Jump to content
Sign in to follow this  
chiliwili69

IL-2 in-game benchmark

Do you want a IL-2 benchmark?  

66 members have voted

  1. 1. Do you want a IL-2 benchmark?

    • Yes, it would be nice to have a in-game IL-2 benchmark (automatic track run, calculate Min/max/Avg, fps log)
    • Yes, but it will be enough with a Flight Record track which came with IL-2 installation (so we can use Fraps or external IL-2 app to count fps)
    • No, I don´t think it is important


Recommended Posts

Along the past 18 months ti has been useful to have a IL-2 benchmark, specially for the VR but also for 4K monitors. We have learnt what hardware/settings is good for a pleasant experience.

https://forum.il2sturmovik.com/topic/29322-measuring-rig-performance-common-baseline-for-il-2-v3005/

This IL-2 benchmark was created initially by Balapan, but the problem is that sometimes new versions make old flight records incompatible. So we have to create a new track.

It would be helpful to have a betterway to do that, either with a becnhmark section or with a Flight record generated by the Dev team.

Edited by chiliwili69

Share this post


Link to post
Share on other sites

Running a track doesn’t put the same demand on your system as live gameplay. You can test this for yourself by just running a mission live with recorder and Fraps at the same time, then run the replay with Fraps. In the past when I tested this there was a significant difference. I haven’t done it recently. Probably the replay isn’t running active physics and AI. 

  • Like 1

Share this post


Link to post
Share on other sites
6 hours ago, SharpeXB said:

Running a track doesn’t put the same demand on your system as live gameplay. You can test this for yourself by just running a mission live with recorder and Fraps at the same time, then run the replay with Fraps. In the past when I tested this there was a significant difference. I haven’t done it recently. Probably the replay isn’t running active physics and AI. 

Moreover, FPS doesn't tell everything about performance. Thing is, the AI puts a high demand on the CPU, resulting in occasional stutters that do not always affect FPS. Such stutters and slow-downs are retained in tracks, but a mission recorded on a beefy system would run just fine when replayed on a weaker system, even though that weaker system would struggle to run the very same mission live with active AI. Despite that, a test track would be great to have, but we have to bear in mind that it's only good to measure graphic performance, not overall game performance. A built-in test function would be a different story, it could also benchmark your system according to how many AI objects you can have concurrently for a smooth gameplay. But please correct me if I'm wrong.

Share this post


Link to post
Share on other sites

I imagine it might be possible to create a live but standard mission that could be used as a benchmark tool in the game. 

  • Upvote 1

Share this post


Link to post
Share on other sites
6 hours ago, SharpeXB said:

a live but standard mission that could be used as a benchmark tool in the game

 

Yeah! a standard mission played with autopilot leaving the view fixed just straight ahead.

But this standard mission will need to have always all the planes in exactly the same initial position.

Share this post


Link to post
Share on other sites
On 10/13/2018 at 8:48 PM, Jade_Monkey said:

This is the kind of stuff a smaller studio cannot afford to spend time developping.

What about a benchmark mission as suggested by SharpeXB? Could also be done by any of our talented mission makers. ;)

Share this post


Link to post
Share on other sites
On 9/26/2018 at 11:32 AM, sniperton said:

Moreover, FPS doesn't tell everything about performance. Thing is, the AI puts a high demand on the CPU, resulting in occasional stutters that do not always affect FPS. Such stutters and slow-downs are retained in tracks, but a mission recorded on a beefy system would run just fine when replayed on a weaker system, even though that weaker system would struggle to run the very same mission live with active AI. Despite that, a test track would be great to have, but we have to bear in mind that it's only good to measure graphic performance, not overall game performance. A built-in test function would be a different story, it could also benchmark your system according to how many AI objects you can have concurrently for a smooth gameplay. But please correct me if I'm wrong.

 Frame times do give you a better indication then FPS, if a wanting to judge if a game runs smooth. 

https://techreport.com/review/21516/inside-the-second-a-new-look-at-game-benchmarking

 

Edited by Dutch2

Share this post


Link to post
Share on other sites
On 10/19/2018 at 8:10 AM, sniperton said:

What about a benchmark mission as suggested by SharpeXB? Could also be done by any of our talented mission makers. ;)

I wonder if that’s not possible because the AI will do something different every time the mission is run. And the AI is what causes some of the load. Live gameplay is more demanding than a recorded track and I figure the AI is part of the reason

Back during EA of BoS 1CGS asked us to post performance numbers in order for them to develop the presets. I’m sure if a benchmark track had been possible, they would have created one back then as it would have been really useful. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×