Newest Nvidia driver 378.49 causes memory leak

[zap737]

"What FSUIPC version are you on ?"

4.964

"The release notes of the latest version said VAS was being misreported in previous versions. That may account for the VAS difference."

I must have missed that ..
which release notes?
What exactly did it say about VAS?

I am on 4.964 as well, so we can rule that out.

One thing is for certain is if you change views to external and cycle through other views, your "starting VAS" goes down by about 200 MB.

Without FTX Vector, if I just load the aircraft at KDTW, starting in the VC, and not touch anything, VAS starts a 1910036 and decreases slowly to about 1850000 and stabilises, just sitting there.
If I cycle to external views through to the top down view then back in cockpit, VAS goes down to 1744404, a decrease of 165632 from 1910036. So one's definition of "Starting VAS" needs to be defined quite precisely.

[Martyson]

"The release notes of the latest version said VAS was being misreported in previous versions. That may account for the VAS difference."

I must have missed that ..
which release notes?
What exactly did it say about VAS?

[zap737]

"The release notes of the latest version said VAS was being misreported in previous versions. That may account for the VAS difference."

I must have missed that ..
which release notes?
What exactly did it say about VAS?

I can't find it, but it said it was misreporting VAS. Doesn't matter as we both have the same FSUIPC version now.

[Martyson]

"The release notes of the latest version said VAS was being misreported in previous versions. That may account for the VAS difference."

I must have missed that ..
which release notes?
What exactly did it say about VAS?

I can't find it, but it said it was misreporting VAS. Doesn't matter as we both have the same FSUIPC version now.

FSUIPC release notes said nothing about VAS being misreported in previous versions that I could find.

[Kosta]

Vaughan,
You said you are using recent drivers and have no VAS problems. I might have found something out, but I need a little help from you.
Can you please make screenshots of your P3D graphics settings and zip them somewhere for DL? (screenshots I can best compare to mine)

I think it's some setting that is causing problems, really. Because:
I have two profiles that I now tested with 378.78.

One profile, that came from one user, ended up after 3-4 hours straight test-flight (NGX, BIKF HDG180, FL310) with 1GB free VAS. While other profile, my profile that I used with older drivers too, in a VAS error somewhere mid flight.

So obviously it is some setting. It would be very good to know which, if I could narrow it down.
I thought maybe, if I get your settings, that might be possible.

[marcon8122]

Vaughan,
You said you are using recent drivers and have no VAS problems. I might have found something out, but I need a little help from you.
Can you please make screenshots of your P3D graphics settings and zip them somewhere for DL? (screenshots I can best compare to mine)

I think it's some setting that is causing problems, really. Because:
I have two profiles that I now tested with 378.78.

One profile, that came from one user, ended up after 3-4 hours straight test-flight (NGX, BIKF HDG180, FL310) with 1GB free VAS. While other profile, my profile that I used with older drivers too, in a VAS error somewhere mid flight.

So obviously it is some setting. It would be very good to know which, if I could narrow it down.
I thought maybe, if I get your settings, that might be possible.

I've found something tested again with f22 without doing anything default flight discovered that with latest drivers i've no vas increasing in 2 hours if i keep water to high if i move to ultra the vas leak appear. With 376 drivers seem that water in ultra doesn't cause the leak. This is what i've found i'm going to do a long haul flight with 378 and water in high settings.

[Kosta]

Ha, thank you! That helps! Next test: water only to medium, rest same...

[Martyson]

KDTW-KDFW
2 hours
at 34000 feet
Nvidia driver 378.78 (just started using this driver)

Vas increased and decreased during the flight
This driver is using vas just like all the previous Nvidia versions on my setup

Start of flight vas (FSUIPC vas remaining):
2066000
End vas :
1954280

Vas used :
111720



Today my starting VAS is shown in the first screenshot = 2453808

[Kosta]

Great, thanks too! I see some key settings comparable to lower settings, I have my eye now on tessellation factor, shadow quality and water quality. Right now flight with lower water is proceeding. Another hour or so.

I'm quite confident that some sliders cause leaks now. And also explains the differences people have.

[Martyson]

It would be interesting to know what the default P3D settings are when first default tests are run ... before changing settings or adding addons.

[Kosta]

It would be interesting to know what the default P3D settings are when first default tests are run ... before changing settings or adding addons.

I don't get it, what do you mean?

[Martyson]

It would be interesting to know what the default P3D settings are when first default tests are run ... before changing settings or adding addons.

I don't get it, what do you mean?

After a fresh install of P3D:
http://www.prepar3d.com/forum/viewtopic ... 2&t=123246

What are the default sliders / options are set to just prior to testing a default aircraft / screnery / flight for an OOM (VAS) problem(s)?

Were VAS problems (OOM) foundl?

Have pilots previously reported the same OOM's problems after they did a fresh install (default settings) of P3D?

[Kosta]

It would be interesting to know what the default P3D settings are when first default tests are run ... before changing settings or adding addons.

I don't get it, what do you mean?

After a fresh install of P3D:
http://www.prepar3d.com/forum/viewtopic ... 2&t=123246

What are the default sliders / options are set to just prior to testing a default aircraft / screnery / flight for an OOM (VAS) problem(s)?

Were VAS problems (OOM) foundl?

Have pilots previously reported the same OOM's problems after they did a fresh install (default settings) of P3D?

I rarely test clean P3D. Only if I have really heavy problems, then I deductively test, until I find the reason. There is always one.
I don't know now by heart what default settings are, honestly. Deleting prepar3d.cfg will reset though... not keen on doing it, as it also has some other effects.

I can only say that now, with medium water, flight is stable.

Now I am also thinking, it might be some water addon, like REX, as it changes textures and also animation.

I have textures backup, before REX installed itself, I will try restoring, and then try Ultra.

Current flight seems stable btw.

[Martyson]

"Have pilots previously reported the same OOM's problems after they did a fresh install (default settings) of P3D? "

Not on your flights .. or mine ..Do you recall what other pilots have said about OOM / VAS tests after a fresh default P3D install on their system using Win7 or Win 10, e.g.?

[Kosta]

Come on, don't talk about default P3D settings, as the sim should be able to cope with it's own settings. Even with sceneries. There shouldn't be a memory leak, which IS there with water on ultra (at least on my setup, currently).
I will, as I said, test without REX water.