SQS tiled floor example

[Madison Lastrega]

These 36 k floor tiles are elegent and add depth to my project, I can build simular for others if you need .... just give advanced notice as I am really getting back into my own project.   ML

[Verdite]

Looks great Joe. Reminds me of some of the floor tiles in Might and Magic VII, 3d, polished and used at the same time.

[Holy_Diver]

The "water ripple effect" is just graphics circa 2000. Use Ex to make everything appear to be solid. Whatever 36k means. These tiles should use 2 triangles. Triangles equal electricity. Equals heat.

[HwitVlf]

That's a big improvement over flat. There's also a subtle "depth perception" effect when tiles aren't 100% flat that adds a lot to the immersion when you're moving.  What is the polygon count on a new tile?

One thing I was found that made a nice effect too, is to raise the bricks on one side of the map Piece a little- not the mating-edge mind you, just in from the edge a bit. You can rotate an off center map Piece like that in 4 different directions to make some very natural, non-tiled looking combos.

  Hey Ben, hauling 3 corpses around!? That's a bit creepy    M&M 7 sure looked nice for a pre-3D game.

[dmpdesign]

HolyDiver, any chance of bumpmap support in the near future?

ML before you go making higher poly count floor tiles ( I admit they look nice), it would be a good idea to see if SomEx may score you bump/lighting support to pull off the same effect with much less strain on the processing.

[Madison Lastrega]

Bumpmap support would be awesome..
Right now they are 36 kb in size and boast a poly count of 364 :) Thats with a ceiling.    My column pieces are 41 kb and have only 412 polys ..... I am trying to keep them small, yet pretty.....   Holy is right I checked the ripple effect is the graphics of the era.  Bumpmaps would be a great addition to lower them even further......   ML

PS Great to see you again DMPDESIGN  :) I have missed you :)

[Holy_Diver]

HolyDiver, any chance of bumpmap support in the near future?

ML before you go making higher poly count floor tiles ( I admit they look nice), it would be a good idea to see if SomEx may score you bump/lighting support to pull off the same effect with much less strain on the processing.

Technically yes. The plan is to extend the MDO format so that every third vertex holds skinning data for two vertices.

I originally thought it would be cool to interleave the vertices and skinning data so it would be 1:1 where the extra space would hold bump mapping numbers. But I decided instead it would be better to automatically generate the bump mapping numbers and cache them along with CP files in temporary folders.

None of this interferes with SOM since none of its MDO (and MSM) files are animated. And the bump mapping data is in separate files. The only real upside is you can still open these MDO files up into one of the old Edit tools.

Each vertex is 8 numbers. So a skinning slot is 4 numbers. That's 3 weights where the fourth is 1 minus the other 3. And the fourth number is really 4 bytes that assign the weights to their anchors. Unfortunately it won't be possible to load this format directly into video memory, since it will have to be unpacked (and interleaved with the bump mapping data) but unpacking should be very simple.

This going to be worked on relatively soon, but I haven't thought about how to accommodate the additional textures that would be involved. There is already a way to do that on a per texture basis. But it doesn't tie into the MDO and MSM files. In theory you can pack arbitrary file references into those formats.

Bottom line though SOM won't know anything about this stuff. It will only work with extensions. MDL won't be used anymore for visuals. It will just be an animation format. There might have to be some fallback junk in them for backwards compatibility.

[dmpdesign]

I have plenty of textures that are already mapped for normal/specular whenever you think you may want to try it, I would be happy to donate test textures for you to try your efforts against....most are tiling as well.
If you crack that barrier, I think it would plant SOM quite near some of the higher end mobile games in quality of output graphics since most of your iphone games and such do not use dynamic lighting, but make great use of scene lighting with normals etc.

[Madison Lastrega]

Any textures you wish to donate would be great, as I lost a bunch and have been cheating, using open source minecraft textures.  Only because I don't want to make new ones      To save time.  yeah I still use DMP classic black boot :-)

ML

[Holy_Diver]

I have plenty of textures that are already mapped for normal/specular whenever you think you may want to try it, I would be happy to donate test textures for you to try your efforts against....most are tiling as well.
If you crack that barrier, I think it would plant SOM quite near some of the higher end mobile games in quality of output graphics since most of your iphone games and such do not use dynamic lighting, but make great use of scene lighting with normals etc.

There's no real barrier to crack here. Part of implementing shields and other things is drawing the graphics for SOM. So under those circumstances it's trivial to pile on additional effects.

That said. I am not a fan of normal mapping. Combined with LOD it can be effective. But too often its implemented where if you look at something sideways it turns into a tunnel Wile Coyote painted on a wall. At that point the effect does more harm than good. Since SOM is tile based you don't have the luxury of making tiles that must be setup so that they can never be seen from the side.

It's good for subtle textures and impressions, would be great for King's Fields scrawling on walls. Etchings and scratches in armor. But I don't think it should be overused the way Dark Souls does for instance. Unless it is coupled with LOD. And not the automatic geometric LOD that Dark Souls uses to such crappy effect.

At any rate. I don't have any specific conclusions about what to do with what kinds of effects like these. I may even opt to make them more difficult to implement. But the one thing they have in common is you need the "tangent space" data in the vertex buffer for them to work. So that much can be facilitated.

The rest may come down to whether authors are willing to leave their comfort zone or not. I suspect not.