i7 Keep up With 285?

Forged results....rotf, I simply switched from 3ghz to 4ghz as per your request and you were proven wrong yet my results are forged according to you. :wanker:

Here, take a long long hard look.....I even circled it for you in case you missed it the first time.



Now compare those settings to these-




Only difference is clock speed, you can go :wanker: somewhere else now.

 

Personally I think something else is going on. What I don't know (perhaps some ram timings or speed?) - but the results you are showing going from 3GHz to 4GHz aren't the same as most other people doing the same thing testing the same games...

Something is different from what you are doing than what others are doing. Perhaps at 3GHz there is something else that needs to be reset (again, perhaps ram timings? or something else perhaps)...

You can go by your own personal experiences, but there are plenty of others (including Hilbert) who have tested 3GHz to 4GHz and aren't getting anywhere near the gain you are getting on a single GPU machine... So something is different.

 

Listen, I will say it again once more, the only thing going on here is a simple clock speed change from 3ghz to 4ghz. Its that simple. As for ram speed etc, yes, that typically changes when you adjust your FSB but you should also note that I made the timings tighter on the ram when I ran 3ghz. Besides, ram speed has very very little to do with game performance.

And I typically do, I have yet to let myself down either.

 

Perhaps something along the lines of enabling driver level AA to force override FC2's in game settings to artificially inflate/deflate the results? :banana:

 

I guess we can go round and round with this all day Lavans as you will just keep making excuses. The driver level of AA was set to application controlled during the FC2's benchmark.

Your just having a hard time accepting the fact that you got it handed to you in this thread and you just cant stand it, now go crawl back under the rock that you came out from under.

 

:3eyes:

 

Are you just ignorant or what? I dont think its any big suprise that when FSB speed is adjusted that it also affects the ram speed unless you go with a different divider besides a 1:1 ratio. Regardless ram speed as I stated earlier has very very little impact if any at all on gaming performance and as I stated before I tightened up the ram timings to suffice for the lower speed when running at 3ghz.

Man you and Lavans are just halarious, talking about desperate. I will be more than happy to redo the benchies running a different ram divider if it will make you happy.

 

I was just pointing out the contradiction... I do totally agree that ram speed and timing have very, very little affect on FPS. What I am stating to you is that others (including myself) have also done a lot of benchmark/testing by overclocking the CPU only and do not see the gains you are seeing. My suggestion is that something else is going on with your computer spefically to cause such differences from 3GHz to 4GHz - something beyond just the CPU frequency change.

What that is I don't know and I am happy you are getting good results... But that does NOT match other people's results including myself.

That said, it has been over a year since I did that kind of testing, well before my SLI'ed 285's... I also don't have a quad or i7 - and these are important factors. But I can tell you that my experience in overclocking the CPU only yields practically no gain (apart from minimum FPS, which is important) in 99% of the games out there when you are testing at higher resolutions.

 

In my case when I had my Q6600 and Two 8800GT's my results was like RobsGTS. I had to OC my q6600 to 3.5ghz for my SLI speed problems to go away. Because I had bad speed issues even with the right profiles setting before hand.

I had the same result with the GTX260 core216 I have now when I had my q6600. so I have to agree with RoboGTS.Because I seen it with my own eyes. Some games less then others But I did see the gain in other titles as well as big gains too it just depends on the game.

Only way to kill your hardware ocing is if you do not know what your doing like using way more voltage then what is really needed or your water cooled and water spill in the cpu socket or leaked etc. In all my years ocing I never killed any thing from ocing since I dont go and set it to the dangerzone first. I work my way up.

Also come on guys do we really need to give each other the finger when one doesnt agree with another. Its like kids fighting over a pack of chocolate or something

 

+1 Im glad someone that has owned similar hardware can back me up on this seeing as Lavans and Dave find it so very hard to believe.

 

Wait, I'm confused... You have SLI'ed 275s? If so I can understand your results better... Going by your specs it looks like you have a single GPU?

 

overclocking a processor really only helps if you have two graphic cards in SLI, or if the card is being bottle necked by the processor which is rare.

robs, when you're the only one arguing to the teeth over a forum discussion, it's time to step back. so far, all your results are vastly different from the rest of the community in terms of overclocking. also, bringing in your own results is questionable because of the above fact. it's better to bring in reviews that someone couldn't call questionable.

not likely. a P45 chipset, like the one he has in his listed specs, only supports CF

 

No, I stated in an earlier post that I used to have sli'd 8800GTX's and I gave an account with my experience in owning that setup.

I own a single GTX 275 right now.

 

Wrong, with todays high end GPU's overclocking the CPU is a big help and I have posted the differences to prove it. When you own a Dual GPU setup overclocking becomes even more beneficial.

I suggest you look again.

 

Lost Planet DX10 1920X1080 4XAA 16XAF all in game options maxed out.

3ghz.



4ghz.

 

man this thread makes me laugh:wanker::banana:uke2: :3eyes:

 

I've seen benches where an i7 system, vs a C2 system using the same video card, pulls higher FPS.

infact, the i7 cpu's seem to make SLI systems perform even faster.

 

RobsGTS, I'm just trying to understand why in a game (FEAR) I know for certain is GPU bound, how you got such great gains on your system...

I never once claimed you were lying or trying to fudge the results in your favor - you are being very defensive and I am just trying to understand what the difference is. Quoting Hilberts results specifically in FEAR is very valid, especially on this site. His testing on that article was done with a GTX280 which last time I checked was a high end GPU. He says FEAR is GPU bound and provided benchmarks to show that you gain nothing in terms of average FPS - yeah your benchmark clearly shows a very large difference. So what is the difference? Did Hilbert make an invalid test somehow? Are you better at overclocking or something?

Just trying to understand.

 

In all honesty I have no clue as to why my benchmark in Fear is different than Hilberts, but I will tell you one thing, I have been benchmarking fear since back in the days that I owned an X1900XT and my results from overclocking my old amd 4000+CPU from 2.4ghz to 3ghz always resulted performance gains in Fear. Why hilberts review states otherwise I have no idea.

The only thing I see is that im running vista 64 and in his test he is running vista 32bit. I dont know why that would have an affect on performance though.

 

After reading this thread for a bit and noticing the fight between some people about whether or not CPU overclocking really does effect performance, I decided to run a little benchmark session of my own just using FC2.

I also did this for the OP as he and I share the same CPU and graphics card. This can give him a bit of an idea as to what to expect out of overclocking his CPU at least.

The benchmark was done with my graphics card at 756/1656/1512 at a resolution of 1680x1050 and the 190.15 beta drivers. I also did benchmarks with 4x AA and 8x AA to prove a theory of mine that with 8x AA the game becomes more GPU limited and the CPU plays less of a part. It turns out I was right as you will see below.

Note: I added in the 3.8GHz tests as that's the clock speed I currently keep my CPU at and found a lot of people that own i7 920's use that speed as well.

2.66GHz



8xAA



4xAA



3.8GHz



8xAA



4xAA



4.0GHz



8xAA



4xAA



As you can see there is little to no difference at 8xAA. To me it proves that with higher AA enable the game becomes more stressful, obviously. This means that that the CPU is no longer the bottleneck and the GPU is. When the game is less intensive on the GPU the CPU is what becomes a bit of a bottleneck and will benefit from a higher clock speed. I ended up learning that from my benchmarks.

Anyways I hope this answers some questions for the OP in regards to if a i7 bottlenecks a 285.