Gideon
2[H]4U
- Joined
- Apr 13, 2006
- Messages
- 3,567
Thankfully AMD doesn't have to carry the torch of being 'first' here
Pros and cons to being first
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Thankfully AMD doesn't have to carry the torch of being 'first' here
It better be running in early alpha midrange card.
I hope so.Will find out in a few months. I feel AMD can likely stay competitive with Nvidia on that front, just expect the ride to be a bit bumpy.
GigaRays is not a industry wide standard, it is the way NVIDIA presents their RT perfomance
Fairly subtle for April Fools, but GDDR7 and PCIe 5.0 are dead giveaways.
The dual BLOWER fans is the real dead giveaway...!
I kinda like that idea, hmmmm . . .And yet, no one would be surprised if AMD launched a dual blower card.
I think all those people thinking big navi was going to bring better DXR than nVidia are gonna be eating crow later this year. We all know who's the best:
The lyrics to that song were made for this thread.
Need huge grain of salt for that.
The TOPS numbers for NVidia are from the Tensor cores, so of course are higher, since AMD is not expected to have any Tensor cores.
The RT Perf numbers from AMD, seem to indicate that there is no dedicated RT cores, which seems a little more than unlikely.
RT type cores are not rocket science. AMD will have dedicated RT core equivalents, and IMO they will be of similar capability of Turing RT cores. AMD obviously had Turing the benchmark and will not want to lag significantly.
We won't have much idea about Big Navi until we see some comparative benchmarks...
You wrote this in march... are you getting ready to eat crow or feed us crow lol?
I guess well see when Big Navi hits the review line. Exciting times with these gpu launches.
The BVH parts are RT dedicated hardware/silicon, but I understand what you're saying, it's not seperate from existing parts/tied in. It's been known/suggested (based on AMD whitepapers and patents) that it was going to be part of the shaders vs a seperate area of the die so, not a huge surprise, but good to know non the less. Guess we'll see how it works out sometime in the near future, but I suspect AMD's first try at RT may fall short on NVidia's second gen. Hopefully it's at least on par NVidia's first attempt otherwise it'd be pretty disappointing. I'll have to check the specific presentation you are mentioning as I haven't seen that one yet. I guess even if it can only run RT at medium instead of high or w/e, it'll still help developers because they have common hardware/target.Did you see the HotChips 2020 presentation on RDNA2 by MS/AMD engineers? (check Digital Foundry utube,they did a vid on it) It has zero dedicated hardware RT silicon,and a small amount of dedicated BVH hardware,the shaders are doing double duty,juggling all the usual tasks and RT computation.
The BVH parts are RT dedicated hardware/silicon, but I understand what you're saying, it's not seperate from existing parts/tied in. It's been known/suggested (based on AMD whitepapers and patents) that it was going to be part of the shaders vs a seperate area of the die so, not a huge surprise, but good to know non the less. Guess we'll see how it works out sometime in the near future, but I suspect AMD's first try at RT may fall short on NVidia's second gen. Hopefully it's at least on par NVidia's first attempt otherwise it'd be pretty disappointing. I'll have to check the specific presentation you are mentioning as I haven't seen that one yet. I guess even if it can only run RT at medium instead of high or w/e, it'll still help developers because they have common hardware/target.
I sure hope so! Has anyone heard what they are doing on the 8th for zen3 and 28th for rdna2? Is it just an announcement of when the release will be or will we be getting any sort of benchmark info, or?Agreed on all points,we will know by the 28th of October.