Josh

I have a GEForce 7200, and the performance is less than 10 FPS, usually. Realistically, you will need to get a better card to use the engine. I recommend a GEForce 8800 or 9800. You can get the 8800 for less than $100.

Try updating your graphics drivers. The engine will run on your card, though it will be pretty slow.

I am looking for a javascript carousel/slideshow with the following characteristics: -Displays HTML content, not just images. -Allows me to add links for next/previous and each page. -Plays nice with jquery menus. (Doesn't appear on top of them.) Any recommendations? Thanks.

If the C# headers are truly ready, I will upload them in the SDK.

You can thanks Rich DiGiovanni for those. I have a whole set of train cars, but haven't found a use for them until now. I'll need to clean them up and offer them in the store.

I recorded some clips from Dave's latest version of his scene he is working on. Somehow he managed to get more detail and faster speed. The octree optimizations in version 2.32 help a lot here. The renderer is really good at dealing with lots of small objects strewn across a scene, even if it did take some trouble before we got it working completely right. So here's a short video we'll be using to showcase the capabilities of Leadwerks Engine: Please share this on Facebook, Twitter, YouTube, etc. We want to get lots of views on this one. Thanks! Suggestions are welcome, as I will be doing a second cut with some small improvements.

I don't mind the lack of ownership so much, but they are still charging for the service as if you are buying something. That doesn't make sense. What if you had to pay a fee to "turn on" every TV show you wanted to watch? How many shows would you end up watching? Probably not many. You just wouldn't bother. How many games do I expect to buy in the next year? Unless I have to get anything specifically for research, maybe 1-2. And they will probably be on sale on Steam. So my expected spending on games over the next 12 months is probably about $60. If I could get a wide selection of streaming games, I would be willing to pay $29.99 each month for that. My expenditures then rise to $360 for the next year. That's six times what I would spend on individual games. Getting the customer to spend more on products would be an advantage for them. The per-game fee inhibits purchases. What they have is a cable service for games, but they are selling it with an old outdated business model. They should think of it more as interactive cable television.

Yep, that's the reason I switched the default install directory to "C:\Leadwerks Engine SDK". The fix is simple, but it's easier if it just doesn't come up.

They don't need publishers. They only need game developers. I think it would have the potential to be a lot more profitable for developers. Currently, game development is a game of working two years and investing millions of dollars in a product, and then either recovering or going bankrupt in the first two weeks after release. Long-term recurring revenue streams are much better. Developers could spend their time adding new levels and features to an existing game, instead of going back to the drawing board every two years they don't go bankrupt. Because the incentives would be to create games with high replay value, quality would increase. I can't imagine a more direct way to reward good games than to pay the developer based on how much people play it.

I signed up for a free year of OnLive. I received a response email within 24 hours, so I think they just are choosing applicants based on connection speed and location. I've been extremely skeptical of this service, but I can report that it works, and it doesn't suck. I can play UT3 on my netbook. There is some slight lag, but it's not bad. Fast mouse looking can amplify it, so I slowed down my mouse speed to compensate. I had no trouble getting headshots with the sniper rifle, and the gameplay experience was very good. As far as technical performance, OnLive has delivered. As for the product itself, I don't see any reason to jump onboard yet. Wi-fi connections aren't allowed, so the fun of using my netbook anywhere in the house is mitigated. Without that, there's no reason not to use my desktop PC. The lineup of games is very limited. No Crysis, and it looks like the servers have the games running on low to medium settings. This puzzles me, because the quality settings the servers run with should be trivial. That's the whole point of OnLive. With the games and settings they are offering right now, you could get the same results with a five year old PC. The pricing model is where OnLive fails. After the first year, you pay $4.95 a month for the service. On top of that, you have to "buy" each game you play, but you don't really own it, and the game may be taken down after 2013. With a revolutionary new delivery system like this, the choice to use an outdated purchasing model is baffling. What if you had to pay a monthly cable TV bill, and then "buy" each show you wanted to enable? It doesn't make sense. OnLive should charge a higher monthly fee, around $29.99-49.99 for access to all games on their servers. OnLive would take their cut of the sale, and the rest would be divided up among the game providers based on the percentage time players spent in their game. This would encourage developers to make games with long-term replay value. It would create a continuous revenue stream for developers, instead of just making money off one-time purchases. It would lessen the complaints people have when an old game if retired off OnLive's servers. And finally, it would deliver a lot of value to the customer. Which would you rather pay for, a $200 XBox plus $60 for a game, or pay $40 this month and play every game? The choice for a lot of people would be obvious. Hopefully OnLive will come to see this. I am surprised it isn't already obvious to them. As of right now, I don't know who would use this service. Who is the target market? People who can't afford a decent PC, but still have $50 to spend on a game they don't own? Maybe some Mac owners will pay for it. I think they should be targeting console players who want the ultimate graphics, but the OnLive console isn't available yet. Still, OnLive has delivered from a technical standpoint, and they may become something amazing and unique. I look forward to seeing what it might grow into.

Sync your SDK. Cubemap reflections have been added into the official shaders.

What shaders do you have selected?

That's the plan.

The specular intensity is controlled per pixel in the normal map alpha channel.

Read the first post in this thread.

Please let me know if you have any other problems.

It's uploaded. Run the updater to get the new shaders.pak file. Plain cubemapping: texture0 = cubemap vertex shader= mesh_cubemap.vert fragment shader = mesh_cubemap.frag Diffuse texture + cubemapping: texture0 = diffuse texture1 = cubemap vertex shader= mesh_diffuse_cubemap.vert fragment shader = mesh_diffuse_cubemap.frag EMBM: texture0 = diffuse texture1 = bumpmap texture2 = cubemap vertex shader= mesh_diffuse_bumpmap_cubemap.vert fragment shader = mesh_diffuse_bumpmap_cubemap.frag

You can easily use a flat normal map for that. Cubemapping is in the default shaders. It's using refraction cube mapping, which is how the skybox is displayed. If you are want reflective cube mapping, that can be added pretty easily. I actually need to do it anyways for another project, so I will add it in right now.

Can you please provide more information on what you are trying to accomplish? You should not have to edit shader files at all. You would never want to have specular reflections without bump mapping.

Thank you.

Hilarious and inspiring, at the same time. The sickle and hammer is one of the focal centerpieces that really ties the whole thing together. Signs of former industrial might, now decayed. What mysteries lie in the ruins of the zone? I wonder if we can get GSC to use Leadwerks 3.0 for STALKER 2? Wouldn't that be cool?

I decided to focus on the term "scripting" instead of advertising Lua. The third-party script we use should be incidental. The point is it is scripting for Leadwerks.

Of course.

To use occlusion culling, individual instances have to be drawn without instancing. This makes occlusion culling slower than not using it for most objects. Only expensive-to-render objects like lights and animated meshes are worth using occlusion culling on. In version 3.0, a more large-scale occlusion culling system will be used to discard chunks of scenery. But you wouldn't really want me to implement this right now would you, considering how many problems we had with the release of 2.32 initially? Do we really need more objects flashing on and off right now?

You have 20 mb now.