• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Microsoft Trademarks Direct Physics - HAVOK Rebranded?

Raevenlord

News Editor
Joined
Aug 12, 2016
Messages
3,755 (1.18/day)
Location
Portugal
System Name The Ryzening
Processor AMD Ryzen 9 5900X
Motherboard MSI X570 MAG TOMAHAWK
Cooling Lian Li Galahad 360mm AIO
Memory 32 GB G.Skill Trident Z F4-3733 (4x 8 GB)
Video Card(s) Gigabyte RTX 3070 Ti
Storage Boot: Transcend MTE220S 2TB, Kintson A2000 1TB, Seagate Firewolf Pro 14 TB
Display(s) Acer Nitro VG270UP (1440p 144 Hz IPS)
Case Lian Li O11DX Dynamic White
Audio Device(s) iFi Audio Zen DAC
Power Supply Seasonic Focus+ 750 W
Mouse Cooler Master Masterkeys Lite L
Keyboard Cooler Master Masterkeys Lite L
Software Windows 10 x64
As you might recall, Microsoft bought HAVOK from Intel back in 2015, promising to "add Havok's IP to its existing tools and platforms, including DirectX 12, Visual Studio, and Azure." Well, it would seem we are seeing the fruits of that particular seeding, with Microsoft having trademarked "Direct Physics".

With Microsoft having previously talked about integrating HAVOK with its DX12 API, that is probably the most probable scenario for this trademark. A tighter, in-DX12 integration could possibly allow for the physics workflow to have increased performance under the API, which is something we can all get behind of. However, this also begs the question as to what exactly happens to HAVOK licensing in the process. I myself wouldn't expect Microsoft to put its HAVOK tools and libraries behind a DX12 implementation wall - the number of companies who license those libraries aren't few in number. So my guess is that Microsoft is simply rebranding the HAVOK middleware for integration under its DX12 API, which could mean opening up its libraries to any game that makes use of DX12.





This is where things get interesting, because this would allow would-be licensees of HAVOK to take one of two paths for acquiring the tool-kits and workflows on the physics middleware: either license it, or get it for free by developing their games under DX12. If so, and I would wager this is the scenario playing out at Microsoft, the company can still monetize its HAVOK IP to anyone who doesn't want to implement DX12, while at the same time, increase the likelihood of developers adopting DX 12 as their development API of choice. This would naturally help Microsoft in gaining the upper hand against Vulkan - arguably the better choice from a developer's perspective due to its greater cross-platform flexibility. At the same time, Microsoft could ensure more games are developed under its poster-child and ecosystem-foundation (both in consoles and PCs) DX12.

As a reminder, many high-profile games make use of the HAVOK libraries and workflows, such as 343 Industries' Halo 5: Guardians, Bungie's Destiny, and more recently, Nintendo's The Legend of Zelda: Breath of the Wild, just to name a few.



View at TechPowerUp Main Site
 
Last edited:
I hope it will run great on all hardware, regardless of manufacturer.

Remember Half-Life 2? It was one of the first notable games using Havok and the physics kicked ass:

 
I hope it will run great on all hardware, regardless of manufacturer.

Remember Half-Life 2? It was one of the first notable games using Havok and the physics kicked ass:


Yeah, I loved the physics on that game. Really made the world come to life. That gravity gun and the slicing disks of doom...
 
That's a real kicker in Nvidia's PhysX nuts.
 
Well freaking FINALLY. Took them a freaking decade. This is what I've been waiting for since release of Half-Life 2 which showed how much physics can impact gameplay and immersion. Even if it was just for gimmicky manipulation of objects and puzzles, it felt amazing.

You can't incorporate HW accelerated physics into gameplay if not all users can utilize it, which is why PhysX remained at "gimmick" level of integration with desperate gimping of CPU physics to make it look better and more meaningful. But with standardization, we can see that in near future.

Now I'm waiting again for the return of damn DirectSound 3D again... Audio has been gimped and neglected for WAY TOO LONG. This generic flat sounding software audio is making me vomit...
 
Microsoft should have bought Havok instead of Intel but Intel wanted Havok more. Hopefully this will lead to a big transition to more physics-based games. It sucks that there's really no open source equivalent which means Windows is going to solidify it's position as the gaming platform for a long time because of this.


Edit: Just looked at when Havok's last stable release was: September 14, 2011. *insert Intel hate here*


Red Faction Guerilla was the best example of what Havok could do.
 
Last edited:
Microsoft should have bought Havok instead of Intel but Intel wanted Havok more. Hopefully this will lead to a big transition to more physics-based games. It sucks that there's really no open source equivalent which means Windows is going to solidify it's position as the gaming platform for a long time because of this.

Well, OpenCL exists so one could make OpenPhysics based of that. Maybe one day. Still, this is a good start. It's so annoying everything has evolved, but physics are still stuck at ragdolls and gimmicky non essential effects.
 
Last edited:
Problem with Bullet is that literally NO ONE is using it. AMD made huge fuss about it for HW acceleration and then nothing happened from that. So lame.
 
I hope it will run great on all hardware, regardless of manufacturer.

Remember Half-Life 2? It was one of the first notable games using Havok and the physics kicked ass:

This 14 years old engine had 100x better animations than the latest Frostbite engine running ME:Andromeda. Amazing.
 
I hope it will run great on all hardware, regardless of manufacturer.

Remember Half-Life 2? It was one of the first notable games using Havok and the physics kicked ass:


This makes me simultaneously proud of Valve for making such an awesome game engine, yet entirely annoyed because they can't finish the story line they worked so hard to create.
 
Anything that moved the needle further away from PhysX can only be a good thing.
 
Well, OpenCL exists so one could make OpenPhysics based of that. Maybe one day. Still, this is a good start. It's so annoying everything has evolved, but physics are still stuck at ragdolls and gimmicky non essential effects.

that solution exist since 2010.
 
Problem with Bullet is that literally NO ONE is using it. AMD made huge fuss about it for HW acceleration and then nothing happened from that. So lame.

the are several games using the engine. some of them is even big triple A tittles like GTA V. but none of them actually use the GPU accelerated feature. AMD partnered with Bullet to make vendor neutral solution but they did not help Bullet in "promoting" game developer to use the gpu accelerated features in their games. they announced their partnership in the beginning and then nothing. AMD just hope that game developer will automatically choose Bullet over nvidia PhysX like it is a natural thing to do.
 
Anything that moved the needle further away from PhysX can only be a good thing.

it is annoying to many people that gpu PhysX is exclusive to nvidia only but outside that PhysX has been providing a very good alternative to many game developer over havok . one thing that make PhysX quite popular in the past is they were much cheaper to license than havok (probably still is). and right now CPU based PhysX is more open than Havok.
 
Bullet is really the only open-source option:
https://en.wikipedia.org/wiki/Bullet_(software)
There is other open source physic engines available besides Bullet.

Here's a list:
Newton Game Dynamics is an open source physics engine for realistically simulating rigid bodies in games and other real-time applications.

Box2D is a free open source 2-dimensional physics simulator engine written in C++.

Open Dynamics Engine (ODE) is a physics engine written in C/C++. Its two main components are a rigid body dynamics simulation engine and a collision detection engine. It is free software licensed both under the BSD license and the LGPL.

Tokamak Game Physics SDK is an open-source physics engine. At its beginnings, Tokamak was free for non commercial uses only. Since May 2007, it has become open sourced under a BSD License. Now it can be used under BSD or Zlib license.

There's even a open source physics abstraction layer available that allows switching through the above physics engine including Bullet. https://en.wikipedia.org/wiki/Physics_Abstraction_Layer

Edit: Click on the physics engines I listed above to find out which commercial games used a open source physic engine btw.
 
Last edited:
None of them are popular though which was my point. Also, they're not actively supported (Bullet is):
Newton: 2015
Box2D: 2014
ODE: 2015
Tokamak: 2008

I think by Direct Physics simply existing, the open source community will coalesce around a standard for developers to target on Android, Windows, OSX, and *nix platforms. Bullet looks like it would probably be the best starting point for that because it is already OpenCL accelerated. I think if Khronos Group and SIGRAPH pushed Bullet (or some other OpenCL physics API), it would take off.
 
Checking Github shows activity for Newton. https://github.com/MADEAPPS/newton-dynamics

Same for ODE on Bitbucket. https://bitbucket.org/odedevs/ode/

Found a couple more going by the names SPlisHSPlasH, and PositionBasedDynamics.
SPlisHSPlasH is an open-source library for the physically-based simulation of fluids. The simulation in this library is based on the Smoothed Particle Hydrodynamics (SPH) method which is a popular meshless Lagrangian approach to simulate complex fluid effects.

PositionBasedDynamics is a C++ library for the physically-based simulation with position-based constraints.
http://www.interactive-graphics.de/
 
Just a fluid library isn't enough. They need an all inclusive physics library for better quality control and uniformity.

Like I said, I think Direct Physics will be the spark to make it happen. PhysX has always been too niche for open source developers to get organized.
 
This makes me simultaneously proud of Valve for making such an awesome game engine, yet entirely annoyed because they can't finish the story line they worked so hard to create.
Just want to give you props your user name :rockout:Oh and you're not alone with wanting that ending
 
it is annoying to many people that gpu PhysX is exclusive to nvidia only but outside that PhysX has been providing a very good alternative to many game developer over havok . one thing that make PhysX quite popular in the past is they were much cheaper to license than havok (probably still is). and right now CPU based PhysX is more open than Havok.

The shift to PhysX happened when Intel bought Havok and basically took it off market. The fact Unreal uses it by default means any game running Unreal will use PhysX as well. Havok is gonna have a tough time returning, especially if NVIDIA would finally open it up for hardware acceleration for other vendors (making it Direct Compute based instead of CUDA). That could again make it a tough competition for Havok. I frankly don't care if it's Havok or PhysX, for as long as both, AMD and NVIDIA gamers can experience the same level of physics details. When that happens, the interactivity in games will explode beyond anything we've imagined so far. Physical manipulation of the worlds will become an integral part of gameplay, meaning we'll be able to finally see some proper game puzzles and ways to defeat enemies using environment and objects in it. Imagine Deus Ex that already offers loads of ways to finish a task, adding a physics layer to that. Or in Hitman games where we'll finally truly be able to make executions in most creative ways. It's gonna be maaaaad when that happens. Soon I hope...
 
This could really help. It will either kill PhysX, or force Nvidia to unlock PhysX. PhysX was promising great physics in modern games. It ended as a limiting factor for physics in systems that didn't have an Nvidia GPU.

Problem with Bullet is that literally NO ONE is using it. AMD made huge fuss about it for HW acceleration and then nothing happened from that. So lame.
The reason why I was opening champagnes when Nvidia was buying Ageia. I knew that if AMD had bought it, it would just disappear. But Nvidia chose to use Ageia's physics in such a way that I am cursing them in every post.

This 14 years old engine had 100x better animations than the latest Frostbite engine running ME:Andromeda. Amazing.
It looks more amazing if you consider the hardware used in that presentation.
 
Last edited:
Back
Top