investigating Framerate decreasing up to 16% with 2.8.0

[Limaro]

New Version, new hope. I've separated ST from MT, for clearer view.

Beirut 4_2 ST:

2.7.18     74.9 FPS
2.8.0      65.8 FPS
2.8.1      66.4 FPS
2.8.3 ST   68.0 FPS
2.8.4 ST   65.4 FPS
2.8.6 ST   65.9 FPS
2.8.7 ST   67.4 FPS

Beirut 4_2 MT:

2.8.3 MT   73.8 FPS
2.8.4 MT   70.3 FPS
2.8.6 MT   71.2 FPS
2.8.7 MT   71.4 FPS

Beirut 4_3 ST:

2.7.18    100.9 FPS
2.8.0      88.1 FPS
2.8.3 ST   92.3 FPS
2.8.4 ST   89.3 FPS
2.8.6 ST   88.6 FPS
2.8.7 ST   90.0 FPS

Beirut 4_3 MT:

2.8.3 MT: 103.4 FPS
2.8.4 MT  102.1 FPS
2.8.6 MT   99.9 FPS
2.8.7 MT  104.5 FPS

Small Reminder, my 4_3 replay is recorded at good weather in Beirut. There, it seams that we gained some FPS again using MT. But always to remember that I changed CPU meanwhile. But with cloudy day (4_2) it's still a big loss.

I'd still whish a nice benchmarking tool for DCS to get easy reproducable performance values.

[Limaro]

Very last Update.

2.8.8
Beirut 4_2 MT: 72.1 FPS
Beirut 4_3 MT: 100.2 FPS

As you can see no interesting changes. I guess we'll have to accept this loss.

I just hope, that we do not get another drop with 2.9.

[darkman222]

What does 4_2 and 4_3 stand for? DCS 2.8.8.43?

So you gained 30 fps between them then?!

I changed to a better, or the best CPU on the market ( I13900KS) So I cant compare results anymore either. But most of the time DCS runs smoothly. Except for the heavy scripted servers like the 4ya servers.

My expectations for 2.9 go in both directions. I wont be surprised either way.

Edited September 25, 2023 by darkman222

[Limaro]

Am 25.9.2023 um 11:31 schrieb darkman222:

What does 4_2 and 4_3 stand for? DCS 2.8.8.43?

So you gained 30 fps between them then?!

I changed to a better, or the best CPU on the market ( I13900KS) So I cant compare results anymore either. But most of the time DCS runs smoothly. Except for the heavy scripted servers like the 4ya servers.

My expectations for 2.9 go in both directions. I wont be surprised either way.

 

These are my two Tracks I use to measure to have a reliable comparison. Described in Initial Post but meanwhile a little far away from this one

You can still compare performance, just use a second installed instance of old Version.

Edited September 26, 2023 by Limaro
typo

[Limaro]

Only one word can describe my feelings: Wow!

So this is my newest and last data. Only compared to last tests with multi threaded 2.8.8 and my tracks I use (attached in very first thread)

Trackfile Beirut 4_2 : 2.9.0  84.9 FPS (2.8.8 72.1 FPS)
Trackfile Beirut 4_3 : 2.9.0 120.0 FPS (2.8.8 100.2 FPS)

Unfortunately the 4_3 does not ply until the end, as the F16 crashes now... ich guess something changed in the responsiveness of the inputs, so the flight ends a little earlier.. but the values speak clear language.

EDIT: 2.9.0 4_2: 70.5 FPS, no further tests. as in comment after this post, I missed the default settings changes

Edited October 19, 2023 by Limaro

[Gripen 4-1]

11 minutes ago, Limaro said:

Only one word can describe my feelings: Wow!

So this is my newest and last data. Only compared to last tests with multi threaded 2.8.8 and my tracks I use (attached in very first thread)

Trackfile Beirut 4_2 : 2.9.0  84.9 FPS (2.8.8 72.1 FPS)
Trackfile Beirut 4_3 : 2.9.0 120.0 FPS (2.8.8 100.2 FPS)

Unfortunately the 4_3 does not ply until the end, as the F16 crashes now... ich guess something changed in the responsiveness of the inputs, so the flight ends a little earlier.. but the values speak clear language.

 

Be careful: In 2.9.0 antialiasing is OFF by default; just in case you tested with AA on in the previous versions

[Limaro]

vor 1 Stunde schrieb Gripen 4-1:

Be careful: In 2.9.0 antialiasing is OFF by default; just in case you tested with AA on in the previous versions

you are right. However, I am now happy using TAA, looks good for me and performance is much better.

Edited October 19, 2023 by Limaro