Oi. Even with the test version 3 or the latest stable build...flying around New York or London just eats away my CPU so bad. Even with all my apps off, zero traffic.

AutoFPS makes it doable at least. Time for an upgrade

 

19 hours ago, mmcmah said:

I think that there might be a bug in -test4. I started in IFR and then switched to VFR after the climbout. Despite having enough FPS headroom, the program would increase the TLOD. However, as soon as I switched the focus AWAY from MSFS, it started increasing the TLOD. Once I clicked back in MSFS, the TLOD would freeze until I shifted focus again.

Great find! I messed up the new TLOD on ground logic for sensitivity mode whereby it was using the wrong automation method code when above 1000 ft, thus behaving like it was in FPS tolerance mode. Thanks for describing it so well, which made it easier for me to find and fix.

I'll put out a fixed test version shortly.

22 hours ago, Reset XPDR said:

I've thought a bit more on this and have the idea that instead of using the 5 second FPS rolling average like is used in all other modes. Following the initial rapid course TLOD adjustment, I could use a much longer longer FPS rolling average, say 30 seconds or a minute, for the remaining time on the ground so that it needs a much longer FPS change trend before adjusting TLOD again, meaning panning around shouldn't trigger a TLOD change so often. I'll try coding it up and see how it goes.

I trialed this yesterday and it did not work out as having such a long rolling average caused overshoots and resultant hunting which made the issue worse, so back to the drawing board on this one. I'm thinking I might just try doing an initial rapid adjustment on the ground to a TLOD that still gives a bit of FPS headroom (kind of like TLOD Min Extra but a variable amount rather than a set 50) and then only change TLOD again on the ground before takeoff if there is a significant drop in FPS way below target FPS so as to restore performance. I will tinker with it and see what works best.

New test version v0.4.2.17-test5 is now available here.

Changes since v0.4.2.17-test4:

• Fixed logic error in FPS sensitivity automation method code causing FPS tolerance to be inadvertently used at times. (Thanks @mmcmahfor identifying this bug)

3 hours ago, Reset XPDR said:

(kind of like TLOD Min Extra but a variable amount rather than a set 50) and then only change TLOD again on the ground before takeoff if there is a significant drop in FPS way below target FPS so as to restore performance.

sweet

15 hours ago, Reset XPDR said:

Fixed logic error in FPS sensitivity automation method code causing FPS tolerance to be inadvertently used at times

Was this only in VFR mode? I've only been using IFR mode lately, even in GA. I'm having issues with stutters below 1000ft on takeoff and approach in VFR mode. 

On 8/23/2024 at 12:22 AM, Reset XPDR said:

Note, it is not 10000 ft AGL that it necessarily triggers at, it is whenever the terrain elevation is above the trigger alt you have set, which from your post seems to be 5000 ft. ie. if you are 10000 ft AGL over ocean (ie. terrain elevation of 0 ft), it will not trigger but if you are at 1000 ft AGL over a 5500 ft mountain (ie you are flying at 6500 ft MSL) it will trigger. The app calculates terrain elevation by subtracting AGL from MSL altitude. Does this make sense?

I still don't entirely understand this logic. For now I've set it up like this:

I'm going to fly over some remote mountaineous areas and see if I can understand the logic better. 

Edited August 25, 20241 yr by Cpt_Piett

Why when i use this app the buildings and objects in approach loads very slowly when i am arriving them?

1 hour ago, Cpt_Piett said:

Was this only in VFR mode? I've only been using IFR mode lately, even in GA. I'm having issues with stutters below 1000ft on takeoff and approach in VFR mode. 

I still don't entirely understand this logic. For now I've set it up like this:

I'm going to fly over some remote mountaineous areas and see if I can understand the logic better. 

Its getting complicated for me too. 

26 minutes ago, GSalden said:

Its getting complicated for me too. 

Good to hear I'm not alone 😉

Another thing: just tested lossless scaling with AutoFPS. Don't know how useful this combo is, as AutoFPS doesn't take into account the total FPS that lossless gives. E.g.: I'm using lossless in x2 mode hence 63x2=126 fps, whereas AutoFPS shows 63 fps. Hence, AutoFPS is not able to use the increased headroom for higher TLOD. 

Which is okay, since I don't use the lossless mod. I just wanted to test it. 

Edited August 25, 20241 yr by Cpt_Piett

Same here 😄

23 minutes ago, Cpt_Piett said:

Good to hear I'm not alone 😉

I am sticking with 4.2.16  It works perfectly.

1 hour ago, Bobsk8 said:

I am sticking with 4.2.16  It works perfectly.

That's probably a good idea if you don't need the added features. 

Still testing .17-test5. Everything seems to be working well, I haven't noticed any strange issues. But I still don't understand the new mountain TLOD logic. 

The mountain logic kicked in:

Haha. I am looking at this in shock as any TLOD above 200 just kills my CPU into a stutter fest when I am near the ground haha.

Well, the new mountain logic seems to be working. I still don't understand how it's working and what settings to use. But it's working 🙂 Performance so far is excellent - been flying for hours today with the latest version - GA and airliner. 

Screenshot below is near Chamonix, France. 

Edited August 25, 20241 yr by Cpt_Piett

This ap is just awesome.  I'm using 4.2.16 with fps @30 with AUTO TLOD & OLOD 50 to 200 and alt of 1000 to 5000. The ap adjust my setting to give my a smooth simulator without changing my settings. Does the new version allow these simple settings?