Hello all, I’ve been watching and reading as much relevant information to my situation as I’ve been able to find and occasionally I get questions that are almost too niche. I was watching Jayztwocents recent video about ryzen and RTX and the benchmarks he showed essentially concluded that 2080ti bottlenecks on ryzen 2700x and the 2080 meshes with it better. I can understand that for monitor gaming, but I’m curious if the same would be true if you use as a vr headset as your main monitor. To my understanding vr is more gpu intensive, and it’s certainly hotter on my gpu than regular monitor gaming is. in that situation would the better card be better? allowing you super sample higher and potentially have less visual discrepancies or would the cpu still create an issue?
Am I thinking about this the right way or am I way off track?
(will get to the VR vs Monitor later in this post)
@8:59 Adored mentiones the overlay displays double or half the real FPS value. So grain of salt to results (have not seen them).
That out of the way: There seems to be again an issue where the new Nvidia cards have trouble on Ryzen. Probably another case of drivers beeing rushed out.
Rise of the Tomb Raider in particular had that issue link to video. A lot of reviewers went and benchmarked RoTR with AMD graphics as a result.
So, with those two grains of salt:
At 720p, wich in itself is a very odd benchmark to do, HWUnboxed found 13% faster.
Here is the graph:
Taking the top and bottom result out of the equation, the difference is reduced from 13.3 to 12.5 percent.
Personally I don´t find this graph particularly useful as I don´t play on low settings and have 2560x1080 as main monitor resolution.
Concerning video cards and display types:
Depending on resolution and target refreshrate, the load on CPU and GPU varies. On midrange gpus (580, 1060 6gb, RX V56, 1070, 1070ti), 1080p or 1440p makes the CPU a non-issue when you are not playing on lowest settings.
On 4k no CPU (assuming not complete fuckups on game or driver side) is a bottleneck.
VR headsets are very taxing on the GPU but having the CPU loaded to 100% will lead to noticeable stutters. The graphics are simple but it is essentially dual 1080p (or higher) monitors at 75Hz or higher target.
topic renamed and category moved
2 Likes
Okay first tbh, I’m not entirely sure what you’re telling me in your post so forgive me if I completely missed the point. Are you saying a stronger gpu could lead to more issues because it loads the cpu too much? *** because I understand the principle but I wasn’t sure if it applied to all applications of the gpu’s job, I wasn’t sure if there was a way with the 2080ti if i could set a frame limit and use higher post processing to attempt to keep the frame rates high and nice with the best resolution possible.
I’m not trying to achieve 4k obviously, somehwere I had 1440p in my head but you are correct about them being 1080p but at 90 fps. What’s happening to me now is i have a 1070ti and I play a lot of racing games. I will stutter where the graphics move much slower than the game sound, often into corners where I’m left trying to blind maneuver the turn. I’ve played with graphics settings, and if i go too low i get motion sick so i haven’t been able to play with that much. I’ve played with supersampling and oc’ing the card. oc’ing and supersampling blended seem to get the best result until the heat kicks in. I’ve invested a chunk into custom watercooling but while i wait for the parts I’m just trying to learn what to expect from my gpu. but I was thinking that a better card would obviously be an upgrade for what I want, I just don’t want to create another bottleneck
***edited in after initial post.
Sorry, I really wasn’t sure where to post it since I wasn’t trying to directly ask what’s better, but more to help me understand how to judge for myself whats better
What I tried to say (and reading it again, I made a mess with words), was that there is an issue with some benchmarking software giving inaccurate readouts.
There will always be a bottleneck. RAM, HDD, SSD, Chipset, CPU, GPU, Cooling, etc. If all is well rounded, some shitty software/game will ruin your day anyway.
Could monitor CPU core load (the individual cores) and the GPU load. I have the feeling windows spreads the load accross the two CCXs of the 2700x.
Using Taskmanager or Process Explorer, you could try setting the CPU affinity like this:
awesome thank you. i’ll take a look at my cpu again. tbh i threw the 2700x in there and moved on assuming all would be fine and dandy.
my immediate reaction is that it’s my gpu (though i’m still trying to figure out how to monitor individual cpu cores) but even just in oculus and steam home w. super sampling and a light oc i’m sitting in the low 70’s for gpu load and low 20’s for cpu. I’ll have to test it over the next few weeks though, i’m building my first watercooled system this weekend so i’ll use this card as a chance to learn how hard I can push a card during gaming and what kind of results to expect,
