baffmeister

I have that game installed and was quite impressed with it. The main issue for me is the graphic quality at low altitude. I get really awful "mid screen shimmer" and haven't been able find any adjustments that fix the problem. Like Streakeagle, I was impressed with the flight modeling and consider it the best of the four flight sims I've tried. Will probably check out the update at some point.

No problem, it's not a big deal but for some reason I totally dis-remember what tweaks I may or may not have done to Capuns Viggen FM.

Quite sure that's not an FM I did. A long time ago a tweaked FM was sent to Capun and I recall the actual FM never found it's way into the package but for some reason, my name did. I do have an unreleased Viggen FM that I was working on with actual canard aerodynamics but it had a few issues that needed attention. Will try to resurrect that project and send something to Jonathan for consideration.

Thanks for that piece of info! Getting flash/muzzle positions adjusted has driven me bananas in the past. I've seen some flexible gun set-ups that seem to be impossible to get right so there is a chance you have a model issue that may not be repairable with ini edits. Sorry, not much help. PS: The gun angle settings you have "should work", assuming no model issues. You might want to try using the same node for both pitch and yaw, similar to Veltros example.

I can't imagine TK ever giving up the rights to the SF code either. There was a massive amount of programming done to get the game to its current level. It has to be considered a valuable asset and a potential source of revenue for future games. At this point, I would be willing to purchase updates to SF if they included new features and kept the game viable with the newer graphics cards and latest directX. No new planes, terrains or campaigns required.

Malta Defender.

Happy308, here is another table you could try adjusting to get the desired results. It may reduce the turn capability a bit at very low speeds but I don't think it's very significant. I consider it as a "stab trim for take off" setting. From the stock MiG-21Bis FM: [LeftStab] ParentComponentName=Fuselage ModelNodeName= ShowFromCockpit=TRUE DetachWhenDestroyed=TRUE DamageRating=DISABLED MassFraction=0.013 HasAeroCoefficients=TRUE LiftSurface=TRUE CL0=-0.0192 CLa=0.2833 CD0=0.0005 CDL=0.0001 Cmq=-4.8380 Cmad=-0.4951 Cyb=-0.0723 Clp=-0.0042 Cnb=0.0443 CL0MachTableNumData=21 CL0MachTableDeltaX=0.10 CL0MachTableStartX=0.00 CL0MachTableData=1.497,[1.549,1.616,1.578,]1.000,0.728,0.580,0.494,0.447,0.394,0.231,0.283,0.297,0.283,0.264,0.245,0.231,0.218,0.207,0.199,0.190.....Try reducing the three numbers in brackets for both stabs. Reducing the three numbers in brackets will reduce the download on the tail over the mach 0.10 to mach 0.30 range. With a bit of tweaking you should be able to get the behavior you're looking for. At one point I was playing around with that auto-rotation effect but ultimately, like you, I found it quite annoying and unnatural during the landing phase. It was difficult to stick a nice landing. I think in RL neutral stab trim settings for take off and landing are probably the norm. Not sure to what extent ThirdWire is using that effect to generate more realistic rotation/take-off speeds for the AI, you might want to watch a few of your AI wing men take off to see if there are any issues.

For fans of MigBusters improvement to the ThirdWire F-104 FM, here is something you will want to check out. I should have picked up on this a long time ago as I fly the type relatively often but I think Saisran gets the credit for noticing that something was not quite right. [a post somewhere on this forum, I think.] Anyway, I decide to do an F-104A fighter sweep mission to practice some boom and zoom, which I suck at, but I'm going to really really try harder to keep the speed up and not get tricked into a turning fight. So, I was at least keeping out of trouble but noticed a certain lack of performance regarding the supersonic behavior at low altitude. Took a quick look at the data ini and didn't have to look far. The first thing I saw was this: [Fuselage] ModelNodeName=Fuselage HasAeroCoefficients=TRUE CLa=0.1991 CD0=0.0241................ Yikes! Way over the top! CDL=0.0001 Cm0=0.0007 Cmq=-2.8225 Cmad=0.4620 Cyb=-0.1991 Cnb=-0.3795 That CD0 value is a carry over from the original ThirdWire FM. I doubt it's an accident, I assume ThirdWire put that large number there to counter some of the over performance with their original F-104 FM. In the SF-1 series, the F-104G was player flyable but a realistic F-104 FM would probably be too much for most people [still is for me] so they went with something easier but not bad in a relative sense. After a quick look at some tables in a NASA document called "Quest for Performance", here are a couple of total CD0 values that some smart guy calculated: Quest for Performance Total CD0 values F-105D = 0.0173 F-104G = 0.0172 Here is what ThirdWire is using for those two types: ThirdWire Total CD0 values F-105D = 0.0159 F-104G = 0.0312........Yikes!.....Again! ThirdWire CD0 values always seem to be lower than the reference values I find but don't really think it's a big deal in a relative sense. The excessive CD0 value being used for the F-104 is a big deal with the improved MigBuster FM though. With that FM running closer to book values, you will want to adjust the CD0 values closer to book values as well. Here is what I suggest is a "Legitimate" fix for MigBusters "Legitimate" improvement to the ThirdWire F-104 FM: To fit in with the other ThirdWire airplanes you will probably want a total CD0 value that is less than actual. Using the ThirdWire F-105 as guide, a total CD0 value of 0.0158 for the F-104 would be appropriate. [Fuselage] ModelNodeName=Fuselage HasAeroCoefficients=TRUE CLa=0.1991 CD0=0.0099................Changing fuselage CD0 to 0.0099 gives a total CD0 value of 0.0158, including wings, vertical stab and horizontal stab. CDL=0.0001 Cm0=0.0007 If you're interested in testing the more realistic CD0 values: [Fuselage] ModelNodeName=Fuselage HasAeroCoefficients=TRUE CLa=0.1991 CD0=0.0113.................Changing fuselage CD0 to 0.0113 gives a total CD0 value of 0.0172 CDL=0.0001 Cm0=0.0007 These changes put a lot of "Zip" into the Zipper. I find I can work the Mach 1.0 to Mach 1.3 zone much better and with better energy retention. I have also seen some occaisional super cruise behavior. I think the real F-104 could do that to a certain extent but not sure of the details. It also stays supersonic much longer after coming out of afterburner. If heading downhill you can save a bit of fuel while maintaining supersonic speed. I still find it a difficult plane to do air to air with. After unloading the Aim-9B's and not hitting anything, I'm left with a gun and a plane that zips around really good but doesn't turn very well. If you get a gun angle on the opponent you probably won't have it for long so good gunnery skills required. Still, it's 10 tons of fun!

Thanks Caesar, I think your impressions are similar to mine. Some improvement but hard to identify precisely. It may just be a bit more precision when maintaining a sustained G load. It looked like you were doing that very well in the video. I'm quite interested in trying to improve some of the ThirdWire FM's if I can do it in a legitimate fashion. The CDL table is all I've come up with so far but stay tuned for an interesting fix to one particular airplane!

Yep! Dell can be a bit weird at times. As delivered the Inspiron 580 has an Intel core i3-550 3.2GHZ with 4MB cache. The 4GB of RAM that came with it are DDR3 so no issues there. The max RAM speed is 1333MHZ. The chipset is Intel H57. I recall shortly after purchasing the computer Intel came out with a much more capable chip set. For some Dell weirdness, even though the Inspirons were shipped with Windows 7 64 bit Home Premium, Dell put a max limit of 8GB on the RAM. This is probably locked away in the bios somewhere, which hasn't been updated by Dell since about 2012. Some people claim they have installed more than 8GB of RAM but it seems to be a hit or miss affair. Other weirdness, the manual that came with the computer says it will accept 1,2 or 4GB modules but the "experts" at the Dell help website say it will only accept 1 or 2GB modules. There is a very good independently run computer store nearby so I may see if I can cut a deal on 4 used 2GB modules. Current install is 4X1GB.

Thanks. I'm going to continue to attempt some fine tuning on the anti ailiasing but I'm probably close to optimal for my system at the moment. I played around a bit with the max pre-rendered frames setting and "1" still seems the best but not as dramatic an improvement as I saw with the previous card. After increasing the object texture to unlimited I ran into an over heat situation on the graphics card [frame stutter] so had to disable K-Boost. On exit the card was at 68degC which doesn't sound that hot to me but really don't know much about the subject. With K-Boost off and my current settings the temps are low and the frames are smooth. Wondering if some of the extra texture filtering I was running limited the CPU and caused some issues. With the trilinear and sample optimization off I don't get random "hiccups" any more so maybe the CPU was maxed out from those settings but the GPU still had some head room. Just a theory. I did finally go online to update the computer so it's now circa 2015. Only took 6 hours! Will try the 353.30 drivers. Also, might increase the RAM from 4 to 8GB for a final update to the old but very reliable Win7 64 bit I-3 Dell Inspiron.

I would be interested in knowing your game settings and Nvidia settings. My last attempt at optimizing some settings for my GTX 750 had some suprising results. Still mainly testing with the Centurion F-5E package which, when in cockpit view, seems a bit more resource intensive than the Mirage F-1. At this point the stock ThirdWire aircraft are working very well. Anyway, the suprising thing is I seem to be getting improved performance by INCREASING some of the in game settings. Setting the horizon distance to unlimited seems to be too much but setting object texture to unlimited seems to have improved performance somewhat. Other in Game settings: Object Detail=High Cockpit Texture=High Cockpit Mirror=On Cockpit Reflection=Off Shadow=Low Effects Detail=High Terrain Detail=High Terrain Texture=High Ground Objects=Unlimited Water Detail=High [it shimmers] Regarding the Nvidia settings, some more surprises. I use V-sync to reduce cockpit frame "jaggies" and it may prevent texture tearing as well although that doesn't seem to be a big issue with the older 340 series driver I'm using. The big surprise with the Nvidia settings was the behavior of some of the additional filtering settings that are intended to improve performance. They appeared to make things worse! With Trilinear Optimization and Anisotropic Sample Optimization set to off I get smoother performance and a higher quality picture. Anisotropic filtering is set at 16X and Antiailiasing mode at Application Controlled.[4X?] I'm still using Nvidia Inspector to access the Sparse Grid super sampling but, with things looking good at this point, I may experiment a bit with the conventional Nvidia Antiailiasing-Transparency settings to see if I can get equivalent picture quality with better frame rates. Any thoughts?

Thanks Caesar.

No joy yet. I've been reverting to stock some of the add-on water features but nothing I've tried so far has made a difference.

Jeanba, If you want some help with the airfield flattening send me a PM. Like a lot of other folks I would like to see this terrain get released so don't mind helping out if you're interested.

At the top of the Terrain editor window you will see a drop down menu called "View". Click on that, then select "View Height Field" from the drop down. The height field should appear.

Things change, in the past an Nvidia card seemed to be the best way to go with the SF game. From my recent experience with a new GTX-750 and some internet info, Russouks' comments might be worth considering for any SF players shopping for a new card. [my previous Nvidia card was very impressive, considering the lightweight system I run.]

Thanks Blaze, I was able to isolate the problem area by turning down the in game graphic settings and then increasing them one at a time.The problem for me was the in game water detail setting. Any setting above low would cause the black screen lock up. I did the tests with the Desert4 installed without Centurions' update pack. Might be worth mentioning that the Desert 4 terrain has some modified settings within the Desert 4 data file. The modified settings are a water shimmer fix but it may cause ships to appear to float above water line, if I remember correctly. This modification isn't causing the black screen though, I get the black screen with stock TW terrain data settings as well. Will try to isolate the issue so I can run this terrain with the same settings I use for everything else. Cheers!

Thanks for the offer. I think I will keep trying a bit longer. The problem is a bit odd, the terrain loads to 100%, the screen goes black, the hard drive keeps doing things for awhile then stops and I'm stuck with a black screen. No CTD. I tried playing around with the naval and water map statements but that didn't help. Anyone have any ideas?

Thanks KJakker, will try those settings when I get Desert4 working again. Tried a re-install, didn't work. [what have I done this time... ]

Thats interesting! Could somebody try that fix with the Desert 4 terrain. I somehow porked my Desert 4 install awhile back and haven't got it going yet. [might have to reinstall it]

Some test pilots here? Some time ago I built an expanded 1 deg symmetrical CDL table for the stock ThirdWire 4 deg table used on some of their aircraft. Assuming the flight engine generates a straight line variation between the CDL data points, my thinking was the stock 4 deg TW table would generate a very jagged drag bucket. After testing the new 1 deg CDL table I thought it offered a bit more precision when trying to maintain a sustained "G" but the improvement was less than I expected. Both the Mirage 3 and Mig-21 use this CDL table so thought some of you might be interested in trying out the 1 deg version. My game set up is very light weight with just an old swizzle stick being used so maybe someone with better gear will notice more of an improvement. It's an easy edit to the data ini, just extract the appropriate data ini and delete this table from both the inner and outer wing sections: delete: CDLAlphaTableNumData=15 CDLAlphaTableDeltaX=4.00 CDLAlphaTableStartX=-28.00 CDLAlphaTableData=49.000,36.000,25.000,16.000,9.000,4.000,1.000,0.000,1.000,4.000,9.000,16.000,25.000,36.000,49.000 and add this table: CDLAlphaTableNumData=57 CDLAlphaTableDeltaX=1.00 CDLAlphaTableStartX=-28.00 CDLAlphaTableData=49.000,45.550,42.238,39.052,36.000,33.053,30.242,27.555,25.000,22.556,20.244,18.058,16.000,14.059,12.247,10.560,9.000,7.560,6.248,5.061,4.000,3.062,2.249,1.562,1.000,0.562,0.250,0.063,0.000,0.063,0.250,0.562,1.000,1.562,2.249,3.062,4.000,5.061,6.248,7.560,9.000,10.560,12.247,14.059,16.000,18.058,20.244,22.556,25.000,27.555,30.242,33.053,36.000,39.052,42.238,45.550,49.000 Any volunteers?

My observations regarding that issue from past experience. When leveling an airfield using the TW terrain editor the leveling process makes changes to both the HFD and TFD files. Your screen shot looks like what I saw in the past when I added just the new HFD file but not the new TFD file. Not 100% sure if the terrain editor you're using actually makes the required changes in the TFD file. The updated Desert 4 terrain by Piecemeal/Centurion exhibits the same issue with some airports. Some of the airfield texture tiles were changed but I don't think they were re-leveled using the TW terrain editor.

I recall seeing that issue before and fixed it by using the stock Desert terrain values. I don't know what any of it means but you could try these: [HeightField] DetailScale=1.0 DetailFrequency=1.0 MaxHeight=1000 MinHeight=0 [HeightOffset] LowDetailMesh=-1.5 WaterMesh=0.8

Found what I think is a very good article explaining the Nvidia graphic card settings. Here is a link: http://www.tweakguides.com/NVFORCE_6.html My new Zotac GTX 750 is working "OK" now but the overall picture quality is less than I got with the GT-240, even with slightly higher screen resolution. Also, I had to resort to some other software to get improvements. Here's a brief explanation of some problems I had and the fixes. PROBLEM: Random frame rate stutters. This was really annoying as I would sometimes get stutters even when there was next to nothing going on in game. After some internet research it seems to be an issue for some people with the newer Nvidia cards. It probably has something to do with the way the drivers throttle the cards operating frequency depending on graphics load. I think Nvidia has been making some adjustments with the newer drivers to try and fix the issues but I'm stuck with old drivers so had to try a different approach. SOLUTION: Installed EVGA's Precision X overclock utility.[free] It has an easy one button feature that will activate K-Boost. K-Boost will override the Nvidia software and allow the card to run at the maximum frequency all the time. This will increase the power consumption and temperatures a bit but I haven't seen temps over 55degC after exiting the game. [not sure if I can moniter temps in game with that utility] PROBLEM: Really crappy anti-aliasing. The old GT-240 was way better in this regard and actually had better picture quality even with a lower screen resolution. SOLUTION: Downloaded another utility called Nvidia Inspector. [free] With Inspector it's possible to access some graphic settings that Nvidia doesn't include in their normal selection screen. With Inspector, I was able to access something called Sparse Grid Supersampling within the Transparency Supersampling field. Sparse Grid Supersampling seens to require a lot of resources but when set only to 2x, it still looks better than the regular Supersampling at 4X. Even with that set, the picture quality is still a bit less than what I had with the GT-240. The Zotac GTX 750 Zone card I have is passive cooled and I think it's been downclocked a bit compared to other GTX 750 cards. From what I've seen, I think there is room to upclock the card to more typical GTX 750 settings. For anyone interested, the free utilities mentioned are available from multiple sources so just do a search and download from your prefered site.