>>I'm taking flying lessons now, and when I'm flying the the>trusty ole 172 out of the valley, over the mountains to the>training area, it is not the smoothest ride on even clear fall>days. Air is generally not stable even on stable air days. >Mountains have up and down drafts, there are drafts associated>with cloud formation. Fly over a mountain in the great SF.260>and you will "feel" nothing disturbing the plane...no matter>how good it's FDE is...it's on rails in FS.>Of course all "real" flights are not smooth....That's what the turbulence slider in FS2004 is for! :)But then, I don't like simulated turbulence, so I usually leave it off. Even active camera is annoying if head latency is set too high. I use to do (real plane)aerobatics, and I could usually get my eyes pointed in the right direction.......... no matter how rough the ride got. "Rails" to me------------ means going to a movie & having the airplane still on course and altitude when you get home. :)L.Adamson

I sorta feel the same way. I leave it at normal and have no problems. Frankly, if the issue only occurs at the "extremely dense" autogen setting the best "fix" imho is to turn the autogen down a notch or two. Sure you might argue that if the option is there you should be able to use it but then is it a "bug" of just an unrealistic expectation of performance? I mean how is deleting a whole category of autogen a better option than simply running a lower autogen setting? The suggestion that you can't enjoy GA flying at lower autogen settings seems a bit odd considering that even at the normal setting theres more autogen buildings than there ever was in 2002...

Will,I guess your are basically right in what you are writing. On the other hand MS has stated many times, that they are closely watching FS related forums like AVSIM to get known of problems, ideas etc.And I think there were 2 issues ddiscussed in theese forums in deep. One is the CDT thing the other the autogen bug. Whereas the seasonal CTD was discovered just recently, too short a time ago to enable MS to act, the autogen bug is known now for approx half a year and was exactly described by many users. Also the reasons for the performance loss (not releasing autogen objects) were discussed over and over again. So MS should have a clear picture of what is going on. And, again, for me as a computer layman, it does not seem be too hard for them to fix this issue. Wolfgang

Exactly It shouldn't be hard to fix these two particular bugs. Many have investigated exactly what causes the bugs and done much of the work already.Virtually every game released these days gets at least one patch, even if just to fix some minor gameplay issues. FS suffers from crashes and low framerate, and still doesn't get patched. Why? Because it will still sell, even with a few bugs and glitches because it's the only civilian flightsim available basically.

First of all, the CTD that is seasonal is not a "simple" bug. Countless man hours have been devoted to it's discovery and cause.Much of the initial reports were found by folks who used mesh, so it was ititially assumed that the errors were related to the add-ons.Even now, there is a "work around" for it, but not a true fix.The "autogen" bug has also been heavily discussed and heavily debated.Not until recently has there been enough data for anyone to even accurately "reproduce" this bug, and even so, some folks who actually fly a "similar" route, (not the pre-recorded one) find very different results. Microsoft has a very limited staff. The resources of which are dedicated to working on Flight Sim 10. It has been "expected" by the sim community that MS release a new version of FS every two years. And with each "two year" release, the FS community expects more and more done.I suspect that Microsoft would probably patch FS more, if it knew that the program would have longer than a 24 month shelf time. For me, the much more pressing matter is scenery errors. Such as missing bridges and mesh anomolies.The fact is that even though the "discussions" have been in the forum, there have been startlingly few posts that explain exactly how to reproduce it.A bug report needs to be Reproducible, provide a severity rating, and provide a detailed description of what is occurring. And for "patching" purposes needs to be reproducable across the board.When someone mentions a bug, and then there is a 50 post thread with over 1/2 saying that they don't "see it", then it's kind of hard for Microsoft to take it seriously.

> That's what the turbulence slider in FS2004 is for! :)Hmmm...turbulence slider...must be next to the up/down draft slider somewhere...No matter, there is a reason why I still have Flight Unlimited 3 on my 'puter... :)'nuff said on this matter

". Frankly, if the issue only occurs at the "extremely dense" autogen setting the best "fix" imho is to turn the autogen down a notch or two. "The issue occours at all autogen settings, but it's obviously worst at Extremely Dense. It has very little to do with the fact that there are more buildings in FS2004. The problem, like I and others have stated countless times, is the distance at which the new library of autogen objects stay active. After you have flown past an object, it's active for up to 27 miles, while the generic library of FS2002-style buildings have a distance of about 3 miles or so. This is what makes it a "bug". If autogen caused this performance drop and there was nothing obviously and inherently WRONG with the Autogen, then it would only be unexpected expectations for performance. A high-end system SHOULD be able to use Extremely Dense+default.xml, but can't always do that because of this problem."I mean how is deleting a whole category of autogen a better option than simply running a lower autogen setting? "Because when you delete the faulty category of autogen, you completely eliminate the objects which have too high visibility distances. This allows you to use Extremely Dense with very good performance, just like in FS2002.I have compared the settings, and Autogen to Normal and all autogen types enabled gives slightly LOWER, but generally comparable performance to the Extremely Dense setting without default.xml and vector objects. The autogen scenery is not nearly as dense as even FS2002 at this setting however. Any higher setting than Normal can cause framerate problems.Neither is a satisfactory solution - you have to choose between displaying less autogen WITH enhanced 3d buildings, or use Extremely Dense but only with generic FS2002-style buildings. Not because our computers aren't powerful enough, but because of an error in the autogen system.I hate the look of flat textures combined with a small number of 3d buildings. I want either full 3D scenery or no 3d scenery at all (epecially in the case of photo-sceneries). How realistic is it to have a huge flat area with one single building sticking up from it like a sour thumb while all the rest of the buildings are 2d textures?"The "autogen" bug has also been heavily discussed and heavily debated."Most of these "debates" have been by users who are very bothered by this problem trying to prove to those who don't care about/notice the bug that it's real :)"Not until recently has there been enough data for anyone to even accurately "reproduce" this bug"I don't remember exactly when the first threads started appearing, but it was a LONG time ago and it didn't take long after that until the scenery guys had figured out what was going on. I have known what triggers the problem since late September or so. I could have done this recorded flight 3 months ago but I didn't think of it until now and even without a recorded flight, it's easy to reproduce."and even so, some folks who actually fly a "similar" route, (not the pre-recorded one) find very different results. "This is not so surprising. The only autogen that gets "stuck on" is autogen you've actually flown near enough for it to become active in the first place. Then, you need to make a 180 degree turn or pan the view for the problem to show itself at all. The easiest and quickest way to trigger the bug in its most severe form is to simply slew backwards. Framerate will gradually be reduced as more and more objects come into view, get "stuck on" and then slowly move away into the distance, while new autogen objects constantly come into view. This becomes even more interesting if you slew out over the sea, still backwards, because about 20-30 miles out, *poof* framerate is restored.Slightly different routes, different zoom settings and different aircraft can have a big impact on this problem. Still, the effect is almost always seen. When flying a similar route, it might drop from 30 FPS to 18 on one flight, from 30 to 12 on another and from 30 to 6 on yet another but the tendency is still there. There should be no drop at all.

I am not doubting your findings. Your data is some of the most complete on this subject I have seen. And you are to be commended on your work, even so far as finding out how long it takes (27 mi) to turn off the autogen.I have no doubt that this is a bug. If you provided this information to Microsoft, I am sure that it would be helpful.However, Microsoft, like any company will weigh the "costs" of fixing the bug with the benefits of the bug fix.Again, with the aggressive timeline set forth by the industry to have a new version of Flight Sim every 24 months. It is most likely that such feedback will be rolled into the next version vice being issued in a patch.How many "bugs" mentioned by users of FS2002 were later incorporated in FS2004?Only the most critical bugs (such as the CFS 3 performance issues) usually are serious enough to warrant bug fixes.The rest usually gets incorporated in the next version.

"And you are to be commended on your work, even so far as finding out how long it takes (27 mi) to turn off the autogen."I have to say, the actual distances were not discovered by me, but rather by scenery designers. They placed out default.xml objects in custom sceneries to determine the exact distance at which they appeared and dissapeared. I had earlier estimated the distance to "about 20-30 miles" by starting from a coastal city and slewing backwards out over the ocean, but I did not find the exact distances that way."If you provided this information to Microsoft, I am sure that it would be helpful."I have told them on three occasions about this problem. Once to the actual Scenery Test Lead, and twice to the "tell_fs" email alias (one time several months ago and again more recently). I provided detailed information and even links to the relevant threads in the scenery design forum and elsewhere, and offered to help them if they needed further information or tests. I haven't heard anything from them. A "Sorry but we don't consider this issue severe enough to warrant a patch" would also have been accepted, but the truth is we've heard nothing at all from Microsoft. There's no way of knowing if they are aware of the problems at all (though they probably are by now)."How many "bugs" mentioned by users of FS2002 were later incorporated in FS2004?"A LOT of the problems people had with FS2002 were fixed or at least reduced in FS2004, which is great. I never really had any serious problems with FS2002 though. Scenery and aircraft worked perfectly, framerate was excellent, no lockups. ATC had a few glitches but was still better than previous attempts by Fly and Flight Unlimited. So, I'd say FS2002 didn't need a patch.Had there been more serious issues with FS2002 however, I would not have accepted to sit and wait for two years for a new version and pay another $70 to get the issues fixed, I would have demanded a patch for fS2002."Only the most critical bugs (such as the CFS 3 performance issues) usually are serious enough to warrant bug fixes."Why was the CFS3 performance issue serious enough to warrant a patch, but apparently the FS2004 performance issue is not? My guess is, because there's more competition in the combat flightsim market than among civilian flightsims, and also because framerate is more important in a combat game.I eventually managed to tweak CFS3 to acceptable performance - no patch was really needed but still they issued one, probably so they wouldn't loose more potential costumers to IL-2 and other sims. I still ended up returning the game within the 30 days allowed. The low performance sure was a factor in that decision, but other issues, and the fact that there were many better and more realistic combat sims were much bigger reasons for my returning the game. I probably would have returned it even if framerate was silky-smooth. With civilian flightsims, it's different. Sure, there's X-Plane, and some addon developers still work on projects for older sims like Flight Unlimited III and Fly2, but the VAST majority of addon scenery and planes are for FS.

Actually, for most setups the 1.0X zoom setting is close to correct perspective if you're treating your monitor as a window. Wide settings are akin to flying while looking through binoculars.....backwards!TonyDigital-Flight

Hey all,Excuse the denseness, but what exactly do you lose (scenery wise) by disabling the Autogen|default.xml file? Is it just those daft giant chickens sign things and others like it, or do you lose any of the autogen buildings aswell?CheersDavidP.s. Nice work JimmiG :)

You lose the new, interesting buildings like silos, churches, etc. The good ol' rectangular buildings from FS2002 are unaffected.

Thanks tdragger,Lucky there aren't many silos in Glasgow :)CheersDavidP.S. If this current method of displaying the fancy new buildings is causing so many probs, you've got to ask why weren't they simply made to work in the same way that the older buildings are displayed?

<My guess would be that the old buildings seems to be rendered using simple geometry (little detail) while the new buildings are more like custom buildings. You can ask in the scenery design forum but I believe these are separate codes that take different paths in the sim.

Thanks,I don't really have any clue about scenery design or how it works, nor am I really interested in how it all works. The question was just the thoughts of an idle mind :) Thanks anyway.CheersDavid