Microbe Models/Images/Textures

Well, each organelle is going to have a fixed rotation point, and you can rotate it in 6 increments around that point. In terms of controlling it, we can do that however we want, whether through a rightclick, button press, keyboard stroke, etc.
In the next silhouettes update I'll include rotation points.

I hadn't been keeping an eye on this thread... looks like I've missed some very interesting developments!

I don't have much to contribute here, but did want to note that many organelles are effectively cells themselves (mitochondria, chloroplasts), or atleast cell membranes (nucleus, gogli, ER, vacuoles, vesicles).  For some of these (I'm thinking of the golgi, ER and vacuole), it may be nicer to have them shaped freehand to make better use of whatever shape the player makes their cell.  This would also allow more freedom in shape, as the current kernel and vacuole shapes enourage fairly round cells.

I would propose that the cell kernal starts of as 3 hexes in a triangle, 1 nucleus, 1 golgi, and 1 ER, and that these can be expanded via mutation to increase their efficacy.  Vacuoles would start as 1 hex, and could be grown to whatever size is needed.  Mitochondria and chloroplasts look fine as they are to me, and while adjusting their size/shape is scientifically feasible, it doesn't occur much in LAWKI as far as I know.

That's a good point. I definitely support the idea of the ER and vacuoles being determined by freeform shaping. I hadn't thought of doing that at all.
For nucleus visibility (and symmetry) reasons, the kernel has to be at least the standard 7-cell group. (the nucleus has to be bigger than 1 hex because it needs to be visible, and having a 3-cell grouping would disallow perfect symmetry.

I'm very skeptic about letting the user determine an organelle's shape. It opens a whole can of worms in terms of user interface; we'd basically need an organelle-editor inside the microbe editor. Freeform organelles also diminish the "tetris" aspect of microbe editing. I rather like the idea of players taking a couple of minutes to rearrange their organelles so everything fits.

Seregon wrote:

[The kernel] can be expanded via mutation

You mean like, physically increased in size? What would happen to the adjacent organelles?

How about only having certain organelles that are "free-form", whilst the rest all have preset shapes. Free-form organelles would be shaped by the player, in a simple manner, simply by selecting tiles which highlight as they are selected, until it reaches the number of hexes that that organelle requires. For example, say a vacuole requires 7 hexes, I select 7 hexes that must all be contiguous, and when complete that is the shape of that vacuole. I don't think this would require the creation of a further editor within the cell editor.

The final result might not look like a new editor, but from a programming perspective, it very much is. Here are some of the things we'd need to figure out, implement and test:

• Communicating to the player that an organelle is free-form, and how many hexagons it requires
  
• Switching from normal organelle placement to free-form organelle placement and back again
  
• Disallowing non-contiguous free-form organelles (unless an organelle can be broken up into disjoint pieces). This is especially tricky if the player should be allowed to change the organelle shape without completely removing it first.
  
• How should a free-form organelle look? It would either have to be very uniform, or each grid cell must have its own, discrete graphics.
  

Granted, all of the above are solvable. But I question whether the required time is worth the added value. I'm not even sure there is added value apart from making it slightly easier to create a specific shape.

Here's a variation that would be much easier to do: some organelles can change their graphics if they are next to another organelle of the same type. So when the player places two vacuole organelles next to each other, they behave like two separate organelles, but look like one big vacuole. That would only require changing the rendering of organelles (which would still be difficult enough), without the need for new interface logic.

Nimbal wrote:

Here's a variation that would be much easier to do: some organelles can change their graphics if they are next to another organelle of the same type. So when the player places two vacuole organelles next to each other, they behave like two separate organelles, but look like one big vacuole. That would only require changing the rendering of organelles (which would still be difficult enough), without the need for new interface logic.

Let's make it as easy as possible: we do it that way. I think I'll make a vacuole the same shape as the agent vacuole (let's just call them agents [toxins, enzymes], solve that problem) so they're not too large and tessellate well, but not extremely well, so it's not too easy to flood a cell with them.
Also, naturally graphics worries are of secondary importance. I don't give a Belgium if the first release cell editor just spits out hexagons. Making it pretty is (relatively) trivial.
So for the original idea, which was to make the ER/Golgi expand as directed by the player, I'd say keep it given a few rules:

• ER tiles must all be contiguous
  
• ER tiles must not form a ring around any other organelle (so the player can't just cram them in willy-nilly
  
• ER tiles cannot be placed on functional edges
  
• ER tiles have a high MP cost
  

The advantage of having more ER tiles is to produce RpAse at a higher rate, but this will see diminishing returns after a certain threshold.

Does that seem feasible?

I have a question to the proggramers, how many pixels should the 2d organells be?

~sciocont wrote:

ER tiles must not form a ring around any other organelle

That could cause some headaches in the implementation. It also wouldn't help much in enforcing a sensible organelle shape, since it would still be possible to "worm" it through the whole cell, or make a spiral.

What about upgrading a fixed-size organelle for increased RpAse production?

Thriving Cheese wrote:

I have a question to the proggramers, how many pixels should the 2d organells be?

Enough to make it look good. We haven't decided on an absolute size for the individual hexagons yet, so it's impossible to give a size in pixels for the organelles. I would estimate the minimum sensible hexagon size to be 16 x 16 pixels, while the maximum size is probably at 128 x 128. I guess the actual size will be somewhere in between (and not necessarily a power of two).

Nimbal wrote:

~sciocont wrote:

ER tiles must not form a ring around any other organelle

What about upgrading a fixed-size organelle for increased RpAse production?

Thriving Cheese wrote:

I have a question to the proggramers, how many pixels should the 2d organells be?

That's what we were originally headed for, I just wanted to throw around ideas related to Seregon's suggestion. If it's not worth doing, it's not worth doing.

Cheese, make them as large as possible, on square canvases, fitting into the shapes defined by the silhouettes above. We can always shrink images, but we can't make them bigger, so make them as large if not larger than the silhouettes posted above.

Ok, so that got some mixed reactions.  I personally don't like the idea of organelle placement becoming a game of tetris - what you do with you cell should be limited by available mutations more than physical constraints - so i see no reason for an organelle not to snake through a cell.  It may or may not be realistic, but it would look rather interesting.

I can see some of the implementation issues, but they seem fairly easy to solve to me.  In terms of the algorithm, a vacuole or golgi would be formed exactly the same way as the overall cell, by stretching a membrane around the adjacent hexes.  The ER would need some sort of folded texture over that though.  Placement could be as simple as having the organelle by 1 hex big, and automatically connecting adjacent hexes.  Disconnected organelles would simply be treated seperately.  For the purpose of processes/compounds etc. we can also simply treat a 3 hex vacuole as three seperate 1 hex vacuoles.

All that said, I see that this would add some work to the implementation, for something which is really just an aesthetic improvement, and I can certainly understand that would be far from a priority right now, and may not be worth doing at all for the first few builds.  I'd suggest we shelve it for now, and come back to it once we have a first release done, and have some free time to tweak this sort of stuff.

Ok, This is the (hopfefuully) final silhouettes mockup. I'll begin on the actual sprites once you approve of this. The red crosses are rotation anchors- I assumed that it would be easiest to rotate them around the center of one hex than at the corners/ on edges I'll attempt to make the sprites quite large, then we can scale them down to whatever we need.

They all look good, and it is nice to get a visual representation of the different organelles to be found in game. Will Thermoplasts and other organelles be included in a later release?

NickTheNick wrote:

They all look good, and it is nice to get a visual representation of the different organelles to be found in game. Will Thermoplasts and other organelles be included in a later release?

Yes, different organelles will be added as we figure out how we want to run them ingame. the nice thing about the toxin agent organelle is now we can handle a lot of potential organelles as one.

New GUI design. Much easier to implement and, I think, much prettier than my last.

Spoiler:

Spoiler:

I've squared up all of the buttons, taken out the size ring, added a symmetry options button (bottom center) and rearranged a few things.
Tell me what's good and what's wrong.

~sciocont wrote:

New GUI design. Much easier to implement and, I think, much prettier than my last.

Spoiler:

Spoiler:

I've squared up all of the buttons, taken out the size ring, added a symmetry options button (bottom center) and rearranged a few things.
Tell me what's good and what's wrong.

i liked it. its simpler than the other, and i think its better...:)

Yes, I think it looks amazing now. I especially like the icons for the stats. Really, the last point I would have to offer, and it is quite trivial, is to switch the MP bar with the Finish and Revert buttons, with the Finish button being on the opposite end of the red X. Its just that usually in GUIs the accept/start/complete button is opposite the big x/delete/end button.

I agree, it looks great!

NickTheNick wrote:

Yes, I think it looks amazing now. I especially like the icons for the stats. Really, the last point I would have to offer, and it is quite trivial, is to switch the MP bar with the Finish and Revert buttons, with the Finish button being on the opposite end of the red X. Its just that usually in GUIs the accept/start/complete button is opposite the big x/delete/end button.

I can do that quite easily.

Ok, I want the artists out there to look again at the models and textures we have and see what can be improved. A microbe stage animation is in the works, and a release should be coming within a year. We need a uniform visual style. As Myself, Nimbal, and Nick have said, this should be a style of realism, and it should look to some extent as if it were being viewed through a microscope. Here's a quote from another page with texture details:

sciocont wrote:

Nimbal wrote:

I'm not sure if that's what scio means, but I'd like to see a style that imitates looking through a microscope, but with a bit more color where appropriate. There are some YouTube videos that could serve as a reference, for example this one, this one or this one.

I think the defining qualities of this style are:

• Translucency of the microbes
  
• Clearly visible edges that hint at the three-dimensional, round form of the cells
  
• Stuff floating around in the background, out of focus
  
• Jerky camera movement as if moving around a slide on the microscope's tray
  

Those aren't a question of highly detailed models or textures alone, after all, how much surface structure do you really see in those videos I linked above? Achieving this style will probably require some hand-crafted shaders.

This is all good. I'd like to see microscope-like images, though with a bit less transparency and more color than in the youtube videos, since we won't always have a light background to see them against.
In terms of textures, I'd like to see

• Pseudo-randomness: Textures should look like they have a definite order or set of rules, but also should look organic. For this, I recommend checking out Perlin Noise, Voronoi textures (which were the inspiration for the "shatter" graphic art style), and real animal and plant textures for inspiration.
  
  
• Uniform sharpness: Texture should either contain bold, sharp edges, medium-sharpness edges, or soft edges. Don't mix edge types, and always try the sharp edges first, then blur them to the desired degree. If you look in a microscope, you'll see many different levels of focus, but our textures each apply to a single object, which will be either entirely in or entirely out of focus.
  
• Light, desaturated colors: Microbes usually have colors like this because they are small and mostly transparent, so there isn't much pigment in them to build up and create dark colors. When choosing colors for textures, keep this in mind.
  

Look at this color wheel:

The colors I think of are colors like  ████CCFF99C, ████FF9933, ████CC3333, ████6699CC, ████99CC99, ████99FF66, ████FF6633. I want to avoid purples, pinks, and yellows, since these are colors one generally doesn't see in the microbe world. Focus on blues, greens, and  pale oranges.

As for models, make cells look natural- give them some taper, some flash, some intrigue. Make any other models look very close to what they look like in real life.

Keep in mind that your textures may need to be recolored to support many different color schemes, so it's best to make them in grayscale and test them out with a few of the aforementioned colors.

Here is my Wriggler, if you like it. :3

Well, I'm not sure if this is correct to do or not, but I tried at creating a sprite/texture for the bacteria film. I wasn't quite sure what kind of bacteria you were thinking of, so I tried to make something that wouldn't look too distracting.

bacteria film:

Also sorry I'm not sure what "download link" would be- perhaps I should create a dropbox account or something? D:

Sorry if I misunderstood and if I need to make an application or something. Didn't see any guidelines on the faq, I think...

For a download link, you just upload it to a file sharing/hosting website, like Rapidshare, Dropbox, or Mediafire, and then paste the download link here.

(dropbox)/s/2xdro9v0j0lwjwz/bacteria%20film.png

I hope that works out. D: Sorry for spamming.
Can't post external links since I'm a new member, it seems.