Re: Models - Chris Kuhn
Posted: Sat Jul 06, 2013 3:03 am
Should I target it for version 0.5.4 do you think? Or v0.5.5?
Open Source 3D Space Flight Sim: Trade, Fight, Explore
https://forums.vega-strike.org/
Try 0.5.4, we can move it if necessaryDeus Siddis wrote:Should I target it for version 0.5.4 do you think? Or v0.5.5?
Maybe we could have a more relaxed version of Vegastrike that just pretended retro thrusters don't matter that much so we could add all those nice models from people that haven't understood the virtue of breaking. Canonical stuff is all fine but It's a bit swimming against the stream when we could be just riding along...klauss wrote:Now... isn't it a shame that it has no retro thrusters?
Wonder how hard would it be to add them...
It could and should be that consistent as discussed in pages 1-3 of this thread. But only if there is iron commitment to implementing a thoughtful and concise vision.klauss wrote: In any case, the game doesn't need to depict everything with utter consistency,
A little non-sequitur. The limited choice of playable ships (which is sort of close already - you can play on a Quicksilver, but what's the point?) is one thing, total ship models present is entirely another - there are also whole not-really-playable classes. And more per faction than per species. The style is third and almost unrelated subject... they don't have to be too alike because they won't be designed by the same groups in-universe... unless you give all species socialism with centralised design & approval process.Deus Siddis wrote:Concise in the sense that there must be a sane number of ships! Perhaps no more than 12 playable ships, which makes ~4 ships for each of the human, aera and rlaan species and with each such species subset of done by no more than one artist. And once an artist has finished a set, like all human aerospace craft, then all the preexisting human fighter and shuttle models get erased. Rinse and repeat this process until all playable ships are done.
That would all have to go. (Aside from the huge modular warships and transports.)TBeholder wrote: there are also whole not-really-playable classes. And more per faction than per species.
In today's world, we have many nations but most don't develop their own advanced technology, they buy it from one of the two or three most sophisticated producers in that field. Your average nation doesn't develop its own CPU or heavy lift rocket or air superiority fighter. It purchases them from someone who knows what the hell they are doing. So it is not unreasonable to expect much more of this design centralization in the distant future.The style is third and almost unrelated subject... they don't have to be too alike because they won't be designed by the same groups in-universe... unless you give all species socialism with centralized design & approval process.
But notice the low quality of that example art; to have so many models done as you see there, by one artist as you see there, would require similar corner cutting in terms of quality. How many 3D modelers do you see around here these days who could and would make even a few consistent ships that use the current engine-supported graphical features?That said? Consistent art styles are good, to a degree where it reflects common technologies. BTW, look at "Starmada: Fleet Ops" ships by J.Dugan - IMO a good example of "different styles within a common style".
Relative consistency is fine initially but... after the first model is finalized you are either dealing with absolute consistency or you have to throw that work away.klauss wrote: Well, I agree generally. Except I wouldn't require so stringent requirements regarding thrusters, but rather relative consistency. Ie: thrusters that have higher output should be proportionally bigger than their smaller counterparts, and adequate heat sinks.
Rough models aren't exactly concept art, they are the early stages of finished models, so they aren't wasted work. They are finished models minus the welding, optimization, polish, handmade UV mapping, handmade texturing or integration. So you can usually reshape them with much less effort lost. You saw some examples of these a while back.klauss wrote:That would require tons of concept art. Who's volunteering?
Rough models can't replace concept art. Some people find drawing really natural to them, and they can draw amazing detail with little effort. That's the strength of concept art: illustrating in great detail what would take maybe months to flesh out in 3D.Deus Siddis wrote:Rough models aren't exactly concept art, they are the early stages of finished models, so they aren't wasted work. They are finished models minus the welding, optimization, polish, handmade UV mapping, handmade texturing or integration. So you can usually reshape them with much less effort lost. You saw some examples of these a while back.klauss wrote:That would require tons of concept art. Who's volunteering?
I would like to see concept art. But really, if you or anyone produces a fully consistent set of good quality, surely I'll commit to integrating it in-game. I've been saying consistency should be a top priority for ages, so how could I not commit to embracing a new consistent data set?Deus Siddis wrote:Anyway I would be the volunteer; I am the only one left around to attempt it. But I would need commitments from this project. If I meet the style and quality goals we agree upon, then whatever subset of ships I complete must entirely replace the old content that filled the same roles. The old stuff has to go at that point with no more random or lower quality work allowed to replace it. Because I don't want to do however many hours of work just to add to the problem of low quality and/or eclectic content. It would have to mean an end to the low standards.
If I failed to produce the content or meet the new standards, the project would lose nothing and continue using the existing content and standards.
True. This still allows a good leeway, though. To begin with, we'd have pure thrusters vs. reactor (main-) thrusters. Now, a thruster also both produces and removes heat (which may be more or less important depend on the rest of the ship), it may be more or less optimized to more save energy (and heat production) or propellant, use different sorts of propellant or fusion fuel. A reactor-thruster may produce extra energy and allow to change power output/thrust ratio... or just give stable output and use for generation what it doesn't vent into thrust. This all may differ with intended ship purpose and depend on capabilities and circumstances of the specific ship builders. All this is tied to its resulting force/size parameters.klauss wrote:Well, I agree generally. Except I wouldn't require so stringent requirements regarding thrusters, but rather relative consistency. Ie: thrusters that have higher output should be proportionally bigger than their smaller counterparts, and adequate heat sinks.
Exactly. Many "nations", but not many truly independent factions, and with oligopolies spreading across even this much. Remember how it was when this wasn't so?Deus Siddis wrote:In today's world, we have many nations but most don't develop their own advanced technology, they buy it from one of the two or three most sophisticated producers in that field. Your average nation doesn't develop its own CPU or heavy lift rocket or air superiority fighter. It purchases them from someone who knows what the hell they are doing. So it is not unreasonable to expect much more of this design centralization in the distant future.The style is third and almost unrelated subject... they don't have to be too alike because they won't be designed by the same groups in-universe... unless you give all species socialism with centralized design & approval process.
It's somewhat schematic. But it's a part of the style, too. Does the job. Speaking of which... I am puzzled as to how "armor" and "detailed texture and bump-thingies" (other than reflecting a purely decorative paint job) can possibly be compatible.Deus Siddis wrote:But notice the low quality of that example art; to have so many models done as you see there, by one artist as you see there, would require similar corner cutting in terms of quality.
I can see the wisdom in that, but I would be willing to bet some of my time and effort that a combination of real world examples, thoughtful design and reasonably modern content creation techniques can get the job done without hand drawn concepts.klauss wrote: Concept art not only is good for documenting the final intent, but also for your own guidance: high quality models cannot be made without a clear view of the final target.
I'm no modeler, but this I've been told by professional modelers.
It is a bigger commitment than just embracing new art. It is also deleting old art on a large scale and rewriting canon to match.klauss wrote:But really, if you or anyone produces a fully consistent set of good quality, surely I'll commit to integrating it in-game. I've been saying consistency should be a top priority for ages, so how could I not commit to embracing a new consistent data set?
Alright send me the WIP (or post it to the forum), I'll take a look.Etheral, though, was working on a station based on our new modularity requirements. We should check his progress on that, and the style he was going after. I think I had some WIP he sent me somewhere.
Yes, that part is the hardest. I cannot commit to rewriting canon, I'm not involved in canon. But if an existing model gets replaced by a better one, yes, old ones get deleted.Deus Siddis wrote:It is a bigger commitment than just embracing new art. It is also deleting old art on a large scale and rewriting canon to match.klauss wrote:But really, if you or anyone produces a fully consistent set of good quality, surely I'll commit to integrating it in-game. I've been saying consistency should be a top priority for ages, so how could I not commit to embracing a new consistent data set?
I'll dig aroundDeus Siddis wrote:Alright send me the WIP (or post it to the forum), I'll take a look.Etheral, though, was working on a station based on our new modularity requirements. We should check his progress on that, and the style he was going after. I think I had some WIP he sent me somewhere.
The models in the splash screens are some of the most outdated. Editing splash screens is also a small task relative to good quality 3D graphics. Plus the screens themselves are far too low resolution for today's monitors.klauss wrote: Some existing models it'd be a shame to delete, as they're rather iconic. Especially the ones in the splash screens (unless we're also replacing those). This should be discussed at least briefly.
By the way, what would you say about deleting some of the more eclectic or low quality models immediately and reshuffling the better looking existing models around a tighter game balance? It would be a preliminary improvement that could be ready in time for the next release or just beyond it.But in essence, yes, the idea was to replace dozens of low-quality models with a handful of high-quality ones. So deleting comes with the package.