The new graphics project is looking very promising indeed, the examples are all over this board. But with all those promising examples, comes a lot of confusion as well. What scale should be used? What lighting setup should be used (it seems like a new one is posted about every month). Where should models be posted? Who decides who makes what graphic?
Now I know that there are some people who are doing a great job at answering this job, but it's not very clear who they are and what they've done.
So my question is: Isn't there room for an official new graphics team? A group of artists and perhaps some developers that together can make decisions on what the first new graphics release will be like? This seems especially important now that Alltaken has left, since he did a lot of the pushing that's required.
(as a side note, I do not want to be part of such a team, I'm just a concerned fan of the new graphics )
To make a graphic you no need blender. Blender is for 3D objects, TTDX uses pixels not 3D. The only you need is a 256 color palette and any program that suports 256 colors (like MS Paint .p) I personally use Photoshop. And I starting now.
¡I'm Spanish!
Sorry if my english is bad. I know that I have some grammar errors.
SORRY
Athorium wrote:To make a graphic you no need blender. Blender is for 3D objects, TTDX uses pixels not 3D. The only you need is a 256 color palette and any program that suports 256 colors (like MS Paint .p) I personally use Photoshop. And I starting now.
errrr this is about the new 3d graphics project for ottd. Not for pixeled graphics. Look at the Blender thread for more information.
Approve graphics for "official" release (they need to be of a good quality, and meet the style of the others)
work out a scale system for ships and planes especially. personally i think a logrythmic scale would be the best. as it gets bigger, the amount it gets bigger is limited (over a certain size). and thus it maxes out. it would keep things larger or smaller than one another in the correct order, without getting out of hand.
they need to Sort out the rest of the basics, like "grasses" and standards. i actually did most of that work, but it seems to have gone a bit sideways since then. i.e. i see new grass popping up all over the show, when i have made every type of grass accurately to the original game, with full overlayability....
things like this need to be made official, and need to be used to limit the randomness thats going on.
lighting setup is one thing which i think has gotten worse over time. its all getting a bit rediculous. make one and stick to it, most of what the lighting setup looks like is a result of the texture settings, and not the lights, so designing your model to the lights is more important than the other way around.
Yeah this will be a good idea. And the less random suggestion by alltaken is spot on (thats what i noticed about this project when i first joined up). I just need to get off my arse and finish that cataloguing system. Well find time for it i mean as i really have been busy.
The "standards" are very important. Even to the point where i can't finish the cataloguing system till i know what sprites are needed. Eg.
Buildings:
- 4 sprites, 90degress incremental of each other.
- season sprites?
- Overlay sprites (lights and company colours)?
- Different zoom levels?
And there's more that i could think of. I personally think that we shouldn't render more sprites for different zoom levels. It would probably take more power to continually load different sprites into memory than just resize all the sprites for all zoom levels and whacking them in the memory when the game loads. Anyway, i'll be back in a few hours to do a better explanation. Gf wants to go shopping and visit a puppy store..
This is a great idea!
How about using ventrilo or something for the leading group? (Voice communication software)
Perhaps differences in language might make it difficult, but when a group has been put together, I think it would be a good idea to try this.
I have a server running 24/7, I could create an extra channel for this, with password protection of course.
This is a very common way to get something done in modding teams and such.
Nothing against forums, but typing in posts take pretty much time and its difficult to finish a discussion.
IM could work too. MSN or something.
I havent contributed with any graphics here (yet), and Im kind of a n00b to blender, but Im very interested in the well being of this project, and I would gladly help organizing this thing. If im needed that is, the leading group should not be too large, but as small as possible!
How will the this group be chosen? A list of people who WANTS to be in it could be a good idea to start with, just as Mosfet said.
I agree, although I really would like it to continue in the way its been running where final answers are generated from discussion, but then instead of the topic fading into the abis of the forum, someone can slam the gavel*, and we can move on. *(judges hammer)
I agree with Alltaken, and the randomness is a byproduct of decentralisation, although I think it doesn't need to be. The lack of final answers in the last 8-9 months has made this. (I don't mean to critise, as your absence was justified) I asume the finishing of the catalogue system would partly reslove this, and just knowing that the disision needs to be finsilised within the forum, by all, would also keep things running more regualy. The reason I agree in my first paragraph is because some things will/do need filtering, as for example while some people may think a graphic is complete, others may not.
I agree that decisions should be made on the forums by everyone, but as you say, someone to "slam the gavel" is needed. So a small group of leaders should be applied, the "chairmens" to slam the gavel, and then perhaps another layer of less mighty people That can be given different organisation duties and hold close discussions with the others in this layer and the "chairmens" when a larger decission has to be made (using the forums and everyones opinions) as basis.
BTW, if Ben Robbins wants to, I think he would fit very well in this leader group.
if you still need a slammer, then i can slam occasionally perhaps, its the funnest part. but you need to get used to things you think have been slammed (like scale, and lighting) to be brought up again and again.
someone definitely needs to do it though. just make a rulling and call it quits.
The thing that bothers me the most that hasn't been worked out properly are shadows. I just can't think of a way that's going to work without it looking dodgy. Take this beautiful mock-up that Ben did. If we have any kind of shadow that is displayed on other tiles than the one casting the shadow how are we going to manage a situation like the one in the red circle?
I think the best way to go about is to not have shadows displayed on other tiles. Also so it doesn't look wierd have the shadow on the dark side of the building rather light and ground shadows on the same tile very faint or not at all.
Yes Shadows are a problem ... I guess having no shadows at all is the easiest solution for the start
For the sake of just producing ideas, we might have lowpoly 3d objects for every object (eg, the bridge tile would need only 7 polygons for surface/walls/road) to cast/receive shadow and overlay this on the landscape ...
Aracirion wrote:Yes Shadows are a problem ... I guess having no shadows at all is the easiest solution for the start
For the sake of just producing ideas, we might have lowpoly 3d objects for every object (eg, the bridge tile would need only 7 polygons for surface/walls/road) to cast/receive shadow and overlay this on the landscape ...
haha so basically make ottd 3d but only used for shadows. Then display the rest of the scenery with sprites. I like the idea but it would be a lot of work.
Aracirion wrote:@mexicoshanty: I meant simple 3d objects. A cube would do for an office block. It would be a lot of work for coders though.
Yeah i knew what you meant . But you'd still have to compute 3d objects for vehicles, trains, etc. I do like the idea though and if it's done right the same sub system may be able to be extended to render ottd in pure 3d .. haha lets not go there.
Oh and what about shadows cast from planes? The only way to do that one is to develop the 3d shadow system. Shadows from planes already look stupid, imagine them now with the high detail and low zoom.
TTD buildings don't cast shadows except their own walls. TTD light is diffused, and all walls are lightened up pretty much.
A preview mexicoshanty attached looks nice, however I think there is too much contrast between light and shadow. If you take a look at TTD sprites, you can see that they're well drawn, and the contrast is pretty small.
)
. But you'd still have to compute 3d objects for vehicles, trains, etc. I do like the idea though and if it's done right the same sub system may be able to be extended to render ottd in pure 3d .. haha lets not go there.