i7 2600K vs i7 4770K : My results so far...

[Mysterae]

Whilst in between systems of an upgrade, I've been benchmarking to see the fruits of my labours. Now I know benchmarking programs like that I've used aren't the best but they do help compare before and after...

At the moment I'm not all that impressed with the transition from P67 & i7 2600K to Z87 & i7 4770K. "Tock" indeed. I say at the moment because I've yet to put any serious overclock on the 4770K. The benefits may present themselves when moving to 2 or 3 cards, I shall see.

So, firstly I thought I'd compare my current system with one stock GTX670 and a 4.6GHz OC on the 2600K with a 4.0GHz OC'd 4770K using the same graphics card. Green equals the highest, orange being the lowest. Yellow is between the two.

Ignoring the Lucid Virtu MVP for the moment, the 4.6GHz 2600K beats the 4.0GHz 4770K in most of the scores, 3DMark 11 and Catzilla favouring the newer CPU.

Bringing the Lucid Virtu MVP (integrated Intel HD 4600 working with the GTX670) in to the equation shows increases almost across the board, with only Unigine Valley completely ignoring it. Virtu MVP may seem like a good thing if your game supports it, but the experience wasn't what I'd call smooth. Virtu MVP isn't possible when you use SLI either..

Perhaps it was a bit unfair of me to compare such uneven overclocks, so I've levelled the field a little by OC'ing both CPU's to 4.0GHz:

As expected, the 4770K is in the lead in almost all, but it's marginal in some when Virtu MVP is disabled.

Hopefully when I get to overclocking the CPU, RAM and GPU's I'll see further benefits. I'm already thinking about changing my GPU's after today .

Some more scores before the system is stripped of it's watercooling and therefore overclock on the CPU...

Catzilla is showing some issues! Comparing some of the 3 way GTX 670 4GB scores with a single 780ti is giving me some serious food for thought. Should I even bother watercooling these in the new system? Would anyone even buy my "modified" 670's?

Only one GPU and the CPU are watercooled due to moving between systems.

That's all the results for this old system, I'll update once I've got them in the new system for checking before the OC'ing fun begins!

 

[Kaapstad]

You need to run the CPUs at the same speed for a fair comparison.

Forget about Virtu MVP, you are not going to use it when gaming with multi GPUs.

Don't use Catzilla @576 or Valley, they both bottleneck the CPU with multi GPUs.

Use
Heaven 4 @1080p with max settings.
Sniper Elite V2 bench maxed is also very good.
Firestrike Extreme preset.
3DMark11 Extreme preset.
Tomb Raider bench max settings.

There are others but the above benches will work your GPUs and allow you to see the differences between the CPUs.

Don't expect to see a huge difference clock for clock between the 2600k and 4770k.

It is also quite possible that the 2600k will overclock a lot more than the 4770k.

 

[Mysterae]

I had forgotten about Virtu MVP although I was interested in the results.

Catzilla is crap, in any resolution. It's a cool animation don't get me wrong, but some scenes are just missing. I see you are at or near the top of the leader board Kaapstad, nice work . I don't have Sniper Elite. 3DMark Extreme, what, you mean buy it? Suppose someone has too. Missed it when it was in the Steam sale.

 

[Kaapstad]

I had forgotten about Virtu MVP although I was interested in the results.

Catzilla is crap, in any resolution. It's a cool animation don't get me wrong, but some scenes are just missing. I see you are at or near the top of the leader board Kaapstad, nice work . I don't have Sniper Elite. 3DMark Extreme, what, you mean buy it? Suppose someone has too. Missed it when it was in the Steam sale.

Sniper Elite V2 benchmark is a stand alone free download, if you do a google search it will come up.

You can also get it as part of the game.