Anonymous edits have been disabled on the wiki. If you want to contribute please login or create an account.


Warning for game developers: PCGamingWiki staff members will only ever reach out to you using the official press@pcgamingwiki.com mail address.
Be aware of scammers claiming to be representatives or affiliates of PCGamingWiki who promise a PCGW page for a game key.

Glossary talk:PhysX

About this board

Not editable


Bring back PPU support (<2.8.3)

1
Mirh (talkcontribs)

https://www.youtube.com/watch?v=bT7tV_d9l3A

https://web.archive.org/web/20180325124515/http://www.physxinfo.com/wiki/Ageia_PhysX_PPU

There are a few old as the hills demonstrations that were never made to use GPU-supported versions. And I'll grant that resurrecting PPUs could still be cool "just for the lulz" of being able to say that you are offloading every[Note 1] game between 2.3.1 and 2.8.1.[Note 2] But that's it.

You shouldn't expect extra effects[Note 3] and even with the CPUs of the day you'd have been hard-pressed to justify purchasing a card on the basis of performance outside of those specific cases/levels/games designed to showcase the technology (and even this excuse was blown away once CUDA came).

Of note that nvidia's hybrid ban also affected the ageia cards. Also, it would be interesting to know what ForceDeviceType did inside their hklm registry key (disassembly shows it should accept either AG1011 or AG1021 as values).

Notes

  1. Or at least most? Putting aside that it's unclear how much hardware support was opt-in by default instead of explicitly opt-out (or even just how often applications looked for system libraries as opposed to sticking with the built-in ones), stuff like 3DMark06 could have a vested interest into keeping everything on the CPU.
  2. Hoping that late AGEIA didn't start to skimp on entire scene acceleration like nvidia (which can give you a pretty bad time if trying to force bDisablePhysXHardwareSupport=False in normal games). FWIW even games were cutting corners for some reason (e.g. ME1 supposedly with NxPrimarySceneHW).
  3. There is a causal relationship between eyecandy and hardware support in the sense that the later gives devs enough computational legroom to "add things", but it's not that these are ipso facto created by the accelerator.
Mirh (talkcontribs)

The physx gpu runtime had several CUDA branches, that didn't always evolve linearly (I tried to guess the SDK version from the minimum supported driver version when unspecified, though it may be not every dll in a given release is always recompiled):

  • CUDA 2.0: in the beginning there was just PhysX 2.7.3, which eventually expanded by mid-2009 with every version between 2.7.1 and 2.8.1 being accelerated. 2.7.1, 2.7.4 and 2.7.6 were somehow dropped after some month (CM2 allegedly relies on this one version for some reason)
  • CUDA 3.0: everything is back in, except for 2.7.2 (in a "supported set" that is already reminiscent of the present).
  • CUDA 4~4.2: from just the release notes you'd think 2012 is when this happened, but judging from the dlls this had already happened in 9.11 (after more than a year of radio silence nonetheless, and after updating their installer toolset)
  • CUDA 5(?): nothing much to add, except this was around the time they decided not to ship anymore the non-accelerated versions
  • CUDA 6~6.5: like above, you would think that it's only with the big Tesla bye bye that something significant happened, yet dlls were already there in the previous release

Other funny juicy facts are that once upon a time somebody got Curie to reportedly flag physx support... and a guy was in touch for real with nvidia for physx to work on ATI cards (fat chance yeah, even today the best hope is ZLUDA)

There are no older topics