Christiano! You should defintly not stay in a 32-bit OS! If Vista is better than Win7...I don't know, never had Vistax64 but I hadbefore august this year XPx64 and that was a very, very reliable OS. Since I now bought a new Rig (see specs) I now have Win7x64 as many other.Haven´t got the NGX yet, and installed my new FSX Gold edition either yet, but I will soon so I can´t judge. BTW about 32-bit VS 64-bit, theres a lot off post about it, but read post #81 from Ryan in this Thread. / Leffe

Ok. Tonight I will uninstall/re-install the NGX to see if it solves anything.Now that I know it's only 1% of customers having this issue i'm gonna try this and post the result tomorow.

Thank you, Leif. I will do something about this. Thank you very much.

With respect to virtual address space, I remain to be convinced that this is a graphics card issue. Microsoft did use to take up some of the 4Gb address space by copying the graphics card address, however this was eliminated in Windows Vista SP1 and later as far as I can figure out: http://support.microsoft.com/kb/940105 quote: If an application creates its own in-memory copy of its video resources, or the application uses DirectX 9 or an earlier version, the virtual address space contains the WDDM video memory manager's virtualized range and the application's copy. Applications that use graphics APIs that are earlier than DirectX 10 and that target GPUs that have large amounts of video memory can easily exhaust their virtual address space. To address this problem, Microsoft is changing the way that the video memory manager maintains the content of video memory resources. This change is being made so that a permanent virtual address range does not have to be used for each virtualized allocation. With the new approach, only allocations that are created as "lockable" consume space in the virtual address space of the application. Allocations that are not created as "lockable" do not consume space. This approach significantly reduces the virtual address space that is used. Therefore, the application can run on large video memory configurations without reaching the limits.

Even on Vista / Win 7, when running DX9 apps, there's supposed to be a shadow copy of VRAM in the process VAS. (as reads your quote). Truth is I can't see this in practice though

If I had 32-bit Vista and a GTX 580 with 1.5 gig of Video Ram on board this would be a nightmare situation cause the OS would have to map the 1.5 gig leaving me with 2.5 gig left over

I'm pretty it doesn't work like that. First off, it's virtual memory, not physical memory where VRAM is mapped (ok, an increase in VAS will most likely lead to an increased RAM usage)But most importantly, you don't get the entire 1.5GB of VRAM mapped to virtual memory, just the VRAM in use for all I know. I'm going to try with SLI again now and see if there's an increase in memory demand with two cards

Hello, I `am another user with panel-freezes- and out of memory errors. My old Laptop has 2 GB Ram an an ATI videocard with 256 MB. Yes - true. All is working fine in my FSX, ok the frames would be better. But in an approach (at downwind) or after touchdown i often have a panel freeze. OR: When I shut down the FSX after a good flight with the NGX then the FSX.exe show very often a errormassage (unable written at adress xyz..). The NGX advise the FSX an memory - action, which can the FSX not handle? I think so. OR: When I fly an approach after a 2 hours-leg, then the frames going up to 0,1 - 1, i can fly but its very difficult. What does the NGX do in this time? Calculate anything as a circle, talk with the FSX about the windows memory management? I`dont know it. The Pmdg-team know it? Please look on this facts. Another fact: When I go at the beginning of a flight in the IDENT-Page in the FMC- there is often a "?" below the ENG. Whats this? A bug? Sorry for my bad english please - but I also wait for the SP1 since many days. Hope it helps... RegardsHans-Peter Burwig

And one more. Exactly like all the freezes I read about, followed the tutorial to the letter, first time it froze on final about 100 ft above the ground, second time after I turned of the runway. My rig is exactly one year old and built specifically for FSX. I had no other add-ons running at the time of those flights (most of them run on a different pc anyway)Love the airplane but am getting very tired having to plough through all the forum posts to see if there's a suggestion that works.By the way, the installation didn't go as planned, I had to install the Directx runtimes by hand. regards Martin Bunjes

Another fact: When I go at the beginning of a flight in the IDENT-Page in the FMC- there is often a "?" below the ENG. Whats this? A bug?

@Hans_Peter_Burwig Look at this thread, you are not alone: http://forum.avsim.n...__fromsearch__1 Bjoern

Ryan, I don't know how many of us are experiencing the freeze thing is it 1% or 5%. For what I can say there are more reports coming in. While some users have a severe case of freezing (eg every flight) others have it occasionally including me. I remember sometime ago with my old config, I had the notorious "black screen” error with the MD-11. A major game stopper. Couple years down the line the problem is properlu addressed and the people are no longer suffering from it. Now we have "The freeze", is a new problem and it appears that it has to be tackled by everyones input. Once again...

Even on Vista / Win 7, when running DX9 apps, there's supposed to be a shadow copy of VRAM in the process VAS. (as reads your quote). Truth is I can't see this in practice though

Hi Dario, The way I read the following paragraph was that the issue was fixed in Vista SP1 even for DX9 applications. Maybe I am not reading it correctly: quote: If an application creates its own in-memory copy of its video resources, or the application uses DirectX 9 or an earlier version, the virtual address space contains the WDDM video memory manager's virtualized range and the application's copy. Applications that use graphics APIs that are earlier than DirectX 10 and that target GPUs that have large amounts of video memory can easily exhaust their virtual address space. To address this problem, Microsoft is changing the way that the video memory manager maintains the content of video memory resources. This change is being made so that a permanent virtual address range does not have to be used for each virtualized allocation.

Hi Dario, The way I read the following paragraph was that the issue was fixed in Vista SP1 even for DX9 applications. Maybe I am not reading it correctly: quote: If an application creates its own in-memory copy of its video resources, or the application uses DirectX 9 or an earlier version, the virtual address space contains the WDDM video memory manager's virtualized range and the application's copy. Applications that use graphics APIs that are earlier than DirectX 10 and that target GPUs that have large amounts of video memory can easily exhaust their virtual address space. To address this problem, Microsoft is changing the way that the video memory manager maintains the content of video memory resources. This change is being made so that a permanent virtual address range does not have to be used for each virtualized allocation.

I don't know Howard. Maybe it was "fixed" after Vista's SP1 yeah. I certainly don't see here a significant increase in VAS when I crank up AA to increase VRAM usage. I hope I can try a 3 screen resolution soon to do more testsSLI doesn't change a thing either. I have the virtualy the same VAS usage with a single and a dual GPU setup

Actually Eric, the 2GB of video memory may be part of it. Please answer my PM so we can try something here... 2GB of video memory means that at max only 2GB of virtual address space is left for the sim to run in. I know it doesn't seem logical that having a better video card could hurt here, but in this case it does. Old 32-bit DX9 apps like FSX have a hard 4GB limit on VAS even on 64-bit Windows and the mapped video memory counts against the total. I had one guy at support getting freezes with two 1GB cards in Crossfire - he removed one and the freezes stopped. This cannot be a code flaw in the NGX if I can go fly the exact same flights and not see the freeze, that isn't how programming works. If it's a logic problem or error in the code it should affect anyone who takes the same steps as the person getting the freeze. That's not what we are seeing here - I have been completely unable to reproduce any freeze that's been reported to us since release. If you don't believe me you are welcome to come fly the airplane on my system remotely through TeamViewer to see for yourself. I have absolutely no reason to lie about this - do you guys really think we'd have released knowing about this issue? Ask any beta tester - none of us or the test team saw this. My comment to that other guy by the way was in reference to the way he never even asked for help but instead came in here yelling at us instead of trying to constructively help. Please don't selectively leave out the other half of my comment if you're going to quote me.

I just want to say this as I think it raises a nice point.The team and the beta testers have all had the same installer, installing the same code on different machines and everything worked.Now, after the release, the customers are having problems with crashes, freezes, lock ups etc. Maybe something has happened to the installer/the way the coding works between testing and the release. Whether that is the Activation process that got added, the debug code the beta testers had... I don't know. What if someone had exactly the same copy that all the beta testers had when there was not a single crash, freeze or lock up. I am happy to test this, just PM me if you wish. Thanks,

+1. Have a suspicion this has to do with system power consumption driven by graphics demand with this AC. I have several powered USB attached racks and notice that the USBs are dropping in and out consistent with the 'freeze'. I am going to install a higher watt power supply and see if that helps.

Installed a 'Wicked' 850 today, going to see if that gives ample power to my GTX560 and about 15 power USB hubs . . . LOL Joe

In regards to computer specs per PMDG quotes. You look at the surveys being done here. The freezes are occuring with every brand, combinations, cpu's memory etc. Im sticking with there is no certain hardware configuration causing this.

I tend to agree with you, however if i may be obtuse, there HAS to be some commonality between all of us experiencing the freezes. if there werent, we wouldnt have them. its just a matter of ferreting out what that commonality is.. However, i dont believe its hardware related.because i'm a developer myself, i've got the SDK and all its DLLs installed, I also have other addons such as LeonardoSH's Mad Dog and LVLD's 767 installed. i remove LeonardoSH's dll and the PMDG flies fine, the LVLD doesnt seem to effect anything.. Still i know others dont have the mad dog installed. But there has to be a commonality there somewhere..

Have been suggesting the software common denominator (as i call it) theory for a while now but i'm not sure PMDG is of that opinion.

I have tried to scan these forums every few nights since downloading the 737NGX the second day it was out. On about the second flight, I had a warning box telling me I was out of memory. I have a two year old computer built by Jetline Systems. These are computers built to run flight sim programs. It runs Windows 7 64 bit and I have 6mb of ram, and I had never seen this box before or ever had a problem. Figuring I needed more RAM, I sent Greg an email asking him to send me some more. He wrote back sending me a link for http://www.ntcore/4gb_patch.php. This is a small program that asks you for the fsx.exe location and then commands the program to use a full 4gb of ram. Since I have installed this patch, I have continued to fly with my two WS monitors, ASE/ FEX & GEX and all my slides to the right without a stutter. This 737 hand flies as smooth as silk. I have never checked how many frames I am getting. I have them set at 40 and never have had a blip or a stutter since installing the patch. By-the-way, it's a free download. Mike Nicholson