The Open Part Mod - Week 4 Voting Started

[jimmyjack]

On the last page of the Dromoman thread, it shows you the .cfg changes you need to make to work with .15 of IR.

Dam man sorry i was just there to any didn't even notice it thanks for the info.

[Greys]

It's really sad to see yalls so down about this round. In my eyes it's rather obvious that this round had a lower turnout simply because many people find texturing to be the least pleasurable and least rewarding part of modelling, I for one have at least five parts made that I haven't bothered to unwrap yet because I know I'm not going to bother to texture them for a long time after that. So to me this seemed like an expectable results pool; I hadn't until just now considered memorial day and that makes loads of sense too, but I was really excited to see the challege when it was released, and had intended to participate but then life happened several times (as it tends to do). Precisely because of why I expected a low turnout, I think this is exactly the sort of challenges that have the most benefit, and I bet this one had the largest spectator rate.

The last few challenges didn't interest me because they asked for boring generic parts and asked so broadly that it didn't require creativity, this one forced entrants to consider exactly two specific elements of texturing and handed them a real nice model to do it on, in more than one meaning of nice.

I hope to see more real specific challenges like this in the future, and fewer "make a pretty cylinder/box/plank", as an example, focus on using less well known PartModules, and more complex Unity assemblies; ask players to make a single part probe that starts as a stack-end cone, which then uses ModuleJettison to eject at least one shielding panel, ModuleAnchoredDecoupler to separate the probe from the stack, ModuleAnimationGeneric to extend some greebles such as antennas, give it energy with ModuleDeployableSolarPanel, build in some science experiments with ModuleScienceContainer, ModuleScienceExperiment, and ModuleDataTransmitter, and top it all off with a monopropellant engine. Plus all the other things that will be necessary. Some of these would just require doing such as the science, some of them require specific organization of the model in Unity and proper matching configuration in the cfg, in the end it'd take a minimum of 3 mesh objects ignoring the greebles, 4 if you wanted gimbals, more if you had more jettisoned panels, I'd probably go with 4 panels with a sphere probe as the tip partially exposed, then have Sputnik style backswept antennas that deploy and two tiny solar panels that stick out the sides, unfurling like the 6x1 stock panel, no gimbal, maybe going with an ion so it'd be more reasonable to have a kilometer or two's dV, but I'm not sure if the panels would supply enough power to run that. All the science would be self contained but maybe animate a shutter system, so that's 17 or so mesh including the stack base which gets left behind.

That was a bit of a tangent but the thought is to try and force players to focus on something, and prevent them from being able to just kinda throw together whatever; a single part probe such as that doesn't actually contain any difficult elements, and a few of them are the same problem for different reasons; ModuleJettison, ModuleAnchoredDecoupler, ModuleDeployableSolarPanel, and optionally ModuleGimbal require that the part be assembled correctly in Unity, ModuleDeployableSolarPanel, ModuleAnimationGeneric, and optionally for ModuleDataTransmitter require animations (this could be problematic as KSP doesn't like running multiple animations at the same time).

Simplicity is boring, but complexity doesn't have to be hard. I think this time was just bad timing.

[Justin Kerbice]

many people find texturing to be the least pleasurable and least rewarding part of modelling

If any one think so, try some low-poly modelling, EVERYTHING there is about texture (~80-90% of details), you can make a car with only 100 polys, and make everything with texture, even without bump map, it add "life" to the model.

Good examples can be found on turbosquid 3D models sharing/selling website, many models are free, look for some low-poly/game ready, it's amazing what can be done with mostly textures. Here an example (~ 1200 polys), or this one (144 polys)

[Snjo]

I didn't enter this week simply because I was overloaded with coding my mod. I know I need to up my texturing game, so I wish I had had the time.

I don't think we should be afraid of making more challenging parts. It has been stated in the past that the main focus of the challenge is the modelling aspect, but there is no reason why we have to adhere to that if it's more interesting to widen the scope a bit.

After all, when people make mod parts, they will invariably have to make parts that are more than fuel tanks. There's a lot to learn about how KSP works from making something like a winglet, solar panel (or if you are adventurous, a wheel), about how to set up not just the model, but the proper hierarchy of transforms, animations etc.

You could even define sound and particle FX through the part fx node system.

Of course I personally wouldn't mind plugin work too, but THAT may be out of scope

[Greys]

If any one think so, try some low-poly modelling, EVERYTHING there is about texture (~80-90% of details), you can make a car with only 100 polys, and make everything with texture, even without bump map, it add "life" to the model.

Good examples can be found on turbosquid 3D models sharing/selling website, many models are free, look for some low-poly/game ready, it's amazing what can be done with mostly textures. Here an example (~ 1200 polys), or this one (144 polys)

Rewarding in the personal sense, not the work vs results sense. A lot of people don't get anywhere for time spent practicing texture work, they have just as little idea what they're doing when they're done with the part as they did when they started, and while some things may turn out better than others they can't replicate that intentionally.

[CaptainKipard]

Of course I personally wouldn't mind plugin work too, but THAT may be out of scope

I haven't looked into plugins yet. There seems to be some modding info on that. If there were a concerted effort to get potential competitors up to speed on KSP specific things then it might work. It would have to be a more lengthy project. I'd definitely be interested. I've dabbled in C++ and Java, and I'm sure there are others like that so it wouldn't be extremely hard, just very hard.

Why don't you start a thread to discuss the particulars of how to get this off the ground?

Rewarding in the personal sense, not the work vs results sense. A lot of people don't get anywhere for time spent practicing texture work, they have just as little idea what they're doing when they're done with the part as they did when they started, and while some things may turn out better than others they can't replicate that intentionally.

I don't agree with that.

[Justin Kerbice]

@Greys: this is just another skills modders have to learn, after that, it's kind of a personal matter and focus.

I myself learn a lot since I make KSP parts regarding textures, but it still takes me n times more than modelling most of the time.

For the challenge, I heard there is a shortage on plane's nosecones (in Prokjet's spaceplane plus dev thread) so...

[passinglurker]

Personally I hope the next challenge involves animating something cause thats a skill I'd like to learn without have to dig through dated tutorials and snippets trying to figure out what's relevant now and what was rendered obsolete way back in .16 (that and with my recent episode of trying to make a part with a built in ladder I now have a healthy distrust of all tutorials that I can't confirm were made in the last 6 months)

[Ven]

Ok, so after quite a bit of thought I came up with a few Ideas to stir the pot so too speak:

Instead of asking for a single specific part, Ask for a part(s) that perform a specific task

Ex: Move an orange fuel tank 5km on the Mun. Possible entries:

1. A heavy duty skycrane
   
2. A wheeled truck
   
3. A cannon with retro-rockets attached to the cargo
   
4. A crane
   

Pros:

1. Allows for ease of entry for even the newest modder (or for those with a limited time to work)
   
2. Allows the modder to choose to go more complex in their design
   
3. Allows for a greater expression of personal styles
   
4. Your own solution to the problem
   

Cons:

1. No Incentive to go in an interesting direction
   
2. Lacks challenge, because the lack of constraints (also reduces creative solutions imposed by these limitations)
   
3. Depending on the challenge it may be required to have multiple parts
   
4. May require the use of some mods (FAR, Deadly Reentry, IR, Ect.)
   

To counter the lack of challenge and incentive to achieve, their may be certain achievements or awards awarded to the modder for their achievement. their are two catorgoiries for this, Hard and soft awards. These do not effect the final vote.

Hard= fact, your texture resolution, the number of polys. these multiply or devide from your soft score.

Soft=opinion based. Things like appearance and functionality go here. Users vote for these categories.

For example: I make a docking boom that has 3000+ polys and a 128x .Png texture map

immediatly I get a modifier of .8* (poly), 1.5 (texture) and a -4 for using .Png.

next, I get voted for the best looking model, but when others try to dock, the vessel flips and merges with the target vehicle. I earn 30 points (best looking) but loose 10 for being crasy buggy. (20pts)

result: 20 points awarded, and I get a nice sig/patch to brag about. If I win I get to request a custom sig unique to me and me alone (obviously this depends on who's running the competition, so its up to them to offer)

To the point: Players are rewarded for looking good, but are countered to give more efficient users the opportunity to compete

Possible criteria for getting awards:

1. lowest poly count
   
2. lowest texture resolution
   

Cons:

• I need Ideas for banners, flags, sigs and or patches.
  
• and I need a week to prepare.
  

Now to more general stuff:

• the thread title needs to change! varies from competition to competition. OPM suffix? Find catchy names to get people looking
  
• simplify the front page, reduce the size of the initial post(I make a custom banner?)
  
• try to find a way not to switch threads from the work-voting thread
  
• (for me) comment more! be more involved you lazy lurker!
  

So please have a go at this, the fate of the Open Part Mod depends on it.

I appreciate the instructions as well, Cpt. Kipard. They came in handy with the part I just made.

http://imgur.com/a/OobIh

its a hub cover for the Dromoman mod.

That looks great, keep it up!

Edited June 5, 2014 by Ven

[Robotengineer]

I like this idea, also contributors should get points for providing MM configs for mod support. I disagree with some of the cons, I wouldn't say it lacks challenge, but rather provides more opportunity for creativity. Also for awards, lowest texture count while still looking good, rather than just lowest count. I could make a 128 tex and it look like crap but still win. Also on the dromoman extension, does any one know if I can use the winch module and release it under the KAS license?

[CaptainKipard]

And I like very little about it, but I'm too tired to rant right now. I just wanted to say this is too complicated. Let's just step back for a moment. There's no rush. No one says we have to start another thing this week.

[Ven]

And I like very little about it, but I'm too tired to rant right now. I just wanted to say this is too complicated. Let's just step back for a moment. There's no rush. No one says we have to start another thing this week.

yeah it is a bit too compicated, expecially with the scoring system. Unfortunatly I couldn't come up with another way of getting players to optimize their parts.

Also for awards, lowest texture count while still looking good, rather than just lowest count. I could make a 128 tex and it look like crap but still win.

If everyone votes for it then yes, but most likely not. In this case, peoples opinions > the numbers (as they multiply or divide the former by a value) to reward those who can make small textures look good.

[passinglurker]

I think the current system works in an elegantly simple sort of way the only hiccup that can develop is if the winner fails to continue it and make the next challenge. This is only the 4th thread and the first to have a lower turn out so to draw an analogy with KSP lets crash a few more rockets before we go back to the VAB and completely change things up with a new design.

As for optimising parts I bet the first guy to actually list texture res and polycount in his picture for the voting thread will win by a landslide.

[Justin Kerbice]

And what Frizzank think about this ? As the creator, what do you think about this ?

Regarding Ven suggestion: I think optimization should be left aside as this is not the purpose of such contest (and optimizations are usually a "tiny" thing, boring and time/energy consuming, with a small visible result, not really good for someone who begin).

And giving more freedom will make harder to define a true winner mostly because parts cannot be compared.