War Thunder - Anything to be learnt?

I was playing War Thunder the other day and it occurred to me just how well it has been made. Sure, it has flaws and difficulties, but name a game that doesn’t!

The thing that made me link it to Infinity was the way the game seems to run smoothly on most systems, but still looks incredible. I suspect a lot of it is down to clever computer trickery, but it’s mighty effective. The one thing that I would say sets this game apart would be the atmospheric lighting. Sure, some of the maps are nice when you’re zooming along at ground level trying to avoid trees and bullets, but War Thunder scores on the fantastic lighting.
Plus I love shooting the wings off Bf 109s with my Spitfire(s). :smile:

Take note, I-Novae perhaps?

Just my random thought for the day anyway. Anyone else played it?

Good point. I wonder if any of the devs have played it.

I stopped playing it after I got fed up with it’s random damage model, but as you say, it looks good even on my GTX 275 7 QX6800 and the gameplay is good.

I think it looks good because the game is very good at tricking you in your peripheral vision. The backgrounds, namely the realistic contrast that the trees and buildings have comparing to the ground tricks the brain into thinking “everything looks good” and you don’t notice many graphical anomalies such as pop-in and LoD transitions. Doing that right is probably very important, the believable backgrounds and horisons as many developers chose to not give as much attention to those little details.

Basically it’s all about tricking you into thinking that the graphics look good when they don’t by focusing the performance on the areas you will most likely look at! I’m probably overcomplicating it :stuck_out_tongue:

Yeah, I think you’re probably right. While a game will obviously not succeed without good gameplay, making it look pretty definitely helps sell it and keep people playing. I have been impressed with the screenshots we have seen so far of the I-Novae engine, but it will be interesting to see what it’s like actually flying across those environments.

As for War Thunder, I’m curious to see what they do with their engine for the ground level combat when they release the tanks part of the game!

I really like the way damage works in it. It makes such an unbelievably nice change for things not to rely on health, but which part of your plane gets hit. For example, an unlucky hit on your elevator makes your plane pretty impossible to fly, whereas if you’re a good pilot you can escape combat with multiple holes in your wings!

Would love to see this kind of modular damage replicated in I:B - what gets damaged affecting the performance of the thing you’re flying. I seem to remember in the ICP there were occasions where one of the fighters would get an engine blown off and go into an uncontrollable spin… (though I think that was partly preprogrammed).

Edit: readability.

I like the damage model. It feels just amazing to manage to fly those 3 Kilometers with no elevation control and burning and stuttering engines!

I was allready amazed by the studios grafics two years ago when I bought another aviation game form them. The clouds and the details at the ground are amazing.

I just hate Beusomethings blowing me out of the sky with one hit :angry: :wink: [size=10]except when they shower to the ground in pieces by the force of my rockets hehe [/size]

Haha yeah, Beaufighters are a nightmare. I personally find it difficult to hate them though since I’m English and they belong to us… I do seem to be pretty terrible at flying them, but get those cannons lined up and BOOM.

More than once, I’ve had new players play chicken with my Beau. It never ends well for them - except for one lucky fighter that rocketed my cockpit… :angry: [quote=“Lomsor, post:5, topic:62”]
except when they shower to the ground in pieces by the force of my rockets hehe
I blame you.

I far prefer the modular damage model over just a generic total health model. It just annoyed me that it doesn’t matter where you get hit in the wing, the damage your wing takes from that shot is a dice roll.
Also bullets could penetrate one component and damage something the other side without damaging whatever it passed through.

I did manage to win a chicken game against a Beau. I’m pretty sure I knocked the pilot out, though. (To my defence, my plane was so damaged that I couldn’t escape anyway)
Note that I tend to play Japanese, because more often than not kaminazing ground units is the best course of action in this game. Too bad I’m terrible with Japanese planes…

I generally like the damage system, but there are some frustrations to be knocked out of a perfectly healthy Beau by one stray bullet, or to just loose this little piece of metal that allow you to turn. While it may be a good inspiration for I:B, as they aren’t constrained by IRL as much, they could be cautious as to avoid those frustrating bits.

Agreed that they need to be careful if they go with a damage model similar to this - especially if in I:B it could have lasting effects as opposed to War Thunder’s quick instanced battles. But I do think the frustration actually makes the successes even brighter. It’s a tricky balance to find.

I have to say I don’t agree with being able to 1-shot a pilot, but put a few points into vitality and that problem largely goes away.

The most important thing I think @INovaeFlavien and the team could take note of would be the damage having an impact on how your craft performs. I really like that in War Thunder, you can take a lot of damage to one wing which makes your plane list over to one side. It’s still usually possible to manually correct and even attempt a safe landing, but it makes things more dangerous thanks to Physics.

Translate that to I:B and perhaps your port-side engine gets damaged which makes your ship constantly want to veer left until you can repair it… (as an example). It makes things more interesting, but also helps massively with balancing - even a rookie might be able to damage a more advanced ship enough to slow them down. You might start with an advantageous ship, but this way you still have to avoid being shot or you’re going to run into problems!

I’ve not played this game… but it sounds like maybe I should :grin:

Nope. It looks like the tech behind War Thunder has existed for quite some time (~2006). The developers (artists+programmers) just have an eye for how to put things together right. In my opinion, from the look of it now, I-Novae has already crossed over the line from Artistry to Feature Juggernaut in several aspects. I don’t want to start a flame war, but meh…

Too many modern games turn out mediocre after getting tastelessly bludgeoned by all sorts of industry standard, “must-have” techniques. These must-haves are viewed in such a pretentious way almost as if it’s impossible to complete a good game without them, despite the generic features being smeared on like duct tape. Ultimately, this cookie-cutter process subtracts from the real, unique and tasteful potential behind games’ concepts.

The single largest reason for delays in our release cycle has been challenges in building our planetary tech. We have a specific vision for the games we want to build and the emotions they’ll evoke and we haven’t been willing to compromise that vision. Fortunately we’re nearly done. The planetary tech is working really well, we resolved all of the issues we were having that we felt would’ve had a severe impact on player enjoyment, and we’re looking forward to shifting our focus over to building games =).


Haha, my work here is done. You’re absolutely right about it feeling ‘solid’ in terms of appearance though. Everything is tweaked just enough to give a smooth overall impression. And I don’t know if you’ve experienced a night battle yet, but seeing all those flashing rounds can be so pretty it tends to distract me from the imminent death coming my way.

The PBL system does incur some artist pain initially, but there seems to be an industry-wide paradigm shift to similar systems in which artists would have to become proficient with anyways. We are still in the throes of our art pipeline development/optimization. We are learning optimization techniques as we go, but so far the art teams experience has been quite positive in regards to our PBL system. We are starting to see the type of quality we’ve desired for many years.

but there seems to be an industry-wide paradigm shift to similar systems in which artists would have to become proficient with anyways

That’s what I was arguing against.

Well, we really like the results so far and it promises to be a more robust system than what we were using in the past.

K. I’ve been looking at Jan’s screenshots and I don’t like what I see.

For imaged based lighting you should have a depth channel of the environment map augmented relative to a fragments’ world coords for many reasons. I’m not sure if Keith has that covered yet, but it’s an important component needed to properly do most things I find wrong with the preview render screenshot.

I don’t see the correct amount of spectral radiance and irradiance. Since I’ve noticed this issue, I have been thinking about an interesting approach to irradiance, but I think it’d require a sampling technique that is difficult to explain. It’s not a solution to global illumination, but more of a shadow mapping method as close to perfect as it can be without accounting for multiple reflections. I’d be nice fun for me to experiment today :stuck_out_tongue:

By the way, I should mention, that was completely unrelated to what I was talking about before :blush: