Or the location of the light itself is slightly in front of the goggles?
Killzone 2 technology discussion thread (renamed)
- Thread starter Terarrim
- Start date
- Status
Or the location of the light itself is slightly in front of the goggles?
Most of the textures look unsharp, lifeless, colourless, "soulless" and flat to me. The lighting is nothing to scream at, also. Looks like HDR is not implemented.
Both features definitely lost their initial focus and touch (I'm talking about the 2005 CGI trailer here).
It's darker, stone cold, whatever... but it lacks colour.
I'd rather prefer a different approach because although DR sounds like a good concept, performance wise, such an expensive technique should offer better results and I don't see the benefits of it.
Ok listen (well, read) not to sound bossy or anything like that but could you please respect my opinion and whoever else is not a fan of the art direction of this game? I didn't like GoW (there was an option for vibrant graphics, my favourite by a long shot, I played the game with that option active all the time, though) and Lair because of that fact.
KZ2 just looks fine (neither good or great) to me but, imho, it's not the AAA game some people believe it is nowadays. This game is WIP, so we'll see.
If you don´t like the art direction wouldn´t that sum up to pretty much not liking how the game looks in any case, and therefore maybe it would be cool to kind a ignore killzone threads?
If that is the case, why do you comment on Deferred Rendering at all ? It seems your personal preference has nothing to do with the technology, and it should go to the "Console Game" area. You should also try to justify/elaborate your points if you think DR caused the issues you perceive (e.g., color choices). It's hard to take you seriously based on the random comments you made.
EDIT: Also, could you kindly answer Fafalada's question ? I am wondering what it all means to be frank.
Cyan said:
Fafalada said:
Last edited by a moderator:
Although it seems like you can not express yourself in proper terms, I understand what are you talking about.
Deferred renderer has to use one fragment assembler when the final RT is blended, so all pixel effects should be composed from the exact same components.
This in turn makes the final picture look kind of "uniformly lighted".
But, when you talk about industrial/city environments this technique produces very life-like results. So I do not see why it's bad for Killzone 2. We do not want another tropical-island-game after all, do we?
Disliking the art-style is a perfectly valid opinion, but not really suited to a thread about deferred rendering where the choice of colour-schemes is not a product of the rendering method but the colour the artists have slapped on their 2D textures. That's kinda like saying, 'I wish developers wouldn't use Photoshop in creating games because Game X used it and it looks dull and grey.'
i agree, if anything the gun that we see in the real time trailer is very very close to the CGI render.
Agree to, the first fun looks amazing, sadly the other gun and handgrenade not. I expect thought the devs to fix it up (they look to blurry much like moire).
Actually, the official GG guy at the PS3 forums (Seb Downie) did comment when asked about this piece of blog, stating the guy was being mostly right in his claims.
"Apart from some naming conventions and minor things it is a pretty decent breakdown. Once we can talk about the tech a bit more we will."
Really, interesting got a link to the post?
http://boardsus.playstation.com/pla...&thread.id=4265&view=by_date_ascending&page=1
It's the same article/blog of the Gametrailer-poster, but posted in the PS3-KZ2 forum, and that's were GG has their guy(' s) commenting on all kind of questions since a couple of years for feedback or just chitchat talk with the community about their games released so far.
On one page of that thread, I think it's somewhere on page 3, the guy comments with the upper quote about the article.
That's an interesting question. Why waste all the memory, bandwidth, and computation time if you can approximate the effect some other way? I know there's been at least one game that stored pre-determined luminance values on a per-area basis, and IMO the results didn't look terrible (certainly not for a PS2 game, anyway!). As for "guesstimating" the luminance value, I wonder if there's a decent way of doing that based on the amount of screen space a light takes up (something you probably have to do for DR anyway).
You can compute luminance per pixel in your pixel shader and dump it down to the alpha channel.
With that you can compute an exposure value that can be used in the following frame to tone map every pixel in your opaque rendering pass directly in your pixel shader.
With a deferred renderer this stuff is even simpler if you render a few bits in your g buffer storing the logarithm of your luminance, you really don't need to be accurate, no one is going to notice the difference.
Marco
With that you can compute an exposure value that can be used in the following frame to tone map every pixel in your opaque rendering pass directly in your pixel shader.
With a deferred renderer this stuff is even simpler if you render a few bits in your g buffer storing the logarithm of your luminance, you really don't need to be accurate, no one is going to notice the difference.
Marco
Isn't it enough to see that a garden is beautiful without having to believe that there are fairies at the bottom of it too?
that is so true when it comes to this game.
It's enough if the artists think it's enough. But if they suddenly need the fairies for something, and you can't give them, then you are somehow in trouble. It depends on the context.
But if the engine is built so you can never give the fairies no matter what (even if it's techinically possible this generation), then, in my opinion, you should rethink your assumptions because you are artificially limiting their creativity. I always prefer a flexible design if it's possible to be able to quickly adapt to changing requirements by the artists.
Said that, it's a matter of trade-offs: not knowing their trade-offs, so from a limited amount of information I have, what I can say is that in my opinion, those limitations (and there are few pretty important ones) for the sake of more point lights is not worth it.
Again it's not a matter of doing HDR or not. If you don't need it, don't do it and go to the pub.
Sometimes it's better and more appropriate to do a couple of things really well, than a heap of things in an average manner.
Why would they suddenly need fairies if it wasn't in the design document? I realise that plans change and all, but you shouldn't chop off your foot on the offchance that you might lose one of your shoes.
I would guess that the developers have made their decisions based on a lot of thought and planning, and with the artists in mind. Let's judge their decision after the game is out?
Why would they suddenly need fairies if it wasn't in the design document? I realise that plans change and all, but you shouldn't chop off your foot on the offchance that you might lose one of your shoes.
I would guess that the developers have made their decisions based on a lot of thought and planning, and with the artists in mind. Let's judge their decision after the game is out?
- Status