4690K with GTX 1080 Ti?

[kol12]

I'm considering a high refresh G-Sync monitor and a GTX 1080 Ti. I would like to do this upgrade before a cpu+mobo upgrade as I think my 4690K machine still has some go left in it and monitor upgrades always outlast other component upgrades.

My question is what sort of bottleneck the 4690K might be to the 1080 Ti at 2560x1440p 100hz+. Will I be able to get a reasonable experience that's worth the monitor+gpu upgrade for 1-2 years?

How bad could minimum fps drops be affecting the experience?

 

[synphul]

It depends on the game. Eventually you may want to upgrade the cpu to an i7. Some in game testing I've seen of the 1080ti it does stay closer to the 90-100fps range at 1440p over the 1070/1080 by around 15-20fps advantage but that was tested with a 6700k, not a 4690k. Some games have fps drops like bf1 and a few others with the i5 so with the 1080ti you'd be looking at the cpu limiting your fps. You'll be able to crank up the eye candy though.

 

[kol12]

90-100 fps is meant to be a nice spot for high framerates with anything above becoming more unpredictable, as in worse frame drop swings, depending on the game...

I currently have the 1070 and decided the 1080 ti would be the more worthwhile upgrade for this type of setup over the the 1080 for reasons which you've mentioned.

The sort of limiting of fps with this cpu shouldn't detract from a 1080 ti G-Sync setup majorly though should it? At most minimum frame drops would lower than an i7? If I can decent performance with this for a little while I'd be happy.

 

[synphul]

It depends on the game, it's hard to say. Bf1 is one that often comes to mind, multiplayer can add additional stress to the cpu and cause more fps drops on an i5 vs an i7.

GtaV seems an odd one from the benchmarks I've seen. Keep in mind these were done when it released a couple years ago so patches or updates could have changed things, but the i5's generally suffer less fps drop than the i7 4790k with ht enabled. I thought it was a fluke but eurogamer showed slightly lower fps lows for the 4790k against the 4690k at 1080p, gamersnexus showed similar findings with the i5 3570k having similar if not slightly better low fps as the 4790k at 1440p with a titan x (comparable to the 1080ti). Even the best results for the 4790k with the titan x at 1440p very high/ultra settings topped out around 66fps average. Some games just won't get close to 90fps, at least not by the looks of those tests.

http://www.gamersnexus.net/game-bench/1911-gta-v-cpu-benchmark-4790k-3570k-9590-more

http://www.eurogamer.net/articles/digitalfoundry-2015-the-best-pc-hardware-for-grand-theft-auto-5

This bench suggests an i5 or i7 should be capable of fps reaching into the 80's at resolution slightly higher than 1440p with a titan x but with details at normal. Maybe turning details in gtav down from very high/ultra to normal will give a fps bump closer to 90. It's not the 1080ti but the titan x is close enough I think to make a loose comparison.
http://www.techspot.com/review/991-gta-5-pc-benchmarks/page6.html

 

[kol12]

I'm surprised the Titan X doesn't pull a lot more frames in GTA V, they must have some very taxing graphics options enabled... Interesting that the i5 sits above the i7 as well, it seems maybe that game doesn't like HT...

It's clear that the CPU doesn't majorly affect the fps either. In this 1080 ti review it consistently gets high fps, usually 100+ fps at 2560x1440, but that is with the Core i7 5960X. Maybe there is a point where a lot more CPU power does affect the frame rates a lot more...

I need to find some benchmarks with the 1080 ti and 4690k or similar...

Those games not getting close to 90 fps are due to using absolute maximum graphic details, not a CPU limitation?