- Joined
- Dec 29, 2017
- Messages
- 4,790 (1.97/day)
- Location
- Swansea, Wales
System Name | Silent |
---|---|
Processor | Ryzen 7800X3D @ 5.15ghz BCLK OC, TG AM5 High Performance Heatspreader |
Motherboard | ASUS ROG Strix X670E-I, chipset fans removed |
Cooling | Optimus Block, HWLABS Copper 240/40 + 240/30, D5/Res, 4x Noctua A12x25, 2x A4x10, Mayhems Ultra Pure |
Memory | 32 GB Dominator Platinum 6150 MT 26-36-36-48, 56.6ns AIDA, 2050 FCLK, 160 ns tRFC, active cooled |
Video Card(s) | RTX 3080 Ti Founders Edition, Conductonaut Extreme, 18 W/mK MinusPad Extreme, Corsair XG7 Waterblock |
Storage | Intel Optane DC P1600X 118 GB, Samsung 990 Pro 2 TB |
Display(s) | 32" 240 Hz 1440p Samsung G7, 31.5" 165 Hz 1440p LG NanoIPS Ultragear, MX900 dual gas VESA mount |
Case | Sliger SM570 CNC Aluminium 13-Litre, 3D printed feet, custom front panel pump/res combo |
Audio Device(s) | Audeze Maxwell Ultraviolet, Razer Nommo Pro |
Power Supply | SF750 Plat, full transparent custom cables, Sentinel Pro 1500 Online Double Conversion UPS w/Noctua |
Mouse | Razer Viper Pro V2 8 KHz Mercury White w/Tiger Ice Skates & Pulsar Supergrip tape |
Keyboard | Wooting 60HE+, TOFU-R CNC Alu/Brass, SS Prismcaps W, Jellykey, Lube/Mod, TLabs Leath/Suede Wristrest |
Software | Windows 10 IoT Enterprise LTSC 19044.4046 |
Benchmark Scores | Legendary |
Proprietary hardware drives innovation because it results in profit - the fact that open source or non-licensed versions arrive a few years later is a benefit of this process, not a limitation.I was agreeing up to this point, so hold up. AMD didn't copy it otherwise it would have been just as crappy and proprietary as GSync. AMD made VRR more standard by not requiring stupid special hardware for it and making the standard open. It forced nVidia to support it. I'd hardly call that copying. nVidia tried to corner the market and AMD flipped that on its face. RT on nVidia cards is no different, more proprietary hardware and APIs to get the best performance at the cost of vendor lock, which is a dangerous thing with a company like nVIdia given their history. If I bought a GPU today, RT performance is going to be at the bottom of my list of things I care about. I want the ecosystem to evolve just like VRR did because eventually, we won't need proprietary compute blocks or APIs to do this and it will all become standardized. Until then, I'll let everyone else play the role of guinea pig.
You can see each company taking it's own approach, Intel with their "optimized kernel" and "compatibility kernel", Nvidia with the RTX stack (that is part of the industry standard set of tools NVIDIA offers, despite not being opensource). AMD is going the route AMD always takes, which seems to be seeing what Intel/NVIDIA do and releasing their own version a couple years later. See DLSS/FSR, DLSS3.0/FSR3, Ray tracing support (AMD competing with Ampere not Ada, and losing to Intel). Hopefully we see some kind of standardization which seems to already be the case. You can Run XeSS/FSR OR DLSS etc. in games if you have an NVIDIA GPU, despite each of these running in different ways and using different hardware to different extents. Unreal Engine 5.1 with Lumen - hardware ray tracing takes perfect advantage of NVIDIA RTX hardware, giving you better fidelity due to taking advantage of specific proprietary hardware with no loss of performance over the hybrid software approach, while AMD/Intel can still use the hybrid approach and have somewhat lower fidelity with a somewhat higher cost to FPS with hardware RT enabled, but still share in RT goodness.
I don't believe that there's much evidence for opensource taking over vs proprietary, last time I checked NVIDIA CUDA and Microsoft Office are still the standards, despite there being open source alternatives to each, I use the word alternatives instead of equivalents since I don't think there are equivalent stacks.
So your approach is basically to take a moral stance against a technical question, which is somewhat admirable, but you pay the cost of that in the product you get. Sadly it seems most people prefer higher performance/fidelity rather than a more open standard at the cost of those factors.RT on nVidia cards is no different, more proprietary hardware and APIs to get the best performance at the cost of vendor lock, which is a dangerous thing with a company like nVIdia given their history. If I bought a GPU today, RT performance is going to be at the bottom of my list of things I care about. I want the ecosystem to evolve just like VRR did because eventually, we won't need proprietary compute blocks or APIs to do this and it will all become standardized. Until then, I'll let everyone else play the role of guinea pig.