wraitii

Personal speculation based on thoughts experiments are no solid ground for history or for cultural debate. Further posts about Dorians will be removed.

How about changing the door to a larger, 2-opening one like you'd expect in more modern barns? I know the reference has two smaller doors, but given our limitations this might work better to convey the idea of a corral. Generally speaking though this looks great.

I'm not sure how I like the flags but that looks good to me.

That worked. I think your approximation isn't applicable to our case because we need the result to have no variations over any system, and using float doesn't achieve that. However this might be replicated using fixed, so if you can update, it'd be to consider.

I can't unarchive the file, it gives me something that looks still compressed. I agree that using an approximation instead of an exact result would be sensible for the pathfinder, though.

Two other quick ones: the patch: pastel.patch

The pastel years:

I've merged several of your questions threads together. Please refrain from creating threads for each question, as this clutters the forum.

Please refrain from opening a new thread for every one of your suggestions. This is spam. There is a thread for suggestions, where I have merged some of your posts.

Some pointers about what is actually slow in the simulation: (this is a profiling using Apple's instruments, on a replay, so without any graphics. They are respectively showing what takes time when you update a simulation turn, and the main culprits. There were 3 very hard AIs and I profiled for the first 20 minutes.) If you sum them up, CcomponentManager iterating over entities is about 1/4th of the time. The rest is mostly pathfinding (sqrt calculation and actual pathing). Note that I've found visual actor to be fairly slow too (particularly updatevisibility), but here it doesn't appear since graphics are disabled. This means that to optimize, beyond the pathfinder, we could mainly: -find a brilliant new way to compute integer square roots. -change those std::maps for entities and use something a lot more efficient -have fewer messages. (note: JS took about 75% of the processing time in that, particularly comparing strings)

If you're using Blender, you can use the "Decimate" modifier to automatically remove triangles without affecting your mesh shape too much.

I just committed an animated (basic flying anim) falcon/hawk/raptor with 2 variations based on Unarmed's photos and a few other public domain photographs. If someone wants to improve on that (dirty work), here are the files. hawk.7z

I agree. Along with a proper options setting that actually saves settings, this will make most users adopt fancy graphics (since that's mostly no the game's slow part).

There is a good and productive discussion on running and charging on Trac. At the very least imo, this is the time to settle what we want for good, possibly write a reference document or something. Same goes for building capturing, it's been planned for ages but has never been started imo mostly because there is no clear consensus. It's also time to bring the design committee back to decide on such matters when there are issues. To already break from what you mean, a long-standing issues that we have forgotten about: the FOW is still updated when it should not be. There's also an effort on civ differentiation, we need, likewise, to push on that and decide what we want so we can start doing changes. Many small art tasks remain uncompleted here and there. I know sanderd has started working on formations, but that's tied to pathfinding as per the usual discourse.

Wiki-ed for your majesty. I've taken the liberty to change a few things here and there, add a few images, and fix a few typos.

I hadn't noticed this post… This correlates with my findings and it's extremely clear, so I think you should put it in the wiki, potentially replacing BasicAnimationImplementation which isn't really all that useful. Some notes: You haven't mentioned the "identifier" bone though, which in Blender is the name of the armature object (in your hierarchy, tiger_armature). I think it would be good to have a sample skeleton file follow your Blender example. I haven't found it necessary to bake anything. What I do is call "Pose->sample keyframes" for the frames I want, and I export it like that with a start/end. Seems to work. Then I revert the sampling before saving. Little less nice but it doesn't create a new action and it's fairly easy to revert anyhow. Some notes from http://trac.wildfiregames.com/wiki/AnimationSync might be useful somewhere, or tell the user to refer to it. Edit: you regularly mention "scaling" objects in several tutorial/tips… Imo we should try to tell users to refrain from scaling and just scale in edit mode. It makes everything nicer.

Some major points behind this whole debate: -animals in 0 A.D. fit three roles: be food source, be small nuisances to villagers so you have to worry about them, be decorative. -most sharks don't attack stuff just like that. The two main exceptions are bull sharks, which tend to attack in shallow water, and pelagic shark (such as the whitetip) which are very opportunistic predators because of food scarcity. -we do not have amphibious units. Sharks, even large great whites, can hardly sink vessels. There are recorded cases, but it's oustandingly rare, and extremely unlikely, particularly given the size of our ships. Therefore, sharks don't have melee attacks because that would be unrealistic. If we ever have amphibious units, it is a possibility that we'll add melee attacks to sharks. We'll see. However, for example, it might be good to have pelagic sharks and coastal sharks, the latter attacking only if amphibious units get in high seas. This would be a relatively realistic way of telling the player "units can't just swim forever". About lions and such: those don't really attack humans either, but could if they feel threatened and so on. For the purpose of the game and fitting point 2, they probably should.

4.2 is not supported by every graphic card (faaaar from it, actually), 2.1 is a version that is, on the other hand, pretty universally supported.

I'll make a scaled down (perhaps colored down) version for a female variation. I just scaled the textures down, looks very good. I'll probably work on my walking animations a bit (feet feel like they're sliding too much) but it's getting quite commit ready. (also, it looks positively menacing) edit: committed the base version after a few tweaks. It now has 3 idle anims, an attack anim, a death anim and a walk and run anim, making it one of the complete animals. I'll update the wiki. edit2: and of course I forgot, though I thought of it, to credit both you guys in the commit message… Ah well, I'll do it for the art repo. edit3: fixed.

All mapped up (textures still 1024x1024) I'm done with anims so it'll be committed fairly soon.

Thanks Enrique, and great model Micket! I've made a few more textures variations to see, but I think a slightly dumbed down version of the below croc is the more nile-y. The basic dark one (shown above) could be a good saltwater crocodile though. I've started animating, it's looking alright (I'm really no animator). Run animation is attached as a gif. (I've made the teeth cartoonishly big and you still basically can't see them).

Enrique: have you posted that texture somewhere?

Hm, actually, it's the older model, I didn't notice your new one. I'll just switch model and have it use the armature I've done. Shoulda checked that before, sorry.

Likely, yes. Hadn't thought it might be a member, had figured some unfindable FOSS project .

For the record, we have a crocodile model (from Historicity according to trac, no idea what that is). It needs both a texture and to be rigged/animated. I've given a shot at tweaking it slightly and rigging it, but the legs are giving me trouble. I've also optimized the shark model and changed the rigging/animations, I'll commit that after A15. I've also added an entity for it (no attacks, but it will make them roam the map and flee if attacked which looks a lot nicer).