Mike,I hope it won't sound argumentative but I have hard time believing what some people (not you) write on this forum in terms of "miracle" fixes or tweaks. I believe some of it is sort of self-induced hypnosis - you do something, you think it should help (since theory is sound) and therefore you "see" the results. I am afraid I probably would not be a very good customer of FS-GS since it might be difficult to convince me that something looks so much better or flies so much smoother than yet some other approach. When I evaluate a tweak I really make an effort to be rigidly objective and not fall a victim of late night hour or desire to find a magic potion. Of course there are many good tweaks mentioned on these forums but I quickly realized none of them provides a significant jump in performance - instead it is a tedious process (that I go through at the moment) to go one by one of those tweaks and see the performance/quality to change by ever so little and make determination that you are on a right/wrong track. I was just reading recently some old posts and one was from a guy who claimed that compressing his FS9 folder made his flight so much smoother, increased visibility, frames rates jumped by 10 fps, sharper scenery, etc, etc. Of course I don't believe a single word that he wrote though I know he was absolutely honest when he was writing it. It took me sort 2-3 installations/uninstallations and over half a day to convince myself that Omega drivers really offer something better than stock ATI drivers. I obviously am in perfect agreement with you about that frame rate counter being irrelevant (to some degree) - once I start more flying and less tweaking I will never look at the counter again.Michael J.WinXP-Home,AMD64 3500+,Abit AV8, Radeon X800 Pro,WD 36GB Raptor,1 GB PC3200

Hi Mike,While I'm convinced that FS-GS offers a great service, my problem is that I feel no real need for such a thing at this time. I have no complaints about the sim as I see it running on my system.Like Micheal I'm rather thorough about trying any tweak. Some that are offered I know to be snake oil (as the one Michael exampled above) while others I find merely interesting, and still others get my firm attention to test. Rarely does any of them singularly make much difference (though I did recently catch one about on-board sound that had me re-installing my Turtle Beach within a couple minutes of trying the tweak).All three of us know that there are so many variables to making this sim run well. And FS-GS is very adept at putting them all together for their customers. If I do buy the service at some point in time it will more for curiousty and the desire to learn more than to enhance my already very enjoyable simming experience.My previous post about trying 6X AA was too brief (the "Boss" had me by the ear insisting that we go somewhere!!). When I tried 6X AA in my standard test flight it was as good as it always is (using 4X AA) while I was straight and level. But as soon as I started my standard rate turn (which I always do) I saw the slightest bit if hesitation in screen loading (I'm not sure I would even call it a "stutter"). But the sim wasn't quite as smooth as it usually is in that turn. The big thing though is that I really didn't see that much better IQ with 6X vs. 4X. Perhaps it's my aging eyes, or my monitor, or any number of other factors, but I simply could not see the value of 6X AA here.Now AF is a different matter entirely for me. I can easily see the difference between 8X AF Forced and 16X AF Forced. I can even retain the same level of AF but use Performance or Quality (thereby allowing the drivers to "sense" the AF) and see a difference.All this proves that we're each different. And makes me glad I'm not a game developer (with all those customers wanting and expecting different things).Cheers,Greg

>All this proves that we're each different. Amen Brother.Michael J.

Good morning guys,wow, thanks for this VERY useful and interesting thread! I've been in the tweaking land again lately (hadn't touched a thing for a long time) and this thread comes as a balsam for my tortured soul :-lolAnyway, I've been scratching my head over the fact that I saw absolutely no change in quality/fps when I was setting different levels of AA and AF for my 9600XT and 4.9 CATs (without the .net thingie). I'll try the Omega drivers and see where that leads. I can't believe that pushing the AA slider from 2x to 6x would have no effect at all!? Thanks again!Regards,Jurehttp://www.globecargo.org/images/VAA_Captain.jpg

Hi Jure,And a very good morning to you to! :)First, I must say I have the '.net thingie' installed and fully updated and have witnessed no impact on either general system performance or within FS."I can't believe that pushing the AA slider from 2x to 6x would have no effect at all!?".....now, as ever, I can only cite my experience but I will try to illustrate what I am seeing and experiencing with the following:My normal startup situation is not the MS default, but, for the sake of this demonstration, I have selected KSEA and prepared a reproduceable situation since others can then recreate what I have on screen and can then compare results.I use RealAir's Cessna and am viewing it from outside. I doubt whether the differences between this models's textures and MS's default will have any impact on frame rates. Did I say frame rates? Yup, I have twisted my arm (ouch!) since there is no other way to show you the results.FS Settings:Scenery Complexity: extremely denseAutogen Density: dittoTerrain Mesh Complexity: 100Sight Distance: 80 milesCloud Draw Distance: 40 miles3-D Cloud % : 100Detailed Clouds checkedCloud Coverage Density: MaximumMy clouds are the defaults overwritten by Chris Willis's HiFPSv2.zip (cumulus dxt3, cirrus dxt3) and the Sky Texture package from FSWater.Traffic 75% (Ultimate Traffic)I use FSAutoStart and for this exercise had only Explorer and FSScreen running and a total of 12 Services. FSAutoStart was idling in the backgound to allow me to dip in and out of the sim without reloading and unloading all the services and programs. Anisotropic Filtering in DirectX remained unchanged throughout at 16x (forced, quality)FSAA 4xhttp://forums.avsim.net/user_files/92354.jpgFSAA 6xhttp://forums.avsim.net/user_files/92355.jpgAnd finally,FSAA 6x (Traffic 0%)http://forums.avsim.net/user_files/92356.jpgI appreciate that these images represent a somewhat artificial situation in that I executed each capture before any AI traffic started moving. Additionally, being a static image (apart from slowly moving clouds) the displayed frame rates are relatively unchanging and vary by only +/- a couple of fps. This would be somewhat different during takeoff and, as you gain altitude, the fps count would rise progressively.Before any one cries "It's a fix!" when they compare the fps count between the first two images I was just as surprised as you may be feeling. Each time I captured an image I tried to hit the button at a moment which would indicate a true average. The fact that the first 2 images have precisely the same count is just a lucky coincidence - you do have to trust me on this one.However, despite all the foregoing I think they do serve to illustrate what I am seeing on my system in a moderately dense situation where much anti-aliasing is going on.Lastly, an admission.... When I examined each image in PSP and enlarged to 250% I have to agree there is not a great deal of difference in image quality between AAx4 and AAx6. I have to assume, therefore, that the improvement I am witnessing is a reflection of the fact that I am observing a dynamic image on screen.As I stated, this is not my preferred startup situation as the frame rates (Ugh, those wretched words again) are a little on the low side for silky smooth operation during takeoff and landing. I have defined the right hand rocker switch on my yoke to switch between 75, 100 and 0% traffic and this does help a great deal in a situation such as this. Having said that, although I am aware of slower screen updating, frame updates remain regular and consistent so a/c are still flyable even at 100% AI traffic.Hope this goes some way towards convincing you (not forgetting Greg and Michael ;)).Cheers!MikeP4 2.4GHz (400FSB), 1Gig PC2100 DDR Crucial, ATI Radeon 9800 Pro 128MB (Omega 2.5.76), SB Audigy (5.12.0001.0443), Hyundai ImageQuest Q17 17" TFT LCD 20ms Monitor (1280x1024x32), Gigabyte GA-8IRXP MoBo, Ultra-Quiet PSU 400W, WinXP Home (SP2), DirectX 9.0c, AGP Aperture = 128MB

Mike,thank you very much for your detailed demonstration!I think I don't need convincing, I believe what I read from you guys alright. :) It's good to know that the ".net thingie" doesn't impact the performance. Thanks for that info. I try to install as few thins as humanly possible, keping the system clean. I recreated your situation and these are the results: Pushing AA and AF past 4x and 8x respectively makes no difference whatsoever on my system regarding visual quality. I eventually lose a few FPS at AA6x and AF16x, but the image is virtually the same. BUT, lowering AF to 2x I can see the difference alright! At last... I didn't try that when I was testing last. I do realize that these are static images we're discussing. As for the fluidity of the sim, I played with the AGP aperture a lot and that did affect things. Unfortunately, I cannot use AGP app. at 256/512 because my secondary monitor is then useless. ATI is aware of this, I noticed.I gained the most (got rid of the blurries) by setting TEXTURE_BANDWIDTH_MULT=400! I have no clue whatsoever why had I left that at default? Must be getting old...Next thing, of course, is AI traffic. Like you, I use UT and very rarely lower it below 100%, although that helps with fps the most. Two other things for better fps are Aircraft shadows (off) and terrain density setting, I leave that at 60% usually. Mike, I am not hunting for more and more fps at all, I believe I hit the sweet spot on my system. Got rid of the blurries and the sim is fairly fluid. But I did get curious reading this topic and like Michael, I had no clue about the Omega drivers. What still puzzles me, is why there is no impact when I push AF fully to the right, past the 8x setting?And last, just for the record, at AA4x and AF8x at KSEA (your settings and a few things running in the background) my fps were around 35. Can it really be that processor is so very important? That itself is a valuable lesson for me. Seeing that you have a superior graphic card... hm... good to know what to upgrade first when the time comes! Anyway, a great thread and I learned a thing or two! Thanks a lot! :)My system:Athlon 3000+1Gb RAMATI (Sapphire) 9600XT 128Mb CAT 4.9ASUS A7N8X DeluxeWin XP ProRegards,Jurehttp://www.globecargo.org/images/VAA_Captain.jpg

Mike,Thanks for posting some pics. Regardless of the frame rate issue I absolutely see no difference in image quality between your AA 4x and 6x. Maybe you or Greg however could show some more screenshots showing apparent difference in IQ between the AF 16 perf - forced. I am really trying to learn what this forced AF is all about. Also I noticed your cloud draw distance is less than your sight distance. I noticed that from other people too (and per default setting). I thought it shoud be the opposite - draw distance should be larger so all the clods drawing is done beyond the visual range.Michael J.WinXP-Home,AMD64 3500+,Abit AV8, Radeon X800 Pro,WD 36GB Raptor,1 GB PC3200 http://www.reality-xp.com/community/nr/rsc/rxp-higher.jpg

Hi Michael, I just found the following post by 'Ricky', a moderator/builder at the Abit forum. It might be of interest, as more and more folks are using Via chips... The thread is here: http://forum.abit-usa.com/showthread.php?threadid=63868The content of the Ricky's message (quote): VIA 4in1 4.53 Users and XP SP2 Please Read... Introduction Identify your chipset Drivers - which ones do you need? About VIA Hyperion 4in1 drivers Special note - WinXP SP2 users Microsoft Windows XP Service Pack 2 (SP2) users - please note: AGP driver recommendation: To deliver a quality user experience, a universal standard for AGP drivers, UAGP35, is being adopted by all chipset vendors with AGP 3.5-capable hardware. VIA have worked with Microsoft on this issue for a long time and continue to do so. There are many VIA AGP3.0/3.5 devices which are supported in SP2. VIA's AGP driver contained in the latest Hyperion 4in1 drivers will work properly with Windows SP2, however, the default Microsoft driver is recommended. Please follow the advice below, depending on whether you are installing WinXP and SP2 on a clean system or installing SP2 on top of WinXP SP1. Clean system install: (1) Install Windows XP (2) Install SP2 (3) Install VIA Hyperion 4in1 4.53 driver but untick the AGP driver so that it is not installed. System Upgrade: If your system already has Windows XPSP1 installed and the VIA Hyperion 4in1 4.53 driver then: 1) Install SP2 2) Run the VIA Hyperion 4in1 4.53 driver setup again to remove AGP driver. By removing the VIA AGP driver you will then be using the recommended Microsoft default AGP driver from SP2. More information: Microsoft has initiated a standardized way of programming and manipulating the AGP driver. In summary, Microsoft are concerned that vendor-specific drivers may define different page table entry (PTE) formats for GART, choose different register layouts for their AGP target (motherboard) devices and invalid GART entries could cause AGP-related system instability, such as memory corruption. Detailed information can be found here. VIA continuously works with Microsoft to provide the best possible experience for users of both products.****************** End quote *****************Hope it's usefull. Specially that PTE stuff sounds quite dramatic! I wonder whether it's also valid for the latest and greatest display adapters, XP-SP1 and/or Win2k-SP4? Anyway, isn't it a bit ironic, to be advised to use MS's default drivers for AGP?! Maybe, the next 'Hyper-Onion 4in1' changes this situation again? Cheers, good luck and kind regards Jaap PS: Greg, this is result number ten when doing 'a yahoo' for 'unian drivers': Italian Police Smash Ukrainian Road Transport Racketeers ... Make www.unian.net your home page! ... which extorted money from hundreds of Ukrainian truck drivers in northern Italy, the La Repubblica web site reported ...www.unian.net/eng/news/news-37033.html - 33k - Cached - More pages from this site Sincerely Greg, are these the drivers you recommend for my FlyTendo? Extorted Ukranian truck drivers?! Hmmm, I bet your next proposal is a bicycle dynamo instead of a PSU!! Suuuure... Hehe

Thanks, how do I find out if I have this particular driver installed?EDIT: I do have VIA 4in1 4.51 driver installed (not "53") so does it apply to me as well?. And when I inserted the Abit CD I see no option to uninstall the driver, only to install it.Michael J.

Hello Jure... long time no see here. Glad all is well with you and that you seem to have your sim humming along nicely.Jaap... bad news for you, my friend. Now that you've revealed the true identity and motives of the "Unian" organization, sad to say they will soon be arriving in your life to break your knees and otherwise make life miserable for you. And you've cost me a great deal of money in referral fees.. which I won't forget. :-) Getting serious now... Mike, I ran some additional tests on this 4X/6X AA matter. My previous tests above were not given enough time for thoroughness (she was pulling my ear rather emphatically). I loaded up a flight as in your screenies above (including the texture package from RealAir and using as close to the sky and traffic conditions as I could replicate). The one major difference is that I don't use UT. I do use GA-Taffic, and I set my traffic level in the sim at 40%.First of all, my eyes just can't see the difference. My performance hit wasn't as bad as I orginally posted (though my personal test flight scenario is not the flight you used in Seattle). I lock my frames at 25, and during my testing using your flight in Seattle I saw 20-22 FPS using 4X AA, and 19-21 FPS using 6X AA. What I did see (and I don't like) is more flucuation in FPS using 6X AA. I use the framrate average tweak in my FS9.cfg not so much so I can see the average framerate, but because I'll study the % change during testing. This to me is far more important to fluidity than just the framerate and is the sole reason I add the line to my FS9.cfg.I took some screenies and then loaded them up in Photoshop and viewed them at 250%. Like you I could see a difference between the 4X AA and 6X AA.But in the sim I still can see no visual difference between 4X and 6X.Michael... I'll soon post some screenies showing different levels of AF.Greg

Hello Greg!Indeed, long time no see. All sorts of things have been happening in the thing we call real life. :) I want to thank you again for your advice on cooling, my temps went from 66-70 Celsius at full throttle down to 39 MAX!! (yep, I got the thingie at last, :-lol )Now, that's a difference and shame on AMD for putting such crappy coolers on their processors. Regarding the fps and AA... maybe we should also tell the screen resolution, I run at 1152x864 so that's a part of the reason for my higher fps at KSEA. And I did notice fluctuation in fps getting higher at times. Interesting observation, Greg! I'll wait for the screenies, am very interested in difference between lowest and highest AF. Regards,Jure

As always any info from Microsoft (and specially install/uninstall procedures) should be treated with outmost suspicion.I finally figure out a way to uninstall VIA 4in1 AGP driver (you have to select the normal installtion and then there is option for unistall) but then my AGP stopped working. I tried to uninstall Omega drivers and then reinstall to no avail - Omega was telling me I can't use AGP. So here we go again - install procedure from Microsoft that don't work. If SP2 constains AGP drivers the certainly nobody knows about them.Michael J.WinXP-Home,AMD64 3500+,Abit AV8, Radeon X800 Pro,WD 36GB Raptor,1 GB PC3200 http://www.reality-xp.com/community/nr/rsc/rxp-higher.jpg

Hi Greg,Ain't this fun?! :)Just to clarify a point:You stated, "I took some screenies and then loaded them up in Photoshop and viewed them at 250%. Like you I could see a difference between the 4x AA and 6x AA."I think you may have missed that one or maybe your typing finger slipped. In fact, I could NOT see any obvious difference between the static images at AA filtering levels of 4x and 6x. However, I do see a subtle but pleasing difference in the running sim. I am assuming that this is a reflection of the dynamic nature of the anti-aliasing algorithms as opposed to what we see when executing a screen grab. In any event, the fact that I (and I do stress this is on MY setup) see no obvious drop in performance when using AA 6x persuades me to carry on using that setting.I would be the first to admit that I am no expert and while these speculations of mine may just be pure fantasy I remain grateful for your patient indulgence. However, as they say, the proof is in the eating....or something like that..and I for one am enjoying the taste.. LOL!MikeP4 2.4GHz (400FSB), 1Gig PC2100 DDR Crucial, ATI Radeon 9800 Pro 128MB (Omega 2.5.76), SB Audigy (5.12.0001.0443), Hyundai ImageQuest Q17 17" TFT LCD 20ms Monitor (1280x1024x32), Gigabyte GA-8IRXP MoBo, Ultra-Quiet PSU 400W, WinXP Home (SP2), DirectX 9.0c, AGP Aperture = 128MB

What follows are test screenies using different levels of AF. Keep in mind that ATI and nVidia both "optimize" their drivers for AF. Again, my feeling is that those optimizations offer us little in this sim. Using Radlinker one can set "Performance" or "Quality" mode (these two modes are also available in the old ATI Control Panel). These modes both use the optimization algorithm. Radlinker also offers the option of "Forced" AF (at whatever level you choose). Forcing AF means the optimization algorithm is shut down and the video card will force AF at whatever level you choose.In looking at these shots pay particular attention to the urban sprawl halfway up each frame. Study it carefully from the right side of the frame to the left where it continues out onto the peninsula.4X AF Performancehttp://forums.avsim.net/user_files/92430.jpg8X AF Performancehttp://forums.avsim.net/user_files/92431.jpg16X AF Forcedhttp://forums.avsim.net/user_files/92432.jpgAs can be seen the area of urban sprawl midway up the scene becomes more defined as AF is increased. Also note that the first two shots were made using "Performance" mode, so the algorithm was giving me what "it thinks" I should have. The final shot was made in "Forced" mode, so it is happening in true 16X AF.Hope this helps,GregP.S. Mike... since you are using the new ATI Catalyst Control Center (and all it's associated "thingies") can you confirm whether or not it offers "Forced" AF?

Hi Greg,Excellent demo!"P.S. Mike... since you are using the new ATI Catalyst Control Center (and all it's associated "thingies") can you confirm whether or not it offers "Forced" AF?"...... I didn't think I was using the 'new' ATI Catalyst Control Center. I was under the impression that this was part of Omega's built in Radlinker module. With previous driver sets you had the option whether or not to install Radlinker and I never did. This latest driver set removes that option and you have to accept RadLinker whether you want it or not. I've just checked the available menu options and I notice that there is a "(Re) Install RadLinker Omega Edition" item. I clicked on this and was offered the option of either Repairing or Removing RadLinker from which I infer that RadLinker is indeed installed.This is what I see:http://forums.avsim.net/user_files/92433.jpgSo, I guess the answer to your question is yes.MikeP.S. I note in an earlier post in this thread that you have your fps locked at 25. Your system specs would strongly suggest to me that this figure could certainly be set higher. As you know, I have mine set at 31 and during a flight the frame rates are frequently brushing that figure. The bonus, is seen, in much smoother sim performance generally. I have not been aware of any downsides in terms of any significantly slower texture loads when fps dip to the low 20s or high teens. Go on treat yourself!