To all VAS and OOM testers that use FSUIPC for monitoring...

[Guest]

I am seeing quite a few posts from people who are saying their VAS dropped during a flight and that VAS was incredibly low at the end of a flight. Take note that the FSUIPC VAS monitor does NOT show you (how much) VAS (is being used) but the amount of free space that's left! When VAS is growing, the number that FSUIPC shows will get lower! So when FSUIPC shows a number of for instance 1.6 GB you shouldn't say "VAS was 1.6" but "VAS was (4.0 - 1.6 = ) 2.4"! 

 

Because there is a lot of talk about VAS and OOMs on this forum right now, I thought it would be a good idea to make sure everyone is talking about the same. Imho it would be best if everyone that uses FSUIPC for monitoring would post the actual VAS by doing the math and NOT the number that FSUIPC is showing UNLESS you make it clear that you are talking about the space that's left and not VAS. Or use something like Process Explorer.  ^_^

 

As it is now it sometimes gets very confusing because people are talking about VAS that's dropping while it is actually rising or people are saying there is no problem with VAS while there actually is!

 

Because there are various topics about all this I decided to post this in a separate topic so I won't have to crosspost is all over the place.  ^_^ (I'll just post the link all over the place LOL)

[fulcrum_64th 9]

How does one get the fsuipc monitor on?

[RelaxX 11]

Sorry, would someone please define (explain) what is VAS and how exactly this can be measured?

[Guest]

How does one get the fsuipc monitor on?

 

 

Go to the FSUIPC logging tab. Enter 024C in one of the 'specific value checks' boxes (0 = zero, not the letter O). Change the 'type' from S8 to S32 and check one of the available options where you want to show the remaining available VAS (I would go for the upper title bar because showing it onscreen may result in stutters).

 

 

 

Sorry, would someone please define (explain) what is VAS and how exactly this can be measuered?

 

http://en.wikipedia.org/wiki/Virtual_address_space

 

To measure it, see above (to show available remaining VAS) or use Process Explorer (http://technet.microsoft.com/en-us/sysinternals/bb896653.aspx) (to show current actual used VAS) and set it up like this: go to View > Select Columns > Process Memory tab and check Virtual Size. Press OK and click on the Virtual Size column which will put P3D on top (unless you are running other programs that are using more VAS  ^_^ ). 

[Ray Proudfoot 8,954]

It's worth mentioning that for those running WideFS setups you can download VAS_Monitor.zip from the FSX Utility section of the AvSim Library.

 

This gives an instantaneous readout of the current fps and remaining VAS.

[spirit_66 2]

I am seeing quite a few posts from people who are saying their VAS dropped during a flight and that VAS was incredibly low at the end of a flight. Take note that the FSUIPC VAS monitor does NOT show you (how much) VAS (is being used) but the amount of free space that's left! When VAS is growing, the number that FSUIPC shows will get lower! So when FSUIPC shows a number of for instance 1.6 GB you shouldn't say "VAS was 1.6" but "VAS was (4.0 - 1.6 = ) 2.4"! ...

Why shouldn't I say that the VAS dropped? One start with a given amount of VAS and the monitor shows what's left for use. If it drops steadily then there will be the moment when it's not enough anymore - bummmm - OOM!

Spirit

[Guest]

Why shouldn't I say that the VAS dropped?

 

Because that would be like saying the level of water in a glass drops when you pour more water in it.  ^_^

 

Apart from being wrong, it simply becomes confusing if some people say VAS dropped when VAS actually went up. That's all. In one post someone might say VAS went up and in the next post someone might say VAS went down while they actually meant the same… that's what I'd call confusing!

[Yuji Y 0]

Or maybe ask Peter D. to change from the remaining VAS to VAS in use?   Or can it cause even more confusion? B)   Anyway the gauge is now so important as framerate monitering in P3Dv2x. I do not like to resize the geen sceen every time I start the sim either. 

[FINXPAS 5]

Just indicate the source of your VAS number in your post. Fsuipc = remaining . Xyz= used.

 

Either way, we're all watching the same thing. Water filling a 4gb room as the sim has less air to breathe and eventually drowns.

[Piotr007 166]

VAS is one big killjoy. We have to get rid of it!

 

[MikeT707 1,181]

I just downloaded Process Explorer and am having a hard time figuring out how to use it to check VAS levels in P3D.  Any assistance would be appreciated.

[fppilot 4,370]

 

 

Or maybe ask Peter D. to change from the remaining VAS to VAS in use?

 

When I read the initial post in this topic that is the first thing I though of as well.  If the direction is a common err in discussion it may indicate some consideration for this proposed change.

 

 

Water filling a 4gb room as the sim has less air to breathe and eventually drowns.

 

Too bad no one has designed a drain.....

[Jim Young 2,356]

 

 

I am seeing quite a few posts from people who are saying their VAS dropped during a flight and that VAS was incredibly low at the end of a flight. Take note that the FSUIPC VAS monitor does NOT show you (how much) VAS (is being used) but the amount of free space that's left! When VAS is growing, the number that FSUIPC shows will get lower! So when FSUIPC shows a number of for instance 1.6 GB you shouldn't say "VAS was 1.6" but "VAS was (4.0 - 1.6 = ) 2.4"!

 

This is incorrect Jeroen or I'm not understanding what you are saying.  For example, I did a flight from Langley AFB to Andrews AFB in the Raptor with Vegetation maxed.  The FSUIPC log stated I had 2.37 GB's of VAS when the flight started.  At the end of the flight it showed I had 0.23 or 240MB's of VAS remaining and the OOM warning signs went off when I had about 0.320 GB's left (less than 1GB).  It clearly shows that something was depleting VAS fast.  I know for sure I did not have almost 4GB's of VAS remaining at the end of the flight.  I had less than 1GB.  So we need to get this straightened out before we go further.  Bob Scott's excellent gauge does the same as Pete's and shows how much VAS is left.  If it were the other way around as you described, then we should not be receiving out of memory warnings from Bob and Pete's gauges when the VAS gets down to around 320000 KB's.

 

Hope you understand and thanks for starting a post to get this all clarified.

 

Best regards,

[simmerhead 885]

I just downloaded Process Explorer and am having a hard time figuring out how to use it to check VAS levels in P3D.  Any assistance would be appreciated.

 

 

Google "Word Not Allowed process explorer" and you'll find a simple explenation on Word Not Allowed's blog. For some reason I can't link to his site here?

[geolpilot 88]

l

 

It clearly shows that something was depleting VAS fast. I know for sure I did not have almost 4GB's of VAS remaining at the end of the flight. I had less than 1GB.

 

Like I suggested elsewhere, log the FSUIPC VAS to the FSUPIC logfile and graph it in a spreadsheet. Picture shows clearly then exactly what is happening. VAS is being depleted / used / memory space is dropping! I call that depletion in VAS just as firehawk does i.e the byte count is going down (well not always hint hint)!

 

I am in the process of doing very thorough VAS testing, taking all sorts of variables into account like varying AGN density settings, AGN veg off, AGN buildings off, location changes, LOD settings,  FTXG on/off, addon Landclass on/off - many, many, hours of flight testing and graph plotting.

 

Getting near the end and really hoping to post the results soon, because I see so many conflicting stories on VAS that my tests can reconcile so well, and some totally wrong conclusions that some are jumping to, that I hope my testing will put an end to. Will be a bit of a thesis, just hope after all the hours I am putting into this it will be read!

 

What I can say definitively now is that veg autogen does not cause a steady memory leak, but it does cause an intial VAS hit / depeletion (unrecovered) just after takeoff, and that initial VAS loss is bigger the higher the autogen density setting (Dense and Very Dense are not good to use, especially with high LOD, unless you are flying in the total wilderness -sounds contradictory I know, but there is a perfectly good explanation - veg stabilises in VAS usage, something else that you will not get in wilderness areas does not :ph34r: )!

 

It will actually come as a bit of a surprise, as it did to me, what is actually causing the steady VAS countdown in specific situations (hence not all will experience it until they really get to fly a lot).

 

But I am staying mum for now, as the scientist in me is stilll screaming "double check, reproduce!" cos the results have serious implications on how one uses P3D 2.1 in various situations / locations (clue above).

 

Maybe up in 24 hours (has to be - I'd better get back to real work soon!)

 

Rob