Why is this project unable to recruit and why don't developers stay?

[Induane]

The Castle isn't my work, its our other 3d dev.  Jekkar did several textures for the interior of the castle, but I didn't throw in those shots.  I did the taven in those shots, most of the texture work in the tavern is Jeks.

And, for what its worth, I applied to join twice and was ignored.   Jekkar also applied and was denied I believe. 

Here are the missing shots Jekkar referenced You'll see some Jekkar texture work in there as well: 

http://hellemans.cyanox.nl/hell/screenshots/castle/castleindoor3.JPG

http://hellemans.cyanox.nl/hell/screenshots/castle/castleindoor2.JPG

[Karyuu]

I bet that was "back in the day" I know you guys are busy with PT now, but our application process and prospect management has changed dramatically since a few months ago. So if you would like to give this another shot, you're very much welcome to. I'm handling art prospects now instead of Talad, so you'll no longer have to deal with weeks and weeks of unresponsiveness.

[Induane]

The more I look at it I just feel like I don't have the available time to dedicate to more projects simultaneously.  Most of what I learn contributing to PT applies to PS since they use the same underlying engine, which is why my PlaneShift documentation site should stay relevant.  I'm always working on new stuff and tutorials for it so I'll continue to try to contribute in that capacity instead.  As for Jekkar, his last attempt was fairly recent anyways(3-4 months, and his was not ignored, it was denied due to not enough of a portfolio or something similar.  Maybe he'd apply again, I'm not certain you never know.

On my end I have more personal reasons than just the ignored applications.   In retrospect, while I like PS, I have serious moral reservations with the licensing of the art, and from speaking with Talad, I get the feeling it is run like a job with deadlines, etc.  He is fanatical about results, which is fine because it works for you all, but I'm not sure I want a 2nd unpaid job style job.  I'd rather just have a nice quiet place to model and commit to.  That and I was told outright that I wasn't good enough anyways so I'm probably carrying a bit of a chip on my shoulder from that even though its not a grudge that should be held against PS, nor in general.  I'd be lying though if I said I wasn't a bit upset about that.

In any case that isn't that important anyways.  I've a few little things I've made I'll be putting in the contribution thread.  Its nothing in any of the lists but I figure if its wanted it will be used, otherwise it won't.  Those models are all very PS specific so they belong here since its who I made it for.  Maybe I'll brush that stuff up this weekend and post it all in one hunk.

[Karyuu]

Thanks for the further explanation, Induane. So there's one reason some people may not be interested in contributing too much - our license. Which is ultimately something we can't do anything about, so we must agree to disagree (Not that it prevents us from expanding the team at all, again - in case someone gets that impression.)

You are right about the project being run with deadlines, and while this can seem stressful, it does work out. At least, we are constantly pushing out new stuff, without pause. So yes, if someone has a more laid-back "I'd like to do this in my own time and turn it in sometime when its finished" mindset, it's not going to work too well for PlaneShift - at least definitely not with the art team. But, (addressing the general public now) if you can do a texture a week, a model in a couple of weeks, some buildings for a level in a couple of months, there's no problem.

Either way, I'm very impressed with the posted screenshots. And I'm definitely interested in hearing how long some of that work took, so I can get an idea of what's a comfortable pace for you guys?

[Induane]

Sueastside did the Castle Model in a week, not counting texturing.  The tavern, not counting textures interior and exterior was about 6 hours of total work.  I guess my issue is that I'm not sure where to sqeeze out more time.  I work 50 hours a week, have 2 kids, play in a band, and volunteer with Kiwanis.

My daily schedule:
6:00 am - Up and
Work all day
Go home, play with kids, often go to park.
9:30 - kids go to sleep.  Go for my run then work on PT.
11:30 - work through a chapter in my C#/Mono book.
Whenever - go to bed.

Exceptions are Wednesday and Friday evenings where PT and C# are replaced with band practice.

Weekends I tend to spend at the PC but doing finishing work instead of modeling.  This is laying out things, getting exports setup correctly, etc...  I try to have some downtime on the weekends to recharge the old noodle.  I chat, play PlaneShift, and write art tutorials.

One thing I am very interested in is fixing up the bronze door region.  Its a brilliant model but it has several things lacking in export that can be done to reduce crashing, improve load time, and up fps.  Also, PT has a custom shader which could be used for the waterfall to make it awesome.  Its developer said they'd be happy to share if someone asked for it (Talad Probably).  Unfortunately when I asked Talad, he said no because the BronzeDoors region was "Done."

[Karyuu]

No more going to Talad for art issues, guys - you'll have a much more successful time using me as an in-between, and I'm definitely interested in that shader! Do you have a demo of it up somewhere?

Unfortunately we won't be able to give someone an entire map level to tweak for various things, but if you have ideas on what to improve and how to do so, I'm all ears. We just can't give out assets outside of the dev team.

[Induane]

Well, I offered to sign a legally binding NDA but to no avail.  I had hoped that would be enough.  I'll make a shader demo when I get a chance - the best example I have of it now is old.

Requires Theora Codec to view:
http://vaalnor.mine.nu/Downloads/Video/waterfall.avi

The water in the pool, and the waterfall itself is using that shader.  Its not a good example of how realistic it can look though because the colors there are so exaggerated. 


What to improve in BD Region:

Convert all thingmesh objects to genmesh objects.  Thingmeshes use more resources and are slower in general.  They are also being obsoleted in CS soonish.  I can do this quickly in b2cs, really anyone could.

Implement waterfall shader.  I am familiar with this shader which takes many special things to setup correctly. I can do this as well.

Convert terrain to terrain engine.  The reason that this was not done was that cherppow had some issues with terrain export in CS and the basematerial wasn't easy to get done right.  The basemap problem was solved by our devs who wrote a basemap generator to build a basemap from the textures splatted to the terrain.  I have extensive experience with terrain export, basemap tweaking and optimization and the like.

Early versions seemed to be using the frustvis culler - I don't know if this is fixed or not, but it should REALLY be using Dynavis.

Culling in general.  All the meshes seem to be set to cull.  This is very bad, particularly on high poly objects.  On high poly objects, calculating culling is very expensive so it is better to turn culling off completely for high poly meshes and add a null mesh or invisible super low poly mesh inside the main mesh which is set to do culling instead.   This can lead to drastic FPS improvements.  I've a lot of experience in setting this up adequately since I have a tendency to overmodel at times

Lighter2.  Once converted to genmesh, I'd recommend lighting with lighter2 since it does lightmaps on genmesh objects.  Highly subdivided meshes should still be set to vertex shading for better performance while using the lightmaps for the larger genmesh objects.  I am quite good at this kind of setup so I can be of help here too.

Grass.  There is really only one grass texture there for the most part.  You really want 3-6 at least blended together in various ways to improve the look so its not so monotonous. 

Ladders.  These need adjusted so you can climb them more easily. (Have not checked if this was fixed yet).

Weird Black parts, crashing, etc:  A combination of terrain engine, genmesh objects, and lighter2 should really fix most of these problems (ones that don't stem from the particle system).

Clouds.  This applies to all maps really.  Some upper level slow moving clouds would be nice.  We have a nice little cloud shader for this as well.

[Karyuu]

Awesome, thanks for taking the time to write all this out. I'll see what I can do to help you help us with this Having some moving clouds would be a cool thing - wonder if that's something Xillix and Jeraphon see as possible within the stalactite, though. Nice water! I'd love to have that in the BD region, and I imagine it would work great on small fountains as well.

[Induane]

Both the water and waterfall are done with the same shader.  There are still issues with it we're trying to work out, but when its totally done it will be in crystalspace anyways for use.  It has some issues with working when totally vertical when viewed from an acute angle for instance.

Also, at the minimum the water in the BD region should be at least set to double sided so they don't vanish when you go under.  This is a 2 second fix.

As for clouds, I have seen it discussed, and I'm pretty certain that they decided that there could be, considering it rains.

[Jeraphon]

Having some moving clouds would be a cool thing - wonder if that's something Xillix and Jeraphon see as possible within the stalactite, though.

Yes. It is known that Yliakum has rain. Thus, clouds are not impossible. (It would also make the skybox a lot less dreary.)

[Induane]

Here is a quick ugly vid of the cloud shader in action.  As before, requires Theora codec.

http://vaalnor.mine.nu/Downloads/Video/clouds.avi

[Vengeance]

Not every project is a fit for every potential developer.  Whether it is license, goals, design, decision-making structure, whatever--it doesn't matter.  Ultimately this is a good thing because it means we aren't all working on one thing. :-)

It's a Bazaar, not a Cathedral, remember?  :-)

I do agree the screenshots are impressive.  I'm a bit disappointed in the "closed" nature of not having a public server though.  ;-)

[stfrn]

I have no idea what developers have been leaving, and I've been on the team for a year and a half. We have had 0 people consciously quit the team, 1 (DaveG) mysteriously disappear, and 1 Settings member (Kerol) switch to a GM Team Leader position because he felt he was better suited for it.

What about me? 

Why would someone stop contrbuting to a free project? Well, because they no longer have the free time, or the interest, or their personal life changed, or they got abducted by aliens, or all of the above.

[Induane]

I do agree the screenshots are impressive.  I'm a bit disappointed in the "closed" nature of not having a public server though.  ;-)

muwahahah we have a server running all the time now - even made Draklar a client last night - well... sueastside did.

[zanzibar]

Not every project is a fit for every potential developer.  Whether it is license, goals, design, decision-making structure, whatever--it doesn't matter.

I don't think that's the issue that people have with Planeshift.  From what I've seen, they aren't unhappy so much as they're unsure.  They don't feel that they know what the license, goals, etcetera truly are despite what they've been told by people like yourself.