winz

Imho it should be done by ED, they should also include all 3rd party models and other graphical assets in DCS World via patches, so you don't have to own a 3rd party jet just to connect to a server/play a mission where it's one of the flyables.

My master degree in software engineering has a clue and fully agrees with GGs opinion. Doing voice recognition would be a total waste of assets for ED. There are products that will allways do a better job, because their business is based solely on making speech recognition.

Satellite images are great when you fly high. When you get down they look terrible, they don't have the necessary resolution and look blurry, have flat residential areas, phantom roads/rivers, missaligment with 3D ground object, shadows based on the time of the day they were took.. etc, etc....

Most likely poor choice of words on my side. I didn't ment to imply that Lockon/DCS was unmoddable or ED was restrictive in modding. But afaik most of modders knowledge was based on reverse engineering EDs products and some areas were problematic (i.e. cockpit controls). Yes ED did provide basic tools for i.e. model export or some basic support for the enthusiastic modders, but it never got big and ED never provided some coherent API for 3rd party developes, nor did it seek to form a payware addon market (as you mention). That's what I ment with 'closed'. As you say, I have an option to choose the products I want to buy and those that I won't. And that's why I have no problem with 3rd party developers. I would be satisfied even If only one of those product is worth my money, because that's one more sim I wouldn't otherwise had. Tbh, I think more then one good product will come from 3rd parties ;)

I don't care, as long it's not ED lowering the bar on it's own products. Everything else is just a bonus opposed to the situation where ED would not open its platform for 3rd party developers. If it's good, I'll buy it... If it's not, I'll just move along. Having a choice is never a bad thing.

The article is only stating, that it can handle - read it will not CTD misserably trying to do so. But there is not a word about how effectively can the engine handle such big map. AFAIK the cryengine uses precompiled AI paths in navigation, so that might be one hiccup.

But DC doesn't define who you are facing. In bare bones, DC is just a persistent combat theatre with dynamicaly generated tasking. Your opponents could very well be human players (PvP servers). DC and CA are not mutually exclusive, in fact they could very well coexist without problems. :) I in fact think that ground-air war integrated high fidelity sim in a dynamic persistent theatre is a simmers wet dream :D Sure that MP will allways feel better than SP. But that has nothing to do with DC and/or CA ;)

@WildFire There is one thing preventing them from charging $ - BMS is based on a leaked source code, and they have no legal right do distribute, or sell it. That's the reason BMS still requires an original Falcon exe. Tbh, we don't know what parts of the original code were improved and what parts just had to be scrapped and reworked from scratch. i.e. the FM in BMS feels so much different from original F 4.0 that it's hard to believe it's based on the original Falcon code. Sure, there are parts that are just updated, but that's true for EDs product aswell, i.e. the graphical engine. Considering the amount of updates done to BMS and the level of robustness (stability) achieved it's fair to assume that their knowledge of Falcon source code is exceptional. They are in different position, than the standard modder groups i.e. IRIS, who work around a 3rd party engine without any control over its features. BMS can implmenet features they require i.e. they redesigned the whole control scheme. They have control over the engine. By moving to DCS they would loose all this control. +they would be moving to a totally unfamiliar platform and would have to take all the baby steps all 3rd party developers are taking. And I don't see what would they gain...a newer graphical engine? DC, CA has nothing to with camaradines. You feel camaradines because you play MP, period. DC can work, and is working, in MP enviroment aswell, and would give you exactly the same sense of camaradine. And no, I haven't played it, but I don't see how it's valid. Basicaly CA is giving you the option to control groud troops + play as JTAC. It will not generate you tasks as guard tanker, escort bomber, capture field etc... Tbh those tasks are entirely possible with current system, so I cannot see how CA will affect/improve that. CA will allow players control the ground battle.

Does it? For me CA is not comparable with DC, because it improves a different aspec of sim. Combined arms adds a human factor to ground troops, that's it. It's a great addition for DCS world, especially for organized MP (i'm kinda curious how it will feel in SP, will be fun to experiment with, for sure). But it wont allow you any larger scale battles, then the ones you currently experience in DCS Missions. You're also tied to predefined scenarios, as you were before. Yes, they will be more dynamic and random thanks to the added human factor in ground troops. It will shine in MP the most. DC is great for persistence and replayability. Especialy for the SP folks (yes, we do exist). And no custom mission can achieve the level of unpredictability a DC can. Sure the mission aren't the most realistic, but lets be honest, 90% of the mission we have for DCS are not that realistic either.

If you really have a phd in computer science, you should know that there is more to an engine then graphics...especialy a simulation one, and especialy when CPU becames the bottleneck - lowering graphics doesn't have as such big effect as when GPU is the bottlenech. Also, you have no ideas about what more the DCS engine is doing compared to FC2 I.e. the handling of scenery is different, then in FC2. The graphical engine is improved, and even if it looks worse on low settings then FC2 on maxed settings (what a surprise), doesn't mean it doesn't use aditional rendering features that cost some FPS. The water was improved...etc..etc Also, if you really have a phd, you should know that your comparasion is very shallow and lacking serious research. And funny you mention textures in your other post, because on my screenshots the DCS ones are much more sharper, even in distance. BTW, talking about BF, do the same 'research' on BF. Fire up BF2, crank it up to max, take screenshot and note FPS, then load BF3, use the lowest possible settings, and compare screenshot and performance... I'm very eager to see the results :)

My take 1920x1080, windowed DCS - All max, except water on medium, 4xAA, FC2 i5 2500k GTX560ti 8GB Ram FC2 is locked at 60fps max. I just can't help to notice how dull the FC2 scenery is compared to DCS:W

Yes, they should develop a new engine from scratch. It should have the on-foot freedom of Arma III, the in-flight terrain fidelity of outerra, the scenery complexity of the FSX custom sceneries, it should keep the DCS level fidelity of fly and system modeling, should have the penetration and armor engine of steelbeast pro, adaptable AI, the boom effect of crytek3, dynamic campaign, full working ATC, should perform well on 4years old CPUs, and use all the modern hocus-pocus APIs available today, and shouldn't take 10years and ED going bankrupt to develop And maybe I should just replied 'not this kind of thread again' but naah.

E8400 is running at 3GHz. Your quad is running only at 2.5Ghz, while beeing basicaly the same core. And if it happens to be the Q8300 version you'll also have less L2 Cache. And as dcs is single-thread mostly, it cannot take benefits of the additional cores you have

It's the simulation of the airframe + systems your system cannot handle. Most of the things you've done affect mainly GPU, not CPU. There is no way to reduce simulation fidelity the same way you can reduce graphical fidelity. Your CPU either has the power to handle it or not. Your doesn't.

Why do you think specs are only tied to graphics? In-depth modeling of aerodynamic is very cpu intensive. Not to mention modeling the various systems and equipment + dozens of AI units (you know, things like pathfinding, are not-so trivial in a the huge enviroment of DCS). Just because you don't see all the things that tick under the hood doesn't mean they don't have their costs. The more fidelity you want, the more you have to pay in terms of required CPU power, there is no magic way around it. And sims (since the dawn of gaming) allways required all the CPU power they could get.

If I only had 30min - 1 hour of gameplay Diablo would be the last game I would play :) And what about other Blizzard games? SC2 is beeing patched and improved regulary. Diablo 2 received it last patch in october last year. If you look on most build guides for diablo 2 you'll notice that they utilize items only added in later patches. Blizzard will get a small % from every real money transaction, it has no point for them to spawn/sell the stuff themself when a)The revenues would be realy realy small compared to the amount of $ they got from regular transaction cut b)When people are more then eager to farm these things and spawning these things would only reduce the price of items and there is no guarantee that a buyer will buy the one auctionied by blizzard. c)The nerd rage would possible kill teh internets I just checked few builds for my beloved necro and all of them looks like this Strength: Enough to use your gear. This should be a fairly low number, as this build will utilize Enigma. Dexterity: Most players will want to keep this at base, as this build does not require max block. Hardcore players, however, may wish to increase this to buff their blocking, though it is not strictly necessary for survival. Vitality: All your spare points will be poured into Vitality. Necromancers gain 2 Life per point of Vitality. Energy: Despite the steep mana cost of Bone spells, this should be left at base. Most of the games used tree system because it was popularized by D2 and WoW. But if you look i.e. on GW2 it uses system similiar to D3. For me personaly GW2 is the biggest MMO release of this+past few years. I just feel that the new system is much better then the old. We will just have to agree to disagree. Yes, the need for using 3rd party generators/trainers/cheaters certainly makes the old system look better, and at this point you are at the same position as in D3, you can choose your build without any costs. And I don't agree with the allways online DRM. With this I agree, seeing a dagger wielding barbarian is really sad. Can you give me one expample when blizzard has stated several times that a feature is beeing implemented and wasn't it actualy added it into the game?

1)You say that gold farming is boring (yes it is), but never-ending boss runs are not boring? c'mon wipe the nostalgia out of the eyes, running the same boss over and over again, half asleep, is boring as hell. Blizzard wanted to address this issue by spreading the loot wide. I actualy found my first and only legendary in act I :D Yes, itemization has it's problems, but lets look a waaaaay back, on Diablo II 1.0, not Diablo II + LoD v 1.13d we have today. Blizzard has a history of supporting, ironing, improving their games for years. Itemization is one of the things that most likely will be ironed out. rly? This is pathetic 2) http://us.battle.net/d3/en/blog/6018173/Game_Design_Update-5_28_2012#blog 0.7% of people playing at lvl 60 are using the same build, that's pretty diverse if you ask me. Enlighten me, how many valid buidls and valid distribution of stats were there in D2? Weren't all the caster the same? -strenght, so you can carry the gear you are aiming for. -dexterity - 0! -energy - 0! -vitality - ALL! The biggest problem I had with D2 system is that it punished players too severely for not be familiar with game mechanics from lvl 1.. So, if you start a new char, you either a)Used a cookie-cutter public build b)Knew the game inside-out after playing it for few years, and done your homework on the build to make it valid. c)Had no idea what you were doing, wasted +30h gameplay time before you discover that your attributes distribution is not 'the best one' (I mean vitality for a caster, how intuitive is that...), your skills could be spent more effictiveli..etc. And you either stick with the build, or you replay it another time, discovering after another +30h that your build is again wrong... I love the new system, it allows me to experiment to a way never possible with D2 system. And when experimenting I'm actually playing the game and not doing a math homework on how many dmg will combination of skills + synergies + items do. 3)This kinda bothers me aswell, it looks weird. 4)I'm kinda happy that weapon stat finally have a value for a caster. It's not like droping a +40intellect dagger + offhand for a +str 2h sword does not sacrifice your damage. Yes dps is the main attribute for..dps (int does play a huge role in dmg aswell) but i.e. weapons speed plays a huge role arcane consumption rate. 5)PvP will come in a content patch

On one hand it looks spectacular and you can clearly see the amount of work it got into it, and the aim for realism. On the other hand, it still has the weird FSX feeling especially in the physics model fluency

An Russian An-30 has crashed during landing on Caslava Airport in Czech Republic. The airplane skid off runway and caught fire, 7 people injured, no fatalities. Photos : http://zpravy.idnes.cz/foto.aspx?r=krimi&c=A120523_120924_praha-zpravy_zep

I'm loving it, it feels like diablo, it plays like diablo, it sucks time just like diablo... and man it looks and sounds gorgeous. The only problem is that I'm too old for this...my body just can't handle sleep deprivation so well, as it could 12 years ago and sleeping in work is a lot more complicated than during classes on college. :D GF realized this is a force she cannot fight with, so I got a diablo t-shirt from her :D

Low contrast, finite pixel size, loss of details during rendering process, no real glare. These are all hw limitations that exists and make it impossible for an object to be spotable the same it is IRL. The fact that our eyes are much much more sensitive to moving object and we run our sims around 30fps, more times less (not smooth for a non-blurry image) doesn't help either. Washing the image even more with HDR doens't help either. :) The only thing developer can do is artificaly increase the rendered aircraft size (so it will be larger, then it should be at the given distance) to help player spot it.

Yes, look in your installation folder\Config\View\Labels.lua There you can set the color for side + information to be displayed for various classes (weapon/ship/aircraft/ground unit) and distances. Tips: Minimize the information provided by labels, turn them off on mission start, and only turn them on when you are really really desperate (i got a button on joystick keybinded to turn labels on/off), i.e. when the mission designer wants you to kill each and every unit on the battlefield.

That's a great find. What would be great if resources would be persistent between missions.

Several years old, riiight - http://www.readwriteweb.com/hack/2011/06/cpp-go-java-scala-performance-benchmark.php You cannot use hotspot JVM on a game, because you don't have the time to gather statistics on what part should be compiled/optimized and what parts are fine left interpreted. You have to compile all executed code, because interpretation is much slower and would lead to stutters and performance issues. And I have no reason to believe that static compiled code is more effective in Java then in C++. +JVM adds overhead no question about that. +Native <-> managed code execution is another overhead. You have another layer in which you have to convert everything from unmanaged low lvl to managed objects + transfer all function calls, are you really going to convice me that this doesn't decrease performance. C++ allows you to do low lvl stuff Java will not. No, it doesn't outweights. It's easier, faster, more effective to develop for java, but a skilled C++ developer will have no problem with multithreading. The main difficulties in C++ lies somewhere else then multithreading. Multithreading using libaries is a 2 line matter in C++, synchronization is a little more complicated then in Java, but still managable without problems. The main problem of C++ is that a programmer has to know its ins and outs before he can effective with it on a high level. Blowing your foot of in C++ is stupidly easy. And memory managament is a mess. If you put another layer between you and GPU, then you're going to use it less effectively. My believe that C++ is faster is based on my personal experience from working as developer for several years. Let's assume that my previous experience is all rubbish and let me ask one simple question - If Java (and managed languages in general) are so great for game development, why is nobody using them? In the end, it's much much much more effective to code in java then in C++, and developers would be stupid to not to use it if it would be really an option.

Well, that's the problem. You kinda have to use C++ for a high-performance title (that DCS is), because it's faster then Java (and .NET for that matter as well). A developer will trade all the nice things you get with Java for a few FPS and lower ram requirement. Java might be fine for a little home brew project, but for a big AAA title it's still a no-no. The only mainstream game written in a managed language that comes into my mind is Magicka and it's performance is miserable considering the basic game engine and not so complex graphic.