Topic on Glossary talk:PhysX
This page shows the changes between two versions of a post by Mirh in the topic "Bring back PPU support (<2.8.3)" on Glossary talk:PhysX.
You can see other versions of this post at its history page.
Line 7: | Line 7: | ||
You shouldn't expect extra effects{{note|There is a causal relationship between eyecandy and hardware support in the sense that it gives devs enough computational legroom to "add things", but it's not that these are ipso facto created by the accelerator.}} and even with the CPUs of the day you'd have been [https://www.anandtech.com/show/2393/4 hard]-[https://web.archive.org/web/20080824053807/http://www.thenextbench.com/hpg/blog/article?message.uid=10771#M442 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 [https://web.archive.org/web/20190806042912/https://www.pcgameshardware.de/Mirrors-Edge-Spiel-4510/Tests/Physics-acceleration-via-the-GPU-673911/2/ was] [https://web.archive.org/web/20090130082728/http://www.pcgameshardware.com/aid,674175/Reviews/Mirrors_Edge_benchmarks_with_Radeon_HD_4870_plus_Ageia_card_-_Alternative_to_Nvidia_Physx/ blown] [https://www.geeks3d.com/20080820/physx-performance-gpu-vs-ppu-vs-cpu/ away] [https://web.archive.org/web/20090901181947/http://www.firingsquad.com/hardware/physx_performance_update/ once] [https://techgage.com/article/nvidias_physx_performance_and_status_report/2/ CUDA] [https://hardforum.com/threads/cryostasis-cpu-vs-ppu-vs-gpu.1376657/ came]). | You shouldn't expect extra effects{{note|There is a causal relationship between eyecandy and hardware support in the sense that it gives devs enough computational legroom to "add things", but it's not that these are ipso facto created by the accelerator.}} and even with the CPUs of the day you'd have been [https://www.anandtech.com/show/2393/4 hard]-[https://web.archive.org/web/20080824053807/http://www.thenextbench.com/hpg/blog/article?message.uid=10771#M442 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 [https://web.archive.org/web/20190806042912/https://www.pcgameshardware.de/Mirrors-Edge-Spiel-4510/Tests/Physics-acceleration-via-the-GPU-673911/2/ was] [https://web.archive.org/web/20090130082728/http://www.pcgameshardware.com/aid,674175/Reviews/Mirrors_Edge_benchmarks_with_Radeon_HD_4870_plus_Ageia_card_-_Alternative_to_Nvidia_Physx/ blown] [https://www.geeks3d.com/20080820/physx-performance-gpu-vs-ppu-vs-cpu/ away] [https://web.archive.org/web/20090901181947/http://www.firingsquad.com/hardware/physx_performance_update/ once] [https://techgage.com/article/nvidias_physx_performance_and_status_report/2/ CUDA] [https://hardforum.com/threads/cryostasis-cpu-vs-ppu-vs-gpu.1376657/ came]). | ||
− | Of note that nvidia's also [https://web.archive.org/web/20111124045719/http://physxinfo.com/news/338/ati-physx-ban-even-ageia-ppu-is-affected/ affected] the ageia cards. Also, it would be interesting to know what <code>ForceDeviceType</code> did inside their hklm registry key (disassembly shows it should accept either <tt>[https://overclockers.ru/lab/show/22639/Obzor_fizicheskogo_uskoritelya_Ageia_PhysX AG1011]</tt> or <tt>[https://www.techpowerup.com/forums/threads/rare-gpus-unreleased-gpus.176929/page-33#post-4174965 AG1021]</tt> as values). | + | Of note that nvidia's hybrid ban also [https://web.archive.org/web/20111124045719/http://physxinfo.com/news/338/ati-physx-ban-even-ageia-ppu-is-affected/ affected] the ageia cards. Also, it would be interesting to know what <code>ForceDeviceType</code> did inside their hklm registry key (disassembly shows it should accept either <tt>[https://overclockers.ru/lab/show/22639/Obzor_fizicheskogo_uskoritelya_Ageia_PhysX AG1011]</tt> or <tt>[https://www.techpowerup.com/forums/threads/rare-gpus-unreleased-gpus.176929/page-33#post-4174965 AG1021]</tt> as values). |
− | + | <references group="Note" /> |