After first announcing it last month, UL this weekend provided new information on its upcoming ray tracing-focused addition to the 3DMark benchmarking suite. Port Royal, what UL calls the "world's first dedicated real-time ray tracing benchmark for gamers," will launch Tuesday, January 8, 2019.
For those eager for a glimpse of the new ray-traced visual spectacle, or for the majority of gamers without a ray tracing-capable GPU, the company has released a video preview of the complete Port Royal demo scene.
Access to the new Port Royal benchmark will be limited to the Advanced and Professional editions of 3DMark. Existing 3DMark users can upgrade to the benchmark for $2.99, and it will become part of the base $29.99 Advanced Edition package for new purchasers starting January 8th.
Real-time ray tracing promises to bring new levels of realism to in-game graphics. Port Royal uses DirectX Raytracing to enhance reflections, shadows, and other effects that are difficult to achieve with traditional rendering techniques.
As well as benchmarking performance, 3DMark Port Royal is a realistic and practical example of what to expect from ray tracing in upcoming games— ray tracing effects running in real-time at reasonable frame rates at 2560 × 1440 resolution.
3DMark Port Royal was developed with input from AMD, Intel, NVIDIA, and other leading technology companies. We worked especially closely with Microsoft to create a first-class implementation of the DirectX Raytracing API.
Port Royal will run on any graphics card with drivers that support DirectX Raytracing. As with any new technology, there are limited options for early adopters, but more cards are expected to get DirectX Raytracing support in 2019.
3DMark can be acquired via Steam or directly from UL's online store. The Advanced Edition, which includes access to all benchmarks, is priced at $29.99.
Yeah, ray tracing truly is a
Yeah, ray tracing truly is a technological marvel, however… when will it look good. Instead of looking like everything is chrome plated?
Handling multiple reflections
Handling multiple reflections (and refractions) from irregular objects is a lot of the point of ray tracing, so one should expect a torture test to include much shiny and many points of light. I rather liked the complex curves on the vehicle with their irregular reflections of the surroundings. More subtle effects of light through fabric or glass can come in the next release, after the budget cards can all handle this test.
Any CPU, or GPU, can be used
Any CPU, or GPU, can be used to accelerate Ray Tracing and Ray Tracing was orginially done on CPUs but that took a very long time relative to any gaming frame times.
It’s the gaming folks that have the hardest nut to crack as only so many rays can be fully traced in 33.33ms(30 FPS) down to the 16.67ms(60 FPS) frame times and below that for higher than 60 FPS gaming. Even with all those Gigarays/s that still only a small amount of total rays for things like Ray Traced Shadows or Ray Traced Ambient Occlusion in addition to any reflections and reflactions that make use of plenty of Rays in order to accurately represent a nearer approximation of reality.
Animation folks do not necessarily need any real time ray tracing as they can take their time time and turn up the rays sample rates so high that no extra filtering is necessary. But Games only have a very finite range of millieseconds in which to get that Ray Sampling work done and more gamers are making use of high refresh rate monitors so that’s going to be in the 60 FPS and higher Frame Time that come in at or below 16.67ms per frame.
Animation folks can make use of way more than 2 GPUs on the same Workstation as there are no sub-millesecond time constraints so any Ray Tracing can be easily spread across as meny GPUs as can be slotted into a Workstation’s MB and that wotk can even be done on clusters of workstations if needed.
Nvidia’s Hybrid Raster mixed with “Real Time” Ray Tracing is still making use of plenty of Raster oriented workloads and just a minimum of Generated Rays. So limited is the available amount of Rays that can be traced per Milliseconds of frame time on Nvidia’s consumer RTX SKUs that developers have to make use of some Nvidia software Tool Chains in order to limit the amounts of game geomentry that can be Ray Sampled to only the necessary parts of a scene or certian meshes/Mesh-Parts in a scene.
So depending on the RTX Turing SKU that’s still a limited amount of rays available and Nvidia has to make use of the other Turing Tensor Core IP to run the AI trained denoining on that limited per frame time Ray Tracing output that has to be denoised and then mixed in with the Raster Pipeline output currently.
Nvidia and the games developers have their work cut out just balancing things as the Frams Fly By in order to get the maximum effect at high frame rates higher than 45 FPS. And those Folks that expended plenty of money on a high refresh G-Sync monitors are not going to be happy at frame rates below 60 FPS and they will be looking for less rays and the frame times that are as short as half of that 16.67ms(60 FPS) Frame Time into the 120 FPS Frame Time(8.33ms). So that’s even less available Rays that can be Generated at 120 FPS.
It’s good to see some benchmarks but they need to be able to dial back some settings on that benchmark and have that benchmark target a specific range of frame rates via dialing back any settings that make use of Rays. That’s so any folks that have spent their money for expensive High Refresh monitors don’t go all ballistic at having their monitor purchase not be fully utilized because the cost of Rays are made up by lowering the frame rates too much using RTX/Turing for gaming.