Extra Prim Options for Extra Prims?

Ok I don't know if this has been brought up, or if its possible. It probably would involve a lot of work.

There are a lot of ways to make extra prim types, you can make a rounded edges cube with lots of cut spheres etc. But the main problem I've found with this is lines, even if you use co-ords, do it as neatly as possible, you still end up with flashing line/joints/meets etc.

The problem with adding a prim such as this, I believe, is the polygon count. It takes more polygons to create than a normal prim, therefore drawing more sim/user resources (I don't know which, I'm not techy )..

Correct me if I'm wrong here.

But what do you think of the idea of having prims of different values in the create menu?

They could add extra prims to the create menu, but have them take up more prims? So you wouldn't have the flashing lines/joints, and it would involve less work, but it would still take up the same amount of prims (or whatever amount of prims is appropriate in terms of the polygon count)

I'm really not sure if that made any sense, if anyone got the jist of it and can word it better, please do

Anyway this post is way too long already, AND it doesn't make any sense, AND I may (meaning probably) have got the wrong idea about polygon counts and such.

Anyway, thoughts?

I like this idea a lot Zap, now that I think I understand it. Your idea is that there should be premade alternative prim shapes in the create menu, but instead of each one being one prim, as we have been used to, they might be 5 or 10 or 15. The advantage being that although they would take up the appropriate amount of resource to reflect their complexity, they would be Linden-made and therefore have no joins showing.

I don't know if it is possible either, but I can think of a few shapes that would be very useful. A pear or tear drop shape is one, rounded corners on a cube is another. Heart shape and diamond...

Maybe they need to go back and reassess some of the prims already available...twisted torii? 20 prims lol!
Cali

Yes Cali! TY hehe!

and *bump*

OK, I think you're talking about something like custom Lego blocks where it's two or more conjoined pieces.

A nice idea, but somewhat redundant since we can make the pieces already seperately. It would save a bit of time though...

I think I'd rather have new shapes or new cuts/hollows/sizes/etc. of shapes than what you suggest. But it is a nice idea.

I've been saying something similar recently.

On a laptop with a really nasty keyboard atm though, I'll go into detail later.

I do see what you're saying, but have you ever tried making a cube with rounded edges? Try, and then count the flashy lines

I imagine, pre-created prims, wouldn't have such problems.



Sounds great AJ! Let us know

Oh right, yeah. Now I've got a decent keyboard...

Okay, I'll start by addressing the rounded corners thing. I don't think the problem's so much polygon count as the devs having the time/inclination to impliment it. Having a separate prim for a shape with rounded corners doesn't seem the right way to do t either - what if you want rounded corners on a prism? I think it's be far better to have a setting for the edge roundness of relevant prims, that'd let you control how rounded the corners are too.

While I'm on the subject of prim settings, I think it'd be good to replace the box, cylinder and prism shapes with a sigle extruded ngon (okay, needs a better name) prim with a variable number of sides (say 1-16 excluding 2 or something). I'd prefer a variable number of sides for the hollow seting as well.


Now, prims of different values.

This, I think, is an excellent idea: at the moment a phantom, untextured, fullbright cube costs the same on the prim limit as a physics enabled, textured, shiny, bumpy, fully twisted torus - this is rediculous. Some of those things have negligible effect, but I thought I'd go to extremes anyway.

What I'd suggest is generating the cost of the prim from how it's various settings effect both sim and client performance: phantom prims cost nothing as far as physics is concerned so just cost relative to their complexity to be rendered; static yet solid prims cost a larger ammount relative to thier complexity including the extra processing by the physics engine; fully physical prims cost even more, but still relative to their complexity. Whether texture options should effect cost or not I don't know since I don't know how much processing you'd save by, for instance, filling the area with solid colour rather than applying a texture and reflection.

This would reward competant builders by allowing them to create far more complex builds by keeping good practices and would keep the maximum degredation to sim performance that can be caused by any particular plot the same (preventing neighbours from lagging your home). I reckon that sounds great.

The only problem I can think of now (apart from all the extra coding that'd need to be done by what seems like an already very busy team) is that some builds would go over the limit as soom as it's introduced. There are ways 'round that though.

Wow AJ! Hmmm..



I agree with you 100% there, should be an option for all prim types with the possibility.



I think I'd have to disagree there though, a lot of people hold off learning building because it seems complicated at first glance, I don't think this would help matters. I quite like how things are with this.



I agree with your thoughts on this, think something needs to be done about twisted tori anyway

We currently have buttons for shapes which are adaptations of the same prim, this could be done with the extruded ngon shape to give the same buttons. I don't think there's much of a problem there.

I built my first ever house, with 2 layers of prims, one brick texture, the other wallpaper, because I didn't know there was a Select Texture option. There weren't any prims left for internal walls, or even furniture, so it was one, big, empty, purple room.

I've been doing some building with people who've been building a couple of weeks or such, and they didn't know about the hollow or cut options.

I just think implementing some sort of ngon shape would make it all the harder to teach yourself building, which is how most of us learn now a-days.

I'd say it'd have negligable effect - it's not hard to understand the concept, people wouldn't have to worry about it to get the standard shapes. Plus it'd make things easier/better for people who do get it.

Like you said about people not knowing there's hollow and cut options - it can be totally ignored if it's not needed. It could be implimented as things are with the only thing changing being an extra input box in the options.

Would you suggest removing the taper option in favour of adding pyramid and cone prims?

Ooh, just thought - the same variable could be used to make the torus, ring and tube the same prim.

I can see yor point, and may totally not be getting what you mean. Anyway..

ATTENTION LL!!

Whoa! What I've been describing is an almost total overhaul of the prim system and there's some unaddressed issues still (how stretching effects rounded corners for instance). I think it's worth fleshing the idea out a bit more first. (Not necessarily now, I'm trying to work. )

EDIT: Just to clarify, I can see the addition of a "scale rounded corners" checkbox and rounded corners variable reducing the actual number of prim types to just two.

Hehe, just want their view on things, or what their thoughts for adding new prims etc are..

You think you'll get any kind of useful response?

No, but we'll quietly mock them later now SHHHHHH

Hehe.

Wow.

I'd previously suggested that prims should cost something like (#flat faces)/24 + (#curved faces)/2 + (1 if hollow).

I hadn't thought about a bonus for non-physical or phantom.

[the calculation was set so that building a room out of hollow cubes or flat cubes both cost the same, and a hollow sphere and hollow cube cost the same... I may have blown part of the calculation, of course, so let me pre-answer any nitpicking about the details of the calculation with "you're right, I'm an idiot, next point?"]

Existing prims could be grandfathered in, even if re-rezzed, but once you make any changes to a prim's shape, link, unlink, maybe even change the contents, it gets counted under the new rules.

Interesting... why would you suggest that hollow spheres and hollow cubes should cost the same?

And since I'm posting; a twisted torus has but one curved side.