Will a gtx1080ti make much perf. difference over a 1070?

Post a reply

Smilies
:D :) :( :o :shock: :? 8-) :lol: :x :P :oops: :cry: :evil: :twisted: :roll: :wink: :geek: :ugeek: :!: :?: :idea: :arrow: :| :mrgreen: :3: :wub: >:( :blergh:
View more smilies

BBCode is OFF
Smilies are ON

Topic review
   

Expand view Topic review: Will a gtx1080ti make much perf. difference over a 1070?

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Rachael » Tue Feb 20, 2018 10:41 pm

As has been repeatedly stated - as long as GZDoom remains CPU-locked, it will not make a difference. It might make a difference with texture and vector point transfers when the CPU and GPU are actually communicating, but otherwise no. The card is going to be fiddling around waiting for the CPU most of the time.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by invictius » Tue Feb 20, 2018 3:39 pm

I take it that restricting the pci-e bus to 4x won't make any difference at all?

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Tue Feb 13, 2018 6:41 pm

This is strange, I can no longer reproduce it, it is now smooth at vid_maxfps 60 with vid_vsync 1 (but still jumpy with vid_vsync 0.) Even in windowed mode. Perhaps earlier another process was interfering.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Medicris » Tue Feb 13, 2018 2:01 pm

In my experience, it's a lot better to set the max FPS to something that's not exactly 60 to avoid that odd hitching you're speaking of. Try 61, 62, 59, or what I ended up using for games, 70. It helps that 70 is double 35, as explained below.

70 seems to be the best compromise between how visible the tearing is. It's a lot worse and when it's in roughly the same spot or the tear slowly and visibly moves downward, like with max 61 or 62.

The physics simulation actually runs at ~35fps at all times, but the engine interpolates these as a part of its unlocked framerate feature. Input polling is still technically 35fps in the "physics" simulation, but it accepts and displays all input inbetween those frames. As a result, whether capped or uncapped, your mouse input will feel responsive.

Edward wrote a more technical post on this, visible here: viewtopic.php?f=4&t=47296&start=15#p874957

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Mon Feb 12, 2018 5:02 pm

Same thing with full screen, vid_maxfps 60, vid_vsync 0. I even set minimum processor state to 100%, but it didn't change it. It's not a problem though, just setting vid_maxfps 200 solves it.

I wonder, are physics fps and input polling tied to video fps?

Re: Will a gtx1080ti make much perf. difference over a 1070?

by dpJudas » Mon Feb 12, 2018 4:39 pm

vid_maxfps 60 doesn't lock the update to the vertical sync of the monitor. If you want it to run at the monitor refresh rate, use vid_vsync 1 together with either vid_maxfps 200 or vid_maxfps 0.

You didn't mention if you run it fullscreen or windowed. If it is windowed, even with vsync off, the vsync is actually still on in Windows. The DWM presents a "screenshot" of the desktop every vsync. This image consists of the latest finished frame at the time the DWM does its composition.

If you're running it fullscreen it should be smooth with vid_maxfps 60 (only if vid_vsync is 0), but there might be CPU speed stepping in play here that makes it speed the CPU up and down between frames due to a low workload. You can try go to the power settings in Windows and force it to performance, if you want to check if that's what is happening.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Mon Feb 12, 2018 4:17 pm

Perhaps stutter is a wrong word. It's like every few frames it 'holds then skips'. For example, say you log the travel distance at regular time intervals and you're traveling in a straight line with constant speed.
Something like 1 2 3 4 4 6 7 8 9 9 11
The velocity averages out properly, but it like stays in one spot for 2 frames, then the third frame jumps forward to make up for the hold in place. My guess is that when a new frame isn't ready it just shows the old frame again, but I have vsync off in game.

I can see this when I set vid_maxfps 60 and on Heretic in the start of e2m5 looking straight up at the high pillars while walking forward. However totally smooth with the default vid_maxfps 200.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Graf Zahl » Mon Feb 12, 2018 2:08 pm

For me as well. It sounds a bit odd that a supposedly steady frame rate causes stuttering. That's normally the sign of something not working right.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Jerry.C » Mon Feb 12, 2018 12:02 pm

Hypersonic wrote: I was comparing vid_maxfps 60 and vid_maxfps 200 on a 60hz display, and noticed that with vid_maxfps 60 there were subtle but noticeable stutters

For me it is the other way around: With vsync on everything is smooth as butter, as long as the map does not overtax the engine, but with uncapped frame rate it fluctuates strongly enough that it can be perceived as some slight stuttering.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Thu Feb 08, 2018 9:25 am

Thanks for the info on the stat parameters, just noticed that if you type just stat it'll show you all of the possible parameters as well which is very useful.

I was comparing vid_maxfps 60 and vid_maxfps 200 on a 60hz display, and noticed that with vid_maxfps 60 there were subtle but noticeable stutters, while vid_maxfps 200 was totally smooth. A similar phenomenon occurs with Quakespasm https://sourceforge.net/p/quakespasm/di ... /a180cafc/

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Graf Zahl » Thu Feb 08, 2018 1:31 am

'stat rendertimes' and 'stat renderstats'.


Regarding performance, compared to a Geforce 1060, my old 550 ti performed at the same level if the postprocessing effects were disabled and shadowmaps were off.
However, with postprocessing on and shadowmaps the 1060 runs circles around the old card, here the more than quadrupled performance really shows.

On a map like Frozen Time, which is about pushing polygons to the GPU there's no difference to be seen.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Wed Feb 07, 2018 10:11 pm

Does GZDoom have something like Quake 1's r_speeds variable? There were some other performance variables as well listed under '1.8 Technical Graphics Commands' in http://www.gamers.org/dEngine/quake/QDP/qconcmds.html

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Rachael » Wed Feb 07, 2018 9:42 pm

My 860M, which is a really weak GPU compared to both listed, barely even sweats with "vid_scalefactor 2" at 1680x1050. GPU is not a limiting factor and like Graf said, a low clock CPU is your biggest enemy here.

As far as GZDoom is concerned, there's literally absolutely no difference in performance between an 1080 and 1070.

Sorry, but you're benchmarking the wrong game for the difference between the two GPU's.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Hypersonic » Wed Feb 07, 2018 9:37 pm

Seems best to first test with a very low resolution to see if you're CPU/Ram limited, factoring out the GPU performance. Then ramp up the resolution to see if the framerate starts to lower significantly to determine if the GPU is a limiting factor.

Re: Will a gtx1080ti make much perf. difference over a 1070?

by Graf Zahl » Wed Feb 07, 2018 12:31 pm

You really should read what this is about before posting. Hint: CPU-limited renderer cannot work fast enough to fully use the card's power.

Top