Killzone 2 technology discussion thread (renamed)

Status
Not open for further replies.
First off, I apologize for going totally OT for a second here but I would just like to offer you, Terra, a different perspective from what the other posters were trying to communicate. While I don't necessarily agree with their methods/wording, I do understand where they are coming from. I don't think there is anyone in here that is blaming you for your inexperience with the graphic technology field, rather I believe their responses were derived from two things; their reverence to these great devs who are willing to share knowledge with us "common" folk and also to your tone when replying to said devs.

I believe what they were trying to say was... instead of doubting the dev's knowledge by implying their disagreements were essentially agreeing with yours(which in some eyes seem rude and insulting), you could have asked where the disagreement stemmed from in the first place(which would have still resulted in that excellent post made by shifty). Perhaps the other poster's felt a certain condescending tone when you replied in this fashion. Implying that their disagreement is in fact false due to it being ultimately agreeable with your line of thought may be interpreted by some as: "You are wrong, Mr. Dev and I was right all along". That is my take on the situation anyhow.


Back on topic... Do we know exactly what "Killzone 2 Devs use Deferred rendering" implies or means? Sorry if I missed a few things here and there from the multitude of links... but it's not written in stone that KZ2 will be exclusively using deferred rendering(shading) is it? I think most people who have had experience with both would agree that picking one over the other is largely situational and really depends on your specific scene/environment. If I was a betting man, I would put my money in the mixed approach, perhaps with a bit more emphasis in deferred shading(which will explain why they are making announcements about it, perhaps new ground breaking techniques?). Just like Fran said, having a rich environment presumably means lot's of light sources (favoring deferred) AND also have a lot of variety in materials(favoring forward) therefore a mixture would technically be the best solution.

To me the question is, did the KZ team conquer many of the issues of deferred shading with clever tricks, which are also part of sony's EDGE? From my understanding, deferred shading is a fillrate/bandwidth intensive method... which "parts" of the ps3 actually excel in so this may be a possibility. It's also definitely good to hear that MSAA might be possible with methods that aren't restricted by API's... rather than the big misconception floating around on the forums this past generation with many believing that hardware MSAA was not available with deferred shading due to hardware limitations.

Either way, it's good news since "theoretically" deferred shading scales better than forward rendering as scene complexity increases... therefore it is safe to say that the KZ team is aiming quite high in regards to scene complexity with this title if they are making a big deal out of deferred rendering.

If I have been to abrasive then I do appolagise for upseting people. I't wasn't my intent to cause any one to be angry or to slur devs far from it. I was genuinly confused at what was being said which is in why some posts you could see me putting question marks, maybe they were taken as sarcasm but they were not I was genuinly confused, until as I said Fran and then shifty came up with the explanation.

In future I will endevour to be less abrasive and I hope I can add a little something to the community to.
 
Anything about the tech presentation ? I think it was today.

I was at the presentation yesterday. Some interesting ideas. Overall, dropping HDR, 4X MSAA for 2X, using only 12 (6) taps on 512x512 shadowmaps for the main directional light, dropping specular color for materials, dropping directional lightmaps, dropping shadows and per-pixel lighting on particles, only one lighting model for the entire world, for the sake of more lights (actually, for the sake of lighting performance non-dependent on the geometry but only on fragments lit which would be desirable) didn't seem worth it to me.

But I still hold to the idea of Deferred Rendering, the architecture is so simple and natural it hurts it's not a win this generation. I'm sure I can go deferred without sacrificing flexibility for the artists when DX10.x like hardware is standard on consoles too.

Ah, The Art Of Cryisis presentation was absolutely ace, the most interesting I've seen in years and I was the only coder in it. Those guys know their stuff.
 
Was there anything on secondary illumination? Did the deferred lighting engine demonstrate the indirect illumination that the conference description mentions? LBP's been looking fantastic in that respect, but KZ hasn't convinced me of any such abilities in what I've seen.
 
Was there anything on secondary illumination? Did the deferred lighting engine demonstrate the indirect illumination that the conference description mentions? LBP's been looking fantastic in that respect, but KZ hasn't convinced me of any such abilities in what I've seen.

If I remember correctly, they have a IBL solution for dynamic objects based on SH and computed in parallel on SPU, which is written to the albedo component in the G-buffer. Using the SPU this way is a pretty cool idea.
It looked like they don't support different time of day, since they pre-compute a single component directional light occlusion term (basically static shadow term). It's interesting how they use this term to stencil mask the deferred lighting computation of the main directional light, which gave me a couple of ideas.
 
I was at the presentation yesterday. Some interesting ideas. Overall, dropping HDR, 4X MSAA for 2X, using only 12 (6) taps on 512x512 shadowmaps for the main directional light, dropping specular color for materials, dropping directional lightmaps, dropping shadows and per-pixel lighting on particles, only one lighting model for the entire world, for the sake of more lights (actually, for the sake of lighting performance non-dependent on the geometry but only on fragments lit which would be desirable) didn't seem worth it to me.

But I still hold to the idea of Deferred Rendering, the architecture is so simple and natural it hurts it's not a win this generation. I'm sure I can go deferred without sacrificing flexibility for the artists when DX10.x like hardware is standard on consoles too.

Ah, The Art Of Cryisis presentation was absolutely ace, the most interesting I've seen in years and I was the only coder in it. Those guys know their stuff.

KZ2 is not very technically advanced then?
 
[maven];1039526 said:
No, they're just making different trade-offs than other developers.

Dropping HDR is a huge lose for me the game looks good now but it would look even better with more dynamic range for the dev's to use.
 
Last edited by a moderator:
If it makes sense to drop HDR (i.e. the art style and the game itself warrants that the loss of it isn't noticed unless such is mentioned in forums ;) ) for otherwise unfeasible technical possibilities, then I think it is a wise decision.
 
Anything about processing the geometry in SPEs or number of polygons per character, world... ? Finally the lighting engine is Wordlight from Second Intention ?

Not using HDR isn´t the end of the world if you use something that finally looks better!.

Anyway, my doubt now is if they use wordlight what does this mean then?: "Worldlight uses a diffuse PRT solution to provide soft light and shadows resulting from direct and indirect illumination by an HDR skydome..." ( taken from Second Intention web ).
 
Last edited by a moderator:
If it makes sense to drop HDR (i.e. the art style and the game itself warrants that the loss of it isn't noticed unless such is mentioned in forums ;) ) for otherwise unfeasible technical possibilities, then I think it is a wise decision.

The trailer clearly shows HDR like effects. So similar to Heavenly Sword, I'm pretty sure he just means they aren't using floating point pixel formats for performance reasons. And not that they have not have a HDR solution.

kzhdrqm9.jpg
 
Status
Not open for further replies.
Back
Top