{"id":31677,"date":"2019-10-31T21:42:30","date_gmt":"2019-10-31T18:42:30","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novaya-statya-geforce-rtx-bolshe-ne-nuzhen-testy-trassirovki-luchej-na-uskoritelyah-geforce-gtx-10-i-16\/"},"modified":"2019-10-31T21:42:30","modified_gmt":"2019-10-31T18:42:30","slug":"novaya-statya-geforce-rtx-bolshe-ne-nuzhen-testy-trassirovki-luchej-na-uskoritelyah-geforce-gtx-10-i-16","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/news\/novaya-statya-geforce-rtx-bolshe-ne-nuzhen-testy-trassirovki-luchej-na-uskoritelyah-geforce-gtx-10-i-16","title":{"rendered":"New article: Is GeForce RTX no longer needed? Ray tracing tests on GeForce GTX 10 and 16 accelerators.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

After NVIDIA demonstrated real-time ray tracing on GeForce RTX series graphics cards, it's hard to doubt that this technology (in a reasonable combination with rasterization algorithms) represents the future of computer gaming. However, Turing architecture GPUs with specialized RT cores were until recently considered the only category of discrete GPUs that had the necessary computational power for it.<\/p>\n

Tests of the first games utilizing Ray Tracing (Battlefield V, Metro Exodus, and Shadow of the Tomb Raider) have shown that even GeForce RTX accelerators (particularly the entry-level RTX 2060) experience a significant drop in frame rates in hybrid rendering tasks. Despite initial successes, real-time ray tracing cannot yet be considered a mature technology. It will only be possible to declare that the paradigm shift initiated by Jensen Huang has finally occurred when not only the most advanced and expensive devices but also mid-range graphics cards achieve the previous performance standards in next-generation games.<\/p>\n

\"New<\/p>\n

Ray Tracing on 'Pascal' \u2013 Pros and Cons<\/strong><\/p>\n

However, even now, while not a word has been said about the future successor of the Turing architecture, NVIDIA has decided to spur progress. At last month's GPU Technology Conference, the green team announced that accelerators based on Pascal chips, as well as the entry-level representatives of the Turing family (the GeForce GTX 16 series), will gain real-time ray tracing capabilities on par with RTX-branded products. The promised driver is already available for download on NVIDIA's official website, and the list of supported devices includes models from the GeForce 10 family, starting with the GeForce GTX 1060 (6 GB version), the professional TITAN V accelerator based on the Volta chip, and, of course, the newly released mid-range models on the TU116 chip \u2013 the GeForce GTX 1660 and GTX 1660 Ti. The update also applies to laptops with corresponding GPUs.<\/p>\n

From a technical standpoint, there's nothing extraordinary here. Graphics processors with unified shader blocks could perform Ray Tracing long before the Turing architecture emerged, although they lacked the performance needed for this capability to be utilized in games at that time. Moreover, there was no unified standard for programming methods, apart from closed APIs like NVIDIA's OptiX. Now that there is the DXR extension for Direct3D 12 and similar libraries in the Vulkan API, game engines can access them regardless of whether the graphics processor has specialized logic\u2014as long as the driver allows this possibility. The Turing chips have separate RT cores for this purpose, while ray tracing in the Pascal architecture and the TU116 processor is implemented in a general-purpose computing format on an array of shader ALUs.<\/p>\n

\"New<\/p>\n

However, all that we know about the Turing architecture from NVIDIA itself indicates that Pascal is not suitable for applications that support DXR. In last year's presentation focused on the flagship models of the Turing family\u2014the GeForce RTX 2080 and RTX 2080 Ti\u2014engineers provided the following data. If all resources of the best consumer graphics card from the previous generation\u2014the GeForce GTX 1080 Ti\u2014were dedicated to ray tracing calculations, the resulting performance would not exceed 11% of what the RTX 2080 Ti could theoretically achieve. It's also important to note that the available CUDA cores of the Turing chip can simultaneously be utilized for parallel processing of other image components\u2014executing shader programs, handling non-graphical calculations in Direct3D during asynchronous execution, and so forth.<\/p>\n

\"New<\/p>\n

In real games, the situation is more complex, as developers selectively utilize DXR features on existing hardware, while a significant portion of the computational load still consists of rasterization and shader instructions. Moreover, some effects created using ray tracing can also perform well on the CUDA cores of Pascal chips. For example, reflective surfaces in Battlefield V do not involve secondary ray reflections, and thus, represent a manageable load for high-end graphics cards from the previous generation. The same applies to shadows in Shadow of the Tomb Raider, although rendering complex shadows formed by multiple light sources poses a more challenging task. Global illumination in Metro Exodus, on the other hand, is a struggle even for the 'Turing' architecture, and Pascal cannot be expected to deliver comparably adequate results in any measure.<\/p>\n

Regardless of how you look at it, there is a significant difference in theoretical performance between representatives of the Turing architecture and their closest silicon counterparts from Pascal. The advantage of Turing lies not only in the presence of RT cores but also in numerous general improvements characteristic of next-generation accelerators. Turing chips can perform operations on floating-point (FP32) and integer (INT) data in parallel, feature a large amount of local cache memory, and have separate CUDA cores for lower-precision calculations (FP16). This means that Turing not only handles shader programs better but can also compute ray tracing relatively efficiently without specialized blocks. After all, the resource-intensive rendering using Ray Tracing is driven not just by the intersection checks between rays and geometry elements (handled by RT cores) but also by the color calculations at the intersection points (shading). By the way, the aforementioned advantages of the Turing architecture fully apply to the GeForce GTX 1660 and GTX 1660 Ti, even though the TU116 chip lacks RT cores, making tests of these graphics cards with software-based ray tracing particularly interesting.<\/p>\n

But enough theory; we have already gathered performance data for 'Pascal' (as well as the lower-tier 'Turing') in Battlefield V, Metro Exodus, and Shadow of the Tomb Raider based on our own measurements. It is worth noting that neither the driver nor the games themselves regulate the number of rays to reduce the load on GPUs without RT cores, which means the quality of effects on GeForce GTX and GeForce RTX should be the same.<\/p>\n

Test bench, testing methodology<\/strong><\/h2>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Test Stand<\/th>\n<\/tr>\n
CPU<\/td>\nIntel Core i9-9900K (4.9 GHz, 4.8 GHz in AVX, fixed frequency)<\/td>\n<\/tr>\n
Motherboard<\/td>\nASUS MAXIMUS XI APEX<\/td>\n<\/tr>\n
Memory<\/td>\nG.Skill Trident Z RGB F4-3200C14D-16GTZR, 2 x 8 GB (3200 MHz, CL14)<\/td>\n<\/tr>\n
SSD<\/td>\nIntel SSD 760p, 1024 GB<\/td>\n<\/tr>\n
Power Supply<\/td>\nCorsair AX1200i, 1200 W<\/td>\n<\/tr>\n
CPU Cooling System<\/td>\nCorsair Hydro Series H115i<\/td>\n<\/tr>\n
Case<\/td>\nCoolerMaster Test Bench V1.0<\/td>\n<\/tr>\n
Monitor<\/td>\nNEC EA244UHD<\/td>\n<\/tr>\n
Operating system<\/td>\nWindows 10 Pro x64<\/td>\n<\/tr>\n
GPU Software for NVIDIA<\/td>\n<\/tr>\n
NVIDIA GeForce RTX 20<\/td>\nNVIDIA GeForce Game Ready Driver 419.67<\/td>\n<\/tr>\n
NVIDIA GeForce GTX 10\/16<\/td>\nNVIDIA GeForce Game Ready Driver 425.31<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n\n\n\n\n
Gaming tests<\/th>\n<\/tr>\n
Game<\/td>\nAPI<\/td>\nSettings, testing method<\/td>\nFullscreen Anti-Aliasing<\/td>\n<\/tr>\n
1920 \u00d7 1080 \/ 2560 \u00d7 1440<\/td>\n3840 \u00d7 2160<\/td>\n<\/tr>\n
Battlefield V<\/td>\nDirectX 12<\/td>\nOCAT, mission Liberte. Max. graphics quality<\/td>\nTAA High<\/td>\nTAA High<\/td>\n<\/tr>\n
Metro Exodus<\/td>\nDirectX 12<\/td>\nBuilt-in benchmark. Ultra graphics quality profile<\/td>\nTAA<\/td>\nTAA<\/td>\n<\/tr>\n
Shadow of the Tomb Raider<\/td>\nDirectX 12<\/td>\nBuilt-in benchmark. Max. graphics quality<\/td>\nSMAA 4x<\/td>\nOff.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

The average and minimum frame rates are derived from a time array of individual frame renderings, which is recorded by the built-in benchmark (Metro Exodus, Shadow of the Tomb Raider) or the OCAT utility, if it's not available in the game (Battlefield V).<\/p>\n

The average frame rate shown in the charts is the inverse of the average frame time. To evaluate the minimum frame rate, the number of frames rendered in each second of the test is calculated. From this array of numbers, the value corresponding to the 1st percentile of the distribution is selected.<\/p>\n

Testing participants<\/strong><\/p>\n

The following graphics cards participated in the performance testing:<\/p>\n