How advanced the graphics engine will be?

Hi lirpakkaa,

I guess just because something has a big explosion doesn't mean everything around it gets massive splash damage. These machines have been replicated and refined so I would imagine they are able to withstand big explosions.

I suppose we have to also think about the awesome to realism scale. If it's more awesome that explosions are big and do little damage than so be it . I guess all of this will be tweaked and re-tweaked during the alpha/beta

Thanks mate, very kind of you to say so.

Cheers

Particle effects for smoke and debris would be pretty awesome.

But honestly, so long as my thousands of robots get an ok frame rate, I don't give a flying F* what they look like.

I'm a LOT less interested in any of the fancy DX 10+ technology than I am about fundamental things like, how many robot arms can I blow off and leave scattered on the ground, or, 'will we be able to see incremental damage on units' etc.

Pretty graphics are one thing, being able to DO things with the models / units is quite another.

I'm sure you guys are going to make every effort possible to make an appealing graphics and visualization package and focus on gameplay; however, for this game to be successful the effect of an asteroid slamming through a planet needs to be appealing and awe-shocking again and again as it is to watch the death star explode at the end of Episode IV and VI ; )

So far, I have complete faith this game will be amazing and I will support whatever you guys deem "possible."

Thank you

Which death star effect? The original one or the "upgraded" one?

Haha good question! In my nostalgic mind I like the "starry" explosion of the original that makes you feel like you are watching Microsoft 'starfield' screensaver on crack, but the upgraded one was pretty epic too...

I have a feeling the "going for awesome" will yield something we have never seen before... I'm greatly interested in playing and seeing it for the first time!

I feel dirty saying it... new one.

http://www.youtube.com/watch?v=z2jPCQbl0Lc

There doesn't need to be an equatorial ring from a spherical explosion. Please. Let that travesty against physics die.

I dunno. This could be an entirely valid case of 'Awesome > Reality'.
Expanding spheres are a little dull.

This will have thousands of units up to potentially 10's of thousands. That alone will be difficult on even the best computers. Not to mention you will have asteroids crashing into planets. The graphics can not be very high end. Impossible under current technology unless they created the most awesome engine ever.

I will be more than happy with graphics similar to Supreme Commander 1 if they get it to the scale they promise. It will be very difficult.

I think a hard part will also be getting a great A.I.

I don't really care about those "wizbang stuff", but it would be good to know if the engine support organic animations, it is quite important for modding.

I don't think advanced shading is going to hurt performance, quiet the opposite. I mean, theres a reason most of this stuff was developed for console application in the first place.

Also just because you have lots of units doesn't mean much. The polycount is hardly the bottleneck it once was, and the texture footprint on graphics should be negligable considering its procedurally generated stuff that possible resuses the same stuff over and over again. Only thing that really comes to mind as performance issue are like, 10000 little figures creating shadows...

Well, just make it look better than supcom1/2 and your job here is done. Oh, this will be hard.

Making it look better than SupCom 2? Not that hard. But beating SupCom 1 might be a challenge, that game never went easy on the GPU.

Besides: Many units and fancy graphics don't cross out each other. Ever heard of LOD? Either you have few units with a lot of details on the screen or you have a large number of units, but with low details on the screen. The graphics engine will switch between those two states anyway. And: The number of units and physics go on the CPU while the fancy graphics put load on the GPU, also server and client will always be separated applications.

Except for particle systems, almost all fancy features only cause load on the GPU, allowing for great graphics while having thousands of units at the same time. Only thing you need, is a proper graphics card.

Well I think you underestimate the CPU requirements. To do thousands of units all acting independently while trying to have intelligent AI. Not to mention when an asteroid crashes into a planet there will be a huge amount of work for a CPU. Think of it you not only have to process thousands of units getting kill, all the buildings being destroy, etc.

I don't mind better graphics as I have a 690 But the CPU will be highly stressed. So they can not make it too CPU straining unless. Well except for optional settings....

True, CPU usage compared to other games is massive. But it hasn't changed much since,... forever.

Even TA already had (correct me if I'm wrong) hit calculation with every fired shot. Every single one. And the machines that were able to run this would be considered ancient by now.

Compared to a xbox360 or PS3 wich doesn't even have dedicated gpu, modern PC cpus are monsters. And PC have dedicated GPU's with their own ram. It's really not all that bad. I can't remember a game that used up more then 40% cpu on my machine. I did develop some swarmy-mass-collision-space-alien stuff with cryengine2 on this thing. So I know how many shaded polys this baby can push around. Don't undererstimate the PC.

But I still think this isn't a really helpfull discussion. I'm just going to assume that the guys working on PA know what they want and what they are doing. None of us can really weigh any decent options against one another, they probably can't either at this point.

Let's just wait for some alpha/beta.

Dont forget when PA is out the next generation of cpu's is here which (atleast they tell us) will be alot faster unlike the ivy core (3th gen).

I have yet to find a game that will use my i5 760 to its fullest (which is the first generation of i5 cpu's). So far i only saw 100% load when i was rendering stuff with inventor

The original post saddens me. It conjures up memories of the "DirectX 10" debate that SupCom 1 had (it was a DX9 game, and every buzzword-addicted idiot wanted DX10 support, as if using a different library would magically make the game look several orders of magnitude better).

Really, I couldn't care less about SSAO or HBAO, Tessellation, Global Illumination and Deferred Lighting, FXAA, Parallax Occlusion Mapping, High Dynamic Range (HDR) ...

The aesthetics need to be spot on, meaning a few things need to be nailed:

• colors and textures
• clean shapes, no visible clipping issues
• no flickering of any maps/textures (shadows in supcom were pretty crappy)
• no noticeable z-fighting (supcom had some of that, as well)
• overall lighting (a little less bloom render in some games would have gone a long way)

Only then should fancy things like SSAO/HBAO be even considered to round off the overall feel and plasticity of the graphics.

Tessellation (actually a stupid name for the feature) would be great in a flight sim or an outdoor FPS. It shouldn't be used in a RTS (PA might be a slight exception because of the strategic zoom).

RTSes profit from properly applied geometry instancing, foremost. That's a feature I'd look out for, but I'd even prefer it if no tech buzzwords at all were used to describe the engine.

Planetary Annihilation should know only one buzzword: AWESOME.

How are people at this point not understanding that "We are not going realism, we are going for awesome" translates directly to:

We don't give a hoot what the real physics really are... we want to create a spectacle that is amazing and unique that will give the most FUN possible to gamers according to our vision.

I don't know about you guys, but listening to these guys in interview and their gaming preferences... this will be great. 'nuff said.

Cut the crap with the realism and let them make "awesome"

You still have a few hours to retract your donation if you don't want to invest in the idea... .but they have CLEARLY stated how they going about making this game and you would be very very foolish to put money down and not 100%-understand what's going on here.

Would you buy stock without first fully understanding what a company success rate is and what their vision is? I hope not....

btustorm; You might want to give this a read;

PSA - Realism VS Awesome

Mike