P3D V3 FMC memory leak back?

[kriggelb 0]

I did not read the entire thread (although I've been following it since last November), but I can say that not only approach procedures containing 'intercept radial XXX to FIX' do bring the VAS leak.

Also SID's containing it will OOM after 10-11 hours flight after a long and steady VAS leak.

As an example try KDFW-VHHH out of either 17C-17R-18L-18R via LOWGN7 SID (do not enter any arrival/approach procedure for VHHH). Same VAS leak will be observed.

 

I had no problem not entering approach into FMC before T/D (that was the first workaround advised by PMDG), but not being able to use any SID containing that intercept radial to fix thingy is another problem far less ignorable than the first one.

 

just my 2 cents,

Happy flying!

Hello denthibbe,

I just did the flight (KDFW-VHHH) out of 17C via LOWGN7 SID and I had no issues at all.

I currently fly with cycle 1604 and the respective SID contains the following text:

"SID LOWGN7

RNW 17C HDG 176 INTERCEPT RADIAL 157 TO FIX NAVYE FIX NAVYE FIX JGIRL AT OR ABOVE 5000 SPEED 240 FIX CORTS FIX BIGGD FIX CMORE FIX LOWGN"

I use FSUIPC to monitor VAS (remaining amount) and I was able to fly with a steady value of 1,2 to 1,3 GB remaining during cruise until TOD (more than 12 hours of flight).

Just a little feedback - until now I've never encountered a SID-leak.

Kind regards,

Chris

[xkoote 57]

I flew a second flight into Dubai. Again I started running out of VAS. This time I selected the VFR to runway 12R. I departed from a different airport (LOWW) and again I started getting beeps about 3 hours in. I didn't open a ticket yet because I have not tried other airports or done enough flights to exclude maybe scenery conflicts or maybe that 1605 is the culprit. 

About 2 hours out, I saved and restarted and had 2.4 Gb of free VAS. By the time I landed less than 2 hours later I barely had 1 Gb.

So I'll have to try other airports or something to see if it is indeed only when going to OMDB, or if maybe going back to 1604 gives better results.

[CaptKornDog 284]

Let us know your results.

[PinkPony 171]

I flew a second flight into Dubai. Again I started running out of VAS. This time I selected the VFR to runway 12R. I departed from a different airport (LOWW) and again I started getting beeps about 3 hours in. I didn't open a ticket yet because I have not tried other airports or done enough flights to exclude maybe scenery conflicts or maybe that 1605 is the culprit.

About 2 hours out, I saved and restarted and had 2.4 Gb of free VAS. By the time I landed less than 2 hours later I barely had 1 Gb.

So I'll have to try other airports or something to see if it is indeed only when going to OMDB, or if maybe going back to 1604 gives better results.

This is hardly a example of controlled test scenario. As a result your OOM cannot be attributed to a leak, which we have identified is caused by a specific string of in missed approach coding.

 

A FSUIPC VAS warning "ding" after merely three hours is a strong hint at your VAS issues being soecific to your configuration. Furthermore, your selected VFR approaches do not contain a missed approach procedure and which further points a configuration issue.

 

I sugget you go back to the beginning of this thread, and pay particular attention on how to set up controlled test scenarios (discussed in detail by the likes

of Kyle, Michael and Dan).

 

Specifically, loading a vanilla P3D with all other addons and addon scenery etc. disabled, and setting up the PMDG at a random default airport. Selecting and executing a approach for one of the approaches you believe may be at fault.

 

From the lack of information provided by you so far, you could be running a 6.5 LOD value with 4096x textures, and every Orbx vector feature enabled...for all we know

[Boeing or not going 578]

Everyone has a different setup. A lot of vas issues are caused by not disabling other addon scenery so if you are flying over an Aerosoft airport in cruise for example, there are some textures that get loaded into memory (and not dumped if you like me still are stuck on fsx...)

 

Disable all other addon scenery. Also RAAS adds to your vas. So does obviously a repaint if the painter uses 4096 textures. Another addition to vas is the use of the 5.1 surround sound set in the pmdg ops center.

 

David Lee

[xkoote 57]

Yes, Leo, that is why I said I didn't open a ticket.

[kamaz74 81]

Everyone has a different setup. A lot of vas issues are caused by not disabling other addon scenery.

I understand your skepticism, but I can corroborate with the other posters in this thread that loading SIDS or approaches into the PMDG 777 which contain heading to radial intercept maneuvers cause an immediate 120MB/hr vas leak. At first, I thought FTX Global was causing my problems. I even reinstalled Windows 7 x64 Pro and installed vanilla FSX SP2 and the PMDG 777 (latest version)(No other addons were installed) and the issue is still there.

 

Here is the route I used which first presented the leak:

 

NTAA VAIT3W VAITE J21 MA DCT 15S156W 12S163W 10S168W 07S175W 05S179W 02S175E DCT KASKU DCT MUBIT DCT PNI A222 UNZ M501 NOMAN V531 BETTY BETY2A ILS07L VHHH

 

If I dive into the NTAA.txt file in the PMDG/SIDSTARS folder, we find the following under SIDS:

 

SID VAIT3E RNW 04 HDG 042 UNTIL 500 HDG 269 INTERCEPT RADIAL 309 TO FIX VAITE FIX VAITE

SID VAIT3W RNW 22 HDG 222 UNTIL 3.0 FROM FIX TAF SPEED 205 HDG 357 INTERCEPT RADIAL 309 TO FIX VAITE FIX VAITE

 

Now, with the VHHH.txt, we find the following under APPROACHES:

 

APPROACH ILS07L FIX CI07L AT OR ABOVE 1700 FIX FI07L 1700 RNW 07L HDG 073 UNTIL 300 TRK 028 INTERCEPT RADIAL 058 TO FIX DEDEE SPEED 230 FIX DEDEE SPEED 230 FIX LINGI FIX D333N FIX SAMPU 5000

 TRANSITION LIMES FIX LIMES SPEED 210 FIX OVERFLY TONIC SPEED 180 HDG 040 INTERCEPT RADIAL 073 TO FIX CI07L

 

Now, the final test I did was I loaded up at NTAA and entered takeoff runway 04 then direct VAITE intersection. I didn't load the troublesome VAIT3E SID, only the runway. Then, I loaded the troublesome ILS07L approach into VHHH into the FMC, finished configuring the FMC and departed as usual. As I climbed to my cruise altitude, I started to monitor VAS leakage. It was at a consistent 120MB/hr. I watched for several hours as my VAS climbed from 1.88GB to 2.57GB. I then went into the FMC and deleted the approach as best I could leaving only the arrival runway. My leak stopped immediately and my VAS usage held steady at 2.57GB for the remaining 6 hours of the test. 

 

About my test system, I'm using a freshly installed version of Windows 7 x64 Pro on a vanilla copy of FSX SP2. I do not have any addons other than PMDG 777. PMDG 777 version 1.10.7025. I am still using cycle 1601 which I will update here shortly and report back.

 

Hope this helps! I have opened a ticket and am really looking forward to a fix for this but the lack of a solution in the 7 months this thread has been active is concerning. 

[ivo09 23]

Now that people have identified what is causing the leak, shouldn't it be  simple to fix for pmdg?

[kamaz74 81]

I got a reply from a staff member on my ticket. They are working on it currently. 

[pcubine 143]

Now that people have identified what is causing the leak, shouldn't it be  simple to fix for pmdg?

 

See post #96 of this thread especially paragraph #1.

[Boeing or not going 578]

 

I understand your skepticism, but I can corroborate with the other posters in this thread that loading SIDS or approaches into the PMDG 777 which contain heading to radial intercept maneuvers cause an immediate 120MB/hr vas leak. At first, I thought FTX Global was causing my problems. I even reinstalled Windows 7 x64 Pro and installed vanilla FSX SP2 and the PMDG 777 (latest version)(No other addons were installed) and the issue is still there.

 

Not denying a vas leak at certain situations like you explained. I hope PMDG fixes but that error in the fmc has been known though in detail for months.- David Lee

$NGXCABINLIGHT

$GAUGEFIFTEEN

$FLAPSTCLLIGHT

$MLGHB

$WINGLIGHT

 

I learned files that have $ are for loading through memory. But having this tweak used at every airport I fly to and during a flight can help find textures fs is searching for. I found UK EGNT looks for a texture file named EGBB. Fixing these missing texture errors by placing a named dummy texture in the main fs texture folder has increased stability and performance on my computer. - David Lee

[pcubine 143]

 

 

I learned files that have $ are for loading through memory.

David

The five texture files that you refer to in your post and quote of the April 22, 2016, where did they come from?

[Boeing or not going 578]

David

The five texture files that you refer to in your post and quote of the April 22, 2016, where did they come from?

 

You add lines to the fsx.cfg file, http://www.fsdeveloper.com/wiki/index.php?title=Missing_textures and every time  a texture is being called for but not loaded, it puts up a message on your screen. There are a lot of airport sceneries that have missing textures. - David Lee

[B777ER 2,486]

Soooo... If I load up the FMC and input say the KSFO ILS 28R, all I need to do is delete the missed app legs via the FMC? Or do I have to manually edit that approach text file? If I have to manually do it I think I'll wait to the very end of the flight and then input the rwy.

[pcubine 143]

Soooo... If I load up the FMC and input say the KSFO ILS 28R, all I need to do is delete the missed app legs via the FMC? Or do I have to manually edit that approach text file? If I have to manually do it I think I'll wait to the very end of the flight and then input the rwy.

I have tried to delete the missed approach out of the FMC on two occasions. I don't remember exactly what went wrong but I didn't work. I believe I was left with the last fix in the missed approach and could not get rid of it. Plus it caused a couple of other problems with FMC. Both flights were 11 hous of just waiting for something to happen. So I just edit the Approach in the SIDSTAR folder/airport file to delete everything after the runway number.

 

This what I would delete out ILS 28R at KSFO - HDG 284 UNTIL 420 TRK 251 INTERCEPT RADIAL 281 TO FIX VIKYU FIX VIKYU AT OR ABOVE 3000 HOLD AT FIX VIKYU RIGHT TURN INBOUNDCOURSE 281 LEGTIME 1.