Havok 6.0 and Open GL 3.0

Wonder when we will get these upgrades

San Francisco, CA (September 16, 200 - Havok today announced the release of Havok 6.0, its most broad suite of games middleware components to date, increasing its portfolio of software to five products. The addition of Havok Cloth™ and Havok Destruction™ to the award winning Havok Physics™, Havok Behavior™ and Havok Animation™ products represents a substantial widening of the Havok product offering. Each product forms part of an integrated suite of runtime SDKs and tools, representing the most comprehensive set of cross-platform modular tools available today, enabling game studios to deliver more compelling gameplay, character interaction and physical effects in less time.


and from Last month:

11th August, 2008 – SIGGRAPH, Los Angeles, CA The Khronos™ Group announced today it has released the OpenGL® 3.0 specification with strong industry support to bring significant new functionality to the open, cross-platform standard for 3D graphics acceleration. OpenGL 3.0 includes GLSL™ 1.30, a new version of the OpenGL shading language, and provides comprehensive access to the functionality of the latest generations of programmable graphics hardware.


Should be interesting to see how long it takes for things like HDR lighting and other Direct x 9 and 10 goodies start showing up in SL.

It took them years to upgrade from Havok 1 to 4 (no 2 or 3 in the middle).... I would not hold my breath. I would not be surprised if they drop Havok for an open source version either as they seem to be on a open source kick these days.

DirectX doesn't work in Linux so I doubt anything from it will be added to SL. Unless you mean Direct X like options through OpenGL.


SL has to stay about 2-4 years behind modern technology so it can keep a good balance of system requirements. Most people run on computers 2-4 years old and many even older.

LLs use of Havok is all server-side. If they want to move to this, it doesn't really matter what people are running at home - it's just a question of if the sim host hardware can deal with it.

A lot of work was done in the H1->H4 project to make it easier to move to future versions. If they do decide to go for the new version, it shouldn't take anywhere near as long.

Well I have heard plenty of the term "shouldn't" thrown around before without success. However I still doubt H4 to H5 or to H6 is even on their agenda. H4 brought much needed stability. What would H5 or H6 add? If nothing really useful to SL then why bother?

OpenGL 3.0 will require support from your video card maker in the form of NEW drivers, as well as LL taking advantage of them. However, the latter won't be worth doing until the former has long become ubiquitous.

and they arent using ogl just because of the linux users

its open source, and theres the mac people too

also were getting close to another age of screwball computers around every corner, starting with the pandora, which is a game console / sub micro laptop (that imo will do neither very well) thats approaching SL's minimum requirements

things of this nature are growing and expanding the ideas so theres a whole new market of 400mhz 3d accelerated celphones, 1ghz EEpc's and blah blah blah, and none of it is using microsoft and direct 3d (out of the box) its mostly linux and open gl

could be exciting ... could kill a lot of companies, either way it should be fun

no back to the op

soon (oh sorry let me translate that out of linden speak, soon = years, if ever)

Well, I'm pretty sure LL is already eyeing them with 'soon' in mind for implementing them.

theres only a few people to pester and the top dog to talk to is the master of physics
Dan Linden

hes the only one i think that would have total insight on beta feautres seeing as he leads the beta team with other lindens.

i recently got the latest havok sdk and demo.

its incredible
i mean down right incredible

they finlly eliminated the blasted prim hover and spring effect.

as the phrase goes in eve, but adopted by LL:

Soon™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™

2032

I think a lot of the things that people think are so cool about new physics systems such as the latest Havok or the next Havok are things that are done by a physics engine running in the client. Things such as explosions with thousands of realistic moving particles, dust that flies up from the ground when the ground is disturbed, skirts that twirl like real fabric.

That sort of stuff would be nice eye candy, but we can't even get textures to appear without a lengthy period of being gray. Who needs gray cardboard explosion particles and gray dust and gray swirling skirts?

The servers having new Havoc code that would let the server be able to calculate cool physics effects intended for viewing on the machine that is running the code wouldn't do anyone any good at all.

Not to say that improved physics systems on the servers wouldn't be good to have, but their benefit will show in things like better or faster collision calculations and other such which is good but not flashy and neat like the physics demos of thousands of objects bouncing and deforming as they collide that one sees in promotional videos.

So should we wait for Havok 14 or Havok 15 to be released before we upgrade to from Havok4 to Havok8 and break heaps of stuff suddenly again?

And yet if LL were to announce the upgrading of Texture Engine Mk1 to Texture Engine Mk4 there would be an equal outcry of "NEVER! It'll break too much stuff!"

Eventually everything we have today will not work at some point. I can't see SL using todays tech 20 years from now. People have to realize virtual products don't last forever.

So we should start stagnating now as there's no point in growing further?

We must not be on the same page I guess. I made a comment "why bother". Maybe that is your focus. That just meant why bother RIGHT NOW as there is a million other things far more important that need work/upgraded/fixed. I am not suggesting NEVER update Havok but there is no reason to NOW or even in the next few years. Upgrade stuff that needs upgrading. Newer version of Havok have nothing to offer right now. Wait until they do then worry about it... Is that clearer?

We won't see the SL viewer use OpenGL 3.0 until the GPU manufacturers deliver stable drivers that implement it. NVidia currently has OpenGL 3.0 support in beta (in a verion of the same 177 build that breaks the SL viewer), and so far as I know ATI doesn't even have a beta yet. I would guess that any use of OpenGL 3.0 features in the viewer is at least a year off, and even then the viewer will have to be able to cope with systems that do not support the 3.0 features or support them badly.