Nice article on 6600k vs 6700k vs 6900k for gaming.

[guskline]

Very nice article on results from comparing machines with 6600k vs 6700k vs 6900k as it pertains to gaming and the results.
http://techbuyersguru.com/intels-core-i5-6600k-vs-i7-6700k-vs-core-i7-6900k-games

 

[Arachnotronic]

This image summarizes the results:

Basically, if you're going to game, go with a 6700K

 

[guskline]

That's what I did with my latest build!

 

[swilli89]

This image summarizes the results:

Basically, if you're going to game, go with a 6700K

So basically.. it doesn't matter lol. even 71.3 FPS is exact same experience as 80.5 FPS. Save $700 and buy a better monitor/video card.

 

[Arachnotronic]

So basically.. it doesn't matter lol. even 71.3 FPS is exact same experience as 80.5 FPS. Save $700 and buy a better monitor/video card.

Not if you have a 144hz monitor!

 

[cytg111]

I'd like to see that graph with the 6700K having HT disabled.

 

[deustroop]

Well the reporter has a statement in the text inconsistent with the graph data . The graph shows the Skylake 6600K @ 3.6 GHz to be a little slower than the Broadwell-E 6900K @ 3.5 GHz. Average FPS for the 6600K are 88.3 and for the 6900K are 94.1 even at a slower frequency. Yet the text says:
"Even so, as we've already mentioned, the Skylake architecture used by the 6600K/6700K is a bit faster than the 6900K's Broadwell-E design, meaning the quad-cores still have a slight advantage on a per-core basis."

 

[Alqoxzt]

This image summarizes the results:

Basically, if you're going to game, go with a 6700K

Why 8core broadwell oced not exceeding 4core oced ?

 

[CHADBOGA]

My ten year plan for my i5 3570K @ 4.0Ghz is still alive.

 

[Thinker_145]

Well the reporter has a statement in the text inconsistent with the graph data . The graph shows the Skylake 6600K @ 3.6 GHz to be a little slower than the Broadwell-E 6900K @ 3.5 GHz. Average FPS for the 6600K are 88.3 and for the 6900K are 94.1 even at a slower frequency. Yet the text says:
"Even so, as we've already mentioned, the Skylake architecture used by the 6600K/6700K is a bit faster than the 6900K's Broadwell-E design, meaning the quad-cores still have a slight advantage on a per-core basis."

Because the 6700K beats it when clocks are equal. He is talking about the architecture not the 6600K in particular.

Sent from my HTC One M9

 

[deustroop]

The graph is inconsistent with what he claims, like I said, and similarly to your point, there is no data here on the 6700 @ 3.5.
While the 6700 @ 4.4 is faster than the 6900 (most barely) at the same frequency, the 6600 is slower than the Broadwell-E at a lower freq. His data is inconsistent with his statement, is all. Its a technical point but we love technicalities here, what ?

Its always best when statements in the text are consistent with the data at hand.

 

[b-mac]

Obviously if you have the budget I think the 6700k is the way to go. The minimum frames in some games is much higher vs the 6600k. I you have a limited budget however I think the 6600k + better GPU could be the best option. IMO I'd save for a bit longer and pick up the 6700k if you can. My 6700k hits 4.6 ghz pretty easily with a vcore of 1.34 using a cryorig H7. I could probably go a bit higher with the speed or drop the vcore a bit more but I have just been lazy lately.

 

[railven]

Seems like the appropriate place to ask. Just got my rig back up (need to update sig) and got me a lovely 6800k.

What kind of clocks should I expect on the average for OC'ing? Seems Google always nails me extreme "that's a lotto winner" results.

I read 4.0-4.2ghz is a good average?

 

[deustroop]

Seems like the appropriate place to ask. Just got my rig back up (need to update sig) and got me a lovely 6800k.

What kind of clocks should I expect on the average for OC'ing? Seems Google always nails me extreme "that's a lotto winner" results.

I read 4.0-4.2ghz is a good average?

That brings up another interesting point apparent in this graph, by which I mean how minuscule is the advantage of OC at all.

The data suggests that for the average increase in frequency of 700MHz there was an average increase of 2.6 fps.

WoW .

 

[LTC8K6]

That brings up another interesting point apparent in this graph, by which I mean how minuscule is the advantage of OC at all.

The data suggests that for the average increase in frequency of 700MHz there was an average increase of 2.6 fps.

WoW .

That may depend on a lot of other factors, though. If you aren't CPU limited in a game, then the overclock may not seem worth it.

 

[moonbogg]

That brings up another interesting point apparent in this graph, by which I mean how minuscule is the advantage of OC at all.

The data suggests that for the average increase in frequency of 700MHz there was an average increase of 2.6 fps.

WoW .

That's because they made the same mistake every review site does when testing CPU's. They used a single mid range card and cranked the res to 1440p. What else would you expect? Lower res to 1080p or throw in a titan or dual 1080's and watch the FPS scream ahead on the OC'd tests.
When testing modern OC'd CPU's, you need a ton of GPU headroom.

 

[jpiniero]

A $650 card is midrange?

 

[BonzaiDuck]

It simply underwrites my judgment for my latest system build. I'd contemplated an "E" system -- X79, X99 -- for the last two years. I don't think I'll ever need that sort of processing power -- six, eight or ten-cores. It could, of course, change. Nobody can predict the future with any certainty about anything.

I just finally decided to build an i7-6700K system based on the "skinny" like that shown in the OP's linked article. I even cheaped out on the motherboard, opting for a Sabertooth Z170 S (without the duct-plate). I could have had a 16-phase power design instead of Saber's 8+4. Perhaps a Maximus or the Z170 Deluxe. But this all looks about right to me.

 

[EXCellR8]

I bought my Z170 board long before I ultimately ended up picking the 6600k.

Actually, I think I bought the skylake board shortly after I finished the 4790k build. Probably explains why there was no rush to update hardware...

 

[moonbogg]

A $650 card is midrange?

Yep. It has to do with its place in terms of performance, not price. The only reason you thought it was high end was because Nvidia released it before the high end, but charged everyone like it was high end.

 

[master_shake_]

A $650 card is midrange?

Yep. It has to do with its place in terms of performance, not price. The only reason you thought it was high end was because Nvidia released it before the high end, but charged everyone like it was high end.

as of the last release and the speculated or leaked (idk) 1080ti specs i think they have reversed course on that.

the 980/1080 i think amounted to an actual mid-range card in terms of product stack.

which leaves me to wonder... is the 1060 low end now?

and anything under that entry?

 

[iCyborg]

The graph is inconsistent with what he claims, like I said, and similarly to your point, there is no data here on the 6700 @ 3.5.
While the 6700 @ 4.4 is faster than the 6900 (most barely) at the same frequency, the 6600 is slower than the Broadwell-E at a lower freq. His data is inconsistent with his statement, is all. Its a technical point but we love technicalities here, what ?

Its always best when statements in the text are consistent with the data at hand.

6900K is an 8/16 core/thread CPU, while 6600K is 4/4. Even 1 or 2 games scaling beyond 4 cores could easily give the overall edge to Broadwell, while Skylake can still be faster on a per-core basis. There are no inconsistencies.

 

[frozentundra123456]

Yep. It has to do with its place in terms of performance, not price. The only reason you thought it was high end was because Nvidia released it before the high end, but charged everyone like it was high end.

What card can you buy now that is faster?? It is "high end" right now, simply because there is nothing else as fast.

 

[frozentundra123456]

The graph is inconsistent with what he claims, like I said, and similarly to your point, there is no data here on the 6700 @ 3.5.
While the 6700 @ 4.4 is faster than the 6900 (most barely) at the same frequency, the 6600 is slower than the Broadwell-E at a lower freq. His data is inconsistent with his statement, is all. Its a technical point but we love technicalities here, what ?

Its always best when statements in the text are consistent with the data at hand.

Actually, if you want to be picky, it is not valid to compare 6600k to 6900K because the 6900k has hyperthreading. The most valid comparison is 6700k and 6900k at the same clockspeed. So skylake with 4 less cores and less cache is still a tic (probably within experimental error though) faster. But these are all DX11 games which dont really utilize multiple cores well. This *could* change with more DX12 games, although I am not as confident of this as some of the "moar cores" crowd.

 

[moonbogg]

What card can you buy now that is faster?? It is "high end" right now, simply because there is nothing else as fast.

You can buy a Titan X. It was released a while ago. You didn't know? The 1080ti will be very similar and the mid ranginess of the 1080 will become very apparent.