Quote:
Originally Posted by The Kraken
(Post 221334)
What people really mean when asking about the "dots" is this: will LODs only depend on distance, or also on object size, resolution and field of view? Because if the switch to a dot representation only happens when the render size is about one pixel, then there's no issue anymore.
|
Actually it would be very good if the amount of pixels used for a given aircraft at a given distance would scale with the resolution used. For example, up to 1280x1024 a fighter at 5km would be 2-pixel dot, but at 1920x1080 the same fighter at the same distance would be 10-pixel dot, etc (numbers arbitrary for example's sake).
Quote:
Originally Posted by swiss
(Post 221382)
Sure. But here we are talking about the moment where the plane is still a tiny dot.
What do you do with this information?
At this stage you don't know whether it's friend or foe - all you know is there's an aircraft.
(the direction of the plane doesn't help too much, I often penetrate hostile airspace just to stab them in the back on their way to the front )
|
You are partially right, however there are cases when it does make a difference. A lot depends on altitudes, relative positioning and closing speeds, so for example if there's two guys approaching each other head on at 500km/h an extra 2 or 5 km of spotting distance give you a better window of time in which to respond (or a worse one in which to be surprised).
I still remember one of my online kills with a Fw190A from a few years back on my 17" CRT monitor at 1280x1024, where i spotted a lone dot against the fog almost an entire map grid away (about 7-8km) at my 10 o'clock low. I set into a pursuit curve and a shallow dive, adjusted my course by observing the dot's movement and just waited to get close enough for identification before pulling the trigger. As it turned out i had flown an almost perfect curve and doing close to 700km/h by the point he was about to cross from left to right across my windscreen, i looked closer and identified it as a Pony and let him have it. It was very satisfying because it was a kill based solely on SA, speed and surprise done in a single firing pass.
I can't replicate that in my 22" LCD at 1680x1050 however. I tried some QMB missions with icons on to judge the difference and it seems the dots can be spotted a little while before the icons kick in, at a little more than 5.5 km, and that's against the sky. Against the ground they could be sneaking up on me just fine and making it to within 3km or less.
In any case, much will depend on how CoD will deal with making camouflage useful at long range while not having planes blend into the background at point blank like it used to happen with IL2 (the problem described by Sturm Williger).
Who knows, maybe we'll have sun glinting off the canopies and other reflections, or the LOD scaling will be so good and the aircraft standing out from the backgrounds that it won't be an issue at close range but the camo will still be effective at long range. Anyway, i trust they'll do their best to do a good job on this so i'm not terribly worried. Just the fact it will support widescreen resolutions right off the bat will improve things considerably, since it's built for them from the ground up and not "forced" to used them on an engine build for 1024x768 resolutions like IL2 is.
Finally in response to Royraiden, a monitor has a much smaller resolution than the human eye. There was in fact a challenge of spotting planes in real life too and yes, sometimes it's hard even for white or brightly painted airliners to spot each other. However, the superior resolution of the human eye will pick up on movement and can distinguish silhouettes at a much higher distance. Since we can't have that with a monitor, in order to simulate how it works in real life there must be a way to offset the disadvantage.
So, in reality it's hard to spot a single dot with no apparent motion but it's far easier to make out the plane's outline and identify it once you pick him up visually and at longer ranges. On our PCs it's the opposite due to hardware limitations, they give us long range spotting of the dots because they can't give us long range silhouette on a PC monitor.
Apart from spotting issues now,
Quote:
Originally Posted by Feathered_IV
(Post 221335)
One more question Luthier and I'll leave you alone :)
Probably the one feature I really hope to see in CoD is improved crew interaction. Can you comment on AI crews, will gunners call out surface and air contacts? Will navigators give you course headings and corrections, and will bomb aimers guide you in on the bomb run?
The crew members of other sims are really only decoration and it would make the player feel far less alone (and much more immersed in the sim) if their virtual comrades could fulfil their basic functions within the aircraft.
Hope you can answer!
|
Quote:
Originally Posted by MikkOwl
(Post 221380)
Luthier, thank you for these interesting, well written answers. :) The gag is finally starting to come off. You seem like you had an itch to share all this cool stuff your team has been developing over the past few ages.
Can you talk a bit about how (AI) crew members can be interacted with? Classical example is the rear gunner of the Bf110. He is mostly quiet in IL-2, and cannot speak at all in multiplayer. There was talk from Oleg of improving this area. I'll write the points numbered so replying should be a quick breeze. I expect 'No' as answer to most, but I'm curious enough to ask. :) Think of it as possibilities if nothing else.
1. Better contact reports?
2. Feeding information of someone on one's tail?
3. ..Perhaps prompted by the player by a button?
4. Instructions on his general behavior?
5. Does he have morale? Panic? Hot on the trigger?
6. Bail out without permission (panic mode)?7. MG-FF's had to be reloaded by this bordfunker in reality.
8. Report visual damage to own plane he sees? (smoke trail from engine, control surfaces damaged and so on)
9. Gets affected by your maneuvering?
10. Navigation & radio (I'm sure not).
|
Very interesting points, i'd love to see something like that.
Or if we can't have that on release, maybe a possibility to include a scripting language in the SDK for the community to implement similar functions on their own would be even better.
