rasmuskrj

The i7-7700k is currently the best CPU for gaming, and it might even still be after the Coffee Lake CPUs since they're lowering the clock speeds. So if you feel like you have the money then go for it. Whether or not it is worth the money for you is harder to guarantuee.

Naturally it only makes a difference once you are limited by CPU performance rather than GPU performance. And that mostly only happens once you get near 100fps(depending on the game of course).
I'd say that when playing on 4K you would almost always be GPU limited even with a 1080ti, unless you're playing something like CS:GO. You could of course have a lot of stuff running in the background, in which case a 7700k would help. If you have your CPU 100% usage at all times during gaming, this seems to be a likely scenario.
 
Cooling a 7700k is... interesting. I have a custom loop with a EKWB Supremacy Evo cooling my CPU and if I OC it with 1.3V it runs at 80+C while gaming. Through research I've learned that the only thing to do to get this thing to run cool is to delid it. I probably wouldn't expect to be able to OC it with a Hyper D92. It will probably be kinda hot even on stock speeds, but I'd guess still within bounds.
 
All this being said, another option would be for you to wait right now. The new Coffee Lake CPU's will be coming out soon which will give you 2 more cores and 4 more threads, albeit at a lower clock speed. This requires you to buy a new motherboard though.

The 7700K is an absolute thermal beast, to put you into context, I ran a short Realbench run on my "almost 5 GHz" chip (somehow the new Realbench always down clocks my multipler by one, weird). I can get a stable 5 GHz, but with a bit more voltage than I'd like to be honest. I am rocking a delidded chip with conductonaut with a full custom loop (albeit, quite a packed one with 2x 1080s).
 
I also ran the fans at max rpm so that it may be easier to reproduce for others. You can see that compared to idle at ~40 degrees, the load temp is at 75 degrees or so.
 
Of course it is possible that my chip is hot, but I think most 7700Ks are comparable.

Admittedly, my research is anecdotal but I have seen people getting very different temp out of the 7700k while running similar voltages and cooling solutions. I am not even the first one to suggest that the amount of paste and distance between IHS and die, might be slightly different between individual chips and be influencing cooling performance.
 
For instance, I'm putting around 1.3v in the 7700k to ahieve 5.0Ghz, and getting the aforementioned 85C. I have seen plenty of people getting similar temperatures with a custom loop as mine, and I have seen plenty of people getting better temps with an inferior cooling solutions.
There are of course a ton of variables that I cannot know about when looking at these results, so every conclusion should be taken with a grain of salt.
 
My main point is this though: You cannot be sure that you will get better cooling performance on the 7700k with a beefier cooling solution. But you might.

Well seems that I was wrong  Looks like none of them use the reference PCB. But I agree that they look very similar.
 
It's hard to be sure without measuring anything though. I feel like the GPU chip looks kinda bigger on the gaming X, which it probably isnt, but the visual effect might be caused by some proportions not being the same.
 
If you're not intimidated by the task I would probably just try to do it and see if it fits

The Strix cooler for this generation of nvidia cards seem to perform really well. For 3.50 € I'd defintely go with that one.

Holy shit, you're right. Those are some really high prices.

Admittedly, my research is anecdotal but I have seen people getting very different temp out of the 7700k while running similar voltages and cooling solutions. I am not even the first one to suggest that the amount of paste and distance between IHS and die, might be slightly different between individual chips and be influencing cooling performance.
 
For instance, I'm putting around 1.3v in the 7700k to ahieve 5.0Ghz, and getting the aforementioned 85C. I have seen plenty of people getting similar temperatures with a custom loop as mine, and I have seen plenty of people getting better temps with an inferior cooling solutions.
There are of course a ton of variables that I cannot know about when looking at these results, so every conclusion should be taken with a grain of salt.
 
My main point is this though: You cannot be sure that you will get better cooling performance on the 7700k with a beefier cooling solution. But you might.

Yes, they do run hot... which is why de-lidding is a good idea... but that voids warranty of course. And now Intel have said people shouldn't overclock their CPUs, despite paying more for the 'K' version in the first place so that they can do precisely that. Great job Intel, great job. 

Thanks man, well both of those problems are resolved. Under ac at full speed fans for 1 hour playing GTA 5 these are my temps. I couldn't be happier!!!

To elaborate on what @For Science! said, I wouldn't get your hopes up too high regarding the cooling performance of an AIO solution. It seems that its kind of a lottery, how well the TIM is applied to the 7700Ks.
 
For instance, I have a custom loop solution with a Supremacy EVO block, and I still get around 85C in games when I OC to 5Ghz at 1.3V. The GPU increasing the water temperature probably doesn't help though.
My point is that you might not see a big improvement in temps when overclocking if you switch to a beefier cooler. The 212 is a pretty modest cooler though, so you probably will see some improvement.
Just be prepared  

The only pump quieter than the Laing DDC3.2 + Koolance top is the D5 vario, and they are even changing positions over flow rate. Were you only looking at the Laing DDC-1 result? This graph kinda proves my point.

Hi Guys,
 
Just spreading the word that you shouldn't use the current D5s in the Fluid Gaming series yet. Stainless steel used in the D5s are as far away as copper is from aluminium in the galvanic series and so probably should avoid being mixed. On the other hand the SPC pumps are mostly plastic and so there is less of this concern. Apparantly EK is looking into aluminium D5 equivalents so one should be patient.
 
 

 
 

I want to second this (I assume you meant loud ). I have a DDC 3.2 pump that I run at 30% speed and it's practically inaudible. I agree that the D5 is better though, but then again it's larger and more expensive.

Compression fittings: EK-ACF Fitting 10/13mm - G1/4 Nickel <----- buy two more of these
 
Tubing: EK-DuraClear 9,5/12,7mm (3/8'' - 1/2'') 2 meters <-- this really should be enough, but buy some more if you don't want to risk it.

To elaborate on what @For Science! said, I wouldn't get your hopes up too high regarding the cooling performance of an AIO solution. It seems that its kind of a lottery, how well the TIM is applied to the 7700Ks.
 
For instance, I have a custom loop solution with a Supremacy EVO block, and I still get around 85C in games when I OC to 5Ghz at 1.3V. The GPU increasing the water temperature probably doesn't help though.
My point is that you might not see a big improvement in temps when overclocking if you switch to a beefier cooler. The 212 is a pretty modest cooler though, so you probably will see some improvement.
Just be prepared  

I seriously doubt that the GPU block is the problem, unless its defective somehow. But I mean its basically just a block a copper, so it would have to be bent or somthing that could reduce contact.
 
I only have experience with Bitspower fittings but they are definetely higher quality than the EK fittings i have. This implies to me that their blocks are high quality as well.

Compression fittings: EK-ACF Fitting 10/13mm - G1/4 Nickel <----- buy two more of these
 
Tubing: EK-DuraClear 9,5/12,7mm (3/8'' - 1/2'') 2 meters <-- this really should be enough, but buy some more if you don't want to risk it.

Yes I believe the IPC for Skylake and Kaby lake are completely the same. If you can overclock a 6700k to 4.5Ghz(which you almost certainly can) then you basically have a 7700k, without the 4k streaming feature and whatnot(which are dependent on the motherboard chipset anyway).
 
The only real benefit of the 7700k is that you can overclock it even higher than the 6700k, IF you can keep it cool. My 7700k hits 5Ghz with around 1.29v(goes a bit higher when tweaked by LLC) but it can hit 85C during gaming on a custom loop while overclocked. I am not comfortable with that

I agree, this was kind of what I was getting at in my first post

I agree, this was kind of what I was getting at in my first post

Vega is still yet another optimization of GCN, but yes I agree due to the time between node shrinks and the need to stay competitive you don't just simply shrink what you have and not really do anything else.

Wouldn't that be in the realm of "optimization"?
 
I'm not really that knowledgeable regarding the AMD achitectures, but what I'm saying is that whether or not it was possible for AMD to simply improve the Fiji architecture clock rates by 60%, call it a day and compete with a 1080 ti is something that only the AMD engineers know.
But I'm pretty sure that If they'd been able to do that, then they would have done so. Probably much earlier too. 

I'm not saying that it is not possible, I'm just saying that we cannot assume that it is. Correct me if I'm wrong, but I believe that Fiji itself was an optimization of AMDs older architectures. Ultimately, you will reach a point where you will have to change something fundemental in order to improve.

Just an aside on your comment about FPS games...
 
It isn't about what you "see", it's about what you FEEL.  The primary benefit to a higher refresh rate has nothing to do with a smoother picture, it has to to with faster updates and better positional information.  Without getting into too much detail the TL;DR is that the higher your refresh rate the better your shot registration will be, for everyone of every skill level without fail.  If you played on a 240hz TN panel with zero input lag and recorded your statistics over a reasonable (and therefore accurate) sample size and then did the same on a zero input lag 60hz screen you'd see a tremendous difference in accuracy and contrary to what most people think it has nothing to do with your reaction time.  I've made big posts on it in the display forum so I won't go into any more detail, but if you're interested have a look in my post history.
 
But yeah, I don't think we really disagree we just have slightly different ideas of value and that's ok.