papagoose

The success of the entire flight simulator genre is based on the expandability of the simulator platforms. FSX, X-Plane, Prepar-3d...all would be a shadow of their current selves were extensive modding were not allowed. What if FSUIPC violated the Microsoft EULA? I'm told that EZ-Dok functions by modifying FSX memory addresses- I could certainly imagine that Microsoft wouldn't endorse this. The NGX itself is built on the ability to interface *meaningfully* with the simulation. It would be disappointing if PMDG was so against third parties expanding further upon their products. Nevertheless, whether this is actually a violation of the terms depends on the exact implementation of virtual CDU. It is clearly not allowed if they are redistributing a modified version of a PMDG file. However, if you write a program that modifies the original file *which is provided by the user*, or if you merely interface with the original program in an unintended way, then it is much more of a gray area. It is notable that 'jailbreaking' a phone IS explicitly legal, even though it involves third parties actively circumventing protections and modifying the original Apple system files. Nothing Apple can say in their conditions can remove this legal right, although they can (and do) make it technically challenging for those who try to jailbreak. I wonder if virtual CDU may fall into a similar category. At any rate, if no proprietary files are being illegally distributed, then I don't see anything wrong with the situation. Indeed, I would go so far as to say I don't think Virtual Avionics *needs* to contact PMDG at all. I do think they should have talked to PMDG, but I don't think they have to - if they are not distributing proprietary files then everything that they are selling is their own work. Of course, PMDG has the right to update the NGX to break virtual CDU, and that's OK too, even if it results in a lot of disappointed PMDG customers. By the way, I think this talk of PMDG getting a cut from virtual CDU is nonsense, unless they want to distribute PMDG intellectual property or collaborate on development. However, the program is already written without PMDG's assistance. PMDG is no more entitled to a cut of virtual CDU's sales than Microsoft is entitled to a cut if I write a program that runs on Windows. Apple and Google receive a cut of revenue on their app stores because they distribute the apps, so if PMDG were to sell virtual CDU on their website, this would be a scenario in which a commission would be appropriate. If virtual CDU wanted PMDG to make changes on their end, then this would be another profit-sharing scenario. But as it stands, I don't think it's appropriate for PMDG to get a cut from the existing work (again, providing no proprietary files are being distributed). It would also set a terrible precedent for the expansion/modding community.

You could fly YSSY-YMML (Sydney-Melbourne) in a 747, which is about 1.5 hours - United flies this. This flight is the domestic leg of journeys like YMML-YSSY-KLAX. Also, you can do YSSY-YSCB (Sydney-Canberra) in the NGX, this is about 55 minutes- Qantas does this with 737-800's.

While I'm reading this (very old) thread, may as well comment that I believe this kind of functionality is possible with Prepar3d. You can apparently run one copy with panel-only, and another copy that displays only visuals, using simconnect to communicate. The visuals-only computer just gets the position etc. from the panel-only one, so I wouldn't expect issues with using the NGX. At the moment, the monthly subscription gives 2 copies to useEDIT- Oh and the NGX seems fine in Prepar3d :) even runs smoother!

You know, it would be nice if PMDG offered free extended downloads (via support ticket) for say, the week or two after a service pack is released

I believe that is more or less exactly what Prepar3d is :) But after a point, it makes sense to rewrite the engine from the ground up to correct design errors that weren't originally apparent, or to take advantage of new technologies. This is why most companies periodically update their game engines and break backwards compatibility doing so. I find it telling that most versions of FS are fairly compatible with the previous version, this suggests that most of the updates to FS have been chipping at the edges so to speak (although I'm happy to be corrected on this!). As for FSX still having a few more years left on it, my (relatively uninformed) impression is that this is less about it being a good program, and more about market share and inertia- even if there are good alternatives now, it takes time to develop for them, and time for those products to penetrate the market to make addons financially viable. Leaving aside 64 bit, I think the biggest issue with FSX (that no amount of addons or tweaking can change) is that the engine under-utilizes graphics cards- there are some things that would only be addressed by redoing it from the ground up (like Microsoft Flight is, albeit with different design priorities)

AFAIK, the fact that it is a commercial license merely allows you to use it in that way. However, that doesn't mean that you *are* using it for commercial purposes. If a copy of P3D is installed and is being used for personal use, then use of the NGX with this installation would fall under "private entertainment purposes only" as that is indeed how you are using it- the particular license details of P3D are in this case irrelevant?Same deal with a 'developer license' - really, a consumer license would be like a subset of a developer license. In other cases, 'developer' access is more about getting access to development tools like SDKs, emulators, extra documentation etc. It doesn't compel you to actually produce something.Each to his or her own as far as payment schemes go- but I'm of the view that when our choices are otherwise FS9 or FSX (and X-Plane but no PMDG for that), having a new option is great, even if not everyone likes it

I believe LM has posted in other threads that this means you won't find it on shelves next to Call of Duty (or Microsoft Flight!) - they won't be marketing it for personal entertainment, but they don't seem to have any problem with people using it as such

I was almost completely flipped once in a regional jet when default ATC put me too close behind a larger jet on approach!

Both of these are excellent suggestions

Lol speedbird114, don't forget Flight Simulator is a *game* But for everyone having this problem- does normal FSX pushback not work? The NGX is supposed to work properly with default pushback (ctrl+P or something like that) so you don't have to use the pushback command in the CDU

So stability is kind of like- if you overclock the computer too much and push the chip too far beyond its specifications, it will literally make a mistake like 1+1=3. Depending on exactly where this mistake happens, it might do nothing, or it might cause your computer to reboot. These kinds of errors usually happen under high load, which is why programs like prime95 are used to stress the computer. Sometimes you will see an unstable system where prime95 runs, but computes an incorrect result (which causes the program to stop, and it tells you the calculation was wrong). The point of testing stability with a tool like prime95 or IBT is that if the system is stable under those programs which test maximum load, then you are guaranteed that no program will encounter problems due to overclocking. Real world usage is invariably not as intense. In fact, sometimes you will see people talking about computers being game-stable vs. prime95-stable. If your overclock is unstable, you won't be doing any damage to your hardware (assuming your temperatures and voltages are OK)- it will just be an inconvenience. I suppose you could get file system corruption if your computer keeps powering off necessitating chkdsk or reinstallation of Windows. If your computer doesn't crash and your programs seem to run OK, then you could say your overclock is fine for normal use. But unless you get it to behave normally when running prime95 etc. you might find one day that it crashes when you do something CPU intense like video encoding.

And for the record, now that I have my second GTX570, I can confirm that all of my previous problems have vanished. I also saw an increase in FPS of about 5 (from say 20 to 25)- there was also a significant increase in smoothness that improved the simulation beyond the increase in FPS

Also, it's important to see the difference between the 'fail passive' and 'fail operational' systems. 'fail passive' simply means that a single failure will require the pilot to take over the landing, whereas 'fail operational' means that autoland can continue with a single failure. In fact, a single failure in a 'fail operational' system causes it to revert to 'fail passive' (the next failure will require the pilot to take over). LAND 3 and 'rollout' is only displayed for fail operational autoland. On the PMDG NGX, I believe you have to set fail passive/fail operational independently of Honeywell/Collins MCP. That is, to get LAND 3 you need to have the Collins MCP and also have fail operational autoland enabled- which I think just happens when you have the electronic standby instrumentation. You can tell by looking at the lower DU selector (which has the ENG and SYS buttons). If there are only 2 buttons, you have fail passive. If there are 3, then it is fail operational.

From where I'm sitting, the ideal solution would be for PMDG to make an API/interface that allows CDU button presses to be sent to the simulator, and the current contents of the CDU display to be retrieved. Then any third party could write a simple interface that acts as a remote front end for the CDU, without having to touch the internals of the NGX and without messing with Windows display settings e.g. networked external monitor. But it would be great to hear from PMDG about this, I don't know how difficult it would be to implement

OK so making 100% sure, when you ran at stock clocks for a week you NEVER had any issues with FSX specifically? And you tested it thoroughly? Also, you are overclocking the correct way already. Don't touch the BCLK, that is terrible advice. Some users report permanently damaging their systems by doing this with sandy bridge. I agree with Greg about disabling XMP though.

So you should get it working at stock clocks BEFORE you start to mess with overclocking! :(

Ah my bad, I thought you said 15FPS minimum. My rig is similar to yours, I see about 20-30 everywhere. However, considering that FSX is the oldest game on my computer, and literally everything else runs at >60FPS minimum, I think that says something too!

If you ask any other gamer what they think of 15FPS, they will laugh because it is so ridiculously low. That a 5 year old game runs at 15FPS on a 4.8GHz 2600K just says that is is coded absolutely terribly compared to some of the other software that is available. It runs like a console port, only with better mouse support :(

Personally, I hope not. I suspect that there are plenty of ways that FSX could be improved from a code design point of view (to better take advantage of GPU features, for example) but these would all involve big changes to the architecture. So I hope that Microsoft Flight breaks every addon to date, that might be a sign that they have been working on the internals of the simulator! :( But this is all just conjecture

Frank, I'm not sure I understand what you mean relating LLC to 'permanent' overclocks or to the stability. As I understand it, what happens is this- Under load, the processor consumes additional power. Accompanying this is a drop in voltage that some refer to as Vdrop/Vdroop. If you enable LLC, all that happens is that the motherboard increases the voltage to compensate. Therefore, when loading the processor with LLC enabled, you see the same or greater voltage- With speedstep/related power features enabled, under load the CPU will increase its multiplier to the BIOS specified maximum, and adjust the voltage accordingly. When idle, the CPU will drop the multipler (to 16x for the 2500k) and the voltage as well. If these features are disabled, the CPU will sit at its maximum multipler permanently- is this what you mean by 'permanent' overclock? AFAIK, this is completely independent of LLC- There are two ways to adjust the voltage- the CPU offset (which adds or subtracts from the 'base' voltage), and the Additional Turbo Voltage (ATV) which only takes effect when Turbo Boost is active (i.e. 100% load with speedstep etc. enabled). The offset voltage in theory raises everything including idle voltage with speedstep on, while ATV only increases the voltage under load- I have found that with speedstep enabled and LLC disabled, using ATV it is possible to destabilise the CPU by very quickly changing the load. In this case, I think the actual processor load is changing faster than the motherboard can change the voltages *and* crashes occur when the current voltage is too low to handle the new load. The solution is to raise the voltage using the offset, not ATV. Or to enable LLC- Other sources e.g. http://www.overclockers.com/forums/showthread.php?t=682932 demonstrate that enabling LLC can increase the stability of an overclock, this was my impression as well. Because when you think about it, the only real effect of LLC is to increase the voltage to something higher than it would otherwise be without LLC- The effect of LLC is clearest if you do a 'permanent' overclock by disabling speedstep. Then, with LLC disabled, you will see an idle voltage about the same as set in the BIOS, and a load voltage that is a fair bit lower. With LLC enabled, the voltage should not change at all VirginAus737- You should definitely do an overnight run of Prime95 to (virtually) rule out your overclock as being the issue. However, the fact that you can so easily reproduce the error suggests that it is not your overclock that is the problem. I would revert it back to default clocks and see if the "right click on the desktop then right click again it causes the AppHangB1" persists. If so, then we know that more optimization of the OC is necessary. Otherwise, you can focus on the software side

I've noticed it too. I may be mistaken, but I think this is part of the design. On the ND, you can see that the diamond is not filled, it is just an outline to indicate that it is 'off scale'. As you intercept, it will first change to a filled diamond, and then it will start to move down. Offhand, I think when it becomes a filled diamond it appears on the PFD as well

It sounds normal, but how are you testing stability?

Nope. In fact, I did the lighting test to see if they would extinguish along with everything else, but they didn't

I started up the sim with the NGX short panel state, APU running etc. I left it and came back about 15-20 mins later, and all 12 of the master caution warning lights were on (6 on each side), and the master caution was illuminated. But pushing it didn't clear the warnings like it usually does (although it extinguished the main light). I tried starting the engines, cycling the battery/standby power, but I couldn't figure out how to clear them. Pretty sure that no failures were active either. Any ideas what can cause this?

One of the details I love is that when you turn the dome light on, the cockpit illumination from the outside changes!