• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Manual ram overclock

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Infinite66

Member
Joined
Jun 27, 2012
Hey everybody. I currently have a 2x4gb kit of corsair lp ram at 1600mhz. 9-9-9-24 timing. Got the kit before the whole factory fore that drove ddr3 prices through the roof. I say this because now a kit of the same ram on newegg is CL10, not 9. Does this mean my chips are of higher quality? Does this also mean I can overclock this set to say the stock settings of the corsair lp kit of 2133 mhz ram? I know faster ram is just generally better binned chips, but is their build quality any different? Would throwing 1.65v at my ram kill it immediately or is ram overclocking the same as cpu clocking? All down to how good the chip is and silicon lottery?
 
Infinite what makes you thing Cl 10 IC sticks are better quality then CL 9?
Would throwing 1.65v at my ram kill it immediately
Most likely no.

is ram overclocking the same as cpu clocking? All down to how good the chip is and silicon lottery?
Yes, it's is usually dependent on the Ic manufacturer of the sticks.
 
Infinite what makes you thing Cl 10 IC sticks are better quality then CL 9?
Most likely no.

Yes, it's is usually dependent on the Ic manufacturer of the sticks.


Im sorry. I mistyped. The CL10 sticks from newegg are the "same" kit as the one I have. Corsair vengeance lp 1600mhz. But the kit I have no longer exists apparently. (CL9). I know the CL9 would be better. What I was asking is the fact that my kit of corsair lp is CL9 compared to the now sold CL10, does that make my kit superior in any way? OC headroom etc.
 
It may, it all depends on the the IC manufacturer. Memory is tough to judge because you can have two exact same kits and they could have different IC's in them. So they could both be rated at say DDR3 1600 9-9-9-28 timings but one could be Hynix and the other could be Elphida. Therefore OC completely different. Here's a link to Woomack's memory tested list, you may find interesting. Also remember different setups will oc the same ram differently. So for example I have a set of DDR3 2133 7-10-7 G. Skill Ripjaws that I use for benching. They will not run the same timings at the same Mhz on my Fx as opposed to my 4770k.
 
It may, it all depends on the the IC manufacturer. Memory is tough to judge because you can have two exact same kits and they could have different IC's in them. So they could both be rated at say DDR3 1600 9-9-9-28 timings but one could be Hynix and the other could be Elphida. Therefore OC completely different. Here's a link to Woomack's memory tested list, you may find interesting. Also remember different setups will oc the same ram differently. So for example I have a set of DDR3 2133 7-10-7 G. Skill Ripjaws that I use for benching. They will not run the same timings at the same Mhz on my Fx as opposed to my 4770k.

I see. So the architecture of the processor also effects ram timing and speed stability. Hmm. That is something to consider. Only way to figure this out is to sit down and play with it. Maybe this weekend.

As a starting point, would you say it's good to shoot for the speed and timing of faster ram from the same manufacturer? (Corsair). Or just change everything manually one at a time from the get go?
 
if i remember correctly, they are actually 1333 sticks with a xmp overclock to 1600... not sure if that makes a difference though.
It doesn't. There is/was a member here that seemed to push a particular brand of sticks because they were superior than '1333Mhz stick overclocked' or some jazz... he couldn't prove a thing (no supporting links, just speculation... I also talked with a Kingston rep at CES who also bunked that theory).

The reality of the situation is that if the sticker says it runs at XXXX with X timings, that is what it does. I could care less if its a turbocharged 4 cylinder (1333 sticks overclocked) or a V8. So long as I get the rated speeds, timings and subsequent performance, that is all that matters.
 
The reality of the situation is that if the sticker says it runs at XXXX with X timings, that is what it does. I could care less if its a turbocharged 4 cylinder (1333 sticks overclocked) or a V8. So long as I get the rated speeds, timings and subsequent performance, that is all that matters.
LOL heck of a reference ED :D
 
lol haha... my analogies are full of fail...

IRL, speaking of cars, Im a V8 guy on an I4 budget (rather have the V8!!!). :p
 
As a starting point, would you say it's good to shoot for the speed and timing of faster ram from the same manufacturer? (Corsair). Or just change everything manually one at a time from the get go?

Better to start without an expectation. You had the right idea earlier. Treat it just as you would a CPU and go slowly, starting with your kit's rated speed/timings.
 
It doesn't. There is/was a member here that seemed to push a particular brand of sticks because they were superior than '1333Mhz stick overclocked' or some jazz... he couldn't prove a thing (no supporting links, just speculation... I also talked with a Kingston rep at CES who also bunked that theory).

The reality of the situation is that if the sticker says it runs at XXXX with X timings, that is what it does. I could care less if its a turbocharged 4 cylinder (1333 sticks overclocked) or a V8. So long as I get the rated speeds, timings and subsequent performance, that is all that matters.



unless you want to overclock more than rated i would think.

i'm running 16gb of ram.
8gb of the vengence
8gb of g.skill sniper


if i wanted to run them at 1866...

1333 > 1866 is a lot more of an overclock than 1600 > 1866

don't you think that would make a difference?
 
Nope. What the sticker says it what it should do. Period. As far as overclocking headroom past what its rated, that is a silicon lottery crap shoot as well regardless if it starts off at 1333 or 1600. Remember they only bin the sticks to batches. Some could reach higher batches that are placed in lower...

I can be convinced... however that person did not support their sentiment though the logic was there.
 
I can be convinced... however that person did not support their sentiment though the logic was there.

Like Faraday, just couldn't do the math . But seriously, it all depends on the chips that are put on the PCB and some binning. I have some G.Skill Flare that run 1800 7-8-7-24 by the sticker. This ram is already binned with "better" CHips. It'll also run 2400 9-10-9 26 on stock voltage. Each brand has different traits and some of them overclock better than others, even in the same line.
The CL10 the OP mentioned reminds me of my sniper 2133 kit . All timings are CL11 something from 1600 to 2133. But the ram itself will run 7-9-8 at 1800 and 10-12-11 at 2500.
In the end what I'm saying is you don't really know what a set of ram "will" do until you try !
 
Don't forget that OCing RAM is usually finiky. Run memtest to ensure it's stable after your adjustments.

In reality, most people won't notice the difference when they overclock RAM. Primarily only for benching is overclocking RAM.
That's not to discourage you of course, have fun with it! :thup:
 
Don't forget that OCing RAM is usually finiky. Run memtest to ensure it's stable after your adjustments.

In reality, most people won't notice the difference when they overclock RAM. Primarily only for benching is overclocking RAM.
That's not to discourage you of course, have fun with it! :thup:

Yea. I'll treat it the same as my cpu overclocking. Stability tests every speed bump. And yea. I know the faster ram won't get me crazy performance or anything, I'm just trying to see if it's even worth it it to buy an $84 kit of 2400 mhz ram if I can get my current kit to say, 2133 at CL10 or better.
 
No noticeable performance in any application, really. Even if you're stuck at 1600 c9, I wouldn't worry about upgrading.
 
If you can OC it great! If not oh well really...

Now, IIRC, with Intel chips = tighter timings are better, and if AMD = higher clock speeds.
Either that or the other way around.
 
Actually on both platforms count balance between high clock and tight timings. On AMD you push till 1866-2400 and then try tight timings. On Intel ( IB/HW ) you go for 2133-2666 and then try to tighten the timings. All depends for what you need it and how good memory you have. Most AMD FX memory controllers won't run stable at 2400+ while most Intel IB/HW will run without problems at least up to 2666 and haswells even higher but at 2666+ you have to use more relaxed timings so overclocking on air rarely has any point at these higher clocks.
 
Back