The reworked textures also add more material diversity and variety to scenes, which in the launch version had many textures looking extremely similar in how light affects them. Version 1.2 now sees individual cartridges exhibit metallic sheen and emphasise the coloured metals on the jacket. For example, the original remaster's shotgun shells in ammo boxes use matte materials that show little differentiation between the box material and the shells themselves. It's all change in 1.2, with art changes that dramatically change and improve many scenes. The original release had metalwork that appeared to lack much in the way of specular properties, so even with the hyper-realistic path traced lighting, the material looked more like stone than metal. Metal and how it interacts with lighting has changed immensely. A key focus for the 1.2 upgrade has been to re-assess many material properties and get them looking just right. While the original Quake 2 RTX launch used physically-based variants of Quake 2 XP textures, not all of them appeared to receive the same level of love and attention. One might think that Nvidia would simply move on from the Quake 2 project and concentrate efforts on the ray traced upgrades for other titles that are being worked on behind the scenes, but the improvements to the 1.2 upgrade are quite profound - and the most noticeable change comes from upgraded art assets. It's one of the most impressive examples of hardware-accelerated RT and thanks to the new 1.2 patch released a few days ago, a phenomenal game now looks a whole lot better. The DXR API is still not a perfect fit for GPU SIMD execution, but at least it's not dead stupid anymore.Quake 2 is over two decades old and yet the id Software classic is one of my favourite games of this year, radically re-invented from a visual perspective thanks to the ray traced remastering from Nvidia's Lightspeed Studios (based on original work by Christophe Schied). These thread sorting units finally make me want to do some HW ray-tracing. But please, give me access to this magical HW block also for traditional shaders! The DXR API is still not a perfect fit for GPU SIMD execution, but at least it's not dead stupid anymore. Mesh shaders are great, but they are still lacking the ability to select the shader like ray-tracing does. We need a fine grained way of spawning new GPU work from shaders. I think these GPUs are starting to be there too.Īlso RTX 4090 is so fast that we desperately need better API support GPU-driven rendering. But CPUs are now fast enough to make this a minor annoyance. CPUs have to pay similar costs for branches too. As a result, we could write CPU-style shader code with branches, instead of compiling (hundreds of) thousands of permutations.Įven with hardware like this, it's not free to shuffle SIMD data around. They could be used to make GPU dynamic branching faster in all shaders. This is super important for ray-tracing and explains why Intel's mid range GPU is so good at ray-tracing, but also explains why RTX 4090 is such a best in RT apps.īut these hardware blocks are not just a great fit for ray-tracing. These new hardware blocks shuffle the registers of multiple SIMDs in a way that each SIMD can run coherent threads. Now that RTX 4090 is massively CPU bound, could we spend 1% of that perf to get rid of shader permutations? With latest hardware: Intel Thread Sorting Unit (TSU) and Nvidia Shader Execution Reordering (SER). Let's discuss about shader permutation hell. Sebbi made some very important discussion points regarding the thread sorting hardware, I will list them here for quick read.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |