Friday 2010-08-20 Dev. update and Discussions

[mazex]

Originally Posted by kedrednael

Verry verry nice screens.

But I have a question..
Will there be thermal?
I heard clouds can move because of the wind, Will new ones be formed because of thermal, will other ones disolve again?
Avarage cumules clouds like the ones in the game have an avarage lifespwan of just 40 minutes I think.

motion of clouds
In il2 I can glide a spitfire with -1 m/s altitude, and thermal can have +3 m/s without mountains or factories so will we be able to rise on thermal while gliding in SOW BOB? At least above fires? pleasee.

,happy birthday Oleg.

Well, Oleg has to check the gliding capabilities of his WWII birds then. No way a Spitfire would have sink rate of -1m/s power off... Something like 5 m/s would feel more realistic. Anyway, that is in a straight glide and in a turn required to catch a thermal I guess a Spit would have >10 m/s sink rate

[philip.ed]

Regarding clouds...this'd be awesome:

http://www.youtube.com/watch?v=pLfHDul5XGw

[Blackdog_kt]

Originally Posted by philip.ed

Regarding clouds...this'd be awesome:

That's pretty much amazing to be honest.

As for the update per se, if the rest of the game is finished (FM/DM/systems, interface, sounds, AI, campaign/mission engine and editors), the eye-candy is sufficient for me that i would buy it and fly it on the spot.

Sure, there are things that can be corrected and even nitpicking has its merit sometimes, but even if there are some things that could be improved i wouldn't call this an "ugly" simulator visually by any stretch. Is it perfect visually? Well, nothing is. Is it visually appealing enough to enjoy it if the rest of the package is there? As far as i'm concerned, it definitely is and then some.

[ChrisDNT]

Originally Posted by philip.ed

Regarding clouds...this'd be awesome:

I would gladly pay BoB 10-15 Euros more, for this technology implemented in the game.

[kedrednael]

Originally Posted by mazex

Well, Oleg has to check the gliding capabilities of his WWII birds then. No way a Spitfire would have sink rate of -1m/s power off... Something like 5 m/s would feel more realistic. Anyway, that is in a straight glide and in a turn required to catch a thermal I guess a Spit would have >10 m/s sink rate

well, there were gliders in WW2 as well, like the me-321 and the G11 or the CG4.
And I'm currentle playing Rise Of Flight, it would really help if you had some thermal there because the best rate of climb with the planes is like >2 m/s or something (You don't have a vario meter in ROF) So maybe it would be of use in SOW as well

[katdogfizzow]

screenies look great...getting excited


those nimble clouds are incredible..wonder if its possible

[BadAim]

I love the cloud video! "Coming in 2007", and people are complaining about SOW being late! All of the dumb ass crap you guys can come up with will be blown away by Oleg. I have faith backed up by evidence. Flame away.

Edit: I've been drinking again, so I will grant that what I've just posted might not make sense in the morning....... yet I still have faith in the boyz from Moscow.

[airmalik]

Originally Posted by BadAim

I love the cloud video! "Coming in 2007", and people are complaining about SOW being late! All of the dumb ass crap you guys can come up with will be blown away by Oleg. I have faith backed up by evidence. Flame away.

I was quite impressed by that video as well so I did some research. Turns out the company that made this product got bought out by the makers of Second Life back in 2007.

The real-time volumetric clouds took only a month to create according to the developer:

I have to say that I’m really pleased with how this turned out. Given the task of finding a way to grow and render fully volumetric, full-sky clouds on regular hardware in real time, with the target ‘look’ being the clouds in these TG2 screenshots (which are rendered offline), all within one month, it was a pretty tall order.

The volume itself is based on a simulation – so you can ’seed’ the atmosphere with humidity and other parameters, and clouds will naturally ‘grow’ into the kinds of interesting shapes you see there – no artists required, yet you still get the ability to dictate the placement and shapes of clouds, rather than placement being random like noise-based methods.

http://www.stevestreeting.com/2006/1...-cloud-system/

The atmospheric lighting (WindLight) is equally awesome:





Pre:


Post:

[Dano]

It's probably fairly easy (in relative terms of course) to make a cloud simulator that looks accurate, it'll be making it use a limited amount of resources and run acceptably while the computer is simulating everything else needed that becomes an issue I'd imagine.

I'm sure Oleg could give us an almost completely authentic sim with absolutely everything modelled to within a few % of reality if he wanted, but he'd need either a massive team of staff or many many more years to do it, concessions have to be made at some point.

[airmalik]

Originally Posted by Dano

..I'm sure Oleg could give us an almost completely authentic sim with absolutely everything modelled to within a few % of reality if he wanted, but he'd need either a massive team of staff or many many more years to do it, concessions have to be made at some point.

To get the high fidelity games we expect in 2010 and beyond, without unlimited money, time, resources, I think it'll be increasingly necessary to license such technologies from specialists. As good as Oleg's team is, I doubt they can do better than specialists focusing on specific areas of the sim. I'm encouraged by Oleg's use of Speedtree which indicates that he's open to this strategy.

As for the rendering speed of the clouds/atmospheric lighting shown in the posted video, they were optimised from the beginning to be used in real time games. Some screenshots on the http://windwardmark.net/ show frame rates between 100 and 170fps on 2006 hardware. I don't know how that compares to the rendering cost of Oleg's landscape/cloud rendering.