6 minutes ago, harpsi said:

Yes. I am at KSFO and it shows 15-17 / 45-50.

Right, so what your eyes are seeing is 45-50fps.

I tried the LSFG app last weekend with the Fenix A320. It definitely makes a difference but it makes my mouse pointer disappear after some time. Anyone else seen this happen and know how to prevent it?

9 hours ago, jjf said:

I tried the LSFG app last weekend with the Fenix A320. It definitely makes a difference but it makes my mouse pointer disappear after some time. Anyone else seen this happen and know how to prevent it?

It think it happens when MSFS is not in focus. Enabling multi-monitor in the app might help.

Done a bit of skimming through here but now that it's been out for a while, what is the feedback from those with a 4xxx series card. Does this replace FG with a 4080 or does it work in addition to it? 

1 minute ago, Snuffleupagus said:

Done a bit of skimming through here but now that it's been out for a while, what is the feedback from those with a 4xxx series card. Does this replace FG with a 4080 or does it work in addition to it? 

In addition. E.g. baseline FPS 30 x2 = 60 FG-FPS. Multiply by x2 (120) or x3 (180) with lossless.

Edited June 26, 20241 yr by Cpt_Piett

4 hours ago, Cpt_Piett said:

In addition. E.g. baseline FPS 30 x2 = 60 FG-FPS. Multiply by x2 (120) or x3 (180) with lossless.

^ yes BUT there is a slight overhead so (these are just rough examples to make the point)

- without app: "base" fps of 35, doubled to 70fps using native Nvidia FG

- same setup, turn on the app to "x2" setting: "base" fps down to 30, doubled to 60fps using native Nvidia FG, doubled again to 120 fps by the app

- same setup, turn on the app to "x3" setting: "base" fps down to 25, doubled to 50fps using native Nvidia FG, tripled to 150fps by the app.

NB the numbers aren't exact: they just illustrate the basic pattern of the trade-off that happens when using the app, at least on my system.

 

Hmmm bummer. I bought it and spent the last hour playing with it. So far, I get better performance without it and with my 4080 FG alone. With it on and my FG on, it claims it's rock solid at 75fps (my monitors max rate) but it's a stutter mess. Without it, I get 40-50 fps and no stutters. Flightbeam SFO + FSLTL + PMDG 777.

7 hours ago, Snuffleupagus said:

Hmmm bummer. I bought it and spent the last hour playing with it. So far, I get better performance without it and with my 4080 FG alone. With it on and my FG on, it claims it's rock solid at 75fps (my monitors max rate) but it's a stutter mess. Without it, I get 40-50 fps and no stutters. Flightbeam SFO + FSLTL + PMDG 777.

Infuriating 😩  

Obviously I have no special insight but my hunch is (still) that it works better with higher "base" frame rates. My reason for saying this is partly that I can also turn my PC into a stutter fest if I turn on enough FSLTL traffic with ~200 aircraft at inibuilds KJFK in the Fenix, driving the base frame rate down to 15 or so.

If 40-50fps is your frame rate after native Nvidia FG, then this implies a base rate of 20-25fps. That could possibly be problematic. To eliminate this as the cause, maybe see what happens if you reduce your settings until your "base" frame rate is consistently higher than 35fps and see if it makes a difference? If so, then at least you'll know what's going on and you can decide whether the lower settings are worth the extra smoothness 🤷‍♂️ 

Edited June 27, 20241 yr by tfm

6 hours ago, tfm said:

Infuriating 😩  

Obviously I have no special insight but my hunch is (still) that it works better with higher "base" frame rates. My reason for saying this is partly that I can also turn my PC into a stutter fest if I turn on enough FSLTL traffic with ~200 aircraft at inibuilds KJFK in the Fenix, driving the base frame rate down to 15 or so.

If 40-50fps is your frame rate after native Nvidia FG, then this implies a base rate of 20-25fps. That could possibly be problematic. To eliminate this as the cause, maybe see what happens if you reduce your settings until your "base" frame rate is consistently higher than 35fps and see if it makes a difference? If so, then at least you'll know what's going on and you can decide whether the lower settings are worth the extra smoothness 🤷‍♂️ 

Good point.  During my experimenting over the last 1-2 weeks with this app, I've found I don't have to limit the framerate in either XP or MSFS to 30 and then double to get 60.  What I've found is if I set Max FPS to 60 in the global settings of NCP and then set my graphics options to get a base framerate of 35-37 (with the app running, this is important because there is overhead), the app with show 35 / 60 or whatever the base framerate is scaled to 60.  

I have no idea what's happening in the background though if the app is doubling my 35 and then NCP is limiting back down to 60 or what.  I've notice it has performed much better with the base rate of 35-37 than it did with 25-27.

Without the app running that base is more like 40-42 under the same conditions if that makes sense.

1 hour ago, rjack1282 said:

Good point.  During my experimenting over the last 1-2 weeks with this app, I've found I don't have to limit the framerate in either XP or MSFS to 30 and then double to get 60.  What I've found is if I set Max FPS to 60 in the global settings of NCP and then set my graphics options to get a base framerate of 35-37 (with the app running, this is important because there is overhead), the app with show 35 / 60 or whatever the base framerate is scaled to 60.  

I have no idea what's happening in the background though if the app is doubling my 35 and then NCP is limiting back down to 60 or what.  I've notice it has performed much better with the base rate of 35-37 than it did with 25-27.

Without the app running that base is more like 40-42 under the same conditions if that makes sense.

Yes this does make sense and is consistent with what I'm seeing, although I don't cap the frame rate in NCP. 

The overhead from the app can be quite steep and I wonder if this is why some people are getting worse results than they expect: the app won't simply double/triple your existing FPS: it knocks off a fair chunk first, and then doubles/triples whatever is left - which might not always be enough, because the "missing" base frames would have contained information that's now unavailable for the interpolation to work optimally. 

I'm locking my FPS to 30 (using @Noel's RTSS Front-edge sync method) and then I apply LSFG. However, my mouse cursor motion is slowed down, as if it was locked to 30 FPS (while the rest of MSFS works at 60 FPS). Has anybody been able to correct this for MSFS? Changing capture API from DXGI to WGC works for the mouse, but then I get lots of stutters.

My experience is that it’s tricky to find the optimal settings. I agree 100% with what @tfm is saying; base fps matters a lot. In my case it needs to be above 30. Any lower and neither native nvidia FG nor lossless can save the day.

I stopped using lossless for now as it was a rabbit hole getting it to work properly. When I eventually found settings that worked relatively well, I then got tearing as lossless-fps didn’t match my monitor refresh rate. So another round of changing settings. But ultimately I found it hard to deal with the input lag and artifacts. At the moment I’m happy at a bit higher than 60fps with native FG and G-Sync. Of course RTSS-locked ad modum @Noel. I just have to lower TLOD and AI traffic accordingly to avoid stutters.

Just an example of how powerful RTSS-lock in front edge sync is: the sim ran much smoother than running a modest CPU overclock which took way longer to accomplish than changing 3 settings in RTSS.

Edited June 27, 20241 yr by Cpt_Piett

10 minutes ago, Luis Hernandez said:

Has anybody been able to correct this for MSFS

I could be wrong, but I don’t think it’s possible. The mouse will lag behind when so many AI frames are inserted. That’s my basic way of understanding it.

When this recently caught fire, I had understood that windowed mode was required and therefore it wouldn't work with VR, but in one of the Facebook groups, someone had said the recent update had some change that made VR use possible. I can't find anything in any release notes or forums about that. Does anyone know what that guy is talking about? I have been looking for days.

On 6/24/2024 at 4:35 PM, andy1252 said:

Ron, I have a second monitor for things like LNM, Flight recorder or any h/w monitors and I haven't had any problems. Got my main screen set for 30/60, and nothing changes with regard to whatever is on my second monitor whether I have lossless scaling activated or not.

Please could you share a little more information about your configuration to validate it with mine and maybe I will have some light on the way since I have tried everything and I still have problems as long as I have more than 1 monitor connected.

Your video card?
Do you have both monitors connected to the same GPU?
Windows version?
Do you have them in extended mode or separate screens?
Do you use MSFS in full screen or windowed mode?