In a real heavy campaign ("Opération Shõ Gõ") I tried to learn how DCS settings, monitor resolution impact fps in game. This could help campaign administrators.
Experimental conditions:
- Real heavy Philippines campaign mission: 23 active airbases, 77 aircraft, 25 ground orders, 75 flotillas orders
- conf.ini: best quality settings
- Monitor resolution: 1960 x 1200 and 2560 x 1600 (native)
- Vertical Synchronization ON (60fps max)
- Fly offline and online (PA-SEOW server/Hyperlobby, only one player)
- Result values are not interesting as absolute values since they are dependent on the player's computer hardware. However as relative values they may be useful to assist administrators design their campaigns for playability.
1. 1st download: MAXI mission = 222 KB
- Indust. Targets activated in ALL locations
- Enable Automatic Artillery Emplacements
- 100% of runway lights
- 100% of idle aircraft parked at airfields
==> 90 Chief_Road, 1492 Static Objects [NStationary], 739 objects [Buildings]
2. 2nd download: MEDIUM mission = 168 KB
- Indust. Targets activated only by GATTACK orders
- No Automatic Artillery Emplacements
- 50% of runway lights
- 50% of idle aircraft parked at airfields
==> 90 Chief_Road, 1290 Static Objects [NStationary], 84 objects [Buildings]
3. 3rd download: MINI mission = 101 KB
- Indust. Targets activated only by GATTACK orders
- No Automatic Artillery Emplacements
- 0% of runway lights
- 0% of idle aircraft parked at airfields
==> 90 Chief_Road, 544 Static Objects [NStationary], 84 objects [Buildings]
Results
I take off in a 4 aircraft flight (J2M3), I slot in the last plane, settings are full (no external views)
I register fps at 5 times:
A: Immediately when I am in cockpit without moving anything, trackIr deactivated, medium view.
B: Without hitting any key, when the 3 aircraft start their engines in front of me
C: When the 3 aircraft are ready to take off. My engine is off
D: I start my engine, the 3 others aircraft are still ready to takeoff
E: The 3 aircraft are airborne, I'm still on the runway, my engine ON
Remarks
- Native resolution is a good choice: When resolution decreases, I do not really increase my fps.
- Better results Online / Offline. I guess the server hosting increases my own processor resources?
- Stationary Objects (AAA, Stationary Units, idle aircraft, campfires...) have a very important impact on fps even they are far away from the takeoff point and even they are inactive (far away from the frontline). They are analysed by the processor in real-time: enemy units and aircraft detection, state (active or destroyed), orientation for AAA...
- objects [Buildings] have not an important impact on fps.
- After takeoff, my fps increase to 60 in every case except when flying over enemy ships' AAA
fps - DCS settings - Monitor resolution
Moderator: SEOW Developers
-
- Posts: 2206
- Joined: Mon 08 Jan 2007 11:10 pm
- Location: Perth, Western Australia
-
- Posts: 97
- Joined: Wed 31 Oct 2007 4:50 am
- Location: Denmark
- Contact:
Interesting post. Maybe it's not directly related but it would be nice to be able to predict what the server and client performance will be before the first campaign test. I understand that it is Map related but this thread would be a good place for people to post some general experiences so that campaign designers could get a rough idea of how much is too much.
One issue I have noticed is that while static objects may not affect performance they do dramatically increase transfer times. My experience is that around one thousand is the max for any kind or reasonable loading time for clients.
One issue I have noticed is that while static objects may not affect performance they do dramatically increase transfer times. My experience is that around one thousand is the max for any kind or reasonable loading time for clients.