Why is P3D handling VAS so poorly?

[GSalden 4,679]

3 minutes ago, tooting said:

I can confirm the new drivers leak like hell using the the fsl out of Gatwick but Ironically dont leak going in... You couldn't make it up.

I am not a technical guy , but to me that sounds impossible.

A scenery uses X mb and not Y mb flying towards-over it and Z mb flying over-away from it seems to me.

There are some guys around here who have good technical knowledge that might explain this..

 

 

[Ray Proudfoot 8,989]

2 hours ago, Afterburner said:

I'm glad to hear that the combination of rebuilding the p3d.cfg and running the 376.33 driver has improved your VAS usage.

I can say that I have seen many instances where users who had tweaked their cfg-file to death started experiencing issues sooner or later, and deleting the configuration file and letting P3d rebuild a new one gave them a relief. Personally, I only "tweak" using the P3D internal display settings (except of maybe the cloud shadow size), and it has worked best for me.

Agreed. I'm not a tweaker. Once I've found settings that I'm happy with I stick with them. The only things I have adjusted recently in FSX was the Ai traffic count in FSUIPC. It was pretty obvious once I thought of it I was too ambitious with my original settings. Although I have yet to fly a route with these new settings I'm reasonably confident I won't get any OOMs.

I too won't be adding any further 'tweaks' via the cfg. Justvia the interface.

[Ray Proudfoot 8,989]

2 hours ago, tooting said:

I can confirm the new drivers leak like hell using the the fsl out of Gatwick but Ironically dont leak going in... You couldn't make it up.

You might care to qualify that statement. Which version are the 'new drivers'? Anyone reading this in a few weeks won't have a clue which ones you're referring to.

[fluffyflops 1,859]

Nope its completely possible.  With the new drivers and all of them now the latest driver back to. 33 any flight I do out of Gatwick in the Fsl the minute I pass 6000 feet the vas half's.  I'd post videos to show it on 3 flights but it has me swearing in it and Baron doesn't like that.

However I can fly into Gatwick with no issues.  Once again I can show a video proving it.

For me using the fsl with a 980ti I definitely have the leak using the fsl on the newer drivers after the. 33 ones

[Ray Proudfoot 8,989]

2 hours ago, tooting said:

Nope its completely possible.  With the new drivers and all of them now the latest driver back to. 33 any flight I do out of Gatwick in the Fsl the minute I pass 6000 feet the vas half's.  I'd post videos to show it on 3 flights but it has me swearing in it and Baron doesn't like that.

However I can fly into Gatwick with no issues.  Once again I can show a video proving it.

For me using the fsl with a 980ti I definitely have the leak using the fsl on the newer drivers after the. 33 ones

By "Fsl" do you mean their A320? And you could prove your theory by flying the same departure with a default aircraft. Doesn't have to be precise. Just take-off and fly the same heading as before. Report back if the same thing happens.

[ZKOKQ 984]

2 hours ago, tooting said:

  I'd post videos to show it on 3 flights but it has me swearing in it and Baron doesn't like that.

 

LOL

[Petraeus 147]

That was my experience with 3.4, Ray. I have now gone back to 3.2, which is much better at purging redundant VAS. OOM's are now extremely rare, although they can still be induced (FSL A320 from Aerosoft EGLL to Aerosoft EHAM, for example), but it's got to the stage where I've stopped looking at the VAS counter..

[Ray Proudfoot 8,989]

Hi Petraeus. I've just bought 3.4 so going back to an earlier version is not an option for me.

Has the VAS handling been reported to LM? Is it worth raising the issue with them?

[Cruachan 705]

2 hours ago, Ray Proudfoot said:

Has the VAS handling been reported to LM? Is it worth raising the issue with them?

Hi Ray,

Take your pick!

http://www.prepar3d.com/forum/viewtopic.php?f=6315&t=124227

------------------------------------------------------

http://www.prepar3d.com/forum/viewtopic.php?f=6312&t=120967

http://www.prepar3d.com/forum/viewtopic.php?f=6315&t=124373

http://www.prepar3d.com/forum/viewtopic.php?f=6315&t=124232

If you, and others, are feeling strongly about all this and, quite clearly you do, then I would encourage you all to add your experiences to the mix. I think the first thread would be the most appropriate. Rob McCarthy, LM Core Lead, is involved so he will need to know that this matter is not yet resolved.

The more voices that are heard the more likely we will see a resolution.

Regards,

Mike

[NZ255 309]

On 07/05/2017 at 11:01 PM, Ray Proudfoot said:

Why? Why isn't P3D releasing VAS as I pass complex areas?

I guess the software doesn't do it, or not to the level you expect

[Ray Proudfoot 8,989]

Mike,

Thanks for the links. I've just completed a flight from ENGM to EGLL in Concorde with Nvidia drivers 376.33.

At ENGM I had 1500Mb remaining. This slowly reduced on climbout to around 1300Mb. After passing the coastline P3D recovered around 150Mb. VAS usage slowly increased during the flight with around 1200Mb remaining as I crossed the UK coastline.

Approaching London it further increased but abeam EGLL for a 09R landing there was 900Mb remaining. Fps were 24 on final approach.

An hour after landing and parked there is 885Mb free VAS. I think I've cracked it through a combination of different drivers and lower settings. I may tweak settings slightly higher and see how I get on.

[Ray Proudfoot 8,989]

2 hours ago, NZ255 said:

I guess the software doesn't do it, or not to the level you expect

It should do anything FSX could do especially as P3D is quite mature these days. Change of drivers appears to be the answer.

[fluffyflops 1,859]

Ive just completed ltn-nce-ltn in the Fsl 320 using p3d 3.35 and the. 33 drivers.  The vas management was great.  I certainly won't be trying out the new drivers again for a long time.

The 2 areas I'm getting leaks for sure is after departing London and around the nova scotia area of Canada when crossing the pond

[Cruachan 705]

3 hours ago, Ray Proudfoot said:

Mike,

Thanks for the links. I've just completed a flight from ENGM to EGLL in Concorde with Nvidia drivers 376.33.

At ENGM I had 1500Mb remaining. This slowly reduced on climbout to around 1300Mb. After passing the coastline P3D recovered around 150Mb. VAS usage slowly increased during the flight with around 1200Mb remaining as I crossed the UK coastline.

Approaching London it further increased but abeam EGLL for a 09R landing there was 900Mb remaining. Fps were 24 on final approach.

An hour after landing and parked there is 885Mb free VAS. I think I've cracked it through a combination of different drivers and lower settings. I may tweak settings slightly higher and see how I get on.

Hi Ray,

Glad to see you are making progress. I note several threads on this topic over at the P3D forums have now been locked. Here is one that is still live. It was started last year but there are some recent posts. I think you will find a few kindred spirits 😉

http://www.prepar3d.com/forum/viewtopic.php?f=6312&t=120967

Just checked and I see it was included in my previous post - sorry!

Regards,

Mike

[Ray Proudfoot 8,989]

53 minutes ago, Cruachan said:

Hi Ray,

Glad to see you are making progress. I note several threads on this topic over at the P3D forums have now been locked. Here is one that is still live. It was started last year but there are some recent posts. I think you will find a few kindred spirits 😉

http://www.prepar3d.com/forum/viewtopic.php?f=6312&t=120967

Just checked and I see it was included in my previous post - sorry!

Regards,

Mike

Mike,

I didn't have the enthusiasm to read through 13 pages. But glancing through there doesn't appear to be much input from LM support.

I can live with my situation. I may even be able to increase one or two settings. I think having the 1050 card benefits me because I can revert to earlier drivers which don't have a VAS leak. Owners of the 1080 can't use those as the card is not supported.