dtmicro

This was already addressed earlier, saying this would cost 300 million is absolutely ludicrous. It's going to be based off a terrain generator to pre-gen the world and the vegetation can be purchased (there is so much vegetation out there already for licensing, you don't need to make it). GTA V was 125 million in development and 125 million in marketing, go look at the Wiki. We are not trying to build GTA V, first it has tons of missions, 100+ car models, even more planes than we'd even need to do. It also has flying and driving, crazy physics, tons of guns, explosions. This project isn't even 10% of what GTA V is (ok well almost 10%, since 10% would be 12.5 mil). It was hand rendered and there are hundreds, if not thousands, of custom people animations doing all kinds of crazy things. There is also what I call the bureaucratic principle, where a company half the size can usually work 50% faster, not twice as fast, but 50% faster. Meaning sometimes a higher skilled better picked dev team of 25 can do the work of 40+ or more, especially with less bureaucratic obstacles. There are plenty of things that have been developed with a budget around 10 million that did NOT even use a terrain generator, where they custom rendered the game. As a matter of fact, the average major game title only had around 25-50 million budgeted for the development side, as documented in the wiki, and most of those games were much more complex than doing a basic flight sim, so those 50+ million budgets are a rarity. Things other games use, that a Flight Sim does not need: 1) Major voice actors in games, in a SIM you may need a couple guys for ATC and what not, but nothing major 2) In regular games, you need a lot more complex sound effects and body animations, you need body suits and actors to do the animations. 3) More complex and ordered missions 4) More explosion variety, a hand rendered world that has to look very good both close up and far away (the SIM will look good close up, but it doesn't need to look GTA V good close up, except maybe on the hangars and runway areas) 5) Sometimes a custom musical score, and often lots of music licensing (with a Sim maybe license one thing, if that) 6) More custom renders because you are providing an "in-your-face" detailed world, we aren't going to rendering boxes of cheerios sitting on Grandma's table like GTA V had to do. 7) Clothing outfits, heck GTA V probably spent 5 million just rendering the clothes. 8) In regular games, there is a lot of pathing AI and enemy bot AI type stuff 9) It goes on and on... All the supposed complexities that people bring up for a Flight Sim are often simplicities compared to regular game development, especially since there are already people from the former FSX community that can be contracted to greatly expedite the product. All the supposed impossible data to get, most of it is freely available for download from the GIS data sites. So still don't see much validity to the arguments that a Flight Sim would cost more than a PC game, non-sense, not if you manage it correctly. Unlike a PC game, we have the resources already available, they just have to be managed correctly.

That's putting the cart before the horse, even though I know some crowd funding games picked the engine right off the bat (I don't think this is a case where you want to do that). There should be a fairly in-depth game-engine evaluation phase that lasts at least 1-3 months, and that means it might be better to come after funding really. I am a strong believer in not tying yourself to the wrong game-engine until you have done tons of research on the one that fits best. That is why I don't see the point in even looking at the game engines yet, someone needs to study their API's first and then do some independent performance testing on how well it handles the mesh fly-overs. The next step is a design document and some animation prototypes showing vividly wild plane crashes in mountains, now this I can actually do myself using Terragen and finding some explosion particle animation generator. The problem is I have a regular job right now as well as I am invested in a side FSX add-on business, so not sure where I'd get the time, maybe in a couple months. Not a bad idea as far as getting them involved, but I think the team would need to be picked very carefully. Picking the team is make or break in this stuff. I would usually prefer to contract out straight to people that we already know can do the best job, hence Rex / Active Sky / etc..., Accusim... I know those guys aren't making tons and tons of money or anything, I can almost guarantee you they'd take a paid dev contract to convert their software to a new platform regardless if they thought the platform would succeed or not. IMO, Rex cloud textures are already good enough, the only thing I would request they adjust is reducing the variety of clouds they offer and getting rid some of the campy cumulus that exist in some of their lists (they have mostly good stuff, but some weird looking stuff as well). Also the clouds can at times look fake, but in reality even REAL clouds outside sometimes look fake, so it's a catch 22.

I know, that's why I mean if a new flight sim ever got funded with real money. The problem I have with Flight Gear is it looks like an FSX-clone (graphic wise but I never installed it). If I were to walk into a Flight Gear dev meeting *(assuming they have meetings, chuckle chuckle)*, then my first statement would be, what the HECK were you guys thinking? Another FSX clone that isn't as good as FSX, and the point of this is? Let's start over, we need a terra-generator and a good game engine, graphics first. This code has to go.

In real life, those planes make big explosions, especially the jet fueled ones. There are always going to be Pro's and Con's to any big project, it's never easy. Funding is always hard to come by, no matter what you are doing. There is usually someone willing to finance something that looks risky and is counter-intuitive to what people think will succeed. There are plenty of rich people that like to fly, as a matter of fact, most people that fly their own planes also happen to be rich. If you can really make the game go beyond what we have today as an actual training resource to make flying safer, then there is a chance it will get funded. You just have to prove the concept that you can perpetually recreate a higher level of simulation with better physics that is much closer to a commercial simulator. You have to have the right people directing the project, and you don't want to make another FSX. If you can make the graphics look real and get the physics close enough, I think you can sell the idea to GA pilots that it is much safer if they practice the exact flight in the game before they do it in real life. I think there is likely at least one rich bored pilot that might want to throw money at the idea. Anyhow, this is starting to go in circles. There are a million reasons NOT to do something, and only ONE to do it. The big advantage for building a next-gen flight sim is that there is an incredible amount of freely available intellectual material from all the previous FSX developers already out there, and this should and can be used to enhance it. The biggest hurdle is forming a team and getting the funding, that I agree. It needs someone to write a basic design document (not technical specs, just design doc), and then some type of prototype demo showing something interesting happening (like a major crash). And BTW, if somehow someone can get financing for this thing, then count me in, hopefully after all these posts my resume will be moved to the top of the pile I'd love to help program a flight simulator, it's been my dream, but I know I cannot do it alone. I thought about applying at LM or Xplane, but there were no jobs hardly.

I still disagree, people keep saying the SIM Market is this or that, but no-one has ever built anything better yet. Of course the market is going to fade after 8+ years of the same simulator. Everything out there looks like FSX to the average joe, why buy another one? So how do we know exactly what the market is for a NEXT GEN Sim, we don't. As I noted just now, you do need to add a few extraordinary features to the simulator to appeal to a wider audience, and good marketing. I believe the re-birth of the Flight SIM not to be about how big the market is, but about creating the ultimate extensibility API so that people can have a LOT of fun just placing objects, making their own islands, and crashing into a farmer's field and watching the entire island burn. Create a Youtube video of it, it'll be like Minecraft with better graphics To the rest of us, it'll just be a flight sim.

Actually I see his point about making it a world sim, and I see your point as well about not doing it. Here is the thing though, you gotta pick something as the KEY focus or else you are just building another version of "The Sims meets Train Simulator meets whatever". I wouldn't pitch it as a world simulator at first, I would simply pitch it as a flying simulator. If you are going to make a VR-like 3D multi-player SIM world where people walk around and talk, well that has already been done about a gazillion times, and you can go online right now and play. So I think you have to establish the game as something before you add that kind of functionality. I prefer sticking to the flight sim idea, and letting the add-on developers take it as far as they want to take it. Since flying is the hardest thing to simulate, it's very simple to add the other stuff later to expand it. Then you can actually drive to the airport in a GTA-Style rampage, blow up a couple objects on the runway, hijack the plane and throw the pilot out, and then have a normal flight. Also you can walk around in third person in the plane and parachute out. See all those ideas are fine, got no problem with them, but I'd start with just making a basic simulator first, the rest can always be added later. That is actually how many low to mid-budget games work these days anyhow, rarely do they Finish the game right from the first release. Look at Elite Dangerous and Star Citizen, heck I bought ED and paid $80 for it, played it for 1 hour and then decided, what is the point. Yet there are plenty of people that love it, and I'm sure to them it's the best thing in the world. The point is though they sold the game WAY WAY before it was finished. The key to being able to do that, is to create your own market. Who cares if there is no market for a flight sim, make the market by showing something off extraordinary, all you need is one extraordinary graphical or event thing to make your own market. But even that said, I don't believe there is NO market for flight simming, I just think the market doesn't exist to create another clone of FSX or Xplane...

People always say that, they said the same thing about Space Simulators all throughout the late early 2000's, no-one wanted them. Then Star Citizen becomes the highest Crowd funded, bla bla... Common sense tells me there should be at least 50% of the interest of the original FSX left, if you build a better simulator. Now take that common sense number, and remember that it should cost less than half the amount it cost Microsoft to develop FSX, because of all the better data and tools out there today. Flight Simulators have an advantage over traditional games, traditional games have a VERY short sales life, most of the sales happen in the first 6 months, except for the absolute most popular games like WOW or whatever. MS FSX has been selling for almost 10 years. People have to remember Microsoft is a multi-billion dollar company, they really didn't want to micro-manage a lot of their game divisions, it wasn't just MS Flight that got canned, but they re-did a bunch of stuff (I forget what, but I remember reading about it). That's how big companies are, unless something is making a 20+ million per year, they lose interest and don't want the headache. The reason no-one wants to move to another SIM is because there is nothing out there that beats our current SIMs by a large enough margin. We know why MS Flight failed, it wasn't a sim, it was a FLY-OVER of Hawaii and Alaska with arcade like plane controls. Then they were charging too much and there wasn't enough content since it HAD NO Add-on database. Also, I do agree that you likely need to add some separate functionality to garner more market potential, but that can be done by adding some side games and war functionality add-on capabilities to the SIM.

Mesh database? That's part of the game engine. There is no low-level mesh work left in today's modern game engines (especially when using a terrain generator and automated GIS data), that is one of the core functions of the game engine is to automatically manage the mesh. The GIS data you use is basically self-documenting, since you can identify the rendered areas by coordinates or regions, since the GIS data already has that functionality. Believable buildings? Buy them and MIP MAP or pay someone to create a generic set, heck the game engine will MIP MAP them for you most likely and off-render them, depends on the game engine. I'm not sure I'd say you need to have all the buildings accurately rendered for the FIRST version, our current sims don't even come to 100th of that level. Otherwise make a deal with Google or the same people that Google used, and buy them from Google. Navigation Databases? Those databases exist already, and you can buy them, or you can just use the data in the public domain to create one yourself. I'm not sure how complete the public domain data is. There is so much free LIDAR data, you can probably just grab it from an earth scan. Airport Database? Same as above, you can just buy one. There are several companies that make flight products that will sell you their database, since you will not be their competitor. Furthermore, this can be created very quickly and automatically from maps that can be purchased anyhow. Designer with Pilot Experience? You obviously have to bring some designers and consultants with real-world flight experience, we do have 10 million here, that's not peanuts The other stuff you mentioned? Building a GPS in the game that works like real-life is not all that hard, you are just overlaying the mesh back into a smaller zoomed out window, then adding the airports, it uses the same coordinate and navigation as the game engine. This is probably one of the easiest things to make of all the things you mentioned. The physics? Well yes, that is one of the hard parts as I mentioned before. Get Accusim or Carenado (or any similar companies) to build something for you, pay them and use them as consultants, give them a partial exclusivity contract that expires in X time. Real Weather and Cloud Systems? I would contact Rex or one of the other vendors and try to contract it out. I'm not going to bother to try to hire a team to do this, since I am pretty sure it can be contracted by giving any of the weather/cloud vendors an exclusivity contract for 1-year.

Nope, just old-school FSX. I've been meaning to install P3D. Just don't feel like messing with re-installing all that junk right now.

Here is my point: Keep in mind the budget for Crysis 3 was only 65 million, even though some games like GTA V were 250 million (but more than half that was marketing). You don't have to create nearly the amount of complex models for a Flight Sim as something like Crysis 3, where they had to create super detailed textures and models of almost everything they were doing. They also probably used body-animation suits in those games and produce tons of animations, all of which is unnecessary for a flight sim. You only need 3 planes to start with, and a terra-generator, very basic animations, the control system, and then the physics. There isn't a whole lot moving around in a Flight Sim, it's mostly just the plane moving. You can also reduce the amount of budget required by buying pre-built vegetation and just converting it, or using whatever the terra-generator has in it. So yah I do think it's about 1/5th the work of the making of Crysis, and therefore that's how I estimate 10 million.

There is no method to do an exact estimate for this type of thing (especially in a hurry), but you can get a ballpark estimate by spending about 3 months collecting and reviewing documentation on how the pieces work and then create a basic specification that breaks down each required task into an estimate of hours. Finally adding it all up and coming up with a number, then add 50% to that number, because almost everyone underestimates. Another method is to find a company or someone that has already done something similar to what you are trying to do, and find out how long it took them in man hours. I do not consider MS FSX a very similar task to what would be required for a next-gen Flight Simulator. The reason is because in FSX they were using very old tools, and very difficult technology, and mainly coding it almost purely in low-level C++. Also they had to build an engine themselves, and all the coordinate stuff. Since 2004, this stuff has evolved greatly, the Geo-Tools are 100x simpler to use, and the coordinate stuff is much much easier to translate into measurable units, as the newer game engines actually do that for you! In most of the engines of today, they already expose basic physics you can use for flying, you just have to tweak the physics. It is still very hard and time consuming, but compared to 10 years ago it takes much less time. Combine that with auto-generating the scenery with a Terra-Generator and you can build something much faster than anyone has ever done before. 10 million is a good generic estimate IMO, you can definitely produce something in that amount of money, will it be complete, not entirely but it will be close enough. I've seen what others have made with 10 million, so I have a general idea of what can be done in that amount of money. As far as the other guy's comments about it would be like XPlane since it would not be finished, I disagree simply because people go nuts over graphics and eye candy, so if you have the next-level graphics and decent physics already coded in, and an easy way for designers to build add-ons, then yes I do believe everyone would eventually switch and leave their other sims. The reason Flight didn't go anywhere was simply because we could not create add-ons, I mean that was obvious and really ignorant of Microsoft to try to do it that way.

Flight had better physics than FSX IMO, it actually had a great physics engine. That's not to say it was accurate, I just liked the way it felt. Sounds like you know a lot more about how the different planes feel than I do. I've been in semi-small planes and to me it surely did not feel like the plane was standing still, there was a lot of sensation of movement, much more than a large commercial airliner generally gives (unless you hit turbulence or something). What I meant though was mainly in other situations besides level flight, such as hard turns with moderate cross-winds, or diving too fast or climbing too fast. Flight felt more realistic than FSX. Also the controls are too sensitive in certain planes in FSX, though I suppose it's partly an issue with my control settings (I did try to adjust without luck). I also notice planes tending to veer left too much on the runways (especially some Carenados), things like that. I think with Accusim you get pretty neat realistic effects, though they don't really match the physics either. Even when I do a perfectly smooth landing Accusim seems to make too much of a sound still, but hey those are minor things I suppose.

I'm not an Orbx fan-boy by any stretch, but they really outdid themselves on this one. Just bought this airport, even though I don't own FSX Global it works fine. I saw the screenshots and then was like, ok I gotta check this one out. Looks great, but that said, they do need to fix the water near the main PAGO PAGO airport (NTSU - that's a bit much!), but I'll forgive it for the incredible trees and mountain scenery they did.

Nah, they all worked too independently and the market was too flooded. Orbx quality could have been done right from the release of FSX, it's just the right people were not formed into a team until Orbx themselves created that team. For such a small market, it is ridiculous how many add-ons there are. You could spend your entire life just going through all of them. FSX was like an entry point for new or wanna-be game programmers so they could get their feet wet. A lot of FSX designers (Bless their hearts) had very little experience in graphics or coding and were just winging it. Except for the plane add-ons (which I think there was a lot of talent in), the scenery and most of the airport designers were a bunch of people releasing sub-par stuff trying to sell it because it was FAST and easy to do. Had they spent real-time on their products (like Carenado and Orbx), then they would have done better. Think how slowly Orbx and Carenado have been to add products, it's been years to even get this much stuff out of them, and even a lot of their products fall short of their own quality control. A lot of them were disorganized that didn't have the right skills. Some had great programming skills but were terrible at graphics, some were great at graphics but were bad programmers. IMO, most of them were just bad at graphics...

The first thing that has to be realized is that the current state of technology is FAR FAR advanced past MS FSX and even MS Flight. FSX was finished development in 2006 (or thereabouts). That means it was using technology available mostly from 2002-2004, that's 12 years old folks. MS flight was using tech available in 2008-2010 or something like that. Software is almost always 2 years behind on tech from its release date, since they don't keep redeveloping the engine every time something else comes out. The problem with the current SIM is repetition and so-so quality PhotoReal textures, as good as Orbx is, it gets pretty repetitive REALLY fast. I tried Southern Ireland and thought, well yah this is some really nice work, but why did you spray the same 20 tiles across the entire area (except for the mountains which are different). A lot of these add-on developers were making mediocre products and they went out of business, and they were first to claim that "Flight Simulation or FSX is dead". What happened was ORBX, Carenado, and a few others came along and took a BIG piece of PIE, and then everyone was left with a smaller piece. Orbx is probably doing fairly well. That's the problem right now, even Southern Alaska (which I think is Orbx best region) is VERY VERY repetitive. It's the same textures across most of the state. You wouldn't have that problem as much with a terrain generator, take a look at some pics over at http://planetside.co.uk/ to get an idea of the variety you can achieve just by varying a few parameters. The real question is can any of the game engines handle any of these terrain generators output, and the answer is yes WITH modifications it can. We are at that point now, but just barely we are there, but it can be done. The whole point is to dump the boring FSX physics model which feels more like you are flying in a car and make it feel like a real plane. Microsoft would say the market isn't there, they may be right, after all they tried to sell MS Flight, but realistically and IMO, they tried to release a new SIM too quickly after FSX (they should have waited 5 more years), and they were still using outdated tech, then they didn't even bother completing it, and tried to control add-on market. Big corporations are often too impatient, any Flight Sim product needs a long-term view of like 5-10 years to make money, not in 1-2 years like Microsoft was trying to do...

nuitkati, on 06 Jun 2015 - 12:06 AM, said: Every new sim development will have to stand toe to toe with a full fledged sim environment with more bells and whistles that even a 20 mil budget will get you, so success for a bare bones systems seems doubtful. At least that is the expectation I glean from the posts here. ---------------------------------------------------------------------------- As an FSX developer, and someone that has experience doing estimates, I can say that you are potentially correct, but the problem with your estimate is that it assumes the bells and whistles need to be coded by the same company that creates the SIM. Then yah, it would cost exponentially a ridiculous amount of money and time, that is like saying FSX should have been released with all ORBX add-ons for no extra charge. The whole point of the discussion was to use a Terra-Engine to circumvent the biggest part of the budget, the scenery design, and then expose that engine to add-on developers (or make a special deal with a company like Terragen to give add-on devs a greatly reduced licensing price). I still put the estimate at around 5 to 10 million though, because of what I stated earlier with the complexity of the extensible add-on model and physics, and graphics take time regardless if you are auto-rendering or not. Also, I do agree that somewhat randomized terra-engine like rendering is the future of SOME forms of gaming, but of course in this case we don't even need randomized, the renders can be based off the GIS data. The goal with a new sim should be to simply build out 1 smaller area of the US (maybe one state) as an example, kind of like MS Flight did, and then to allow the add-on developers to finish everything (the opposite of what MS Flight did). Even with generated scenery, you still are going to need some custom add-ons (airport buildings, sound effects, seasons, color variations, texture variations, etc...). So all the sim really needs to do is have better flight model physics, better base scenery, and a better framework for other developers to create add-ons. If it can accomplish those 3 things, then I do think it would be a success even with a limited area completed. For instance, as a hard example, let's say it needs an ATC interface (which I believe Flight did not even have), and we want to license that to a former FSX add-on developer, hence a convert that moved over to the new cutting-edge sim. All you actually need to expose is an object that contains all the needed data... Then an add-on developer can create the sounds, and the pop-up messages. Really an ATC should be done in the base game anyhow, but at least make the ATC easily extendable. I could picture an add-on developer adding voice recognition as an example. Same concept, and I say 5-10 million is VERY possible if you can find a Terrain generator over the Earth's mesh. Let's say you had 25 developers being paid on average $80k each (maybe some slightly less, some more), that's 2 million per year, so it will take about 3-5 years. That's where I get the estimate from. The one advantage is, if you started developing it today, you can develop it for only the absolute fastest CPU setup (6 core overclocked 5820k), and by the time you release it, the mid-range CPU's should be as fast as the CPU you designed it for. The truth is that many game engines (with literally no coding at all) could have mesh and texture map loaded, then you could SLEW around a mesh with much higher res textures than FSX with very little performance issues, that is if done correctly.

The GIS work is the easy part, as most of this work is easily done by many people that have experience in writing FSX add-ons. The time consuming part is the extensibility API for add-on developers, the physics, animation, and graphics. The mesh is also fairly easy (time consuming to find the optimized method, but still relatively easy), the game engines these days provide you with about a gazillion different ways of loading mesh already, a bit of GeoTools / FS Tools / Warps and some coordinate math and you can load whatever directly into a game engine. The harder and more time-consuiming part would be rendering over the mesh using the third-party tools. A flight sim is just a "box" (plane) moving over a mesh that has rendered textures on it, that is all it is. The game engine does all that for you, you sure are not going to write that (what a waste of time that would be). I personally would recommend the first attempt be done with a Terragen like application by loading the mesh directly into Terragen and rendering over it. Then trying to load that into a game engine, try a few different game engines and see which ones work best for flying over a rendered texture base on top of a mesh. The problem with this method is going to be POLY Counts of the objects and the size of the textures, so you'd have to experiment with many different Terragen like Apps (outerra whatever) until you find one that is easiest to maintain the highest quality looking textures over the mesh and objects with the lowest poly counts. That is where the REAL problem is. The trick is all in the way you MIP MAP the objects and gaussian blur the textures and background, one reason FSX is so slow at loading photo-real or rendered graphics is because the MIP MAPs are too large at a distance. That's why mountains and objects in games like GTA 5 look better closer up and worse farther away, whereas FSX is the opposite (generally speaking). Also, the lighting work is a major pain to simulate real time of day, as well as the volumetric cloud renders. Also, things like ATC are very simple once the core game is already done. ATC is only hard to extend in FSX because of the way FSX was coded.

Very very little C++ Coding is required these days, that is for coding a game engine or a huge blockbuster title with 50+ million. Interfacing with a pre-existing game engine usually does not require C++. Most of the engines expose a scripting language, in the case of Unity they have a couple options, one being a proprietary Javascript like API, and some others. There would likely be a few modules that would need to be moved over to C++ for optimizations on some floating point and some custom Poly stuff the game engine wasn't handling (only cause flight sim needs some special coding), but overall you could actually 100% write the game in C # as a wrapper over a game engine, which is absolutely the way I would recommend doing it. Then you later convert bits and pieces from C# to C++ where mass optimization is needed only if you hit a road block on performance in some area. Even when they build their own engine, most of the core game was still usually created by scripting the engine with whatever scripting language the engine exposes, not inside the engine itself in C++. The days of needing to use C++ for every game is over, unless you want to code your own engine or are trying to port to multiple systems. One reason C# is not used in more games is because it's nearly impossible to port it to a PS 4 or Xbox, Apple, whatever without rewriting it. Whereas with C++ they have a lot of helper libraries and SDK's to aid in the process (though it's still hard). Even blockbuster games that do not use their own engine are rarely done purely in C++ (some of it is). Sure, when they build their own engine or when they change the underlying game engine code, or if they just want it as optimized as possible because of the user base they are targeting. Some games are done in Objective C (Apple), regular C (instead of C++), Java (Android), even Python scripting. Not to say that a lot of times you do not need to add tid-bits of C++ code still for optimizing, you do, but most of the time the CORE work to wrap a game engine is going to be in the scripting of the engine, not the coding of the game. The game would be coded in C #, not C++. The DirectX12 capabilities in any modern game engine are going to handle all of the actual graphics work so that the actual coding is simply going to be interfacing with the engine and then writing extensible add-ons to handle the math. My experience: 2 family members in the gaming industry, software dev for more years than I'd like to admit, etc.. etc... (bla bla bla) Also, if the budget were only 2.5 mil, you're going to produce nothing more than a skeleton game unless someone is just exceptional at managing open source programmers that are contributing for free. The game would need a very experienced high-dollar architect to manage the quality control if using open source coders to interface with, whomever were to take this role would need to understand how to lock the design and be great at version control documentation as to keep everything in order. The budget for a basic flight sim that would be NEXT GEN, needs to be around 5 - 10 million, and that is if the game were 95% developed in C# with only a few C++ interfaces for optimizations added, and then you can leave the rest of the coding to the add-on developers.

Just Cause 3 should be interesting on how they did the dynamics of the engine, it looks cutting edge on flying. http://www.eurogamer.net/articles/2015-04-28-heres-our-first-look-at-just-cause-3-in-action

Either solution works, just have to watch the memory height on the Noctua. The Corsair's have high failure rates and you have to replace the stock fans in order to achieve same temps as Noctua. That said, most people will not get a failure, but it can happen, on the Noctua it's almost unheard of.

1) CPU Speed 2) Video Card 3) Hard Drive and SSD's (though really only helps for load times) 99999) Everything else doesn't matter NEAR as much Either wait for the next gen CPU to come out, or go with the 5820k instead and overclock it to 4+ GHZ over the i7-4790k. I own the i7-4790k and regret it after I realize all the other crap in the background running always using some percentage of half the cores (meaning things interfering with FSX CPU availability). I really wish I had 2 more cores (6 instead of 4), and besides the 5820k can be stablily overclocked to almost the same speed per core as the 4790k. I'd rather have 6 cores running at 4.2 - 4.5GHZ than 4 cores running at 4.5GHZ to 4.6GHZ. I bought the 4790k because of the stock 4GHZ speed vs. the 5820k @ 3.3ghz stock, this kind of screams at you like wow the 5820k is only 3.3ghz, not even a contender, 6 cores or not. Yah well after a little research, basically it turns out Intel was WAY WAY too quick to set the default multiplier on the 4790k as a 4GHZ, as with stock cooling it basically will melt the CPU at 100% usage in many cases (depending on room temp and all that). The 5820k is basically a 3.8GHZ to 3.9GHZ CPU at stock cooling if the 4790k is 4GHZ, because the 5820k was a server cpu they rated it way lower than they should have. Both these CPU's easily overheat, so aftermarket cooling is an absolute necessity. Don't waste your money on anything but the CPU and Video card, everything else is WAY WAY down the list. Furthermore, it's just a better CPU with a better dual bus architecture, so that when doing file copies your Win machine across multiple drives, it won't be as slow. The speed of your memory will have virtually no effect on FSX speed with any modern memory. Buy the cheapest memory that the motherboard and chipset supports.

Too much encapsulation is likely, not a refined enough API for development. Anytime you try to use something pre-built engine-wise in software that hasn't been used in a lot of projects (and thereby NO REFINED API), you will always run into limitations. The problem is these limitations are sometimes not found until you are 50% or more into a project, and they can be absolute deal breakers. I'd think it'd be easier to use any of the other more established game engines, because they don't overly encapsulate anything, as most of the things that are encapsulated are by necessity rather than by lack of a refined API. That is just a guess, I don't know and I really couldn't know, the problem in software is the same as in statistics, it's NOT KNOWING what you DONT KNOW. That is what always kills a project, and unfortunately in too many cases that issue arises. If you at least know what you don't know, then you can be cautious of it, but that's easier said than done.

I've been in software dev for a very long time, here are my thoughts... Anything is possible given the right amount of motivation, luck, timing, and resources. The biggest problem to this idea is I think it puts the cart before the horse. If you really want to get people interested in the idea, then you need a smaller goal as a start. I would find something you can do within the flight software community to get a user base (whether that be reviews, blogs, add-ons, or preferably a combination of all), rather than going straight for the gusto of HEY Let's get a user-base going to throw around ideas of creating the next gen flight sim. Since you seem like a managerial type personality, maybe head over to a recruitment forum and review existing projects, and try to get people to work together on a smaller goal. Then after having finished this smaller project, use what you learned to build onto it and continue your symposium. Not saying to shut down your symposium, just saying try something else in addition if you can. The #1 most important recipe for success for this type of project (if it were to get off the ground), is to design the software in such a way that extensibility is very simple for anyone to code on it. Then various versions of things come into existence, and the one that is most successful becomes the standard and gets integrated into the core "official product". Then you can establish an add-on community with a centralized portal that can manage the flood of add-ons and group them by quality. The problem we have in FSX with all the add-ons, is some are just atrocious and just spam for the most part. It should have been managed better. That said, I'm not against this at all, nor am I for it. I'm neutral at the moment, but I think the approach is coming at a too steep of an angle. You had to start somewhere though, so I suppose doing something is better than nothing, but you may need to refine your approach. As far as an engine for rendering a Flight Sim World goes The hardest parts about a next gen flight sim (well there are several difficult parts), but I'd say overall the most complicated programming is the physics landing and turbulence model and weather engine, though some of that is assisted by various game engines to varying degrees, it's going to have to be HEAVILY modded by an expert in flight physics. The above is the most complex, but not necessarily the most time consuming part, the most time consuming part will be the graphics, the rendering, and dealing with 3D coordinate systems (always fun!)... I don't see how using Outerra would save all that much time, as I'm guessing it is too limited (it looks exceptional, true, but...)... You could just render the world using something like Terragen and Vue D'Esprit probably almost as easily by just having different add-on designers. There are really only about 20 different types of scenery that exist on the earth (excluding a few freak places here and there). Not certain these two products have a successful enough or automated enough way to compress the poly counts and MIP MAPs, but they've been out for so long I'm guessing they do.

Norway is one of the better areas, it's big though so it is hard to find the best looking spots. Found a few good ones though. It's a little disorienting since the names all sound funny Alaska might still have slightly better textures than Norway, but hard to say, Alaska was more consistent whereas Norway is a bit all over the place. Still it's a good one to buy. Going to try Southern Ireland, one of the last 2 areas I have not yet tried (haven't done that one or England), but have all the other ones. Scotland and Alaska regions have the best textures I think...

My monitor is pretty close to Rec709, but even a bit over saturated or too high luminance compared to a perfect D65, and the greens are still muted other than maybe Alaska. I also had to adjust green on a projector and 2 other devices, so not just my monitor. In reality, green is very bright in the mountains almost anywhere in the summer in the US after the monsoons other than the driest desert areas. Especially the mountains of NCAL, Orbx representation of NCAL looks more like the pine forest of Arizona than California, the Redwood forest area and the Northern Coasts are much more of a bright green than what Orbx shows.