Posts : 726 Reputation : 70 Join date : 2012-10-10 Age : 30 Location : Spain
Subject: Re: Building Microbe Stage Sun May 26, 2013 10:13 pm
Ok, i finally got around to it and readed through all of the microbe stage specification. I mostly agree with it. I would rename a few things, but that's not too important. However, there is a thing that i want to note: Saving.
One of the many benefits of having an entity-system based engine is that all data is stored inside the entities, in the shape of components. That means that if we just save all the entities, including their components, we should be able to regenerate the whole world in the exact same state as it was before by just loading that entity list and restarting the systems.
That makes it much better in two aspects: One: We dont have to look for the specific entities that are storing the data we are interested in (the player's microbe's shape and its compounds) Two, and most important: No matter what we do, we wont need another save system again. If we keep storing all the data into entities (which we should anyway), storing all the entities will mean storing everything important in the game. Even if we are in strategy mode and what we are storing are cities and tecnological objects.
On top of that, it avoid a few other problems, like an exploit. Whenever the player is in danger, it could just save, load again and it would be placed in a brand new environement.
Nimbal Programming Team lead
Posts : 258 Reputation : 24 Join date : 2013-03-17 Age : 39 Location : Ratingen, Germany
Subject: Re: Building Microbe Stage Mon May 27, 2013 2:10 am
Eventually, I'd like to save the game state like you describe, but it's not as trivial as you might think (which is why I went for a simplified version). The most important hurdle is that we currently put Lua callbacks into some of the components. These callbacks are not serializable, so we'll need to introduce hooks for the scripts so that they can set the proper callbacks after everything else is loaded.
Daniferrito Experienced
Posts : 726 Reputation : 70 Join date : 2012-10-10 Age : 30 Location : Spain
Subject: Re: Building Microbe Stage Mon May 27, 2013 2:43 am
Yes, those callbacks are not convenient. There are slo a few other things we will need to store separatedlly:
-Which systems are running: For strategy mode, we wont need any evolution-related system, for example. That could be avoided if we just removed all components that the evolution system needs. The first option is the best in my opinion, for a bunch of reasons.
-Things stored directly inside the engines: For example, the btDiscreteDynamicsWorld stored inside the bullet engine. That object stores the world's gravity, for example, but it is not contained inside any entity. If we just restarted the bullet engine, gravity would set back to (0,0,0). This isn't needed anywhere yet, but it might be something we need to acount for.
There is something we could do for lua: Convert everything into callbacks. Normal callbacks just remain the same, and all the code that is outside of a callback will need to be called on a special even called newWorldCreation or something similar. That way, we can run all the scripts when loading a game, and if we are starting a new game, call newWorldCreation so all the cells, cameras, viewports... are created correctly. If we are loading, we will have all of those alredy created and we wont run newWordCreation. Whenever any of the other events happen, like a keypress, the apropiate callbacks will be called, as we DID load all of the lua scripts.
The only downside is that whenever you want to define a callback on an entity, or use/modify a variable in a callback, you will need to find it first (like player = findEntity("player") as the variable player wont necesarly point to the right entity, as that part of code was never runned.
Nimbal Programming Team lead
Posts : 258 Reputation : 24 Join date : 2013-03-17 Age : 39 Location : Ratingen, Germany
Subject: Re: Building Microbe Stage Tue May 28, 2013 8:49 am
I've just transcribed the specification over to the wiki. There might be some formatting errors I didn't catch. If you find some, please point them out (or fix them yourself, if you can).
Also, where do we stand on the naming of toxins / enzymes / watchamacallit? Scio, any thoughts?
Oliveriver Music Team Co-Lead
Posts : 579 Reputation : 59 Join date : 2013-01-21 Age : 25 Location : England, United Kingdom, Europe, Earth, Solar System, Milky Way, Virgo Supercluster, The Universe
Subject: Re: Building Microbe Stage Tue May 28, 2013 10:53 am
After reading through the specification, the previous few pages of this thread and watching Thriving Cheese's video on the Heat-Eater animation, I decided I'd make a video concept for the microbe editor. It's mostly based on Nimbal's concept, but I've added in a few parts similar to the prototype made by FunnyGames (such as the membrane editing, which I've changed slightly to allow more freedom than simply placing hexagons but far less than being able to place nodes anywhere). The UI I designed just for the purpose of the concept video, although if it's good enough we could work on it further to get a final UI design. This is just an animation, though, so I have no idea how easy it would be to program.
THIS IS NOT PLAYABLE. IT'S AN ANIMATION. I wish to stress that. It also looks best if viewed full screen, which videos embedded on the forum don't allow, so it's best to watch it on Youtube itself.
Open spoiler for a detailed breakdown. All are written as if it was playable.
Spoiler:
1. In the shape tab, you edit the shape of the cell by placing hexagons around a central kernel, which contains the genetic information and is mandatory for the cell. Pivot points can be added at places where only two hexagons are touching each other - the speed and angle of the pivot can be edited, resulting in a change to ATP usage. 2. The organelles tab works almost exactly as the specification says. Organelles can be placed inside the shape and vary in size. I could only think of a couple of examples of organelles, so I set the others to 'Locked'. 3. For the membrane tab, I took inspiration from FunnyGames' prototype, allowing the player to move the nodes of the outer shape. However, to prevent a huge membrane which differs from the shape made of hexagons, all nodes must be attached to the original shape. This idea came about from thinking about the cell editor ideas on the ModDB page, which shows a cell which couldn't be made with Nimbal's concept alone. 4. The external tab is used to add outer parts to cells, such as flagella and cilia. This is also similar to FunnyGames' prototype. 5. I wasn't so sure about the decoration tab, so I just added skins that can be applied to the cell and a paintbrush tool for adding custom patterns. 6. At the start you see the cursor turn on and off the line of symmetry. When placing anything with the line of symmetry on, it will be duplicated on the other side (if it costs Mutation Points, they will also be doubled). The line of symmetry can be turned off to allow asymmetrical parts. 7. Tooltips appear when you hover over something and explain more about it. These too can be turned off. 8. If you try to load a genome (cell) which requires more Mutation Points than you have, it won't allow you to. If you fossilize your genome it can be reloaded like this later on. 9. The two arrows at the bottom of the screen turn the cell either left or right. 10. Near the end you see a tooltip appear when the cursor hovers over the 'Finish' button. This displays all the things the cell must have in order to continue. If any one of these is missing, you can't exit until you add it.
Thriving Cheese Art Team Lead
Posts : 321 Reputation : 9 Join date : 2013-01-06 Age : 24 Location : Sweden
Subject: Re: Building Microbe Stage Tue May 28, 2013 11:16 am
Oliveriver wrote:
After reading through the specification, the previous few pages of this thread and watching Thriving Cheese's video on the Heat-Eater animation, I decided I'd make a video concept for the microbe editor. It's mostly based on Nimbal's concept, but I've added in a few parts similar to the prototype made by FunnyGames (such as the membrane editing, which I've changed slightly to allow more freedom than simply placing hexagons but far less than being able to place nodes anywhere). The UI I designed just for the purpose of the concept video, although if it's good enough we could work on it further to get a final UI design. This is just an animation, though, so I have no idea how easy it would be to program.
THIS IS NOT PLAYABLE. IT'S AN ANIMATION. I wish to stress that. It also looks best if viewed full screen, which videos embedded on the forum don't allow, so it's best to watch it on Youtube itself.
Open spoiler for a detailed breakdown. All are written as if it was playable.
Spoiler:
1. In the shape tab, you edit the shape of the cell by placing hexagons around a central kernel, which contains the genetic information and is mandatory for the cell. Pivot points can be added at places where only two hexagons are touching each other - the speed and angle of the pivot can be edited, resulting in a change to ATP usage. 2. The organelles tab works almost exactly as the specification says. Organelles can be placed inside the shape and vary in size. I could only think of a couple of examples of organelles, so I set the others to 'Locked'. 3. For the membrane tab, I took inspiration from FunnyGames' prototype, allowing the player to move the nodes of the outer shape. However, to prevent a huge membrane which differs from the shape made of hexagons, all nodes must be attached to the original shape. This idea came about from thinking about the cell editor ideas on the ModDB page, which shows a cell which couldn't be made with Nimbal's concept alone. 4. The external tab is used to add outer parts to cells, such as flagella and cilia. This is also similar to FunnyGames' prototype. 5. I wasn't so sure about the decoration tab, so I just added skins that can be applied to the cell and a paintbrush tool for adding custom patterns. 6. At the start you see the cursor turn on and off the line of symmetry. When placing anything with the line of symmetry on, it will be duplicated on the other side (if it costs Mutation Points, they will also be doubled). The line of symmetry can be turned off to allow asymmetrical parts. 7. Tooltips appear when you hover over something and explain more about it. These too can be turned off. 8. If you try to load a genome (cell) which requires more Mutation Points than you have, it won't allow you to. If you fossilize your genome it can be reloaded like this later on. 9. The two arrows at the bottom of the screen turn the cell either left or right. 10. Near the end you see a tooltip appear when the cursor hovers over the 'Finish' button. This displays all the things the cell must have in order to continue. If any one of these is missing, you can't exit until you add it.
That looks really awesome, and cool GUI Oliver! Would be awesome with that GUI in the game
Subject: Re: Building Microbe Stage Tue May 28, 2013 11:23 am
Looks amazing oliver, good job, it is so real looking (as in it looks like a game) And great idea to conceptualize it in an animation.
Nimbal Programming Team lead
Posts : 258 Reputation : 24 Join date : 2013-03-17 Age : 39 Location : Ratingen, Germany
Subject: Re: Building Microbe Stage Tue May 28, 2013 11:33 am
Wow! Thanks, Oliver, that looks awesome.
Just a few notes:
There are no external organelles in the specification anymore, they have been replaced by functional edges (makes organelles like the one for engulfing a little easier).
I like the symmetry feature. If I'm not mistaken, the hex grid should allow us to easily support 4-way and 6-way symmetry, in addition to the 2-way seen in your video.
I'm not sure about the pivot points. They enable a whole-microbe "wriggling" motion, but if they use ATP, they should give more than a cosmetic benefit. And I don't see them being used for motion, as that would be very hard to control.
Similar for the membrane. It's relatively easy to generate a smooth "hull" around the hex shape (at least if the algorithm I figured out doesn't produce garbage), so allowing manual manipulation would only add cosmetic benefit at the cost of implementation time.
And finally, the decoration. From a gameplay perspective, I'd really like the player to be able to look at the organelles of other cells to estimate their abilities and value, so an opaque texture over the whole microbe would obviously pose a problem.
*Looks back over the list*
Ugh. Sounds so negative despite my delight at this glimpse into the future. Really, great work!
Oliveriver Music Team Co-Lead
Posts : 579 Reputation : 59 Join date : 2013-01-21 Age : 25 Location : England, United Kingdom, Europe, Earth, Solar System, Milky Way, Virgo Supercluster, The Universe
Subject: Re: Building Microbe Stage Tue May 28, 2013 12:46 pm
I never intended it to be a completely accurate representation, so in some ways I'm actually surprised the list of corrections is so short!
I agree with you on the external organelles - adding them as functional edges sounds much better. 4-and-6-way symmetry would certainly be possible but I didn't realise that. As for the other points, some players may prefer to have more options on membranes and decorations, and the pivot point idea I based on the original cell editor idea image on the ModDB page, but it may be that they're not needed.
EDIT: Just to show the versatility of all the systems working together, here's another cell I created with it - the one from the ModDB page (note the pivot point and kernel in the same hexagon):
Subject: Re: Building Microbe Stage Tue May 28, 2013 4:58 pm
How did you do that? What program are you using for the animation?
NickTheNick Overall Team Co-Lead
Posts : 2312 Reputation : 175 Join date : 2012-07-22 Age : 28 Location : Canada
Subject: Re: Building Microbe Stage Tue May 28, 2013 6:33 pm
Amazing work Oliver! But could I make some suggestions to minimize the UI?
I think it would be better if you reduced the size of the bars on the left side of the screen. Maybe make them even smaller, and have them enlarge when hovered over. Keep the mutation points bar in the bottom left corner, and keep it the same size regardless of whether the mouse is over it.
Also, maybe this is just my OCD, but I think it would be better if the left side of the shape containing speed, ATP, compounds, etc. was a straight edge.
Other then that though, this is great work!
~sciocont Overall Team Lead
Posts : 3406 Reputation : 138 Join date : 2010-07-06
Subject: Re: Building Microbe Stage Tue May 28, 2013 11:19 pm
I think this looks great: the only qualm of mine that I haven't seen listed yet is that the UI is overall much too large, it's taking up about 2/3 of the screen, and I'd rather it take up only 1/3 or 1/4 of the screenspace. Also, some transparency would be great. Awesome work.
Oliveriver Music Team Co-Lead
Posts : 579 Reputation : 59 Join date : 2013-01-21 Age : 25 Location : England, United Kingdom, Europe, Earth, Solar System, Milky Way, Virgo Supercluster, The Universe
Subject: Re: Building Microbe Stage Wed May 29, 2013 3:23 am
@Untrustedlife I'm using a program called Xara Designer (the same program I used to make the website). It's mainly used for graphic design, as its name suggests, but it can also create HTML files and animations.
@~sciocont and NickTheNick I wasn't sure about the size of the UI. I know you wanted it to be as minimal as possible, but in the end I decided to make the UI just for the purposes of the concept and not worry too much about its eventual size.
WilliamstheJohn Regular
Posts : 409 Reputation : 10 Join date : 2012-12-26 Age : 30 Location : Third Rock from Sol
Subject: Re: Building Microbe Stage Wed May 29, 2013 5:32 am
It looks great!
~sciocont Overall Team Lead
Posts : 3406 Reputation : 138 Join date : 2010-07-06
Subject: Re: Building Microbe Stage Thu May 30, 2013 1:20 pm
Oliveriver wrote:
@Untrustedlife @~sciocont and NickTheNick I wasn't sure about the size of the UI. I know you wanted it to be as minimal as possible, but in the end I decided to make the UI just for the purposes of the concept and not worry too much about its eventual size.
That's what I figured.
WJacobC Outreach Team Lead
Posts : 220 Reputation : 17 Join date : 2013-04-05 Age : 25 Location : The United States of America
Subject: Re: Building Microbe Stage Sun Jun 02, 2013 11:49 pm
Okay, first of all I want to applaud everyone on all the great work that's been done. However, it seems the discussion here has dwindled recently. Nick has asked me to help organize the progress here and rekindle the discussions that have brought us so much progress recently.
So, first of all I want to commend Xazo-Tak for his initiative in organization. Also Oliver and Thriving Cheese for their efforts in making thing neat and organized. I can't stress this enough but we need as much of this as possible. The more organized the content, the easier the programmers can find it.
That being said, we also need more content! Cheese has done excellent modeling for microbe stage, but we need as much as we can. Cheese, keep up the fantastic work. And to you other artists, anything you can give us is fantastically helpful, especially if you can model 3D cells. Also, any audio is greatly appreciated. We now are too the point where we need sound effects. Bubbles, colliding cells and any other primordial oozy sounds will keep us moving along with microbe stage.
We've gotten so far recently. Let's not let this streak end. We will complete microbe stage.
Keep up the great work everybody.
WilliamstheJohn Regular
Posts : 409 Reputation : 10 Join date : 2012-12-26 Age : 30 Location : Third Rock from Sol
Subject: Re: Building Microbe Stage Mon Jun 03, 2013 2:03 am
Yes! We will complete the microbe stage! Never forget these three words: Fun, Simplicity, Science! Keep up on awesome work!
Nimbal Programming Team lead
Posts : 258 Reputation : 24 Join date : 2013-03-17 Age : 39 Location : Ratingen, Germany
Subject: Re: Building Microbe Stage Mon Jun 03, 2013 2:44 am
WJacobC wrote:
That being said, we also need more content!
Not to curb your enthusiasm, but I would like to caution against blindly creating assets, especially 3D models of microbes. As stated in the specification, all microbes will have a dynamic shape. That means their 3D models will have to be created on the fly.
Graphics and sound are currently under-specified. Before creating content that has a reasonable chance of actually ending up in the game, someone needs to write down a detailed description of how stuff should look, keeping in mind the constraints given by the current specification (hex-grid for microbes, compound clouds, toxin clouds, etc.). Same goes for audio. Someone will have to identify everything that should have a sound and figure out a way to make the important stuff sound important while keeping the "unimportant" sounds interesting.
I can't say when I'll be able to make time for that if nobody else steps up. There's still much to be done on the programming front, not to mention other hobbies I'd like to pursue.
Oliveriver Music Team Co-Lead
Posts : 579 Reputation : 59 Join date : 2013-01-21 Age : 25 Location : England, United Kingdom, Europe, Earth, Solar System, Milky Way, Virgo Supercluster, The Universe
Subject: Re: Building Microbe Stage Mon Jun 03, 2013 11:02 am
WJacobC wrote:
Okay, first of all I want to applaud everyone on all the great work that's been done. However, it seems the discussion here has dwindled recently. Nick has asked me to help organize the progress here and rekindle the discussions that have brought us so much progress recently.
So, first of all I want to commend Xazo-Tak for his initiative in organization. Also Oliver and Thriving Cheese for their efforts in making thing neat and organized. I can't stress this enough but we need as much of this as possible. The more organized the content, the easier the programmers can find it.
That being said, we also need more content! Cheese has done excellent modeling for microbe stage, but we need as much as we can. Cheese, keep up the fantastic work. And to you other artists, anything you can give us is fantastically helpful, especially if you can model 3D cells. Also, any audio is greatly appreciated. We now are too the point where we need sound effects. Bubbles, colliding cells and any other primordial oozy sounds will keep us moving along with microbe stage.
We've gotten so far recently. Let's not let this streak end. We will complete microbe stage.
Keep up the great work everybody.
Yes, exactly. I've noticed a lack of meaningful/relevant discussions going on over the past few days, whereas only less than a week ago we had plenty of activity on many threads. ~scio and Nick haven't been quite as active as usual over the past few days, and of course Calli and Seregon have said they'll be away for the next few weeks, but that doesn't mean progress has to stop. We've had plenty of new talented members recently, and at the moment it seems (not that I'm trying to be harsh to anybody) that their discussions are the only ones going on.
Just out of curiosity, what time does the summer vacation (and I'm using American lingo there ) start in the US? Hopefully once it starts many members will have more time on their hands to help. I'm currently quite busy, and the summer holidays (what we call them in the UK) don't start here for another six or seven weeks, so I'm afraid I won't be able to help very much (I only made the microbe editor concept because I was bored during the half term holidays/vacation and thought it might be something worthwhile to do, but now that I'm back at school I won't be able to help so much).
I'll help with collecting sounds, and if anyone else wants to they can post them in the Music List Thread.
EDIT: 100 posts!
ANOTHER EDIT: On an unrelated note, does anyone know what's happened to the podcast?
WJacobC Outreach Team Lead
Posts : 220 Reputation : 17 Join date : 2013-04-05 Age : 25 Location : The United States of America
Subject: Re: Building Microbe Stage Mon Jun 03, 2013 11:31 am
@Oliver, summer vacation has started for me and scio I know for sure. Nick said he's still got about two weeks left. Are you already back in school? That's strange. I know nothing about the podcast. I've tried contacting scio over Skype and haven't heard a thing.
@Nimbal all of what you said is true. I'm not sure I can do all that specification. My summer is actually pretty full, and I'm not the most knowledgeable on the current specification. If someone could do that it would massively speed up production.
Oliveriver Music Team Co-Lead
Posts : 579 Reputation : 59 Join date : 2013-01-21 Age : 25 Location : England, United Kingdom, Europe, Earth, Solar System, Milky Way, Virgo Supercluster, The Universe
Subject: Re: Building Microbe Stage Mon Jun 03, 2013 12:03 pm
Well, back at school after a half term holiday/vacation in the middle of the school term. The UK summer holidays/vacation start/s in about six weeks and go/es on for six weeks. Sometimes I envy US students - you have a much longer holiday/vacation.
~sciocont Overall Team Lead
Posts : 3406 Reputation : 138 Join date : 2010-07-06
Subject: Compound Visuals Wed Jun 05, 2013 5:22 pm
OK, It's time to discuss the visuals of compounds. This is going to involve both our programmers and our artists, because we not only need to let the player know what's what, we need to handle what would be >billions of particles in a simple way. The actual movement and behavior of compound entities needs to be properly defined.
Here are the options that I can think of of handling compounds.
Spoiler:
Individual Components: each compound entity is a tiny sprite about the size of this "o" When you or another cell runs into it, the compound is absorbed (unless some feature of the cell's anatomy prevents absorption). Compounds are all moving around the screen in the midground in accordance with the currents of the environment. Pros -Quite realistic, each entity is individually affected by the currents around it. -Small amounts of compounds can be generated and consumed easily. Cons -This would mean we have to handle the physics 1000s of these compound entities at once, which might rob a lot of processing power (each frame, 1000+ entities have to be updated) -This might create a rather cluttered environment. -AI doesn't know where the compound came from unless we store that information, and therefore can't swim towards/away from the source.
Density Cloud: Depending on the amount of a compound that is released from a source, an alpha map representing a cloud of the compound is released from the source. The amount released decides the decay rate of the map. As the compound spreads out and decays, the map becomes less opaque and larger. Both of these properties are affected by the amount released and the properties of the compound. When a cell is within the boundaries of the alpha map, it continuously absorbs some amount that compound (the amount it absorbs per frame will depend on the perimeter [surface area] and the alpha value of the map at each hex cell on its perimeter) - its absorption accelerates the decay of the alpha map. Pros -This is less processor-intensive than an individual entity system. The compound cloud acts as a single entity in the environment, so physics is much easier: we need only keep track of ~200 compound entities, tops. -If AI wants to head to the source of the compound, it just needs to head up the gradient. Cons -It will be fairly obvious that compounds are working this way unless we create a very clever visual system for compounds.
Unless someone has a better Idea, I'm going to assume the Density Cloud system will be used: in previous discussions, I believe Nimbal or Dani have alluded to a system like that described. If you have another system in mind, feel free to post it. Now, about that clever visual system.
I'd propose we still handle the visuals of the compounds with sprites, but the sprites in question will exist within the microcosm of the compound cloud. Let me explain.
The compound cloud is essentially a bitmap that is scaled and faded (lowered in opacity) as it disperses. We can use the alpha values of the cloud to render sprites at random positions within the cloud, then change the sprites' opacity to match (or scale with) the opacity at their position. That way, we get visual representation via shape and color without having to handle all of the sprites individually in terms of the physics. To illustrate that the sprites rendered within the cloud are not individual items, they can "twinkle" byt fading from an alpha of 0 to the max alpha at their position, then disappearing, with a new sprite being rendered somewhere else in the cloud.
Spoiler:
Last edited by ~sciocont on Wed Jun 05, 2013 7:59 pm; edited 2 times in total
WJacobC Outreach Team Lead
Posts : 220 Reputation : 17 Join date : 2013-04-05 Age : 25 Location : The United States of America
Subject: Re: Building Microbe Stage Wed Jun 05, 2013 5:54 pm
I too vote for density cloud system as well. This system will work very well. A few examples I came up with: -When a cell dies all compounds inside it would spill out in a high gradient. -As you evolve, you could mutate organelles that acted as sensors, so that when you entered a cloud a visual representation would begin to show. This could be upgraded to show other toxins that cells have excreted, etc. -You could couple this with toxins that you have evolved to show other cells where food is, or to trick other species into a deadly toxin.
This system excites me. I'm not sure about visual system. I thought just an alpha map with a more intense gradient as the compound grows more sense. Great idea, ~scio.
~sciocont Overall Team Lead
Posts : 3406 Reputation : 138 Join date : 2010-07-06
Subject: Re: Building Microbe Stage Wed Jun 05, 2013 6:00 pm
Bumping to draw attention to the proposed visuals.
NickTheNick Overall Team Co-Lead
Posts : 2312 Reputation : 175 Join date : 2012-07-22 Age : 28 Location : Canada
Subject: Re: Building Microbe Stage Wed Jun 05, 2013 6:54 pm
~sciocont wrote:
I'd propose we still handle the visuals of the compounds with sprites, but the sprites in question will exist within the microcosm of the compound cloud. Let me explain.
The compound cloud is essentially a bitmap that is scaled and faded (lowered in opacity) as it disperses. We can use the alpha values of the cloud to render sprites at random positions within the cloud, then change the sprites' opacity to match (or scale with) the opacity at their position. That way, we get visual representation via shape and color without having to handle all of the sprites individually in terms of the physics. To illustrate that the sprites rendered within the cloud are not individual items, they can "twinkle" byt fading from an alpha of 0 to the max alpha at their position, then disappearing, with a new sprite being rendered somewhere else in the cloud.
Spoiler:
I much prefer the cloud idea as well. Plus, the manner of using sprites for the graphic would be very effective too.
Home 
by 
