Update for 1.10.40?

[Baddest-Juju]

Would really appreciate it. 😊

[TheRealBix]

Yep, not working with latest patch :/

[Cube-13]

Yes please, we love you long time!!

[StochasticTinkr]

For the record, its not as simple as "recompiling". It's necessary to find the correct memory addresses of several things that fallout 4 uses. It appears they've rearranged memory a little bit.

[Baddest-Juju]

Well shit

[Mortercotic]

Damn, hope this can be updated. Really helped my game performance.

[StochasticTinkr]

I've done the legwork to find the new addresses. I have my own fork of this project (reworked significantly). It's not quite ready for release yet though.

If xwise is interested in updating this, the addresses I found are:

        const unsigned __int64 offset_fShadowDirDistance = 0x676BA1C;
	const unsigned __int64 offset_iVolumetricQuality = 0x3903FD8;
	const unsigned __int64 offset_bGameUpdatePaused = 0x5AA50E4;
	const unsigned __int64 offset_bIsMainMenu = 0x5B473A8;
	const unsigned __int64 offset_bIsLoading = 0x5A9D300;

I couldn't find one that is pip-boy specific, but the "GameUpdatePaused" address includes pipboy, menu, console, etc...

shadowDistance hasn't been used in the past, so I removed it.

If I have time, I might make my fork "ready for release". I'll update here if I do.

[StochasticTinkr]

I've gone ahead and updated my own version of this here: https://github.com/StochasticTinkr/PerformanceTuner/releases/tag/0.9.0

It is slightly different than xwise's, but functions similarly.

[Cube-13]

Gravy, will give it a try :)

[Zzyxz]

@StochasticTinkr how can i read those adresses from the memory?
-nvm just found, that Cheat Engine was used... sigh how could i forget about that...

[Baddest-Juju]

Awesome, thanks. Does the dxgi_linked functionality still work?

[StochasticTinkr]

@Baddest-Juju: I didn't touch that code, but didn't test it either. So maybe?

[Zzyxz]

@StochasticTinkr Question regarding the functionality. Did you remove the fEmergencyDropFPS&fEmergencyDropRate?

Also, fIncreaseRate&fDecreaseRate is now only hardcoded as it seems, but why removing them?

Thanks for updating the project and reworking it.
I'm also now looking into this topic, but forgive me, i'm a java/web programmer. so c++ is kinda new to me.(Only had basic instructions at the university.)

[StochasticTinkr]

I'm using a completely different algorithm than xwise.
The way it works now:
There is a "momentum" value.
If it is going in the wrong direction, we set it to +1 or -1 depending on the direction it should be going.
If it is going in the right direction, we accelerate it (multiply it by 1.025).
The momentum value is then capped between -500 and 500.

After the new momentum is calculated, we add it to the current shadow dir distance.
We then use the new shadow dir distance to calculate the GR quality (if you have that feature enabled).

[Zzyxz]

Is it normal that the TargetFPS caps the FPS? I'm running normally at 60 fps and i set targetFPS to 48 and it caps it at 48 all the time.
As i want the shadows to reduce only, when i fall under 48 FPS.

At some places, even with shadowdistance 2000 im at 55 fps and with 8000 im at 52 fps. So i'd like to keep my 8000 distance at 52 FPS.(which runs fine)