Jensenfc

Member
Jul 27, 2003
67
0
0
Well, I thought I read somewhere that you didn't have to bump up the voltage so high if you had your multiplier less. But as I said I have only overclocked once before so I know just the basics to it. I think I will try 445x9 @ ~1.368 volts to start out with the ram 1:1. I shouldn't have any issues with this even if I get a C0 chip. I may have to bump up the vcore a little more but as long as it is below 1.4 I should be good. As I said I probably wont get the CPU till late this week and just wanted to get a feel for what was the best option to get to 4ghz.
 

polarbear6

Golden Member
Jul 14, 2008
1,161
1
0
Well this is a classic debate of Multipliers OC vs FSB OC. Heres my take on this. Well according to Wikipedia(My bible), Multipliers maybe thought as the ratio of the Internal cycles: external cycles or how many internal cycles are done for external cycles. So a 9X multiplier with 400 mhz fsb is like 9 internal cycles are done to produce 400 mb of data calculation or processing for a second(correct me if iam wrong as its only my interpretation,This is how i understood the article), The thing which i dont understand here is whether it is 400 Mega Bits or Bytes of data. so its 9 cycles for 400mhz.Its like 9 trips with 400 cups to get 3600 liters of water"(again its my interpretation, Correct me if i am wrong). now you want to reduce it to 8 trips . Then it will be like 8 trips with 450 cups to get 3600 litres of water.
So the energy or volts required is almost the same or slightly higher in the lower multipliers case as you have to carry more glasses or more weight for each trip. but then again that extra weight carrying is almost equal or slightly higher than the walking done for the extra trip,The 9th trip.

Now lets talk about the NB and Vdimm voltages in terms of our cups of water example.

Another thing we have to see here is that for 9 multipliers 400 cups are required and for 8, 450 cups are required. For that extra 50 cups ( e = mc^2) some energy is required to produce them ( It's called uping the NB voltage).We even need to count how many cups of water e have filed for a cycle/trip And for counting the number of cups(I'm talking about ram voltage) of water taken in each trip more speed is required as earlier we were counting 400 cups and storing the number 400 in the memory Within 1 second(We have to go from storing 1 to 400 in ram within one second) now we have to do it for 450 cups and that also in one second.(going from storing number of cups from 1 to 450)Which requires more speed(as we are storing 450 entries in one second) and hence more voltage(as we have to increase the speed).


As this is totally my way of thinking it can be wrong. So please correct me if I'm wrong.



I think i will open a new topic for this discussion to get more views and comments.

So Roughly the 8 multipliers case and the 9 multipliers case is almost the same in terms of voltage(vcore only) according to the cups of water example i gave. But may require higher NB voltage and Vdimm voltage.

So according to my cups theory higher multipliers is preferred.Hell thats why we have black edition AMD and extreme edition intel. Higher multipliers are therefore always preferred.
Edited for clarity.

I'm praying that someone will read the crap that i posted as it took me a lot of nerve and creativity to post it
 

CoinOperatedBoy

Golden Member
Dec 11, 2008
1,809
0
76
tl;dr.

I have neither that CPU nor the mobo, but I would think that raising FSB to 533 would almost certainly require an increase in voltage to the northbridge, which would increase heat. The benefit is the ability to run the RAM 1:1, which may make a completely negligible difference. Only way to know is to try it. Just read the sticky on overclocking and be systematic. You shouldn't just get your CPU, pop it in, and crank the FSB to what you heard some guy on the Internet say was possible.
 

Jensenfc

Member
Jul 27, 2003
67
0
0
I got the CPU today from UPS and popped it in and fired it up and noticed I had a E0 stepping chip =D I started out @ 9x445Mhz vCore 1.36v with the ram 1:1 @ 5-5-5-15 2.1v. I ran prime95 for a half hour and it was fine. I kept lowing my voltage two steps at a time in the BIOS and I am now down to 1.26250v. I did try stock voltage (1.25v) before 1.2625v and it BSOD 5 min into prime95. Prime95 has been running for about an hour now and it hasn't crashed yet @ 1.26250v. I will let prime95 run for awhile then try orthos to make sure it is totally stable. I think I got a fairly good chip because I was expecting to have to raise the voltage up to ~1.37v. Now I got some playroom in the future to overclock it past 4Ghz. My temps are 39C in a room that is 79F with a OCZ Vendetta 2 cooler. Thanks again for everyones suggestions!
 
Nov 26, 2005
15,188
401
126
Run Small and Large FFT's instead.

I didn't but I was told it is a better trouble shooting way.......

Blend will be about 9 1/2hrs or more to do a full tilt back to the first set of iterations..
 

Jensenfc

Member
Jul 27, 2003
67
0
0
Thanks, I changed it from blend to small FFT for now. I kinda wanted to test the ram too since I did overclock it a little. The ram seems stable so now I will work on mostly the CPU. How long should I let it run on the small FFT and large FFT settings?
 
Nov 26, 2005
15,188
401
126
Originally posted by: Jensenfc
Thanks, I changed it from blend to small FFT for now. I kinda wanted to test the ram too since I did overclock it a little. The ram seems stable so now I will work on mostly the CPU. How long should I let it run on the small FFT and large FFT settings?

till it loops back to the first set of numbers

blend took me 9 1/2 or more...
 

Jensenfc

Member
Jul 27, 2003
67
0
0
Weird, the first time I ran the small test core 1 failed on the last test but it didn't BSOD. I ran the test 3 more rounds it hasn't failed again. I also ran the long test once and it was fine. O-well I better bump up the vCore one more step to be safe. I will move to orthos next and let that run for a bit.
 

Jensenfc

Member
Jul 27, 2003
67
0
0
I just looked and I saw MCH core @ 1.1v and ICH core @ 1.1v. I am assuming one of these is the NB. Why should I bump it up to 1.3v?
 
Nov 26, 2005
15,188
401
126
Originally posted by: Jensenfc
I just looked and I saw MCH core @ 1.1v and ICH core @ 1.1v. I am assuming one of these is the NB. Why should I bump it up to 1.3v?

Bump it up little by little... quads usually need 1.42 to 1.50 to hit 500fsb... a dualie surely needs less

Bumping it up usually stabilizes the fsb... as well as the VTT and PLL
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |