LE3 Can't make up my mind

[tjheldna]

Hello all,

 

I've been in two minds for the purchase. I really want to buy it and am impressed with what I've seen with the demo, however the uncertainty of upcoming features; time frames; and do we need to pay for features that I feel probably should be part of the engine keep stopping me adding LE3 to my cart.

 

The way I'm looking at it is, I have another engine licence which has all features I need apart from mobile device builds which is on the roadmap. With the other engine I have full source, GUI, terrain, saving/loading levels, networking and the license is for life among others. The thing that's keeping me from using it is LE is a whole lot easier for me to get my head around and the community is very helpful.

 

I love LE2, and want to buy 3 and now think the pricing is reasonable for an indie, however with the absence of some of those features like GUI editor; networking; no out of the box way of save/load levels and no formal documentation on what's planned is making me think I really don't know what id be buying? Without a roadmap of planned features and approximate time frames I really have no idea.

 

I must admit I haven't been reading all the posts on LE3 as I've just been too busy as of late. So if there is any documentation on any of these points please point me in the right direction and please change my mind!

[cassius]

At this low price its worth taking a chance.

 

Le2 was developed slowly and my guess is terrain etc within a year. But I am a lousy guesser so what do I know.

[tjheldna]

Part of me thought of that too, you may just be right. However if LE3 will never end up having the features I need, affordability isn't the point. It's all money wasted if I can't use it due to it not suiting my needs. I think I need a confirmation of planned features before I purchase, unless my brain decides otherwise.

 

What is this purchase getting me in the future?

 

Are we going to have a roadmap???

Edited March 3, 2013 by tjheldna

[cassius]

Well Josh has a plan but I doubt he can quote dates with any certainty, but still a stated plan of action with hoped for timeframing would be comforting.

I hope its not too long as I am VERY old and may be in my own bit of terrain before le3 is complete.

[knocks]

I personally judge products by what’s available and not what’s promised, it avoids disappointment.

[Admin]

Our next step will be based on what the Leadwerks 3 users need. Terrain will definitely be included in 3.0 (at no extra cost).

 

Here's the scene load command:

http://www.leadwerks...ap/mapload-r510

 

My end goal is to have graphics better than Leadwerks 2 with the Leadwerks 3 tools. The only question in my mind is whether we go straight back into graphics and release that in Leadwerks 3.1, or whether we do one more update that improves the tools further, and release the OpenGL 4 renderer in Leadwerks 3.2. My guess is we can improve the tools a little bit still, but we're reaching the point of diminishing returns in that area soon, since what we have now is pretty darn good.

[Juggernaut]

Why is the high end rendering effects getting shifted to OpenGL 4 only ? Older cards whose driver only support up to OpenGL 2.x to 3.x

will not be able to take advantage of the dynamic lighting and other shader based effects that are present in LE 2.5

 

Secondly completely level editing from within perspective mode would be a very helpful feature.

[Josh]

If you're going high-end, I figure go high-end.

[tjheldna]

The way I look at it these days is fancy graphics don't do it for me, I want game play! I've played a few 3d games on my galaxy, they all had not much more rendering capability than what LE3 (maybe less) and it looked great on a phone. I think if LE3 had dynamic shadows, that would probably do me for the time being and the rest can come later.

 

I think GUI tools are needed as a priority. I almost didn't buy LE2 due to that reason. The amount of time you need to do the "tweak, compile, play, exit" (loop) with widgets does my head in. There is practically an editor for everything else in LE3 now, why not an editor for the most important thing that links the player to the game (IMO)?

[Admin]

Do you mean simple projected shadows, like what Source Engine uses?:

 

Those are extremely easy for me to add, and they run fast on all hardware. We actually had those in Leadwerks 1: XD

 

If enough people ask, I can add that when I do terrain. It's not hard to implement.

[tjheldna]

Those shadows look good to me.

 

Still want a roadmap though =)

[Josh]

http://www.leadwerks.com/werkspace/blog/1/entry-1055-the-road-ahead/

[Canardia]

Can't you add stencil volume shadows, they are built-in into OpenGL somehow I think?

LE1 style shadows have the problem that they work only on static geometry, but I need them to work on models also.

[Josh]

Can't you add stencil volume shadows, they are built-in into OpenGL somehow I think?

LE1 style shadows have the problem that they work only on static geometry, but I need them to work on models also.

They work on animated characters.

 

Stencil shadows require "closed" geometry and can generate graphical artifacts if the geometry isn't correct. It was kind of a mid-way technique that got abandoned in favor of shadowmaps when GPU speeds increased.

[Canardia]

They work on animated characters.

Ok, so when I add a dummy animation (1 frame, 1 bone?) to each model, it should work?

Approved!

[Josh]

Not sure I follow you...

[Canardia]

You said projected shadows can be casted only on static geometry and animated characters.

So my conclusion was that they don't work on normal physics bodies, like cubes.

So I have to add a dummy animation for each cube so that they receive and cast shadows.

[Josh]

They are cast by any geometry. For performance reasons, I would only make them cast onto brushes and terrain...that's how Valve does it.

[Wchris]

Why is the high end rendering effects getting shifted to OpenGL 4 only ? Older cards whose driver only support up to OpenGL 2.x to 3.x

will not be able to take advantage of the dynamic lighting and other shader based effects that are present in LE 2.5

 

+1 juggernaut

 

https://developer.nv...m/opengl-driver

 

Does the above link and Josh comment mean future LE3 AAAAAAA+ renderer wil only support fermi chipsets ???

If yes I join Juggernaut in askin for a LE2 quality like AAA driver supporting AAA renders on non fermi chipsets.

[Canardia]

They are cast by any geometry. For performance reasons, I would only make them cast onto brushes and terrain...that's how Valve does it.

I would however use mostly dynamic scenery, made of models, but there the old skool level design should be sufficient to keep the performance good: smaller scenes, smart placing of models to occlude unnecessary models from view.

[shadmar]

Do you mean simple projected shadows, like what Source Engine uses?:

 

Those are extremely easy for me to add, and they run fast on all hardware. We actually had those in Leadwerks 1: XD

 

If enough people ask, I can add that when I do terrain. It's not hard to implement.

 

Add me to the ask list. Those are more than good enough.

[Guest Red Ocktober]

i agree... i think adding those would make a world of difference...

 

--Mike

[Wchris]

I love LE2, and want to buy 3 and now think the pricing is reasonable for an indie, however with the absence of some of those features like GUI editor; networking; no out of the box way of save/load levels and no formal documentation on what's planned is making me think I really don't know what id be buying? Without a roadmap of planned features and approximate time frames I really have no idea.

Because it's tjhena's topic I'll also aswer tjhena.

 

You must consider the new features LE3 cost to implement yourself versus features you want.

Let me explain. If your game needs pathfinding it'll cost you more time to do pathfinding yourself then creating a GUI. So LE3 investment on pathfinding is better than GUI because GUI requires less programming capabilities (2D only).

 

For networking, LE2 had A separate DLL that still can be used, so LE3 has actually LE2 networking capabilities. (and there's raknet also). But yes a full networking server handling networked physics could be a smart addition someday.

 

Game levelling management and level state load/save is something missing. But would probably require C++ programmers to use engine commands to store game data to let the engine handle them on load/save; this could be cumbersome.

 

I consider LE3 to be an evolution of LE2 so I logically bought it because I like the friendly community and the easy commands syntax. I'm only annoyed by the regressions (missing opengl3 renderer and terrain)

 

You have until 30 april, but don't miss your chance to get it at $199 .... just in case it's set back to $1000

 

Edit: See Josh's Blog some answers there http://www.leadwerks.com/werkspace/page/roadmap.html?showbanner=0

[gamecreator]

Add me to the ask list. Those are more than good enough.

Same. It would be really nice to hold us over until 3.2.

[tjheldna]

The roadmap Josh posted is exactly what I needed to make what I feel is a more educated decision. Also liked what I saw for the future, so yes I will make the purchase.

 

Thanks for your time to get me across the line!