TPU's Core i7 Overclocking and Feedback

[msgclb]

Guy's, Is it better to run ddr3 1333 or 1600? The 1333 has tighter timings or will the 1600 's speed make up for having more latency? I've got to make a decision tonight.

I actually don't know! I seem to always go for the 1600 or higher. I've seen several articles such as this one. Maybe that will help.

LGA 1156 Memory Performance: What Speed DDR3 Should You Buy?

 

[mastrdrver]

Guy's, Is it better to run ddr3 1333 or 1600? The 1333 has tighter timings or will the 1600 's speed make up for having more latency? I've got to make a decision tonight.

Madshrimps did a nice over look on memory speed on the i7 and came to this conclusion:

Madshrimps

Last in row would be overclocking the memory. Quite frankly, after having reviewed the figures I couldn't believe the memory has so little effect on the performance; only in the memory-related benchmarks such as Lavalys Everest you can really see a difference. But in any other 24/7 application 1400CL9 is almost as fast as 2000CL7. To back me up, I use a quote from a X58 memory round-up in which 8 memory kits were tested.

"... suppose you got an average performing 1333 kit (almost all do cas 7 nowadays), and compare it to the best performing midrange kit. The average performance increase you can expect from your system is 3,59%. I’ll let it up to you to decide whether this is a worthwhile performance increase, it is certainly something you will not be noticing in your day to day pc tasks (you need about 10% speed difference to really note a difference). The conclusion of today must therefore be that any of the tested kits will do fine in your i7 rig, of course, if you do want the last bit of performance, and if you’re an avid benchmark enthusiast (like myself), just go on and pick up that upper midrange kit you’ve been drooling on …"
(~ "X58 Triple Channel DDR3 Memory Roundup! 8 Mid-range Kits Tested"

Again, it's very simple: if you don't want the absolute maximum performance, it's just not worth looking at memory kits in the higher regions of the pricing charts; for most, if not all, daily applications you'll see no difference whatsoever!

 

[Bo_Fox]

Guy's, Is it better to run ddr3 1333 or 1600? The 1333 has tighter timings or will the 1600 's speed make up for having more latency? I've got to make a decision tonight.

Most of the time, if the memory can do 1600, it's usually worth it over 1333 at just a little bit tighter timings.

Most of the DDR3 memory cannot do 1333 at super-low timings of say, CAS 5-5-5 and then only be able to do 1600 at like 8-8-8.

Some DDR3 memory can do 1333 at decent timings at like CAS 7-7-7 and not be able to do 1600 at all, not even at 9-9-9 (just don't try CAS 10 or higher.. I'd stay away from latencies that high).

Overall, I think that 1600 at 9-9-9 is just as good or better than 1333 at 7-7-7. A wider range of memory clocks also allows you greater flexibility in overclocking the CPU with different multipliers.

 

[n-ster]

I thinlk that the soft spot is 1600 cas 8, so get 1600 cas 9 and OC it to cas 8

 

[phanbuey]

I thinlk that the soft spot is 1600 cas 8, so get 1600 cas 9 and OC it to cas 8

Yeah i did that... my 1600 CAS 9 at 1.5v will bluescreen at 1600 CAS 8 1.65v GeIL sticks.

Could have gotten DDR3 2133 for $35 more, but didnt think it was worth it

In my case 1600 CAS 9 is faster than 1333 CAS 7. i5 at 4.2ghz

 

[AddSub]

Wow, I haven't posted in here since page 54. This winter has been brutally cold so far and I've decided to take the advantage of potential low ambient temps and try to break the 4.6GHz barrier on my i7 920 + EVGA X58 setup. I'm aiming for SuperPi stability but the biggest roadblock is the fact that I don't want to run my VTT above 1.325V for any extended periods of time. I've seen lot of folks here and elsewhere (EVGA forums, Overclock.Net, etc) that have been running their VTT at 1.4V or more for benchmarking purposes. What I want to know, has anyone experienced any issues, hardware failures, or performance/OC degradation when it comes to running high VTT voltages over extended periods of time and I don't mean just a few hours to run a few benchmarks? (months?)

Oh yeah, and this exchange here:

Chicken Patty: I don't have an i7 anymore
mudkip: why not?
Chicken Patty: Traded it for a Phenom II 965 setup.

Please tell me you had financial issues or you were forced at gunpoint into downgrading or something along those lines, because otherwise... wow.

 

[BUCK NASTY]

Thanks for the info guy's. I'm going with 1600 and I ordered 8GB of this. I've never tried AData before, but there's a first time for everything. Plus they are black to match the P55 Sli FTW.

 

[Chicken Patty]

Wow, I haven't posted in here since page 54. This winter has been brutally cold so far and I've decided to take the advantage of potential low ambient temps and try to break the 4.6GHz barrier on my i7 920 + EVGA X58 setup. I'm aiming for SuperPi stability but the biggest roadblock is the fact that I don't want to run my VTT above 1.325V for any extended periods of time. I've seen lot of folks here and elsewhere (EVGA forums, Overclock.Net, etc) that have been running their VTT at 1.4V or more for benchmarking purposes. What I want to know, has anyone experienced any issues, hardware failures, or performance/OC degradation when it comes to running high VTT voltages over extended periods of time and I don't mean just a few hours to run a few benchmarks? (months?)

Oh yeah, and this exchange here:



Please tell me you had financial issues or you were forced at gunpoint into downgrading or something along those lines, because otherwise... wow.

No financial issues. Just wanted to try something new. This point of view from people is just sickening. What is so wrong about that? That I can't get super high benches like I did with my i7? big deal. Sorry if it sounds a bit rough, but seriously. The Phenom II is one heck of a damn CPU. I still love i7 to death, but I had my fun with it. Time to try new stuff.

Intel has the gulftowns coming out soon, the EVGA dual socket board also if it's not out already. I like trying new things. (computer related, just specifying for the pervs LOL)

 

[PaulieG]

Wow, I haven't posted in here since page 54. This winter has been brutally cold so far and I've decided to take the advantage of potential low ambient temps and try to break the 4.6GHz barrier on my i7 920 + EVGA X58 setup. I'm aiming for SuperPi stability but the biggest roadblock is the fact that I don't want to run my VTT above 1.325V for any extended periods of time. I've seen lot of folks here and elsewhere (EVGA forums, Overclock.Net, etc) that have been running their VTT at 1.4V or more for benchmarking purposes. What I want to know, has anyone experienced any issues, hardware failures, or performance/OC degradation when it comes to running high VTT voltages over extended periods of time and I don't mean just a few hours to run a few benchmarks? (months?)

Oh yeah, and this exchange here:



Please tell me you had financial issues or you were forced at gunpoint into downgrading or something along those lines, because otherwise... wow.

I still have a couple of i7's, but I did sell one to buy a AMD rig. As CP said, it's for something new. Honestly, after overclocking 11 920's and several 860's I've just become board with the platforms. So, it's back to AMD to play a bit. They are more of a challenge to overclock, which is what I'm looking for. That is the hobby for some of us.

 

[Chicken Patty]

I still have a couple of i7's, but I did sell one to buy a AMD rig. As CP said, it's for something new. Honestly, after overclocking 11 920's and several 860's I've just become board with the platforms. So, it's back to AMD to play a bit. They are more of a challenge to overclock, which is what I'm looking for. That is the hobby for some of us.

really fun if you ask me.

Although like I said in another thread, I feel like my penis shrunk by more than half

 

[BUCK NASTY]

After running an E8400 and then a Q9550 i sorta went back to AMD. I found more excitement in unlocking AMD cores(4 cpu's so far) and seeing how far I could push them. Nice thing is the AMD platforms are relatively inexpensive, so I have 2 identical builds.

 

[trt740]

Hey fellas does QPI link speed make a giant difference and north bridge frequency or is clock speed more important whats your take.

 

[Fitseries3]

yes.

if you do a cpu benchmark at say 4.2ghz with high QPI then do another with slower QPI you'll see a good difference.

ive seen up to 3000pt difference in 3dm vantage.

high QPI is key to better performance but low qpi will give you better stability and is generally easier to get stable.


Interesting fact of the day: X58 NB chip is a neutered 5520 server NB with 1 of the 2 QPI links disabled.

 

[mudkip]

I miss my i7 so fucking much

 

[Chicken Patty]

I miss my i7 so fucking much

We are on the same boat. I don't regret the trade, but I do miss it.

 

[FilipM]

Last 2 days ive tried everything to get this dam thing at 3.6 - even went up to 1.19V (lol) but it didnt help. IOH voltage, QPI, ICH, nothing. And today i started with the bios clean and stable on prime for an hour (just brief, i will test more) at 1.168V

Multiplyer - 21
BCLK - 171
DDR 1371MHz - 7-8-7-21-1T
UCLK - 2913 (+1 click of that atm)
QPI - 6171

---------------------------------------------

CPU - 1.17500V (1.22500 is stock i think)
QPI - 1.20V
PLL - 1.80V
IOH - Auto
IOH PCIE - 1.52V
ICH - Auto
ICH PCIE - 1.50V
DRAM - 1.64V

----------------------------------

Loadline Calibration - Enabled

Spread spectrums - Disabled


Temps are: 77 - 75 - 74 - 72
- mind you, im on stock cooling



Is that any good?

 

[mastrdrver]

What are you using for stressing?

BTW, stock cooler?! I wouldn't bother with overclocking until you get something else.

Also, do you know which i7 you have (see link in sig)? It will make things a ton easier if you know this.

 

[TheShad0W]

Yeah i did that... my 1600 CAS 9 at 1.5v will bluescreen at 1600 CAS 8 1.65v GeIL sticks.

Could have gotten DDR3 2133 for $35 more, but didnt think it was worth it

In my case 1600 CAS 9 is faster than 1333 CAS 7. i5 at 4.2ghz

I got 8GB of the DDR3 2133 ram for £130 (~$200) - The first set I got was better than the second and did 6-7-7-18 @ 1600MHZ @1.7v no trouble at all. Now running 7-7-7-21 @1.68v with all four sticks in. it's pretty decent ram I'd say.

 

[TheShad0W]

Last 2 days ive tried everything to get this dam thing at 3.6 - even went up to 1.19V (lol) but it didnt help. IOH voltage, QPI, ICH, nothing. And today i started with the bios clean and stable on prime for an hour (just brief, i will test more) at 1.168V

Multiplyer - 21
BCLK - 171
DDR 1371MHz - 7-8-7-21-1T
UCLK - 2913 (+1 click of that atm)
QPI - 6171

---------------------------------------------

CPU - 1.17500V (1.22500 is stock i think)
QPI - 1.20V
PLL - 1.80V
IOH - Auto
IOH PCIE - 1.52V
ICH - Auto
ICH PCIE - 1.50V
DRAM - 1.64V

----------------------------------

Loadline Calibration - Enabled

Spread spectrums - Disabled


Temps are: 77 - 75 - 74 - 72
- mind you, im on stock cooling



Is that any good?

Get a decent cooler on there and you'll knock an awful lot off those temps - and that's not a bad vcore for that frequency, so with the right cooling that's definitely a chip with potential

Also I'd test with OCCT rather than prime95 - I've had systems stable for HOURS in prime95 which couldn't handle 5 minutes of OCCT linpack.

 

[Chicken Patty]

Filip I say stay at that clock because of your temps. To me 80 is safe max temp. To go higher you'll have to get closer the 1.2v vcore mark and that'll cause your temps to go over 80. That's a nice clock on stock cooler

 

[FilipM]

@masterdrver, i cannot really tell, im thinking type 4 would suit the most?

I cannot afford OCCT or Intel Burn test atm just because i dont want to go over 80 deg. I might do a Intelburn test, 5 to 10 passes on standard level, just to see if it is stable.

I had the VTT high, like 1.3, 1.35, 1.28, nothing worked then i read somewhere that acutally 0.48V off the Dram voltage works wonders, so set as low as it goes (1.2V) and it proved stable. But also, thats the Ram downclocked but at tighter timings (1600 @ 9-9-9-24-1T originally). Also, low VTT knocks heat off the CPU - previously i was at those temps on 3.4Ghz @ 1.128V.


For now, 3.6Ghz is good, considering the cooling i have. Im still new to this and some times it's like :/ - lol



BATCH: 3927A448 - Manufactured 09/24/09 (Costa Rica)

 

[brandonwh64]

filipm you need cooling like mine LOLOLOL

 

[Fitseries3]

is that a colostomy bag next to your rig?

 

[FilipM]

Na, I'll use the fridge instead, LOL.

 

[Fitseries3]

Na, I'll use the fridge instead, LOL.

a PC would burn up inside a fridge due to the fact that the fridge is not designed to pull heat away as fast as the PC gives heat off.