FSPS FFTF Dynamic?

[F737MAX 5,589]

27 minutes ago, vgbaron said:

Every now and then I'll run it for a week or two - I never have seen any change, one way or the other. I think it might also depend on how well balanced your system is. The "tighter" your setup the less change you'll see

I probably ask a lot from my setup, so I can see as much as a 5-7 FPS gain by using it.

Definitely works for me.
Signed: A happy user.

[Snuffleupagus 327]

I also ask a lot from my setup (9700k @5.0) so that's also probably why I see such a significant improvement. I can say back when I was on the i5-2500k @4.5 and lower settings, I didn't notice much of a performance difference, but it did help with blurries.

[pgde 571]

Works for me. 

[d.tsakiris 418]

Here as well. Just wish they would (could?) implement the feature they had back in the version for FSX, namely that it would switch to unlimited TFR as soon as FPS dropped under the target, to bring back a few FPS.

[G-RFRY 1,710]

I use with FPS locked at 30 and it keeps it there in high detailed airports and no scenery blur, this is were I think it works best when you lock FPS it regains that loss that locking FPS causes.

[Navigator43 25]

On 4/13/2019 at 5:46 AM, G-RFRY said:

I use with FPS locked at 30 and it keeps it there in high detailed airports and no scenery blur, this is were I think it works best when you lock FPS it regains that loss that locking FPS causes.

I agree

[Dirk98 354]

On 6/21/2018 at 7:02 PM, vgbaron said:

I use it but haven't noticed anything good or bad about it. My sim runs the same, with or without it to my perception.

What I mean is, quite possibly if I set up test measurements I could see a difference it something but in actual use - everything looks the same. So if you are looking to see a big difference, I don't think you will except in testing.

 

Vic

Ditto, no harm noticed.

[SteveW 1,792]

Someone mentioned early in the discussion that the P3D system must be working well already because Peter is Robbed to pay Paul. When Peter has nothing, Paul get's nothing. The fill in rate is reduced to provide a lift to fps. If anything is already at or above a limit then effectiveness of the program is reduced.

[Dirk98 354]

If somebody forgot there was a bug with missing autogen when setting internal FPS @Unlimited in P3D, like many of us did. And this tool I think was supposed to deal with it in the first place (or how it got attention that time at least). Not all had been affected by the missing autogen and Unlimited setting though.

 

Edited March 4, 2020 by Dirk98

[SteveW 1,792]

Some might have found that  the autogen reacts and comes back to life if they go in to Task Manager and uncheck one of the CPU tickboxes in Details, Set Affinity. That 'nudges' the jobscheduler to find unblocked cores or Logical Processors and the autogen fill-in continues normally. This thread blocking situation seems to effect systems with setting too high for the speed of the aircraft. The reason is that P3D relies on system resources and pushes those out when running hard and has to wait for them.

[SteveW 1,792]

I notice talk of how Fixed fps on the slider somehow works FFTF differently to other settings. When we use the fixed or locked fps setting, the frame rate is produced at a predetermined frequency, so the ratio is calculable as a constant. No matter whether using locked or vsync or unlimited anything else, there will always be an operating ratio of loading data and building frames. A lower FFTF setting gives less time building the frame but setting zero won't reduce it to nothing. When near to or on the ground we can give up fill-in rate a little because we don't see it so much and allow fps to be maintained with locked, or fps increased with unlimited.

Edited March 4, 2020 by SteveW

[Rogen 1,087]

I chose FFTF Dynamic to resolve burries while running the unlimited fps setting in P3D with a  30 Hz vsync.

Even had a nice test area where at a particular spot I'd have a case of the blurries that nothing would resolve unless using the limiter in P3D (which unfortunatelly also came a cost of loosing too many fps to be useful) and FFTF Dynamic resolved the problem resulting in no blurries over the testing area.

Cheers

[SteveW 1,792]

So by limiting how much time the simulator spends on filling in to the distance, the nearer parts of the scene have more throughput.

[Evros 310]

58 minutes ago, SteveW said:

Some might have found that  the autogen reacts and comes back to life if they go in to Task Manager and uncheck one of the CPU tickboxes in Details, Set Affinity. That 'nudges' the jobscheduler to find unblocked cores or Logical Processors and the autogen fill-in continues normally. This thread blocking situation seems to effect systems with setting too high for the speed of the aircraft. The reason is that P3D relies on system resources and pushes those out when running hard and has to wait for them.

This is only the case when JobScheduler is used. At least that is what I have observed and collected from others. If that is the case, then whole terrain loading is tanked and you can end up with a complete blur with no distinguishable terrain. Missing autogen and gradual blurries are usually caused by running unlimited as it doesn't leave much to FFTF.

[SteveW 1,792]

17 minutes ago, Evros said:

This is only the case when JobScheduler is used. At least that is what I have observed and collected from others. If that is the case, then whole terrain loading is tanked and you can end up with a complete blur with no distinguishable terrain. Missing autogen and gradual blurries are usually caused by running unlimited as it doesn't leave much to FFTF.

Yes that's right on some systems. When core zero is avoided with an AM the blocking can happen on some systems when flying too fast for the settings.