Intel Core i5 7600K Review

[tsk]

Quote

Today we take a first look at Intel Kaby lake processors in the form of a Core i5 7600K processor. A processor that has been fabbed on the 14nm node, a spin-of from SkyLake. The Core i5 series processors are intended to be the more wallet friendly offering from Intel, for a more mainstream to high-end consumer. In this preview we look at the performance of this processor in a wide variety of benchmarks and situations.

I don't have much to add here, kaby lake for desktop isn't all that exciting. Go read the review for the all the results.
The 5GHz overclock is nice though.

Source

[PCGuy_5960]

1.7Volts?!?!?!?

[tsk]

6 minutes ago, PCGuy_5960 said:

1.7Volts?!?!?!?

Yeah that's not correct, he had 1.35V for the 5GHz OC.
Changed pic to AIDA, shows the correct voltage.

[PCGuy_5960]

2 minutes ago, tsk said:

Yeah that's not correct, he had 1.35V for the 5GHz OC.

Oh, OK. Because if it took 1.7 volts for 5GHz, Kabylake would be much more disappointing than it already is

[VerticalDiscussions]

Wait, so the CPU was oced to 5GHz, and only then was it able to barely outperform the 6700K in single threaded performance .-.?

[tsk]

Just now, VerticalDiscussions said:

Wait, so the CPU was oced to 5GHz, and only then was it able to barely outperform the 6700K in single threaded performance .-.?

They were both at 4.20GHz for that test.

[PCGuy_5960]

1 minute ago, VerticalDiscussions said:

Wait, so the CPU was oced to 5GHz, and only then was it able to barely outperform the 6700K in single threaded performance .-.?

I don't think that Kabylake is supposed to be an improvement over Skylake.... I think that it is just Skylake with a better iGPU

[Cyborgsmith]

3 minutes ago, VerticalDiscussions said:

Wait, so the CPU was oced to 5GHz, and only then was it able to barely outperform the 6700K in single threaded performance .-.?

According to Jokers latest video

At the same clock speed skylake actually comes out ahead. 

[A Damn Crab!]

Congrats intel, yet another pointless CPU launch.

 

AMD on the other hand..

[patrickjp93]

40 minutes ago, Cyborgsmith said:

According to Jokers latest video

At the same clock speed skylake actually comes out ahead. 

No it doesn't. Kaby comes out ahead in this test. At 4.2GHz, the 7600K beats the 6700K.

[VerticalDiscussions]

Oh okay, i misread the article then, since the part that said:

Quote

I don't have much to add here, kaby lake for desktop isn't all that exciting. Go read the review for the all the results.
The 5GHz overclock is nice though.

Made no sense to me :p.

[NumLock21]

This is so boring.

The installation on, how to install a xeon phi processor with the cooler, was more exciting than this. It uses a totally different approach, where how you would properly install a cpu and cooler, out the window. 

[Cyborgsmith]

1 hour ago, patrickjp93 said:

No it doesn't. Kaby comes out ahead in this test. At 4.2GHz, the 7600K beats the 6700K.

Ok read the text you quoted.. 

 

At the same clock speed skylake actually comes out ahead.

 

[Matu20]

15 minutes ago, NumLock21 said:

This is so boring.

The installation on, how to install a xeon phi processor with the cooler, was more exciting than this. It uses a totally different approach, where how you would properly install a cpu and cooler, out the window. 

Would you link it, please? Made me interested as well

[Coaxialgamer]

This is so boring i already knew how this would turn out even before clicking on the article.

 

4 hours ago, patrickjp93 said:

No it doesn't. Kaby comes out ahead in this test. At 4.2GHz, the 7600K beats the 6700K.

The two are less than 1% apart. This is WELL within margin of error, and they effectively perform the same.

[NumLock21]

21 minutes ago, Matu20 said:

Would you link it, please? Made me interested as well

I made topic in cpu section.

 

[DocSwag]

So Kaby lake is just skylake + 200-300 MHz then.

 

I'll have to see the prices to see if it's worth buying over a skylake CPU, but there's no way you should upgrade unless it's from i5 to i7.

[Megahurt]

The whole point of kaby lake is increased clockspeed, I don't have a clue why people are comparing IPC, a 5-10% increase in clocks is just the same as 5-10% increase in IPC, at the end of the day, it's the single threaded performance that interests me.

[Sauron]

2 hours ago, PCGuy_5960 said:

Oh, OK. Because if it took 1.7 volts for 5GHz, Kabylake would be much more disappointing than it already is

More like it would be melted lol

[Master Disaster]

1 hour ago, Coaxialgamer said:

The two are less than 1% apart. This is WELL within margin of error, and they effectively perform the same.

True but one is an I5 while the other is an I7.

[Coaxialgamer]

2 hours ago, Master Disaster said:

True but one is an I5 while the other is an I7.

but that is a single thread benchmark , and i doubt an extra 0.5MB of cache does much

[patrickjp93]

7 hours ago, Cyborgsmith said:

Ok read the text you quoted.. 

 

At the same clock speed skylake actually comes out ahead.

 

 No it doesn't. In the source of this article, both chips at 4.2GHz, Kaby wins.

 

3 hours ago, Coaxialgamer said:

but that is a single thread benchmark , and i doubt an extra 0.5MB of cache does much

2MB, 6MB L3 vs. 8.

[LucidMew]

6 hours ago, Megahurt said:

The whole point of kaby lake is increased clockspeed, I don't have a clue why people are comparing IPC, a 5-10% increase in clocks is just the same as 5-10% increase in IPC, at the end of the day, it's the single threaded performance that interests me.

because patrick here says that Kaby Lake is going to have an IPC improvement over Skylake, and that the performance increase doesn't come from prefetch/more cache/clockspeed.

[Space Reptile]

7 hours ago, Cyborgsmith said:

Ok read the text you quoted.. 

its Patrick star , what do you expect 

 

[Coaxialgamer]

22 minutes ago, patrickjp93 said:

 No it doesn't. In the source of this article, both chips at 4.2GHz, Kaby wins.

 

2MB, 6MB L3 vs. 8.

That's 4*0.5MB,  as it's a quad core, and extra cache doesn't improve the capabilities of the core,  just removes a potential bottleneck  .  Plus,  it still doesn't do anything to prove an ipc advantage for kaby.  Those results are still very much within margin of error.