Recommend a Benchmark / Stress Test for Flight Sims?

[BabtridgeMcJoystick 0]

Hi there

 

I thought I would come to you guys for some suggestions on this as you are the experts 

 

I have been asking around on some of the benchmarking / review sites if they would be willing to include flight sims in their benchmarks, as opposed to the usual suspects and 'normal' games. Obviously getting a great frame rate in BF4 takes a different set of technical chops than getting a smooth final in P3D with that beautiful airport addon you just bought, but very few if any sites ever focus on it. I got a reply back from these guys: 

 

https://www.youtube.com/channel/UC9Tn-atYOt8qZP-oqui7bhw

 

And they asked what would be a recommended stress test when it comes to simming. So, what do you guys think? What is the holy grail / computer murdering apex of flight sim performance? FSX with all the trimmings? XPlane with a ton of photoscenery? 

 

I know they all have their little things that make them unique, but if you were looking for a benchmark that would inform a purchase of a new CPU, what would you want to see in sim land?

 

Thanks for reading, I will submit the most common answers (or the only answer, in case no one reads this corner of the site that much because you are all off enjoying your overclocked 4790K already  ) to the channel and see what happens!

 

Cheers

Ross

[dazz 69]

There's FSMark07, FSMark11...

 

http://forum.avsim.net/topic/329116-fsxmark11/

 

Problem is they need to install FSX + SP2 or Acceleration, there's no stand alone benchmark for FSX, and the results will not reflect what most of us see because no one flies FSX without addons, AA and CFG tweaks, but should work fine to compare CPUs

[BabtridgeMcJoystick 0]

Hi dazz

 

Thanks for replying So what I am wondering is, would you be able to suggest a kind of "torture test" combination of addons etc you might wish to see in FSX, so that the benchmark can then be run?

 

Or does FSmark need to be run on vanilla? If not, would simple frame rates a la normal gaming banchmarks not be a good indicator?

 

Apologies for the noobie questions, the opportunity came up and it seemed too good to miss! 

[dazz 69]

It doesn't need to be run on vanilla, you can add texture packages, weather & traffic addons, etc.., but it uses the default 737 which is a pretty big limitation on itself if you're looking for an approximate FPS figure since pretty much any other plane will perform differently.

Nonetheless, if you run the same test (same install & config) on different machines, the FPS comparison should be valid. Setting the fps to unlimited & turning off Vsync is a must for these tests

 

I wouldn't bother with anything other than vanilla TBH

 

And welcome to Avsim Ross  ^_^

[Bryan Lee 0]

For a base test I would recommend FSX on maximum settings (SP2 or Acceleration) and an approach into KJFK airport using default scenery and the default Cessna. Seems like the most taxing area of FS.

 

So basically, all they would have to do is line up for an approach into JFK, save the flight as "benchmark" or something, use a temporary CFG file that maxes all the sliders and there you have a reasonably fesable benchmark