• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Do you gain any Frames?

Z

zod96

Suspected BAD TRADER
Joined
Feb 23, 2008
Messages
2,709
How much of a gain do you usually get from overclocking your video card and cpu? I've overclocked my CPU to 4 ghz and I also have overclocked my 6950 to 840/1300
 
I honestly find little appeal in OC'ing most video cards. 5 fps extra seems to be a norm for most cards, not worth the heat and stress in my opinion. Unlocking shaders and the like? Totally worthwhile.
I will however, push my CPU and OC that, as it benefits far more than just my games.
 
this is not always a simple answer. results can vary depending on the hardware and games and corresponding settings being used.

its funny that the two replies here will oc their cpu but think there's little reason to oc their gpu. I guarantee you I can oc my gtx570 by 15% and see a direct gain of 15% in nearly every case at typical settings most people would use for this type card. I could oc my 2500k by 30% and likely would see almost zero difference at those same settings.

overclocking my gtx260 by 20% kept me playing games at decent settings much longer than stock clocks. of course I needed to oc my E8500 too for those cpu demanding games like GTA 4.

as for the OP, your 6950 will clearly be the limitation even with a 2600k at stock unless you are running games at 1024x768.
 
In some games bumping my 460 to 800 mhz from its factory oc of 763 mhz will net me 15-20 fps. With metro 2033 its the difference between playing or not
 
herfalerf said:
In some games bumping my 460 to 800 mhz from its factory oc of 763 mhz will net me 15-20 fps. With metro 2033 its the difference between playing or not
Click to expand...
if you are getting a 15-20 fps increase from a 19% overclock then you were already getting pretty close to 100 fps.
 
It's entirely dependent on the limiting factor. If it's your CPU holding you back in games, then overclocking the CPU will get you results, not the graphics card. If the CPU is plenty powerful enough (such as cannondale's 2500K), then it's all down to the GPU - not all graphics cards overclock well, sometimes you'll only get 5% extra which is a bit of a waste of time. If you happen to have a card that overclocks well, however (usually the cards that are one step-down from the top of a class) then sure, why not? In rare cases you can see 25% performance increases from doing it, which is definitely tangible.
 
would a phenom ii 965 3.4ghz(not overclocked) be a bottleneck for say a 6950 or 6970 in crossfire?
 
If the game isn't CPU intensive then all you need to do is break the CPU bottleneck speed. I found on the C2D and the Core I7 this was at the 3.6ghz mark. Anything after that didn't gain anything unless the game stressed the CPU more. Only game I played that saw any sort of benifit was WOW. But alot of strategy games can be cpu intensive as well.

As for GPU overclocking, it's not worth it like everyone thinks. It can help in some games but won't help you in a game where you're only getting 25-45 fps on average. Even if you overclock 15%, it doesn't necessarily mean it's going to output 15% more fps which doesn't add to much when you're getting that low of frames anyways. If a game is hangin around 50-55 fps then you can probably hit the 60 fps smoothness more easily when overclocking. That's about it.
 
MendeZ said:
would a phenom ii 965 3.4ghz(not overclocked) be a bottleneck for say a 6950 or 6970 in crossfire?
Click to expand...

Depends on the game, but for now, very rarely. At the moment only strategy games like Shogun 2: Total War are a bottleneck with a CPU like that. The only exception to this rule is people who run 120Hz monitors, but not in 3D mode (so they use 120fps instead of 60).
 
cannondale06 said:
its funny that the two replies here will oc their cpu but think there's little reason to oc their gpu. I guarantee you I can oc my gtx570 by 15% and see a direct gain of 15% in nearly every case at typical settings most people would use for this type card. I could oc my 2500k by 30% and likely would see almost zero difference at those same settings.
Click to expand...
I simply point you to the last review done by [H], regarding the MSI 6950 1GB Twin Frozr III PE, right here. Which concludes with the fact that OC'ing the card had no tangible benefits. You see this fairly often with cards, and like I said earlier, or you see an increase of a few frames. To mention your fps increasing by "15%, imo, is incredibly deceiving if you do not detail the original fps and OC fps numbers themselves. In fact, its the opposite of being helpful when you do this.
I stand by my original post.
 
Stating 15% is fine. It's when people say they got an extra 15fps that it's meaningless.
 
I disagree, you should be stating both. And stating what the original fps were to begin with. Excluding other sets of data that compliment each other, is a hindrance.
For instance:
To claim that, "GUYS I WAS ABLE TO OC THIS CARD AND HAD AN INCREASE OF 20% FPS IN RAGOCANOBLASTER!!!" is stupid, when in reality, the 20% increase sent RAGOCANOBLASTER from an average of 10 fps to 12 fps. It's still an unplayable POS.
 
DOOMHAMMA said:
I simply point you to the last review done by [H], regarding the MSI 6950 1GB Twin Frozr III PE, right here. Which concludes with the fact that OC'ing the card had no tangible benefits. You see this fairly often with cards, and like I said earlier, or you see an increase of a few frames. To mention your fps increasing by "15%, imo, is incredibly deceiving if you do not detail the original fps and OC fps numbers themselves. In fact, its the opposite of being helpful when you do this.
I stand by my original post.
Click to expand...
on average Nvidia usually scales better with overclocking. again in ALL my experience overclocking the gpu has most certainly helped in games. again it helps some game be more playable or allows higher settings to be used.

overclocking the cpu does almost nothing if you have a high end quad core cpu and are running a single gpu. its the gpu that will be the limitation at realistic settings in 99.9999% of cases.
 
When hexus reviewed the GTX570 they overclocked the core by 9% and the memory by 6% and got performance increases of 12% and 14% in the two games they tested. The Fermi cards scale oddly with clock speed increases, that's for sure.
 
yeah my 6600 and 8600gt cards scaled unbelievably good. my 8600gt could pick up 20% with 20% increase on core and another 20% on top of that with 20% on memory in a game like FEAR. a whopping 40% gain is massive for nothing more than a simple oc that did not even touch voltage or impact noise levels.
 
cannondale06 said:
on average Nvidia usually scales better with overclocking. again in ALL my experience overclocking the gpu has most certainly helped in games. again it helps some game be more playable or allows higher settings to be used.

overclocking the cpu does almost nothing if you have a high end quad core cpu and are running a single gpu. its the gpu that will be the limitation at realistic settings in 99.9999% of cases.
Click to expand...

I fall into the not having a high end quad core. I actually saw tangible improvements from my OC of 2.6 stock to 3.4.

The benefit? The swirly red ball in the opening of Dead Island quit stuttering. (quite so much)

:D
 
DOOMHAMMA said:
I simply point you to the last review done by [H], regarding the MSI 6950 1GB Twin Frozr III PE, right here. Which concludes with the fact that OC'ing the card had no tangible benefits. You see this fairly often with cards, and like I said earlier, or you see an increase of a few frames. To mention your fps increasing by "15%, imo, is incredibly deceiving if you do not detail the original fps and OC fps numbers themselves. In fact, its the opposite of being helpful when you do this.
I stand by my original post.
Click to expand...

Of course they were not able to get any tangible benefit from that OC the card they reviewed was a dud and would only OC 5% and 5% is not much of a difference.

that being said in my expierence (having owned both) overclocking AMD cards is usually not worth it as they tend not to OC well (usually 10% at most) and don't scale 100% (so you usually get a 5% performance increase at best) but Oc'ing Nvidia GPU's is usually well worth it as I have never owned an nvidia gpu that wouldn't OC at least 20% and Nvidia GPU's tend to scale close to 100% with OC.

That's not saying one brand is better than the other just that one tends to see more tangible benefits from overclocking.

as for why this is my guess would be that to get the best performance per dollar AMD tends to run their cards out of the box as fast as they will go, where as nvidia for whatever reason chooses not to.
 
The quality of manufacturing with geforces isn't as high as it is with Radeons (by this I mean the GPUs themselves, not the rest of the card) - so I think nvidia allow for a lot more tolerance to prevent sending out loads of unstable cards at the stock specs.
 
You must log in or register to reply here.
Back
Top