Leadwerks Engine 2
Languages Written In:
C/C++, Java, C#, Visual Basic 6, VB.NET, Basic, D, Perl, Python, Delphi, Pascal, Fortran, Lisp
C/C++, Java, C#, Visual Basic 6, VB.NET, Basic, D, Perl, Python, Delphi, Pascal, Fortran, Lisp
- User Reviews
Leadwerks Engine 2 features a unified lighting system with dynamic soft shadows. Our advanced deferred renderer can draw dozens of dynamic lights at good speed. Advanced shader effects and support for huge terrains up to 33 million triangles delivers next-gen technology at an affordable price.
- Object-Oriented Design
- Leadwerks Engine is provided as a DLL for use with virtually any programming language, a scriptable executable, and a BlitzMax module.
- An object-oriented design is combined with a procedural command set to deliver an easy-to-use yet powerful API.
- See an overview of features here: http://www.youtube.com/watch?v=bAYPTWqgF5U
- Deferred Shading
- Advanced deferred renderer draws many dynamic lights at superior speeds.
- Fast instanced mesh rendering.
- Render buffers can be used to create post-processing effects, or render-to-texture effects.
- Render to any graphics context to create windowed applications.
- Particle System
- Depth of Field
- Motion Blur
- Particle emitters.
- Refraction and reflection effects for water and glass.
- Post-filter effects like DOF, bloom, SSAO, volumetric light scattering, and more.
- Multi-world rendering can be used to display 3D skyboxes.
- Unified lighting system with dynamic soft shadows.
- Support for point, spot, and directional lights.
- Deferred renderer can draw dozens of dynamic lights at good speeds.
- Advanced lighting effects like screen-space ambient occlusion (SSAO) and volumetric light scattering.
- Shadow Mapping
- Dynamic soft shadows with per-light control of shadow map resolution, softness, and quality.
- Masked texture effects for tree and plant shadows.
- An optional color texture can be attached to a light to simulate cloud shadows or stained glass effects.
- Customizable materials system.
- Support for normal and parallax mapping with specular reflection.
- Each material can have a number of textures and effects defined, as well as a shader.
- Choose from our library of material shaders, or create new effects with a custom shader.
- Pixel Leadwerks Engine supports vertex and pixel shaders, with support coming soon for geometry shaders on SM 4.0 cards.
- Mesh Loading
- Support for Autodesk FBX, Collada, .obj, .x, .3ds, .md3, .3dw, and .b3d formats. Load FBX and Collada files directly, or convert any of these formats to our own Game Model Format (.gmf)
- An export plugin is provided for exporting GMF files from Ultimate Unwrap 3D Pro.
- Meshes can also be constructed in code by adding vertices and triangles.
- Occlusion Culling
- An internal scenegraph silently manages and optimizes the world. This provides fast performance of culling, raycasting, and physics without complicating the end user experience.
- Hardware occlusion culling performs per-pixel culling of all meshes on the GPU.
- Support for mesh and terrain LOD.
- Forward Kinematics
- Keyframe Animation
- Skeletal Animation
- Animation Blending
- Skeletal animation system allows the user to specify any time in the animation sequence.
- Animations can be blended or transitioned in any way.
- All skinning is performed on the GPU for fast performance.
- Objects can be parented to any limb to add weapons or accessories.
- Terrains up to 33 million triangles are supported with real-time lighting and shadows.
- Alpha blended texture layers with constraints for height and slope.
- Mesh layers allow the user to paint grass and plants onto the terrain, simulating billions of grass instances.
- Modify terrain in real-time.
- Terrain editor with heightmap filters including fluvial erosion and fractal noise.
- Basic Physics
- Collision Detection
- Rigid Body
- Vehicle Physics
- Powered by Newton Archimedes.
- Support for multicore-optimized processing.
- Fast and accurate rigid body simulation with support for spheres, boxes, cylinders, capsules, convex hulls, and triangle meshes.
- Supports ball, hinge, corkscrew, slider, and universal joints.
- Built-in character controller with smooth movement and accurate stair climbing.
- Physics-based vehicles with support for any number and configuration of tires.
- 3D Sound
- OpenAL is used to provide compatibility with Windows XP and Windows Vista.
- Support for WAV and OGG files.
- 3D spatialization.
- EAX effects like reverb and echo.
- Seamless sound looping.
- Leadwerks Sandbox features real-time lighting and physics, and allows the user to place models and build terrain.
- See an overview of Sandbox features here: http://www.youtube.com/watch?v=cmN7OLy6yUY
|License Name||Price in $US||Source Code Included?||Additional information|
|Single user commercial license|
Its okay, but compared to other engines it falls far behind
Its feature set is nice, but if you're serious about game dev, there are much better solutions out there for you. Many like comparing it to the
CryEngine, but its definitely not even close to that level. It's more comparable to Blitz and Dark Basic level, with a few more fancy features.
Performance is something I would consider horrible. Its very slow, so if you are wanting to target a wide range of end users for your app then
its simply not going to happen (even with plenty of optimization).
Its fairly stable, and I didnt run into many problems, which is pretty good considering that there isnt a large team behind the development of the
Support is also very very weak. Getting straightforward answers is like pulling teeth sometimes. The forum community is also quite rude at
times, including the main developer.
Ease of use is okay if you're a programmer. If you're an artist you'll definitely have a hard time with it, and expect to run into a myriad of
problems. But even for programmers it can be problematic as well.
This engine does have plenty of potential to become something really good if a team was assembled to take it to the next level. But seeing how
other indie engines out there like Torque, Unity and UDK are moving forward quickly, LE2 is starting to fall far behind very quickly.
Overall if you're a hobbyist, then its not bad if you just want to play around with it, but dont expect to finish any game within a reasonable
amount of time, in fact there arent any completed and published games from this engine yet, so that in itself is not a good sign.
It does not work on older vid hardware and even on newer hardware it runs lke my grandmother and is pretty unstable.
It's also written in a Basic Language called Blitz Max which is an extremely nooby type language. Professional engines are written in pure C/C++.
It's no wonder the engine runs like my Grandmother on a graphics card that doesn't cost $1,000.
The documentation is also poor compared to what you could get with other engines.
I'm not quite sure who these people are trying to compare this to something like Unreal or Cryengine but they obviously have no idea what they are talking about. This does not even begin to compare.
Basic framework of an engine
Been using leadwerks for a few years now off and on. Yes the graphics are nice and on decent hardware performance is great. However it’s not really an engine more a rendering api and the tools lack any finesse and is missing many features prepared to many of its competitor’s today.
Don’t believe the buzzwords on their site that this engine is easy and fun it’s simply not true. Its pipeline is painfully slow with many developers turning to a 3rd party product for file conversion.
The sandbox editor is a box without sand, there is no support of scaling of objects, despite being tools for this no undo support, and navigating through complex scenes objects is a nightmare. The terrain and texture brushes are extremely limited in settings and no support for custom brushes.
The vegetation system has potential but the high visible popping between lods and imposters is an eyesore and shadow blending is poor and not being able to manually move objects after placement really defeats this tool as there placement on terrain is very questionable.
Programing you have many options but unless you’re a programming guru who wishes to complete the engine before commencing game development this engine is not for you as you will need to develop your own networking, AI, path finding and Gui as none of this is incorporated in the “engine”. No language is well documented, with many tutorials out of date.
Engine development is extremely slow, with the developer more interested in reinventing the wheel all the time leadwerks is simply not going anywhere fast and will always be always one step behind its competitor’s.
Not very good for forests
I recently did a serious comparison between engines for rendering large forest scenes. I tried out Leadwerks, but found it to be completely inadequate compared to some other engines. First of all, the frame rate is very slow, even on high-end graphics cards. I cannot render a large forest with good performance on good hardware, and you can just forget it on a low-cost graphics card. Second, the trees don't look particularly good, because they use only basic shading techniques. There is horrible popping between tree models and impostors that should be handled much better. The shadows really look like crud, with highly visible seams between resolutions and just disappearing at about 20 meters from the camera. I also found the shadows to be very buggy, disappearing at close distances when the camera faces certain directions. There is also no collision detection between the player and the trees, and the trees sort of float on top of terrain slopes when they should be sunk into the ground.
Aside from the tree rendering, I found that Leadwerks Engine doesn't really have many things needed for making a real game. It seems that they spent all of their time trying to make pretty scenes (and failed for anything other than still screenshots), and skimped on everything else. The online community also isn't very helpful or mature.
Overall, this engine is buggy, slow, and incomplete. Not recommended.
Didn't startup at all...
Tried on a Ati mobility 4650, which is en par with a Desktop Nvidia 8800 GT (don't believe? Research some benchmarks then...). Sandbox editor crashed on startup everytime, although everything was installed correctly.
Funny... obviously no mobile chips are supported, because the developer thinks that mobile GPUs must be slow? Good luck in creating Games which can only be played by Desktop users. If you don't know already, the laptop market is huge. Also I am not talking about cheap and slow Intel GMA GPUs...
Mr. Developer, get your engine fixed...
Good for few, bad for most.
Im not gonna go into too much detail here, just a few pro's and con's i feel is important.
+ Fast engine.
+ Has a "decent" support for todays technology.
+ The sandbox is a nice tool, but not wysiwyg, (con here) you have to program all loading rutine in yourself for the entities..
- No asset's protection, and most likely wont be added.
wich means everyone have access to your scene data and so on, wich makes it hard to distribute any games created with this engine.
- Support is only forum based (besides a few c++ tutorials, and some video displaying how to use it in c++)
The con is, it says it supports C# and many more languages, it does support C# via wrapper but its not always working as intended and there is virtually no tutorials on how to use it in C# (though the forum users are very nice).
- No source code access (and there is no FULL source code for the engine, atleast some parts) i heard.
I DO NOT RECOMMEND THIS ENGINE.
I did alot of engine testings, and i have to say the guy at leadwerks is truely unprofessional when it comes to feedback and such, if its not good feedback, he will just ignore or delete the post(s), seen this happen.
I'd say go with bigger, but cheap engines like Unity3D (now support windows) or Torque3D (upcomming engine from garagegames), and you are WAY better off.
Good engine for the price.
It's a good engine for the price, but it's really more suited to Blitmax users. This is understandable considering the engine was orginally written in Blitzmax.
The support for C++ is very poor and many of the C++ tutorials are already outdated and don't work.
I purchased the engine because it all seemed concise and easy to get into. But after attempting to create a basic scene in C++ I quickly realized that things weren't going to be as easy as expected.
If you're a Blitzmax coder then things should be much easier as there's a framework that can be built upon. Although some may find that the framework a little too complicated and it has a considerable learning curve all of its own.
Finally, the developer of the engine isn't very supportive at all. He kinda reminds me of a frustrated yoda with his cryptic one line answers. Although dealing with people over the internet can take its toll, so I do sympathize with the dude.
To sum up:- "the force use you must"
Coming along nicely
Rendering looks pretty. If you want the new editor be ready to fork over $$$. No console path.
Ease of Use:
No source access, which is needed as their are MANY features missing.
Stability & Performance:
Hardware hog. Fine if you aren't developing for a large market.
Josh lacks basic social skills. Does not take constructive criticism well at all. Tends to be very condescending.
Look at Unreal Development Kit or Unity, both of which are free and surpass Leadwerks in everyway.
Pile of junk
- Very poor C/C++ support
- Quite buggy and sluggish engine
- Very poor compatibility with video cards, especially older models
- Very unprofessional developer who thinks he is a professional but is really just an amateur wanting to be a professional. Cannot take any form of criticism even the helpful kind.
There are much better engines out there than this. Most are much more expensive.. but you get what you pay for.
Decent, for the price.
Wow, the features in this are great!
C++, VB, Bmax, LW Script, and lots more.
And with the 2.1 upgrade, the Sandbox makes terrain making too easy!
Ease Of Use:
Depending on the language you use, it can become quite burdensome.
I only used C++ and Bmax, so I will only write for those.
BlitzMax: Very easy, physics were a breeze, no built in networking commands and I hate the Bmax ones so it lacked there (this is being added though). Overall, easy to use.
Lacked proper documentation, syntax seemed off from what was in the docs, couldn't even figure out how to make a friggin camera as the closest I found to that was
And C++ worked 1 time for me, after that, I kept getting errors.
I gave this a 4 because when I owned Leadwerks, the docs were lacking, now they seem to be much more ordered and there is much more on C++.
This is why I returned the product! It was unstable and horrible!
The game itself was stable, sure, but I got the most random error after using this for only 2 weeks.
I got an error, you can search it on the Leadwerks forum, it was written by "Mike". (No I am not the same Mike who reviewed this).
The error I got was
"GL_INVALID_OPERATION" and Leadwerks wasn't outputting my OpenGL version. Happened one day and I couldn't use Leadwerks anymore after that so I had to return it.
Well depends on your card, I got 24FPS max on a physics demo with a bunch of boxes falling from the sky, I also outputted the number of boxes on the screen and on a 8400GS, I got up to I believe, 1740 boxes with physics until it became "unplayable".
I gave this part a 3 because of the error I had.
The forums offer great support, and Josh tends to reply fairly fast to most subjects, great mod and developer. Quick confirmation of order.
This engine is great, I am thinking of repurchasing it as I just ordered a HD4870 1GB card, this should run Leadwerks so easily... and if it doesn't, I am coming back and changing that Performance score to a "0" lol.
Support, like I said is great, as well as the ease of use.
The features can only be surpassed by a whole set of tools for modeling and texturing that come with it, lol.
There has never been a charge for an update either, I was given Sandbox for free, that is a steal compared to what it is worth.
At $150 and if you have a computer that can run this, you would be a fool to pass it up.
Lets see. This engine doesn't work on old hardware and even on new hardware it runs very slow and is somewhat unstable.
Every day some new direction for the engine is planned and with every change comes a new price tag.
Not much to say as I think it'll be best if you bought it then you would realise yourself why you should not have.
The developer is very arrogant, also.
The website and forums keep changing from punbb to phpbb to whatever forum Josh feels like having on any given day.
Oh yeah, the engine is supposedly written in some Basic Language called Blitz Max. I checked it out and blitzmax is a newbs oop language with used to make match 3 games. No wonder the engine runs like a dead horse on any gfx card that doesn't cost $800.
A mixed bag of rendering features, but not an engine
The Leadwerks Engine is initially very impressive but there are some major reasons not to invest any time or money developing with it.
It relies heavily on hardware and needs a higher end graphics card to be usable. This is apparently by design so while I consider it a negative, some may not.
The biggest issue is with the the developer. The problem is simply that every week he changes the entire direction of it's development. You just cant be sure next week he wont decide to work on something entirely new and abandon this project for something else he'll charge you for. It has been something he has done with many of his projects in the past.
Documentation is nonexistent, the developer would point to their Wiki but that simply consists of user's often incorrect guesses at how things work.
It is also developed using a sort of kiddie language, and its easy to wonder about how serious an engine could be written in BASIC. People have complained about a lack of source in other reviews here, but it would be useless to any real developer anyway.
Most importantly, the engine is simply NOT complete, and not close to being capable of being used for a real project. Don't tell the developer that, though. He does not accept anything remotely negative on the engine's forum and openly states his belief that his engine is already better then Unreal and CryEngine. On the other hand, reading his posts can be good for a laugh.
The CSG level editor (a hammer clone) by the same author is a very good product, however, and definitely worth the price.
Good in some cases, bad in others
First, it's good to clarify two main types of engine. Game and Rendering, game as the name would suggest is "game ready", as in, it has a lot of framework required to make a game in it. Rendering is something like OGRE, no framework at all - just graphics.
Leadwerks would best be described as an awkward mix of the two without the community or tools to really push it to the "game engine" side. C4 Engine, for example, has a lot of things for a game intergrated into the editor while Leadwerks would require those things to all be done in code. A button, for instance - in Leadwerks you would have to manually code the button to trigger an "AddBodyForce" command to the door. Even with rendering engines - OGRE, for example - there's complete art pipeline tools with exporters to the native format. With Leadwerks, the answer to all exporting troubles is to buy another program that's actually made for UV mapping to convert to their own format (rather than just including a converter to their own from Collada or FBX like most other engines).
It has a very nice renderer, it must be said. Dynamic lighting based on shadow maps with vaugely similar rendering capability to some of the high end titles. That said, some things are very awkward and the engine works on a "leave it to the programmer" philosophy. As such materials are created manually with text files, which can actually be very awkward at times.
As far as the 'game' side is concerned - well, it's kind of there (after a long while). There are still problems due to the physics solution, but these will (eventually) be fixed. Hopefully.
As far as support goes, there is a fairly good community surrounding the engine - problems are generally solved by either Josh or Lumooja. Although if a problem cannot be fixed right away it has a tendancy to be ignored, which can get very annoying. Suggestions to the engine are either taken on board (in about 0.5% of cases) or simply ignored (99.5% of cases) with no response. The most annoying thing is having a genuinely good suggestion that would make the engine better ignored.
Really, overall - there are better engines out there. Their renderers might not be quite as good. C4 Engine comes close, but it is based on stencil shadows mostly which don't look quite as good - although you get a promised upgrade to all versions of the engine (unlike Leadwerks, where it's free updates until Josh decides to charge more for it). The only advantage Leadwerks has really is price - and if the $200 is worth more hassle to work with it, then go for it by all means. Mind you, that lower prices means it comes without source code. If you're easily annoyed and have a short temper, I wouldn't try it.
overall basic somewhat poor
Something like Abyssal Engine is better than this even if more expensive and doesn't support all the latest FX because it's coded properly in pure C++ and has a much better tool set. Anything in C would be better...
This engine is poor because it is written in BlitzMax. A very amateur language. It's performance seems poor on anything other than the best most expensive 3D cards which would hurt sales of a product built on it.
It uses Newton Game Dynamics for physics which is just a free Physics library that anyone can download and use. Yet the website states that it's "the most precise" system in the world which is just plain BS.
Also there are no commercial games that have ever been published with this engine as far as I know which means no serious professional company cares to use it which is very telling in itself I suppose.
It's basically just a poorly coded rendering engine IMO which concentrates far too much on stuffing all kinds of meaningless "FX" into it rather than having good tools and proper coding conventions.
Leadwerks is a decent engine. The graphics, shaders and lighting is great. Lua scripting is simple, but a user will need to know programming before using this engine. C++ is a positive to know if you plan on working with this software.
A Year Later... A Second Look
A RE-REVIEW... ALMOST A YEAR LATER (July 09)
Leadwerks appears to be out of beta as of the end of July, 09... the author says he has incorporated everything he set to put into the 'engine', and it appears he in fact has, at least based on the results of a weeks long testing of my recently acquired license...
this warrants a review of the critical review I gave Leadwerks almost a year past.
First off, if you license this 'engine', NOW, you won't regret it... unlike a year ago when i posted my critical
review, there has been a host of improvements, and the code appears to be much more stable...
while i still don't consider this to be a game engine (unlike Torque or Crysis or Unreal it's more a
framework from which the developer will base his/her engine on), it is now much easier to implement the
your game engine using Leadwerks, and with a lil coding you will find yourself able to implement a decent
OOP based game engine in either c++ or BlitzMAX... some new under the hood logic, plus a few user
developed wizards, will help get even the most uninitiated coder's engine up and running...
there's much more documentation available than a year ago (not such a surprise), plus the author has spent considerable time whipping up some really useful video tuts... these will probably be the best way to get started...
the framework, while appearing to be fully cross platform capable, is still only available for Windows... this continues to be a sore spot as far as i'm concerned... an OSX version, or at least the ability to compile the code on a Mac would be a BIG plus... i'm sure the Linux crowd feels the same way...
there are indications that this may change... at least partially, in the near future...
as for scripting... regardless of what you may have seen in print, scripting had been abandoned for quite a
while, but... judging from what i've read in recent LW blogs, the author appears to soon be adding a lua
scripting interface to the editor in hopes that it will make it easier for non coders to use Leadwerks to
express their creative intentions... while i probably won't be using it, i think this will be a good thing...
the art assets workflow is marginal (from what little i've seen so far)... gmf is a native format that you need
to use to get models into Leadwerks... i use the UltimateUnwrap plugin which crashes UU3D every time so
far, but at least it's after each attempt, and i do get the model in gmf format ok... there are no textures...
that's no biggeee though, and there seems to be other paths available, and a few converters come with
the license... plus, reapplying the textures is super easy in code...
the sandbox editor continues to be a bright light here... i said this last year, and there has been small improvements with stability, and a few under the hood tweaks that keep this true today...
performance wise... well, it has definitely gotten better... until recently, every time i ran anything done with LW, whenever i turned the view cam quickly, i could see the meshes pop into view... that is gone in the latest release... stuff written using Leadwerks seems to run well (40 fps) on a GeForce 8600 (not the most desirable or recommended gpu for LW), and this is with most of the eyecandy turned up, and a decent amount of unnecessary vertices (result of my best programmer art) in a scene...
to summarize (without getting into the personality of the author, which was one of my biggest concerns a year ago), Leadwerks seems to have found a direction... at least you don't wake up any more and discover that the latest update has changed the api and functions without so much as a mention... i think the core api is pretty much settled on...
if you're a BlitzMAX coder, this is one of the very few viable and modern 3D tools... in my opinion, most of the others will eventually be a dead end for serious game developers... Leadwerks' similarities to Blitz3D will make it very easy to transition to, and you'll become productive with it within a few days of getting the
capable c++ coders will probably look around a bit before choosing... they'll be more comfortable picking the perfect combination of renderer, sound, physics, and net libraries to suit their needs...
finally, non coders (not a put down as a good art type is worth at least 5 programmers :) ) will jump on Leadwerks after seeing the pretty pictures... but after replacing a few of the demo models with some of their own stuff, most will give up and move on in search of a tool with that legendary make_my_game_for_me() function in it... or they'll wind up contracting out the coding...
with all this said i'm raising the ease of use grade i gave it from a 1 to a 4... i was able to get up and running in c++ and in BMax within minutes... seriously...
i'm also upping the stability and performance grade to a 4... i've gotten a few crashes of my exes, but only after exit... and most likely because i'm doing something (or not doing something) i should be (or shouldn't)...
support i'm also upping... from a 2 to a marginal 4... the tuts and docs loom big here...
in short... Leadwerks has gotten a lot better and is much more capable today than it was almost a year
ago... i was prepared to be disappointed... i wasn't...
if you're a Blitzer... and you want 'bleeding edge' 3D... get a Leadwerks license... or wait for BlitzMAX3D...
if you're not a Blitzer, you might wanna take a second look at this... like i did...
i only call em as i see em...
..i cant say much what are real reasons why some people hitting this rendering engine so low...I suppose, everything has its purpose, long as you using it properly and so do this engine, long as you using it in proper way withing given limits (every engine has it)..my experience with this system is rather positive, and i do have almost all stuff outside within range of 2.5K US$..for me, only downside is actual non multiplatform nature of engine, and long as you are happy with windows development only, i see no reason why not to give a try...community is rather helpful and contributing with tutorials..developer itself were quite responsive on pretty much everything I have been asking for, so i dont know why some folks here exposing quite negative thoughts about developer..I dont want to judge anyone but my experience so far is very very good and everything I have been asking for, were replied instantly and if not then 24 hours later(respecting countries/continents we are living at)..anyway..if you looking for puzzles, and similar games and such things..drop this engine because its no use for that..if you wanna make decent looking 3D game, then go for it..just my 2 cents..
Best engine? No, or better not yet!
I try it and i need to say is very easy..
Trust me i am a real NOOB!
But those guys help me a lot and now i can use it and i feel myself "fast".
I wanna finish my game and i feel to have the possibility to do something really great with this engine.
Now there is an evaluation kit, if you are searching for a good (and not so expensive) engine keep a look on this.
Good high-performance engine.
After brief evaluation I bought the full version one week ago.I found the API extremely well organized and powerful. Rendering, physics and systems are perfectly integrated !
Another good point is advanced ragdoll and skeletal animation support that you can only find in expensive products !
Tech support has been very good so far.
I managed to put together my prototype in 3/4 days! Its surprising for such a cheap software. I would certainly use this engine for the real game in place of licensing a multiplatform expensive 3D Engine.
Very feature rich,easy to use and yet powerfull at the same time.
I've been using Leadwerks engine for a year or so now, and I'm very impressed. It has an active development community, all the features I need, and it's completely next-gen.The forums and the wiki are a great source of information and tech support.
If you haven't checked it out yet, I strongly recommend it.
Very nice engine !
- Amazing water shader.
- Up to 33 million triangle terrain system,nice !
- Easy to use OpenGL API.
- To be able to use it with C++ gives you the hi-performance of AAA title games.
- Rivals technology that is comparable with modern game engines that costs thousands of dollars.
- The PhyGen tool makes it easy to make rigid bodies without even coding it in C++ or BlitzMax.
I know for sure that I'll be glued to this amazing engine for a very long time to come.
Leadwerks Engine 2
Well for a newbe like me the Leadwerk Engine represents a real - stable - and powerful solution. With small steps you can create, setup and render you first scene.
For BlitzMax users (like me) the engine is easy and powerful: the ideal 3dengine for BlitzMax.
In any case the engine is available for many languages, so there is not limit to your imagination and skills.
Finally the engine comes with Sandbox, a simple but very powerful world-place editor: only this program is a great added value to a great engine.
Documentation, Tutorials and support
Every week there are new informations (the wiki is growing up) and new tutorials (some are in video form!) for BlitzMax and for C language.
The forum is very responsive and positive, any question posted is considered and a reply is granted: you will never feel lost.
The engine is not a 'full-game-engine': Leadwerks Engine 2 (at the moment) handles advanced 3d graphics, physics and 3d sound. All the rest (IA, scripting, networking) must be handled by the programmer, but this is NOT a fault (in my opinion).
For the tech-fan the graphics engine itself is heavy based on Shader (SM4) so it does the best on recent hardware (but it works well - optimizing what you want to draw - on a GeForce 6600). But nowadays every engine (if you want all the great graphics fx...shadows, bump & co) a 'decent' graphic card is needed.
The engine is growing: the main lead programmer adds new features (like the God's ray) or fix/update the engine.
150 USD is the 'correct' price for this engine (and for the Sandbox you get)
Just seems to go from strength to strength
There seem to be a lot of conflicting reviews here regarding this engine, so as someone who has spent two and a half months programming with it I would like to try and give a balanced view of my experience with it to date.
The engine is openly advertised as requiring a minimum of SM3 support from your graphics card and ideally SM4. Likewise a reasonable processor is required with a dual core being recommended. This is hardly surprising when you look at the specification and capability of this engine. Why anyone moans that you can't run this engine on old hardware is beyond me, show me any game engine with a matching spec that would do so!
Having said that, I'm running with a Pentium4 and an nVidea 7900 card and I can load 300,000 polygon scenes consisting of terrain with multiple layers and over 680 models with directional lighting giving real time dynamic soft shadows on all models along with 23 point and spot lights in the scene and I'm getting nearly 30 fps. Mine is hardly state of the art hardware! I'd love to see what I could get with an 8800 (which are not exactly expensive these days).
So with that myth out of the way ....
I have to say that this is probably the most stable game/graphics engine I have ever used given that the engine is so young. The only official release so far, being version 2.0, proved to be stable full stop! I never experienced any crashes on my system. Subsequent releases have all been beta versions of 2.1 and have suffered from some issues ... but these are clearly released as beta software and as such it’s almost inevitable that there will be bugs. The current version 2.1 beta6 seems to be very stable and I fully expect the final release to be stable.
This is minimal and comprises a wiki where virtually all of the engine commands can be found along with their syntax both for BlitzMax and C, the two main languages being used. The descriptions of the functionality are somewhat minimal and could be a problem for new programmers who have no previous game engine programming experience. However, having said that, posting a sensible question on the Leadwerks forum will usually get a quick response and answer often from the developer himself. There is example code around too, both provided as part of the engine package and available from programmers within the forums.
Evolution of the Product:
This is one of the most exciting areas of the engine, the rate at which the developer has expanded the core functionality and improved areas which have been found wanting. The final release version of the 2.1 build will be a far better engine than the original version 2.0 There is obviously a knock on effect on already written code but the gains have made that acceptable and it has been fairly minimal on the whole. The Developer is hinting that the engine is approaching a more stable era where the current syntax should remain pretty much unchanged in the future although hopefully not excluding additional functionality.
My personal experience has been very good. Given that this engine has been developed and supported by one individual, so be it helped along the way by some other key members of the Leadwerks community, his support has really been first class. All the way through the beta process I have logged issues as I've found them and have found these have been dealt with in a timely and professional way. Solutions or subsequent releases with fixes often occurring within days
Engine Design and Capability:
This engine really is from my experience all it promised to be. It is indeed capable of a superb level of graphical quality and I'm sure with top quality art work capable of competing with a lot of the major professional game engines out there. I'm pretty sure the Crytek engine 2.0 served as a major influence in the design of this engine and it shows. No-one is pretending that it rivals Crytek but it really does get pretty close at times, which for a $150 dollar engine is no small achievement!
The architecture of this engine is where the developer has really hit the nail on the head. It's a joy to program with, having been very carefully thought out. I'm using it in conjunction with C++ and its a really rapid development system! The use of Newton Physics has proved to be a good choice, I am finding it pretty stable and reliable.
From a programmers point of view I initially found this troublesome although I am finding it easier as time goes by. Basically, the engine supports virtually no standard model formats, requiring everything to be converted into its proprietary .gmf file format. The only exceptions to this are the .obj and the .md3 format which the engine can load directly. However, invest in the Ultimate Unwrap 3D application and you have a way of converting just about any major format to .gmf with ease.
There is also a free tool provided to convert from the Blitz .b3d format to .gmf
But in terms of level design you need look no further, as the engine comes complete with the Sandbox editor, which although still in beta release, is proving to be a superb real time level editor with dynamic physics and lighting and shadowing enabled!
This has been a fairly positive review of the engine as I have had a mainly positive experience with it and would like to recommend it to others. I cannot believe that most Indie game designers would not find lots to like in Leadwerks Engine 2. I sincerely believe that it will only be a question of time before a game appears which will truly showcase the strength of this engine!
Leadwerks Engine Review
I am very new to the Leadwerks Engine, but I have worked with 6 different engines and Leadwerks Engine 2.0 is a very easy to use, feature rich and very stable engine. I was very surprised how easy it was to work and now with Sandbox well on it's way, it will be even easier to create games. My only concern is with the support documentation, but this will be enhanced very soon and I am sure it will be of the same high quality as the engine itself. I am very excited to continue working with the Leadwerks engine.
As a first point release it's pretty good, the graphical quality for a SM4 engine is as you would expect very good. But is it a full game engine ? That's the question. Certainly not on the same par as Crytek's technology as some forums members would have you believe.
No multiplayer network support, no AI, the physics engine is Newton and as such is very accurate but slow, just look at some of the videos dotted around the forum. It will only run on windows and there is no console support. As a RAD tool it's pretty good, but as an affordable AAA game engine, there's better out there.
The licensing is expensive, quite a few indie game engines only charge for the source license and the rest of the team, artists etc do not count as a licensable seat, however Leadwerks is per seat. It may seem cheap at face value but digging down it can end up being expensive. Couple with this a non-standard animation import format creating an inexplicable need to purchase UU3d to convert animated models for use in the engine (more expense).
My biggest bone of contention is the lack of source code, when I use a game engine I would like to get the complete source as well, especially if you want to change core engine parts rather than relying on the author. Yes I appreciate that Josh knows people can break parts of the engine, but hey, aren't we all experienced devs knowing the implications of what we take on ;)
The direction of the development seems to change on a regular basis, is there a long term design for the engine ? Josh seems to use a lot of trial and error approaches to get where he's going. Although over the last few months it seems as though LWE has finally found its course.
The forums can be quite intimidating, one or two members are agressively in your face when asking about how to do something trivial. Complete beginners stay away.
It's going to be interesting to watch the engine evolve as long as Josh doesn't end up completely changing his development direction ... again. Let's hope the pipeline cleans up as well, fbx, COLLADA anyone ...
Last edited Dec 27, 2012 at 18:43