[RECALLED] TextureFlow ORBXLIBS - Recalled due to crash bugs for some users.

[Iain Emms]

Yes lets not stop the timer on PNW, the Textureflow Libs will be fixed as soon as possible of that im sure and i know which one i want first of course its PNW with out a doubt and as we all know these hiccups happen in our flightsim world from time to time and in the end its not the end of the world And will be sorted out for us by the dedicated team we have here who keep raising the bar for us at every turn and Thanks to them we have what we have now and the like has never been seen before.

cheers

Iain

[bulldog07]

Your support for your releases are nothing short of spectacular, your ability to accept flaws in your work commendable, something quite a few programmers fail to have.

You are and continue to be, if not the best representation of our community.

Up front and honest makes us all patient and loyal customers.

Thankyou FTX team

[Blitzer]

John,

I am glad to hear that, although reversion did not work quite as well for me, as the GPS problem remains. Is it coincidental, perhaps, but unlikely. I am going to revert back to an earlier driver to see if that might help. If any more information come forward on this that might help me, I would appreciate the assistance. Thank you.

Stephen

I recently posted a problem with crashing when using the GPS - can you tell me how the crash happens for you please?

[Valcor]

Let's not look for ghosts here people

It's 100% safe to revert back to 091028, and I can assure you the libraries don't affect GPS functionality and other issues you may think it has caused. These libraries simply affect Orbx created models and objects used by our airports. We suspect that the crashes are caused by the crashbox for the 3D grass or perhaps the line markings, so it's a simple enough fix.  Of bigger concern to us is the introduction of stuttering, which definitely should not be happening.

As mentioned, we're sort of snowed under with the PNW release, but we will address this asap and indeed we have several of the team (Ian, Graham) working in the background in debugging this.

Apologies once again.

Just to add to that.  I should have mentioned this in my "crashbox" issue bug post.  Other than that problem, I was seeing very positive gains in performance.  So thanks for you continued dedication to improving already released products.

[Martin Henare]

Just a word on Crash detection, while it is a nice feature in theory, in practice its is pretty much impossible to control the size and shape of the "Crash Box" generated around a model.

Basically when a model is created the SDk tools automatically create the Crash Box, the developer cannot specifically do much to determine the actual size. Where it goes all wrong is for say an L shaped building compared to rectangle building. This is be cause the Crash Box shape can only be rectangular. So if you image the L shape with rectangular box generated to fit the whole L shape the space in the inside of the L shape will be inside of the Crash Box and thus if you have crash detection On, you will get a crash without your aircraft physically touching the building.

Now imagine how hard it is to predict for a much more complex shaped model, where a whole series of rectagular crash boxes are slapped all over the model to fully enclose it. And thus you get random crashes all over the place.

One of the obvuious weird ones is the crashes at altitude. Some times the crash box created is some random rediculous shape that extends 1000's of feet into the air. Again the developer has little control one why and how this is.

So what we do in general is turn crash boxes off on all the models. What this means is you can then fly through the models without getting a crash. Its the only solution a devloper can make. we can disable crashes for part of the model but its just too inconsistant.

Which is why we recommend crash detection be turned off in the Sim. It just doesnt work right, except i guess for aircraft handling errors. like hard landing, CFIT, and over stressing the airiframe.

I can reiterate what john has said about the GPS, there should be no reason why a library object would hurt the GPS. The GPS uses data from your Terrain Mesh and other Geographical data. it totally unrelated.

[John Venema]

Please folks, can we not have a tangent discussion about GPS related issues here? I am firmly convinced it has nothing to do with any library changes we have made for the YSTW public TextureFlow beta or the new ORBXLIBS.

The best way to resolve the GPS issue being related is to:

1. Uninstall YSTW and reinstall the RTM version (before the TextureFlow betas)

2. Then install the 091028 version of ORBXLIBS.

That will clearly indicate if any changes we've made in the last 6 weeks have had an effect.

We are doing the best we can, in the time we have to resolve the libraries issue, but in the meantime, please revert back to 091028 and give us some space to deal with it asap ok?

[Valcor]

I would LOVE to uncheck the crash detection within FSX.  I can't stand the way it's handled, modeled or the way it automatically reloads.  Plus, I think many aircraft respond more realistically to hard landings, etc, with it disabled.

However, wouldn't I be losing the damage modeling of high end payware aircraft packages since the check box say's "Detect crashes AND damage"?

[Patrick Scharnowski]

However, wouldn't I be losing the damage modeling of high end payware aircraft packages since the check box say's "Detect crashes AND damage"?

Nope, by my experience the "damages" part only refers to the "aircraft stress" and "collisions with other aircraft" settings. Payware aircraft with special damage modeling usually circumvent FSX' realism settings and have their own setup that you can use to switch damages on and off. I certainly can still fry the engines of the Catalina or break the propeller off the Cub when FSX is set to "Ignore crashes and damages". Only on purpose of course. I would never let that happen accidentally. 

[Valcor]

However, wouldn't I be losing the damage modeling of high end payware aircraft packages since the check box say's "Detect crashes AND damage"?

Nope, by my experience the "damages" part only refers to the "aircraft stress" and "collisions with other aircraft" settings. Payware aircraft with special damage modeling usually circumvent FSX' realism settings and have their own setup that you can use to switch damages on and off. I certainly can still fry the engines of the Catalina or break the propeller off the Cub when FSX is set to "Ignore crashes and damages". Only on purpose of course. I would never let that happen accidentally. 

Wow, that's great to know.  Thanks for the insight.  I am really looking forward to getting the A2A Piper to go with NA Blue and I wouldn't want to miss out on it's fickle nature by having that unchecked. 

[realatp]

Flickering of tire marks at YSTW John.

[Lupin500]

Thank you for listening our feedback and deciding to spend more time and resources to perfect the code.

Looking forward to flying soon over PNW!

UPDATE: JV here - I will recall this library and revert back to the 091028 version for now. Some users are experiencing crash bugs which is just not good enough to allow this code to be released. Right now we will focus on the PNW release and the two USA airports, and in the meantime the guys will test the new ORBXLIBS code a lot more to ensure it's good enough for a release. It's odd, since I don't experience any of the symptoms on my PC, but that's not a good enough yardstick.

Sorry for the inconvenience.

If you have installed the 100222 library, simply re-install the old 091028 library again to revert back to the previous build. If you did not save it locally on your HD, it's available here - http://fullterrain.com/support.html#orbxlibs

We appreciate all the bug reports and specific details, that will help us pinpoint the problem files.

Thanks for your patience, I know a lot of you wanted these new libs badly, but I'd prefer to do it right, rather than too quickly. Lesson learnt.

[BAT21]

I saw another problem after the installation of the new library:

using the A2A Cub the tag "Sim connected problem .... restart FSX" start to come out.

I have a "fresh & clean" re-installation of FSX + ORBX SP3 + OZx + A2A cub only, done few days ago.

I hope this report can help in the debug process.

cheers Marco

[ja-durant]

o.k. John .Done

jacques

[Peterok]

May not be related but after installing the now recalled library I had no orbx AI traffic at YMML. Reverted to the older library and also uninstalled and re-installed AI traffic but still no joy. I will pursue further in the next few days but I suspect the orbx AI traffic & AC are AWOL.  Default FSX aircraft can be made active. Also noticed some increase in stutter and the flickering on markings at Launceston.

[Nyxx]

Let's not look for ghosts here people

It's 100% safe to revert back to 091028, and I can assure you the libraries don't affect GPS functionality and other issues you may think it has caused. These libraries simply affect Orbx created models and objects used by our airports. We suspect that the crashes are caused by the crashbox for the 3D grass or perhaps the line markings, so it's a simple enough fix.  Of bigger concern to us is the introduction of stuttering, which definitely should not be happening.

As mentioned, we're sort of snowed under with the PNW release, but we will address this asap and indeed we have several of the team (Ian, Graham) working in the background in debugging this.

Apologies once again.

NP John, take care of PNW look forward to the fixed lib's

[Spirit Flyer_old]

Trying to revert now. I have serious problems with a crash, non functioning GPS and a litany of other things. I hope this gets me back to where I was.    :'(

Stephen

It was just a driver corruption issue. I reinstalled 185.50 and everything, speed, smoothness and GPS and instruments are all back and running sweetly.

I think the trouble may have started a few days ago when I installed what I thought was a couple of neat things from Nvidia to watch some sample videos that did not install correctly and would not work.

So it was not the texture flow creating the problem after I uninstalled it, but Nvidia viewing software. Thank you for your expressed concerns and the behind the scenes encouragement and help.

Now I am all ready and waiting for PNW again!

Stephen

PS: Thanks John

[justinm001]

Im still using the new libs thatwere recalled. Increased my performance and havnt seen to many issuess. have tested the flights you guys mention and dont get those issues.

She's real smooth now in all places. Not a stutter in sight.

I am however.. not sure about something.. is it possible to new libs could affect how the PAPI lights work ? If its a no I shall keep using it. Unless ORBX strongly recommends not to ?

If not does anyone have any suggestions I could try. I have SP3 and YMML (not installed) and FSX/SP2.  Pretty sure they used to work properly.. recently..

thanks heaps

[Ian Routley]

No the libraryies shouldn't affect PAPI lights.

These do suffer from "effects priority" inscenery dense areas and may not show ... that is (unfortunately) an FSX thing.

The performance with the LIBs is enticing ... and we hope to get a new version out (when possible) ... but after some further work to get them "right"

[justinm001]

yeah libraries seem to be pretty good.

Thanks

[justinm001]

found the problem with the PAPI lights. If your using FTX BLUE and the default FSX YMML airport the lights dont appear to function correctly.  I tested a number of times with FTX YMML on and off and it works perfectly with FTX YMML on everytime... Never works with the default YMML though.

The G/S still works though regardless which version of YMML is loaded in, just the lights that is the problem.

send me that secret YMML beta please, you know you want to !!