• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

If RT is off, does hardware RT silicon just sit "parked", or does it aid the workload in any way?

Xdrive05

Member
Question in the title. I'm talking about Nvidia's RTX and AMD's RDNA2 hardware ray tracing solutions. Of course I understand that using ray tracing hits the hardware pretty hard, but if you DON'T use it, then does that extra silicon just go to waste? Or does it help get non-RT frames out the door faster?

My google-fu is failing me as I can't seem to find a straight answer to this question. I may also be misunderstanding what "RT cores" even means. I guess my same question would apply to "tensor cores" too. Thanks in advance for any help clarifying this.
 

Iamborghini

Member
I am not an enginer but logically they are "OFF" when not used, because they are build for a specific task and need a code to specifiy what it sould do. I think it can do more than just RT but it need to be specified by game devs.
 

ToTTenTranz

Banned
It "sits parked", but the die area that is exclusively dedicated for RT is small, even on nvidia's implementation, because RT still resorts to shader processors for a good part of its pipeline.
 
Top Bottom