Visit the purchase page:JD.com NVIDIA Series Product Summary
The LTX 2070's L2 cache is twice that of the GTX 1070, up to 4 MB, and the total size of its register files (distributed in each SM) has expanded to 9.216 MB, while the GTX 1070 is 3.840 MB, providing more for the execution unit. Multi-data is a reliable way to improve performance. Compared to its Pascal predecessor, the TU106's on-chip cache is definitely a bigger improvement in this regard.
Like its larger Turing brothers, the RTX 2070 has resources dedicated to speeding up certain ray tracing operations. The RTX 2070 has 288 Turing Tensor cores (8 per SM) and 36 Turing RT cores (one for each SM). In contrast, there are 368 Tensor cores and 46 RT cores for the RTX 2080. The RTX 2080 Ti has 544 Tensor cores and 68 RT cores.
Since we don't have any real-world applications that use these processing resources, it's hard to say exactly what the RTX 2070's ray tracing performance is. For its value, Nvidia says RTX 2070 Founders Edition can perform 45 RTX tera-OPS (using Turing GPU) The performance potential of the hybrid rendering is measured), while the RTX 2080 FE is 60 and the RTX 2080 Ti FE is 78.
Since RTX 2070 is manufactured using TSMC's 12-nm FFN process, its 108 billion transistors will not benefit from area shrinkage, if any, compared with the 16-nm FinFET process used to manufacture Pascal. Therefore, the TU106 chip is 445mm, compared with the GP104 chip area of 314mm for GTX 1070 and GTX 1080, the power consumption of 175W for the public version of RTX2070 graphics card and 185W power consumption for the founder version of "factory overclocking".
From 25 games and comprehensive benchmarks, Nvidia Test Labs expects the RTX 2070 to increase performance by 33% in 2560x1440 SDR games compared to the GTX 1070, while HDR games are up 35%. The company recommends partner RTX 2070 graphics cards starting at $499.