As you might already know, the FOnline became a 2.5D game (2D background with 3D characters). Here is a link to video, screenshots and more info -> clickie. Now I would like to ask you, what do you think about Arcanum being a 2.5D game? Would you support this idea? I mean, there is a lot of advantages using 3D models instead of 2D sprites, and using correct shaders, its possible to make them look like 2D sprites, so you (probably) won't even notice that difference between 2D and 3D. Why I'm talking about this? I've been experimenting with 2.D mode in OpenArcanum as well, I did a lot of progress, but I don't have any 3D models of characters nor original animations, so its a bit impossible at this moment. I made a few 3D models before, e.g. Automaton, but they are all high poly, not suitable for realtime rendering, plus as I said before, I don't have any of original biped animations, so they are all pretty useless. Another question, is there anybody who's skilled in 3D modeling? it would be nice if we together create some basic rigged character and recreate all animations, so they could be used for old and new characters, for both Arcanum and OpenArcanum. What do you think? P.S. For those who haven't seen that automaton yet... (it's badly rigged on that video, but its already fixed) <object width="480" height="385"><param name="movie" value="http://www.youtube.com/v/kNBF6kZNq_8?fs=1&hl=en_US"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/kNBF6kZNq_8?fs=1&hl=en_US" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="480" height="385"></embed></object>
2.5D is hugely benefical in that it's relatively easy to create and modify any animations and models you want. No conversion to bitmap images is required, and the potential number of animation frames is much greater than in standard pixel animations (which usually require an entire image for each frame). As far as modifying goes, 2.5D means you can modify textures and models on the fly, and implement them directly, again, without any sort of conversions. I say, if you are positive you can do it, go right ahead. As for modelling, I'd really love to help you, but I have neither time nor talent. I've been trying to do some basic modelling in XSI Mod Tool, but this is about all I came up with:
Honestly, if you're going to go to the trouble of implementing 3D character models (I'm assuming containers and other static objects would be 2D still) then you may as well go ahead and make the entire thing 3D ala titan quest. Generating landscapes is actually the easy part in 3D - sure you may not have the artistic freedom available that you do in 2D, but if you're going for a massive, generated world where do you need that much freedom and will it ever show anyway? I'll use a reference to modern IC engines and fuel types because I feel like it; 2.5D is sorta like running a dual-fuel setup for petrol and LPG - it works, but because both possess completely different characteristics you will always have to compromise to get an acceptable, if less than optimal result. Most modern 3D modellers will let you set up a scene and very easily generate frames for each animation in however many angles you want and dump them as files - the size difference isn't even that large if you use a heavily compressed format such as DDS. With 2.5D you're shifting this conversion process from the modelling application to the game itself - this'll hurt performance, even if it's not immediately noticeable and forces you to do more work than necessary. Not to mention being a violation of the single responsibility principle (SRP), which might sound rather trite but it does explain the reasons *not* to do this quite well. My opinion, go 2D or 3D but don't mix the two - in this day & age there's really no reason to, if a machine will run your 2.5D version there's a 90% chance it'll run a decent 3D version too. Personally I went for pure 2D in my project because it's significantly easier for a lone indie developer such as myself and also fits in with the original style, or so I think anyway.
Thank you in advance. I'll definately give a try, at least :roll: ----------------- Well, there are a few disadvantages, I'm aware of, for example, prerendered images usually looks way better than realtime rendered stuff. Look at FOnline, their 3D characters actually look worse than Fallout's original prerendered characters. Also with 2D stuff, you get better speed, since you're only blitting pixels. However, I don't care about these disadvantages. I still want to switch to 2.5D. I brought this question mainly because I know that many people are skeptic about the "Arcanum 3D" idea. By the way, this reminds me a question which I saw in UberFAQ: Why is the game 2D instead of 3D? ----------------- That's a thing that I actually don't want to do, because I would like to keep things identical as much as possible. If I switch into pure 3D, then I have to also recreate all the artwork back into a 3D models, which is an enormous task, and pretty impossible one, even with a huge team of 3D artists, since it would require to exactly clone everything. So that's why only 2.5D. Its a way more possible and easier task. I can stick with 2D, but there is a lot of critter, npc and monster animations missing in the Arcanum, accorting to my latest findings. Here is the list of missing animations. Really? Didn't know, since I'm not aware about your project. Is that project Arcanum related? Do you have any website, blog or something? I would like to know more
It's the same one I've been pioneering in the Arcanum 2 suggestion forum. No website/blog (just not necessary at this stage); it's in fairly early stages yet and I still have to refactor a couple of things. Even if you just went for 2.5D you'd still have to recreate all the character models, correct? Not to mention Armours/weapons etc. which I'm assuming you'll want to do. If your sole reason for 2.5D is adding in those missing animations, I wouldn't bother, no-one bar yourself seems to have missed them up until now.
Well, having the ability to model any and every armor and weapon and apply it directly to the character (chapeau + nice suit, anyone?) is a fairly nice goal in itself, but it's true that it will require a huge amount of work on the part of the modellers and animators.
Thanks for reminding me. Do you have any (working) name for that project? Just out of curiosity Okay, but still, it will be nice to get some more info about your project, even if its still in early stages, as you claim Are you planning to finish that, right? without website/blog, I guess, you won't get as much support, so its more likely that your project will stagnate because there won't be anybody to kick/motivate you. Nor you get any help. Just saying... Also, you don't need a website/blog, just a subforum here at TA would be nice. Every Arcanum/ToEE related project I'm aware of has a official website/blog/subforum: -Arcanum: Revolution -> FORUM, BLOG, SUBFORUM. -Arcanum 3D -> SUBFORUM. -Arcanum: Dead World -> WEBSITE. -ToEE Engine Recreation Project -> WEBSITE, SUBFORUM. -ArcanumAlive Maybe TA staff would be so kind and will create such subforum for you. I would definately support this idea -> TA officially supporting such project. Correct. A lot of work, but when you look at FOnline, they are using models from Van Buren. We could do the same thing, we could use models and anims from ToEE, tweak them, and use in our projects. However, you are right about these missing anims, most of the people don't even know that they are missing, nor they care about them. I was just thinking that it would be a nice way to go, because it will be possible to bring whole Arcanum into new modding dimension, a lot of new content at wink of an eye. Anyway, I'll stick with 2D, for now. I still wonder what other member here think about this 2.5D or 3D idea.
This is far from my first project, personally I've found that in the initial stages it's significantly easier to do the ground work on my own before even making it public. Normally I wouldn't have made a post w/o anything to show for it, but I got bored one day. Webspace on it's own isn't an issue, I can quite easily set up a basic site/forum in a couple of hours but frankly what would be posted there? Some minor technical detail I did today? Something I planned to do but didn't because life/work/alcohol got in the way? It's all relatively unimportant unless the end user can see some sort of benefit from my ramblings which at this stage they can't. As for information, what did you want to know? It's written in C++, uses SFML for graphics/sound/networking (if I want it) and lua for scripting. The scripting system is virtually complete bar the API (which I've developed a framework for so components can be added/removed quickly) - it uses cocurrent lua threads for objects that share a global state but maintain private member variables so that no singular object can interrupt program flow. Each object maintains a vtable that maps string literals to lua functions so that events may be raised in either C++ or lua using the raiseEvent member function. Currently I'm trying to sort out the data file format, scripts are fine & can be carried between compatible objects but actual data needs to be serialised in maps/save games (currently thinking that they'll be one and the same) etc. Originally I had intended to write my own binary format for each that would be *very* efficient in terms of size (particularly if I used something like Huffman compression; the increase in processing time can be mitigated by processing & caching files first, at a cost of more RAM) but now I'm thinking something like json or YAML would be more suitable. My plan is to get a barebones package going, assemble a demo, release it and open the floodgates from there. Once people have a 'taste' of what I'm doing I'll setup a proper blog/forum and start recruiting etc. I have no intention of doing it all on my own, just the initial parts. If you're not concerned about copyright etc, then go ahead and use the ToEE models (keep in mind they were fairly limited in terms of types of clothes/weapons being dispayed etc, or so I recall) but leave out the pointy wizard hats or I'll be royally pissed.
I totally understand, but at least some sticky topic here at TA would be nice. I think that many people here don't have a clue about your project and its existence, nor they know what's your project all about. I think that you know much better than me, in what kind of information might be folks here interested in. Also there could a plenty of developers out there, who might be interested in your project as well. A lot of people might be able to help you, you know, and not only with programming, but also with making content and other stuff. It's okay that you don't have anything to show, as you said, its still in early stages, but having some official place where to post updates, plans and other stuff would make people think that you are serious about it and thus you would receive much more support, I think. No matter that you get drunk or raped by naked clown, there still be others who would contribute, so your project won't stagnate. Hell, you know better than me. Just saying By the way, the Arcanum: Revolution is in early stages as well, and they have site, blog, forum and subforum already. They're also receiving a lot of feedback & support from other forums, such as ArcanumClub, already. Such idea makes them so interested that they fully support that project. Plus they also managed to assemble a small team (about 5 people, I think), so that project would definitely be successful(ish). Sounds amazing! Is that going to be open-source? Or you better keep source code private and bind whole engine to lua? If I remember correctly, some people here were asking you same question before, probably because they wanted to (at least) contribute(?), so I guess that would be a clever thing to do, to setup private/public svn and let them contribute and see whats already accomplished. Actually I was thinking about using resources from ToEE demo version, not retail game. Also I would need only one player character, which would serve as base for others. I've already had a look at anims for human characters, and I must say that they're way better than Arcanum's one.
I'm positive on the idea of a 2.5D Arcanum. Furthermore, I'm impressed by the animation you've shared, ZeroBot. I cannot speak to the technical obstacles of implementing this, or the merits of choosing 3D models in lieu of sprites, but I wouldn't be discouraged by how much better pre-rendered animations look over real-time animations. Arcanum is a bit dated as it is so this sacrifice might actually blend with its style. You could also likely get away with lower polygon counts in your models for the same reason. My interest in 3D modeling is little more than passing so I am somewhat familiar with modeling in Blender. Should I decide to pursue improving my modeling ability, I'll let you know. Perhaps at some point I can help, but that point is not now.
Using models from the demo is a very good idea, I'm thinking you may have to modify them slightly but either way they would be a lot easier than working from scratch. Beyond that, I have every intention of letting others help, it's just at the initial stage it's a lot, lot easier to do the basics on your own with out anyone else jumping in and confusing everything. Like I said, once I finish the core of the engine I'll setup a public git repo and people can help out in their own time from there. It will be open source by the way
