Nothing much for gaming basically. Intel CPUs have gone nowhere fast since the 2500K and at this kind of resolution all of those chips from the 2500 upwards will perform nearly identically.
CV1 recommanded specs!
- Thread starter atlas974
- Start date
Nothing much for gaming basically. Intel CPUs have gone nowhere fast since the 2500K and at this kind of resolution all of those chips from the 2500 upwards will perform nearly identically.
See here:
http://cpuboss.com/cpus/Intel-Core-i5-4590-vs-Intel-Core-i5-2500K
I will just need a better graphics card. Going to wait for the release of the AMD 390x and grab one of these if it does not cost too much.
Last edited:
This would be the biggest factor I would think, good luck to all of us but do like what I am hearing. Maybe FD gets a little more serious about this as have some very high profile people I would like to show it to at work but not in its present form. Hey FD I can gain support through a massive Internet provider?
You are confused, young padowan. The 970 runs the Oculus DK2 PERFECTLY at 75hz.... just not Elite Dangerous. I play a half dozen other games and they are 100% fluid and issue free. The issue, is with this game.
I don't see how they can list specs for a monitor. You don't see that elsewhere, do you? "In order to runt his 1080p monitor, you will need at least X-video card". The monitor doesn't matter, only the app you're running on it.
So while the specs "in general" might be a good ballpark... there is no way in hell they will work for Elite Dangerous. (unless some strange voodoo magic optimization occurs)
Last edited:
Yep, unless the devs somehow up the efficiency to push out the 400Mil shaded triangles on a 970 (or current cards) then it will never be able to reach the required 90Hz threshold for CV1. Not a display or driver issue, but a program issue.
Now with the coming Win10 and Dx12 (and other AMD/NVIDIA appropriate drivers optimization) perhaps the game, if FD re-code the engine for it, might be able to reach that but indeed as you indicate other than turning down the bells and whistles (and therefore lose much of the graphical appeal) it will be a long time, or much better hardware to reach those requirements.
In a few years it will be moot anyway as said hardware will be easily obtainable
Thanks. I see that the i5-2500K will only support 2 displays, whereas the i5-4590 will support 3. I had thought that this would be down to the graphics card, but perhaps not?
The 2 monitors is for the integrated graphics. Your dedicated graphics should be able to handle 3.
Is the general consensus that if a i5-4590 is suitable for the rift then a i5-2500K probably would be as well (with all the usual caveats)?
I was working on the idea that I would need a new graphics card (and therefore a new PSU as well) to the run the Rift, but if I need a new CPU and motherboard as well then that's practically a new PC!
Im running a i52500k and a 980.
Right now I don't expect I have to upgrade when the cv1 comes out.
so just wait and see what happens when it does if you have a similar system.
http://www.dslreports.com/forum/r28517633-I5-2500K-vs-I7-4770K-Gaming-Benchmarks
Right now I don't expect I have to upgrade when the cv1 comes out.
so just wait and see what happens when it does if you have a similar system.
http://www.dslreports.com/forum/r28517633-I5-2500K-vs-I7-4770K-Gaming-Benchmarks
This is exactly what I was thinking. The only difference being that we still have LiquidVR and the Nvidia equivalent to be included along with other optimisations. Still, it would surprise me if it they suddenly made a 970 or 290 viable.
Your 2500K will be fine.
LiquidVR is mostly gonna rid us of judder and latency problems, there is nothing that is really going to make a difference regarding graphics horsepower with single cards. DX12 and Vulkan will help a lot with "dips" in fps that plague a lot of DX11 (and below) games but again it won't make cards faster per-se.
Still talking 2 years before this is really nailed. It's gonna take a combination of LiquidVR + DX12 but also time for devs to get their heads around it. I know it's exciting that VR is almost here but the industry is really slow to adopt new stuff still so it's still 2 years out before we really start to see major benefits. Frontier are probably ahead of the curve but don't expect to see great improvements until DX12 either, which for me is likely to be the first or second expansion. They're gonna need DX12 for planetary landings.
Last edited:
It has been massively overstated. Far too much hope has been placed in asynchronous timewarp in particular, which is extremely unsatisfactory for use in systems with positional tracking. When used to replace more than the occasional dropped frame, the artifacts introduced can themselves resemble judder, ironically. The news of no free lunches has yet to percolate to the peanut gallery, however, so claims of it being a magical solution remain.
On the other hand, proper kernel mode direct access that works better than the current rather desperate shim will indeed yield some good results for all properly supported GPUs. Addressing an HMD as a monitor is a short term kludge at best.
At the end of the day, judder will only be addressed with enough temporal resolution coupled with low persistence displays, configured correctly to avoid obvious strobing (e.g. taking care that the persistence doesn't get low enough to defeat the user's saccadic masking). This primarily necessitates the use of fast GPUs with reliable drivers, rather than any particular handwavium or cute trick. Luckily, both Valve and Oculus seem to be aware of this, and are planning for it appropriately.
Last edited:
Most of the peanut gallery already read that 2 months ago - https://www.oculus.com/blog/asynchronous-timewarp/
Or when Oculus said it afterwards at GDC.
It seems you have a contact at Nvidia - could you ask them if Valve ever got those VR SLI drivers they talked about months ago btw?
We already know from multiple sources (Oculus, Valve, Roberts Space Industries, 2K Gaming) that LiquidVR gets doubled framerate (with Stereo Rending/Split frame rendering/one gpu per eye with AMD Crossfire.) I'm sure many Nvidia SLI users are waiting with bated breath (been a 9-month long wait so far) on Nvidia's reliable drivers providing the same solution so any info you can provide us with on that would be much appreciated.
The last info we got at that time was that VR SLI was only capable of 50% faster at best -
Oculus, Valve, RSI, 2K AND Nvidia themselves(!) were saying that Nvidia is nowhere near in VR 2 months ago so obviously we'd all love to know if something has changed recently.
Last edited:
Sorry I can't hear you over the "peanut gallery" laughing at your ignorance.
Last edited:
I don't know what you two girls are arguing about, all I know is i'm a bit disapointed in the CV1's resolution specs. boo.
I think it'll be better than you expect, with the bump in temporal resolution, providing that the pixel spacing is better (which they're claiming). Hopefully, both Oculus and Valve can learn the lessons from the practical travails of the early adopters.
I think you're right. I think the quality increase we will see will be larger than just the small increase in resolution. I predict we will see a solid 100% increase in overall image quality. We will see be rez limited, and there will still be an SDE... but I bet GREATLY reduced. I'm still excited.