Moa

If anyone wants to really understand the ED document or is creating their own flight model (I am) here are three books to really help get you up to speed: 1. Mechanics of Flight, 2nd Edn Warren F Philips https://www.amazon.com/Mechanics-Flight-Warren-F-Phillips/dp/0470539755/ 2. Applied Computational Aerodynamics: A Modern Engineering Approach (Cambridge Aerospace Series) https://www.amazon.com/Applied-Computational-Aerodynamics-Engineering-Cambridge/dp/1107053749 3. Numerical Recipes 3rd Edition: The Art of Scientific Computing 3rd Edition https://www.amazon.com/Numerical-Recipes-3rd-Scientific-Computing/dp/0521880688 Mechanics of Flight - gives you an understanding of aerodynamics and dynamics (the physics of motion). It includes a chapter on the equations of flight in matrix form suitable for use in a flight simulator. It goes over the such subtleties as of the differences in sideslip determined from experimental data vs that obtained from calculation (don't mix them up !). Applied Computational Aerodynamics - shows how aircraft performance can be estimated without access to a windtunnel or the aircraft. ED are doing this with their CFD modelling (bravo !). Numerical Recipies - shows the techniques for non-linear fitting where you can take the observed or computed aircraft performance curves and fit to get the fundamental aircraft parameters which characterise the performance. Hopefully that helps jump-start anyone with the inclination to get into this stuff (there have been aerospace engineering students on these fora in the past). Thanks Wags for the technical overview. It won't stop that whining noise here, but still great to see.

They're working in dimensionless units.

Thanks for asking your customers as to what their particular preferences are. This is a good initiative Wags/ED. My personal emphasis was on standoff weapons. F-16C cannot easily defeat SA-6/SA-11/SA-10 with current munitions. SHORAD can be taken on from high alt, but not the better defended targets (and warships). So pretty please get the standoff weapons done. Then the other features can come in a more luxurious timeframe. Edit: The Maverick is the most versatile of the stand-off weapons, get the AGM-65D or G IR versions done and the Viper can do SEAD/DEAD, CAS, Anti-Armour, Anti-Ship, budget reconnaissance with the seeker etc. ps. I had no problem doing the survey in Google - but the same browser session was already authenticated against YouTube.

Good job c0ff. Every coder knows bugs happen. It's the speed you fix them that matters. I hope the rest of your week ends up better.

Wow! that looks fantastic. Thanks for putting in so much work not only to produce such a comprehensive document but also making it so aesthetically appealing as well. Outstanding work.

Thanks for the info ED. It's nice to know how many cool things are planned. If you slip the release date past the 15th April I will not complain, since so many cool things are coming. Stay safe.

The manual is great. The big improvements are: * topics oriented around the procedure you need. * a combination of text and orienting images (eg. in startup) * logical layout with good progression of topics. Great job on the manual. It provides an example that is worth following in other modules.

Always do 'profile guided optimizations'. That is, use a profiling tool to look at your system's performance. For CPU profiling you can simply use the Task Manager built into modern versions of Windows, eg. Windows 10. Watch the CPU load. For GPU profiling you can use the free tool GPU-Z https://www.techpowerup.com/gpuz/ If your current graphics settings have the GPU running at 100% utilization you either must decrease your graphics settings or increase your GPU power. If your current settings have the CPU core running DCS running 100% then you must either decrease your settings (eg. draw distance etc) or get a new CPU. Eagle Dynamics has done a lot of work to offload what it can to other CPU cores (eg sound) but essentially DCS is still single threaded. That means it is better to have fewer but faster cores than many slower cores, at least for DCS purposes. Making DCS truly multithreaded would take a huge amount of effort so don't hold your breath for that (I wrote an unreleased flight simulator in Java that was multithreaded and getting it right was tricky - but worth it, but you have to start the architecture as multithreaded, adding it later would be very hard). Given more resources you would always get a new CPU and GPU. In the real world resources are limited so you have to chose which to replace. Profile your system and see which of your CPU and GPU is being stressed the most. CPUs single core performance increases very slowly these days (increases in performance comes from using more cores on multithreaded problems). So generally it is best to upgrade your GPU for better DCS experiences.

I think the question has been answered. You are not the only one who does cold starts. The beauty of DCS is it gives you the choice: have all the time in the world, then cold start have limited time/distractions/family/pets/etc use autostart or airstart Choice is good. Anyone who thinks they are especially awesome for only cold starting and anyone else is inferior because they don't always should back away from the keyboard and go outside to get some perspective. This community has room for everyone no matter what their preference is :)

My stats-board is all set for multi-server. Once the UI/styling is complete it can take data from other servers. We are a community after all :) As well as a heat map of player deaths I'm also keen on getting aggregated statistical data, such as the actual missile Pk map. Plus, a dynamic campaign needs to know things such as ejection locations etc to generate SAR missions for the rotorheads. Thank you. I'm into realistic Order-of-Battle and 'wargamer' type stuff, and want realistic ATO and maintenance/serviceability and a bunch of other stuff. Plus, I have lots of TN Dupuy's works, such as: https://www.amazon.com/Numbers-prediction-war-history-evaluate/dp/0672521318 Working on coding that model is why I originally learned to code (fortunately it has been very lucrative since). So I'd love to complete that model and incorporate realistic ground attrition and movement rates - but that is further in the future.

Yes, I store everything I can, including some additional information (like position and velocity of each unit for each event which has it - which I may use later). It's still a work in progress. Slmod works as it did before, I just extended Slmod to write to a separate log, which I transfer to a Postgres database and then have a Java/Groovy REST webservice read that and display using HTML/JS/CSS. I know *exactly* what you mean. I too have a rocky relationship with Slmod, although I'm very, very grateful to Speed and yourself for providing a stable basis for doing other things with, like detailed pilot stats. So thank you very much for the times you do grit your teeth and fix it up. I very much prefer working outside LUA. One day I may add socket export to a service where a more robust language can be used. As I mentioned, once i've finished tarting the stats page up I'll be using it as Battle Damage Assessment for mission generation, and close the loop dynamic-campaign wise (I know ED are working on this, but it always good to have choices and healthy co-opertition). Thanks again for your work on SLmod. ps. Slmod works fine with the new Dedicated Server - probably due to your work testing. I didn't have to change a thing.

@Grimes: if you are ever interested in the mods I made to SLmod for 104th stats just give me a bell. Thanks for your work in maintaining SLmod.

It's a LOT of work - mostly because the events logged in DCS, even with SLmod, omit all sorts of information. Rebuilding it takes a lot of effort. Here's my effort for the 104th Phoenix server (work still in progress): http://dynamicscore.org I based on SLMod and then wrote a bunch of custom code running on an Amazon EC2 server (with a Postgresql database). However, once you have stats like this you can do Bomb Damage Assessment, and then link it with a mission generator and Order of Battle database - for a full dynamic campaign (which will be my next trick now I have some free time for modding).

"Perhaps in Wellington the Viking has been bred out of you," 'Vikings', those wimps ? I personally am descended from cannibal natives of the region. Most of our environmental protection is actually at sea - because oceanic species travel far to breed. An we have an 'Air Force' with no force to it. The first duty of government is national defense and the proceeds of the tax slaves should be spent on other things after internal and external security are looked after - but that swindle of the political class is not unique to the Antipodes. On topic: even on light aircraft 'ballooning' from extending flaps and 'sinking' when retracting are very prominent effects you learn to anticipate. So too with the Hornet and its best to be under 200 kts IAS as other posters have started.

+ Baco - "I am running out of runway at landing. what Am i doing wrong, I am touching around 130 Kots, applying wheel brakes, but it just doesn't stop..." How far down the runway are you touching down ? this thing has quite a long roll out. If you touch down in first quarter of the runway you should be ok. Can deploy airbrake with B key once down, and spoilers will be activated automatically. Landing worked ok for me. Yes, the brakes are weak so you need to land as close to the threshold as possible and not float above the runway for a long time.

+Frag - "That thing can fly, I was quite impressed by its rolling capability. Was it THAT maneuverable!? If yes it was quite an airplane!!!!" Roll rate of A-4 is around 720 degrees/second, which is about twice what larger US fighters do today. Yes, it is pretty impressive. Try going down gorges south of the Hoover Dam, the thing can turn. Great mod A-4 team. I have had no graphics glitches and it runs like a dream. Flying it the way kiwis were trained to do, 50 feet AGL (I kid you not).

Fantastic. I was going to do an RNZAF 75 Sqn skin for this, but you've already done it. From the RNZAF 75 Sqn skin it will be easy to convert to an RNZAF 2 Sqn skin too (the Operational Conversion Unit). Mix that with the MB-339 in an RNZAF 14 Sqn skin and you have a mod for a virtual New Zealand Airforce. Hmmm, now there's an idea .... Great work on the A-4 guys. Thank you so much for this.

University of Texas has digital scans of the Tactical Pilot Chart 1:500000 series for the whole World. http://legacy.lib.utexas.edu/maps/tpc/ World aero charts from UTexas http://legacy.lib.utexas.edu/maps/jog/ Enjoy. There is a digital flight planning tool being created for DCS you may also be interested in: CombatFlite https://forums.eagle.ru/showthread.php?t=195724 It looks very promising. For scrolling around the US this is pretty awesome, eg. if you are planning flights on the Nevada map: https://skyvector.com/ The US taxpayer has kindly created charts for everyone to use through the Federal Aviation Administration https://www.faa.gov/air_traffic/flight_info/aeronav/digital_products/vfr/ Direct link to Las Vegas sectional charts: PDF - http://aeronav.faa.gov/content/aeronav/sectional_files/PDFs/Las_Vegas_99_P.pdf GeoTIFF - http://aeronav.faa.gov/content/aeronav/sectional_files/Las_Vegas_99.zip

Grim Reapers have a great tutorial for Mk82 Snakeeyes "FA-18C Hornet: Accurate Snakeye Ripple Bombing Tutorial | DCS WORLD" After trying strafing it also seems the whole bombsight and gunsight needs work. The convergence of the gunsight doesn't seem to take into account slant angle and range (assuming radar and onboard sensors are providing these). Put a pipper on a ground target and pull the trigger and the rounds go above the target to land behind it. Both A-10 and Harrier handle this correctly, but the Hornet needs more work on this - we expect such teething issues for Early Access.

Kuky is correct in this. I can't comment on the Mirage but in the F-16 the leading edge flaps are controlled by a flight computer. While the actuation rate of this flap is 25 degrees per second, there is a 0.136 second lag before the command has percolated through the flight computer. Similarly, the Viper's aileron and rudder slew rates are 80 degrees/second and 120 degrees/second with first-order lags of 0.0495 second. It is well known that other pilots transitioning to the F-16 can often tail strike on their first takeoffs because they command pitch but there is a noticeable delay before the aircraft responds, so they pull harder and then finally the response kicks in. This is a pig to model in-game, but a circular buffer of responses after the lags works ok for me (again, in a different simulation) - although there might be a better signal processing way of handling the lag. Anyway, something to think about - although I'm sure it'll freak out sim users who don't anticipate this lag in their FBW aircraft.

Great job RAZBAM. Just a little note, you may want to use the magic words "Angular Momentum" when talking with CptSmiley, since "Inertia" is resistance to attitude changes (that is, it is the angular equivalent to mass in linear kinematics). The Mirage 2000 has a wing area of 41 square meters. Compare that to the 27.87 meters for the F-16. Since all aerodynamic torques and linear accelerations multiply the dynamic pressure by wing area (and with the torques further multiplied by either span or mean chord) that means the forces are around 47% greater for the Mirage 2000 than the F-16. Given the shorter wingspan of the Mirage to the F-16 (9.13 vs 9.96 m) the roll intertias may be comparable (although I haven't checked). With greater torques and similar roll inertia it means the Mirage 2000 is likely to have a faster roll response rate (faster to start rolling, and faster to slow down) than the F-16. Looking at the F-16 FBW models in NASA TP-1538 (where I'm modelling the F-16 for a different simulation) the FBW is not magic. Its purpose is mostly to moderate down pilot input so that they cannot command departures. A second function is to rotate around the stability axes so that angle-of-attack is preserved when rolling (preventing angle-of-attack being converted into sideslip in a simple aileron roll). Anyway, I don't know much about the Mirage, but just thought I'd throw my two centimes worth in. It looks like CptSmiley has done a smashing job, and a few tweaks (in the corners of the control lag simulation) will give us as good as unclassified civilians could hope for.

I met Les Munro at his home in Tauranga, New Zealand, as part of a project for veterans. He was a great guy. Thanks for creating this thread AussieFX.

SkateZilla and JoeKurr are correct. The lighter part of the sky is caused by scattering of light from the sun (physically this is known as Rayleigh and Mie scattering - from atmospheric molecules and dust particles, respectively). The dark streak is where light from the sun is blocked by a cloud or terrain feature. This means that light is blocked before it can be scattered toward you. Hence there are dark streaks. The term for the brighter shafts on either side is 'crepuscular rays' because they are most often seen when the sun is low in the morning and evening ('crepuscular' refers to dawn and twilight, and applies to animals that are active at that time, as well as light effects; Christopher Hitchens also described himself as crepuscular when he was sick with cancer). Such light rays are an important effect for realistic computer graphics. Implementations such as those of Bruneton and Neyret's "Precomputed Atmospheric Scattering" also produces such effects: http://hal.archives-ouvertes.fr/inria-00288758 See: http://proland.imag.fr/ Silviu Andrei also implemented this, and it looks fantastic: ${1} I've included the links in case the Eagle Dynamics team want to implement this in EDGE (although they probably already have). I'm certainly working on such effects in my own simulator.

Ah, environment mapping. Looks great.

Why build when you can buy? Check out TurboSquid, and you can often pay for royalty-free licenses for external models, which you then need to convert into ED's proprietary format. There is still lots of work to do, but you get a head-start at least. http://www.turbosquid.com/