"You're having to repeat yourself because you refuse to see any other perspective on this issue."I had to repeat because somehow you missed the fact that I actually did test normal flight and slew. I didn't encounter this problem when using slew, I first encountered it over the Portland Or. area after flying south, then flying back north again over the accumulated autogen."What I SEE is perfectly normal and fine for me. I don't SEE the problem you are concerned with."That's what I saw too, at first. I actually said that I got better performance in FS2004 than FS2002, and I really did. Until I noticed the problem, which only happens at certain scenery locations with certain scenery settings (which happen to be the scenery areas I fly in mostly, with the settings I use).Before I found a workaround, people said it's normal behavior and there was no workaround. Then I posted the workaround here and on Flightsim.com and got 60+ replies. Proof that there IS a bug, and deleting the default.xml file is a workaround."flying (or slewing) at lower zoom settings can tax the system (based upon system and sim configs)."Yes, but the difference normally isn't very big. Certainly not enough to cut the framreate to 1/3. The reason it works better is because less autogen is displayed, meaning less autogen accumulates.It's possible that with your zoom setting, the framerate drop is not noticable unless you fly with a fremrate counter. At 0.5-0.6x, we're dealing with a drop from 32 FPS to 12 as autogen accumulates."Because I don't fly that way, and realized early on that we have our systems set-up differently."Because I used Slew, or because of the zoom setting?If the former, I'll make another test flight, with a flightplan to fly (via AP coupled to GPS) with specified points to report the framerate at. It will take about an hour to test this way, but that seems to be the only way.If you mean the zoom setting, then that's not good enough. I've used 0.6x since FS98 and i'm not going to change that because of some bug that causes displayed autogen to accumulate."I don't SEE it."If I steal money blindfolded, I didn't see it. So, did it really happen? Just because you're not bothered by the bug doesn't mean it doesn't exist and shouldn't be fixed by Microsoft.There are MANY bugs in games that most people would never notice, but they still fix them through patches.

Tom, not sure if there is an upgrade path from ME to XP, but even if there is I would advise against it. XP has been know on upgrades to make the HD cluster size too small, which has a very big performance hit. Also, it's really a good thing to install an new OS on a fresh clean formatted HD. Even if your keeping an old OS, it's good to re-install every 6-12 months, they get bogged down with so much crap.As to your programs, depends. I partition my HD's. I run all games on "G". When I built this PC (an upgrade, just new CPU/MB/Memory) 3 weeks ago I just formatted "C". I did copy my Program's and Personal Data Tree's first to another Partition. (I have 2 hd's, 8 total partitions) Sometimes after reloading or loading a new OS, you can re-install the Program in question (to establish registry info) and then drag from the other partition your old program folder, and it often works. I read someone way at the top of the thread, has a GF2-MX and running AA! LOL! You complain and your running a 4 year old business video card! I don't recommend AA at all unless your running a FX5600+ or ATI-9500+....these cards were designed to run AA with a small performance hit. Anything before takes a big dive with AA on.I just loaded FS9 couple nights ago...did a VERY brief flight last night. Mostly medium settings though I pushed a few up high, locked on 20fps. C-185 at KSNA...loaded Internet Weather (clear-no clouds really) and flew. It was unremarkable actually. Short 20 min flight. Missed all the projectAI traffic I have in FS8, and the add-on KSNA..MS didn't even try there. (can I add a nice FS2002 scenenry to FS2004?) I got 20fps the whole time. No problems on landing or anywhere, though I don't think I pushed it very hard. I do believe in overclocking and have done it for years. Don't even get me started on that. He He....Someday I'll upgrade my Video Card, but everytime I look the new ones have issues. The top NVidia's look like crap and have problems in the DX9 stuff it appears, and the ATI stuff looks great, but so many FS users have problems with them! My OC'd GF2 keeps plugging away till I find a good replacement...

After testing this on both my 700Mhz machine and my 3.2Ghz machine, I will concede that there IS a problem here somewhere, although the conditions under which it happens isn't likely to be encountered during normal flying. I spent over 1-1/2 hours flying last night and then typed in a long reply based on the information I had collected, but my DSL connection died momentarily and I lost the whole @#$! message. And there are more settings that I would like to test on both boxes.So, bear with me as I change some settings and see if I can get a better handle on what is happening. I will say, quickly, that I was able to track the accumulation of (apparently) autogen scenery buildup, but it reached a stable level and did not affect my display quality (on the 3.2Ghz box). There were some stutters that I haven't previously seen on the 700Mhz with a geForce 3 card.Please give me some time to play with this a bit more to see if we can figure out what is happening. I'll explain the comment about RAM with some figures when I finish.MDavis

What the hell is wrong with people here? To worried about framerates. I am into Xbox and console gaming in a big way, and none of us are worried about framerates. We just get in and play the games.I have FS2004 running on only an AMD 1.2 Ghz Duron, 512 megs DDR Geforce 4 Ti4200, and have no complaints about this sim at all. Have all the sliders well up and find FS2004 runs well. I find it very playable and smooth.I am not worried about framerates, and just get on and enjoy it.Seems to many here are to worried about framerates to accually just sitting down and enjoying this sim.To many moan and groan, forget the framerates, or wanting the best PC. As long as it goes, and is smooth enough for your needs, who the hell needs the fastest processor, or best specs to enjoy any flight simulator.Mine is low end and enjoy it, and what the hell is all the bugs that's been mentioned here. I don't see them?There are two many here worried their PC is not up to spec, and want the latest gear all the time, so worried about framerates, and wanting to upgrade all the time they are missing out on all the fun and enjoyment, and just moaning about their computer specs and framerate issues.Take a leaf from a console player, don't worry about specs and framerates, just sit down and play the thing and enjoy it.I have even had, at times this sim dropping to 10 fps at times and it's still smooth, so what the hell are all you going on about?I find FS2004 better than FS2002 in framerates anyway, with far more adjustable sliders, so can twink your sim more. Trouble is two many PC owners crank everything to the max and moan why it's not going well when that's not always the best idea, or works well with every computerLook at the instructions and it even says to adjust the sliders for your PC for performance etc, but how many PC owners do this? Either you increase detail and lose framerates, or decrease detail and get better framerates. It's that simple, and need to compomise.Seems sum PC owners lack a bit of intelligence I think. When it comes to framerate issues and preformance problems and maybe don't know what the sliders do in these flight sims, and expect everything on full gives great preformance.If your having problems then it's easy drop your slider settings.

You missed the entire point here. The initial complaint was stuttering, not framerates. I totally agree with the framerate opinion! Low framerates are acceptable if you can control the aircraft on approach and if they are smooth, and high framerate settings that are not reasonably attained are worse than useless. We're fighting stuttering at normal or slightly below normal framerates. Read on.MDavis

Here's what I found after running some tests similar to Jimi's. I'm not trying to prove anything here, just reporting the facts from some rather simple observations, and then making some inferences from the data as I understand it, which may be in error. I have no problem admitting I'm off base if someone can show me why.First, I ran FS9 with the following settings. I think Jimi's settings are a bit out of line with the reality of my everyday flying so I left settings like I fly them.Settings - Display - Scenery: Terrain mesh complexity 60 Terrain texture size "high" Terrain detail "land & water" Water effects "none" (I don't like the effect) Dawn/dusk texture blending ON Extended terrain textures ON Special effects detail "high" Scenery complexity "extremely dense" Autogen density "extremely dense" Ground scenery casts shadows ON Sun glare ON Lens flare ONSettings - Aircraft Virtual cockpit gauge quality "high" All ONSettings - Weather Sight distance 100 mi Cloud draw distance 60 mi 3D cloud percentage 50% Cloud coverage density "medium" (these are settings I like better than "maxed out") Detailed cloudsSettings - Hardware Target Framerate 24 1600x1200x32 Render to texture ON T&L ON AA off (see card settings) Filtering "trilinear" Mipmap 4 (maxing this out gives a rotten display with shimmering) Hardware rendered lights 8 Global max texture size "massive"Windowed mode in FS9Graphics card settings: AA4X, AF8X, all other settings at best qualityHardware listed below.I began at KSEA under rain clouds, thunder, lightning, but 10 mi visibility. I turned on the Task Manager to monitor page file usage, commit peak and the dynamic system memory cache values. I am running a 3.2Ghz hyperthreaded CPU which maxes out the CPU0 at 50%, and flickers CPU1 occasionally at 1-4% when needed while FS9 runs. Take off the the south in the default Cub using 2D panel, zoom at .75% (Jimi's setting), climb to 1,500 ft to make sure I didn't minimize the autogen display. Flew south to the 4 fuel tanks, turned north, flew to the Seattle skyline, back south over the runway to the 4 tanks and repeated 5 more times. Changed views to include interior scan, exterior spot plane, tower, overhead during flight at various times as would be normal. Beginning values were prior to view changes.Beginning page file RAM used was 483 which worked up over 1-1/2 hours to 502 (increase of 4%)Beginning Commit Charge Peak was 385, ending 522 megs (increase 36%)Beginning System Cache 580, ending 618 megs (increase 6.5%)CPU usage stayed around 50% and never exceeded 55% at any time. It was apparently completely independent of RAM demands. Framerates stayed at lock rate with the usual momentary dips into the low 20's when turning into high density scenery. No display anomalies were noted during the flight.Some interesting things did occur, however. Each circuit did increase the dynamic system cache setting, and the commit charge increased during the first 4 laps. Then they both stabilized and did not increase thereafter. This is on an essentially "unlimited" system with 2 gigs (2,096,620K) of RAM and a Commit Charge Limit of 4,039,084K of RAM. In other words, there was all the room FS9 wanted to grab more as needed. Also note that the CPU usage demand did NOT change during the flight. Framerates did NOT change during flight on this test.I re-ran the same test at 1024x768x32 which is more normal for most users and obtained similar numbers.Page File 388 to 416, then flattenedCommit Charge Peak 410 to 478System cache 470 to 491Essentially identical CPU usage numbers 50-55%, same framerates.However, on the 700Mhz box, things were a bit different (see specs below). Running at 1024x768x32 the gradual accumulation of (whatever) did begin to lower the average framerates by about 30% from the 18 fps lock. The same increase in system assets was observed with the same leveling off after 3 laps or so. Actually, there was a greater framerate decrease when turning north toward the Seattle skyline than the total accumulated average framerate loss for each circuit. I won't bore you with the settings on the 700 except to say that the sliders are pulled back on clouds, AI traffic and distance settings (not autogen), and on the graphics card settings to accomodate the slower CPU.So what's happening? I think Jimi is right in saying that autogen accumulates on successive passes over the same territory. I'm not sure why, but it would seem reasonable to assume that the computer is not filling all the available "holes" on one pass, and continues to seek places to add autogen structures on additional flybys. Perhaps this "filling the holes" is enough of a load that it causes framerate stutters? CPU demand doesn't seem to change, so that's not a likely answer. RAM usage demand does increase slightly each pass up to a level of somekind (all holes filled?). It might be instructive to take an initial screenshot on a given area, then overfly it, say, 5 times, then take another screenshot at the same location and count autogen structures to see if additional buildings were added. But this still doesn't answer the question of display stutter which I did observe on the 700 after 3 laps. I did not have access to graphics card usage on either box which could be a factor.Another thing that is apparent here is that, if you run settings "maxed out" with less than 1 gig of RAM, you are not getting "maxed out" performance. Note the increase in the system cache demands above, even in 1024x768x32 resolution, with less than "maxed" settings. That's why I recommended adding RAM if you have 512 megs. Someone will have to convince me that this makes no difference after seeing those commit peak and cache settings. If the system cache had been sufficient to begin with, XP would not have found reason to increase the cache size. Most other programs (check it for yourself) stay at a constant cache setting.No answers here, just questions. I can see where Jimi got his figures, but it doesn't seem to affect high end systems with adequate RAM, unless you raise the framerate lock much higher than needed for smooth performance? I'll try that next if I find time, but, unless I can increase CPU load, I don't expect to see any changes.3.2Ghz P4 (800Mhz FSB), MSI 865PE Neo2, 2 gigs Kingston PC3500 RAM, nVidia FX5900 Ultra (256 megs RAM), Audigy2, (2) 120 gig 7200RPM 8 meg caches and700Mhz P3, Asus CUV4X, 786 megs RAM, geForce3 (64 megs RAM), AC97 onboard sound, 120 gig 7200RPM HD.Mike DavisMDavis

Wow, what an interesting thread. Just my two cents, and to respond to the original author's post: for the first time ever, I've been experiencing problems with FS Meteo, running FS9 OR FS2002. It was just after the last product update. I'm hoping Marc Philbert is reading these forums as I no longer have an accurate E-mail address for him. If anyone knows how I can get in touch with him, I'd appreciate it. At any rate, quitting FS Meteo removes the slide-show like stutters I see while it's running. I never thought I'd see the day where downloading real weather from the Jeppeson link would result in smoother performance, but it does. Marc, help I'm a devoted fan, and I'd like to resume using FS Meteo, but not until it's fixed / patched whatever.Bracing for Isabel,http://southwest.corpmerchandise.com/image...l/200015138.gifAlex ChristoffN562ZBaltimore, MD