So buy a slower 8core CPU and hope and prey things get fixed or get a faster 4 core and get the best experience now with the need to potentially upgrade later?
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
2700k @4.7ghz ok with 1080 ti?
- Thread starter nightrider1470
- Start date
So buy a slower 8core CPU and hope and prey things get fixed or get a faster 4 core and get the best experience now with the need to potentially upgrade later?
Depends if you need absolute power or imperceivable extra frame rates.
Coffee Lake will do everything Kaby Lake does and more, Ryzen will do everything you need it to do and more, plus if you need extra performance down the line you can have it.
Come on gavin stop the nonsense mate, you are as guilty as anyone else. We have already established more than 4 cores is preferable and 4 cores performs great in the here and now.
Not potentially. VR needs CPU horsepower and unless quad cores designs make a big performance jump they can't provide it. You really need the game engine running at stupid high frame rates like closer to thousands rather than hundreds of frames per second.
You're looking at the problem all wrong. VR is very limited regardless of the hardware you pick. Even the headsets are designed around the limitations the fix is a very big and complicated one.
Stop the nonsense? LOL
I am genuinely curious as to what the suggestion here is. I have no idea about VR, but looking at the benchmarks its clear that 8cores don't do well here.
Frame rates aside, look at the frame times. Any slight jump here is going to be exaggerated in VR, no?
http://www.tomshardware.co.uk/oculus-rift-vr-cpu-performance,review-34008-5.html
I am genuinely curious as to what the suggestion here is. I have no idea about VR, but looking at the benchmarks its clear that 8cores don't do well here.
Frame rates aside, look at the frame times. Any slight jump here is going to be exaggerated in VR, no?
http://www.tomshardware.co.uk/oculus-rift-vr-cpu-performance,review-34008-5.html
I CBA to fully explaining the situation but I'll sum the problem up like this. The short term fix is to offload the work that can't be done on the graphics card to the CPU until we have graphics cards that render whatever game engine at cira 1000FPS. So how important will the CPU be while all the fixes are getting worked on?
Nvidia off load more work to the CPU and a quad core has less processing power than a chip with a higher core count.
Last edited:
No you can buy a 7700k if you like you can also wait for zen 2 if you like. You couldn't wait, ironically did you not upgrade from a 4770 or something?
It's the depth of the fixes and the complexity. Plus every fix is a botched work around. First you need a game engine and API that can either run the game at probably unrealistic frames rates or at the very least peel the rendering threads away from the games engine. Not ideal and possibly not viable. Again you'll need lots of CPU horsepower avalible from the system.
So task one talking engine developers to fully support a capable API then talk them into the need for a render pathway over 60FPS with baked in VR support for half a dozen vendors central to the design. A hell of a task in itself. We have started down this path with a colaberation headset although I think we will need at least another version of directory X and Vulkcan and Nvidia seem very reluctant to fully get behind any API. (Mantle could come into it's own in this regard but maybe not for all engines)
Next you need a studio to use said engine and produce a game worth buying. Most developers probably have zero or little experience coding for such a task, its doable but expensive and time consuming.
Sony are in the best position to pull VR togeather the quickest. They have the money, market and level of control.
So task one talking engine developers to fully support a capable API then talk them into the need for a render pathway over 60FPS with baked in VR support for half a dozen vendors central to the design. A hell of a task in itself. We have started down this path with a colaberation headset although I think we will need at least another version of directory X and Vulkcan and Nvidia seem very reluctant to fully get behind any API. (Mantle could come into it's own in this regard but maybe not for all engines)
Next you need a studio to use said engine and produce a game worth buying. Most developers probably have zero or little experience coding for such a task, its doable but expensive and time consuming.
Sony are in the best position to pull VR togeather the quickest. They have the money, market and level of control.
Last edited:
Curious to know what the recommendation to the OP is here.
Don't spend too much on a VR system and don't expect VR versions of games to be great. What Graphics chips come after Navi?
Go back a bit mate I've already gave my opinion for what ever that is worth.
Volta should be with us soon, but I doubt it will bring the hardware support or a 8x~ performance delta.
Navi will very probably have the hardware but not the performance.
Soldato
- Joined
- 1 May 2013
- Posts
- 9,927
- Location
- M28
VR mostly uses Unreal or Unity, the odd Cryengine.
Whatever works best for those engines.
Whatever works best for those engines.
Associate
- Joined
- 17 Oct 2007
- Posts
- 1,778
- Location
- Some where in England
Put it this way I have i5 2500k O/C 4.5 GHZ with MSI 1080 Ti and it plays every single game at Ultra at 2K 1440p RES No problems at all with well over 60 FPS on the likes of witcher and PUGB so you do not need to upgrade, you will probally get 10 or 20 more FPS by upgrading to i7 7700k at most.
Caporegime
- Joined
- 21 Jun 2006
- Posts
- 38,367
he has a 2700K though, it already has 8 threads and it's overclocked to 4.7 Ghz.
he will see a boost to minimum fps but not much overall.
he doesn't need to upgrade for another 3 years easily. 2700K is 4 cored and 8 threads. 2500K is 4 cores and 4 threads. which is why the 2700K is fine. no game needs 12 threads and won't for at least 3 years easily.