• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

**Official 3D Mark 11 scores**

a little bit, but I can only run my board chip at 4.7ghz (down to board being rubbish) so the extra 300mhz jump should yield an increase in speed.
 
Rusty0611 said:
What's the performance bump in 3D Mark 11 like between that and a 3930K? Don't know the difference between the two and too lazy to search :D.
Click to expand...

Clock for clock the 3930 and 3960 are exactly the same in 3dmark11.

Going from 4.7ghz to 5.0ghz on mine adds about 700 to the graphics score. It also gets me over 17000 on physics.
 
Locky said:
a little bit, but I can only run my board chip at 4.7ghz (down to board being rubbish) so the extra 300mhz jump should yield an increase in speed.
Click to expand...

Kaapstad said:
Clock for clock the 3930 and 3960 are exactly the same in 3dmark11.

Going from 4.7ghz to 5.0ghz on mine adds about 700 to the graphics score. It also gets me over 17000 on physics.
Click to expand...

So do you both have any clue to what speed you would require for your CPU to stop being the bottleneck?.
 
J.D said:
So do you both have any clue to what speed you would require for your CPU to stop being the bottleneck?.
Click to expand...

Most of the scores on the hall of fame are clocked at 5.0 or slightly higher. Thats running 4 gpus so it would be around 5.0 - 5.2ghz. If you can go higher it will still increase your score (the higher cpu clock will push up the physics score).

The other thing about 3dmark11 is its just a benchmark where you can see fps getting close to 300. This is way more than I would see in a game with settings maxed out. For gaming I don't go higher than 4.0ghz and don't have a problem with bottlenecking.
 
Last edited:
Kaapstad said:
Most of the scores on the hall of fame are clocked at 5.0 or slightly higher. Thats running 4 gpus so it would be around 5.0 - 5.2ghz. If you can go higher it will still increase your score (the higher cpu clock will push up the physics score).

The other thing about 3dmark11 is its just a benchmark where you can see fps getting close to 300. This is way more than I would see in a game with settings maxed out. For gaming I don't go higher than 4.0ghz and don't have a problem with bottlenecking.
Click to expand...

Ahh I see. Thank you for explaining that. I'm just glad you're not far off as those are nice clocks already. Crazy how far a benchmark will push people to spend a lot of money but it would be different if I was in the top 5. I wouldn't need dice/LN2, I'd just open a window up here in Scotland :D.

I'm relieved that you don't need those CPU clocks for to get the best out of your games. With all these resolution changes (as in higher use of higher resolutions) the GPU can manage more of the load compared to a few years back in multi-gpu systems where I think some were being bottlenecked more from the CPU. I may have remembered wrongly though, little sleep last night.
 
Scores added and not to worry on positions Wazza. You can deffo squeeze more out of that 7850 if you wanted but for gaming purposes, no need :)
 
gregster said:
I am happy with the fastest score in the GFX, if you beat me, it will remain that way :) I am upgrading this way, as I am bottlenecked in some games and only seeing a max of 85% GPU usage.
Click to expand...

I did hear about the possible bottleneck from the 2500k chips with multi cards. What motherboard are you going to pair the chip with?
 
pgi947 said:
I did hear about the possible bottleneck from the 2500k chips with multi cards. What motherboard are you going to pair the chip with?
Click to expand...

Yeah the bottleneck is definitely there. Usage in BF3 averaging low to mid 80's for me whereas with my Ivy chip it averages mid to high 90's. That's an unscientific observation though - didn't take any data.
 
Kaapstad said:
Is this something related to the 2500k or does it affect the 2600k and 2700k for example.
Click to expand...

Well for me the benefit was two fold - i5 to i7 and PCIE 2.0 to 3.0. The PCI-E jump does make a fairly notable difference for dual cards.

(Well around 3% :D)

Somebody on here - maybe chipachap - went from an i5 2500k to an i7 2600k and I believe that that released the bottleneck on his cards too.
 
I'm positive I saw a chart perhaps on ocn that showed a bottle neck on a 2500k but no bottleneck on a 26/700k with ht enabled, ht disabled the bottleneck showed up again, this was on bf3 which must make use of ht.

In short, more cores/threads are better for avoiding a bottleneck on multi card setups.
 
pgi947 said:
I'm positive I saw a chart perhaps on ocn that showed a bottle neck on a 2500k but no bottleneck on a 26/700k with ht enabled, ht disabled the bottleneck showed up again, this was on bf3 which must make use of ht.

In short, more cores/threads are better for avoiding a bottleneck on multi card setups.
Click to expand...

Yeah. The i5 2500k is a good compromise but if you're considering shelling out for two top end graphics card in the future you should also consider an i7 as well as the difference in price is not really that much considering.

*makes note to self who learned the hard way*
 
I too learned the hard way when an unclockable q9300 was bottlenecking a pair of 9800gtx+'s. Higher end for me from now on, looking forward to possible 8c16t on ib-e :D
 
You must log in or register to reply here.
Back
Top Bottom