• 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.

Sid Meier's Civilization: Beyond Earth Benchmark

Associate
Joined
27 Aug 2008
Posts
1,874
Location
London
Yes, I did have full gpu utilization. Your 7950 clocks quite good, so I quess that's where you get the advantage. My 90 is from first batch, so it's not GHz edition. Also using win7, maybe that makes some difference aswell. But even if bench itself ain't cpu bound on this card, the game itself is. What this bench doesn't simulate well, is when player moves view around the world instantly.

Yeah, I recall Civ5 late game, presumably not much different?

Done a few more runs, the minimums look to be a bit of a fluke, most the time get 28/29 depends whether you get an unlucky dip that you see on the frame graph at the wrong time.
 
Soldato
Joined
23 Dec 2013
Posts
3,527
Location
North Wales
Single 280X @1110/1575
1080p
DX11
FX-8320 @4.7
14.9.2 Driver

355f755acbaae4613323ea3d7562bfe8.png
 
Soldato
Joined
29 Aug 2010
Posts
7,839
Location
Cornwall
I realise most people have probably run this now, but I thought I'd let people know of a little update.

I've made a few changes to the webpage for calculating the fps and doing the graph. It still does that but now it also produces a frametime graph a bit like the FRAPS one that Vortez show at the bottom of this page.

The new section will identify the max frametime and the min frametime as well as display the value of the 99th Percentile. This means that 99% of the frames were better than this value. So if you ignore a few extreme high frametimes (such as at the start of Kaap's latest run where there's a low FPS dip near the start) you should mostly not see dips below that.

My understanding is that you want the line on the graph to be as low as possible and as flat as possible. Lower means higher FPS and flatter means more consistent.
To compare it to the FRAPS graph, I believe the 99th percentile is what they call the 1%. I've not shown the 0.1% value and I believe the average will probably be the frametime used to calculate the average fps that is displayed in the original graph.

I realise it won't affect any of the results and probably isn't a lot of use for a benchmark thread, but thought people might find it interesting.

Greg has kindly uploaded it and replaced the existing page, so no links need to change. Other than adding the new info in it hasn't changed, so it doesn't invalidate any results.

For anyone interested I used the following page for my understanding of how to calculate it:
http://www.dummies.com/how-to/content/how-to-calculate-percentiles-in-statistics.html

I believe Excel may have a built-in function for calculating percentiles. I've not got Excel so haven't tried it, hopefully the results would come out the same!
 
Back
Top Bottom