Whats the difference between a core2 @4Ghz, and an I7 @4Ghz

Michaeljcox24 · 3 Apr 2011 at 22:34

Running Assassins Creed Brotherhood I get fps drops, which really shouldn't happen if you look at my rig.

I've just seen a lad running an I7 950 in the same game @4Ghz on the tube, with a lot less fps drops than I get, for pretty much the same card (a 570)

Whats the difference between a core 2 and an I7 running at the same speed?

Randomface74 · 3 Apr 2011 at 22:39

3-4 years worth of r+d

setter · 3 Apr 2011 at 22:42

Id still suspect the 570 will be the main difference, in a lot of games ive played, my current 4ghz i7 920 isnt much of an improvement over the previous quads i had, a 3.8ghz q6600 and a 3.8ghz q9550.

Michaeljcox24 · 3 Apr 2011 at 22:46

It makes no sense. Temps on the cpu never really go up, gpu usage never really go above 50C, usage is never much over 40%

So is it just a case of the game was designed on I7's, and 500 series cards, and anything less, you get fps drops, even if you run equivalent kit.

Roksonixx · 3 Apr 2011 at 22:48

Michaeljcox24 said:
Whats the difference between a core 2 and an I7 running at the same speed?

Trust me i've just ordered a 2500k and its faster at 3.3ghz than my q6600 @ 3.5ghz ever was...

world of difference, not only is the architecture more efficient but they can clock a hell of a lot higher too

also, if your graphics card is only at 50% usage it means your cpu is bottlenecking

my assassins creed brotherhood has gone from jumping from 30fps to 60 to basically staying at 60 since i upgraded...the old quads are good but they don't compare

Michaeljcox24 · 3 Apr 2011 at 22:53

Roksonixx said:
Trust me i've just ordered a 2500k and its faster at 3.3ghz than my q6600 @ 3.5ghz ever was...

world of difference, not only is the architecture more efficient but they can clock a hell of a lot higher too

also, if your graphics card is only at 50% usage it means your cpu is bottlenecking

my assassins creed brotherhood has gone from jumping from 30fps to 60 to basically staying at 60 since i upgraded...the old quads are good but they don't compare

When I upgraded from a Q6600, I noticed that Just Cause 2 almost stayed at a constant 60fps instead of dropping down to 40 when things kicked off.

In this game though, 4Ghz makes no difference.

Annoying.

Roksonixx · 3 Apr 2011 at 22:57

I made a thread about my fps issues in AC Brotherhood and they totally disappeared when i got the i5 ... it really does sound like its your processor because that gpu usage should be at 100% (unless you're using vsync or limiting the fps) because otherwise it's not your graphics bottlenecking you ...

Johnnytoxic · 3 Apr 2011 at 23:00

It's clock for clock difference.

Just because two processors run at 4ghz won't mean that they perform the same. the i7 gets more done at the same speed.

Michaeljcox24 · 3 Apr 2011 at 23:11

Roksonixx said:
I made a thread about my fps issues in AC Brotherhood and they totally disappeared when i got the i5 ... it really does sound like its your processor because that gpu usage should be at 100% (unless you're using vsync or limiting the fps) because otherwise it's not your graphics bottlenecking you ...

When I put vsync off, fps increases to 70 odd, but in the game, it still looks jittery compared to just cause 2.

Really dont know why.

Roksonixx · 3 Apr 2011 at 23:20

turn mouse smoothing off and vsynch off (both in game and in drivers)

do you mean jittery like crossfire jittery? or when you move your mouse ?or as you're moving along?

this is why we hate console ports

Michaeljcox24 · 3 Apr 2011 at 23:47

vsync is on, and I use xbox 360 pad, so no option for mouse smoothing.

With vsync off, graphics are stuttery even more so.

With vsync on, I get drops from 60fps down to around 40 in crowded areas, nothing major but still wondering why with a 4ghz cpu I should get this. Its not so much the average fps I notice, but the drop that I notice.

Johnnytoxic · 3 Apr 2011 at 23:56

Do you have many programs running in the background? Virus scanner maybe

postmanfw · 4 Apr 2011 at 00:35

Sorry for jumping in but how do you know the difference between crossfire stutter or mouse stutter you talk about?
At the risk of answering my own question....when i move along/back or sideways with keyboard all is smooth but when i turn using mouse its very jerky.
Playing on bops and doing 150fps+

Sorry again to op for butting in but had to ask

Edit: Yep answered my own question

Remembered i had updated mouse firmaware og G5 to 1.2 when i was about to get drivers for it.
Swapped to another G5 and the stuttering has now gone completely....phew was about to sell the cards if it was micro stutter as had just built the rig.

On topic though an i7 has a lot more power per MHz than old quads.
I had a E8500 and that did superpi in about 11s at 4Ghz and others were getting it down to low 9s with 4.5Ghz+

New 2600K does it in 10s flat at stock and 7.5s at 5Ghz

Confused Stu · 4 Apr 2011 at 01:32

OK, not at 4GHz, but this does a pretty good job of showing the difference between the generations of CPUs: http://www.tomshardware.co.uk/quad-core-cpu,review-31788.html

The other thing to bear in mind is that you may find it's now the CPU holding you back, in gaming it might be the chipset itself. For example I found much better results for my 5870 crossfire setup when moving from a [email protected] on a P45 to a [email protected] on a P67. Both are 8x/8x for crossfire, but the P45 definitely gave poorer results. Here's a test showing the chipsets compared which may shed some light: http://www.legionhardware.com/artic...ossfire_cpu_scaling_performance_part_2,1.html

Hope something there helps!

Michaeljcox24 · 4 Apr 2011 at 18:09

I thought that having a quad chip at 4Ghz with 12MB cache would at least last me a good 12 months combined with the card I have.

Really can't justify splashing out £500 for the sake of one game, because so many others run very decetly on my rig.

If Mafia 2 can run half decently, and JC2, and this, which I would say has worse graphics, runs at a lower FPS, I put it down to poor optimization for PC hardware.

panyan · 4 Apr 2011 at 18:59

Johnnytoxic said:
It's clock for clock difference.

Just because two processors run at 4ghz won't mean that they perform the same. the i7 gets more done at the same speed.

i have heard this before, but i really dont understand why that is and if is true, then GHz shouldnt be what we rate cpu;s by, it should be something else like:

ghz * work/ghz = actual work/second

i.e.

4 ghz c2d * 2000 work units/ghz = 8000 work units/second

4 ghz i7 * 7000 work units/ghz = 28,000 work units/second

DragonQ · 4 Apr 2011 at 19:07

Bloomfield = ~15% faster than Yorkfield clock-for-clock
Sandy Bridge = ~15% faster than Bloomfield clock-for-clock

Also remember that each generation can clock higher too (Yorkfield = ~3.6 GHz, Bloomfield = ~4 GHz, Sandy Bridge = ~4.5 GHz).

Note that this doesn't take into account HyperThreading - encoding on a Bloomfield or Sandy Bridge i7 should be a lot faster than on a Yorkfield because of this. For example, an i7 920/930/950 @ 4 GHz will just about beat an i5-2500K @ 4.5 GHz in x264 encoding. However, most games can't take advantage of the extra HT cores so it's not very important in gaming.

Cr416 · 4 Apr 2011 at 20:04

Down to the cpu/chipset. I saw this issue on a q6600 with a gtx 280. changed to 2600k i7 with the same gtx 280 and it stopped.

Poor optimisation for the pc leads to a better cpu requirement to run properly.

Michaeljcox24 · 4 Apr 2011 at 22:31

Cr416 said:
Down to the cpu/chipset. I saw this issue on a q6600 with a gtx 280. changed to 2600k i7 with the same gtx 280 and it stopped.

Poor optimisation for the pc leads to a better cpu requirement to run properly.

this about sums it up. its not optimized, so only the latest hardware will run it the way its supposed to run