NVIDIA GeForce GTX 1070 vs NVIDIA GeForce RTX 4080 SUPER

Every spec side by side, from each maker's published figures.

Better figure, where one is clearly better

GeForce GTX 1070 compared with GeForce RTX 4080 SUPER
NVIDIAGeForce GTX 1070NVIDIAGeForce RTX 4080 SUPER
Overview
ReleasedJune 2016January 2024
Performance
Chip makerNVIDIANVIDIA
ArchitecturePascalAda Lovelace
Shader cores1,92010,240
Boost clock1,683 MHz2,550 MHz (better)
ArchitectureMain figures and photo describe the Founders Edition, verified in MSI's retained datasheet. Partner boards can use different clocks, coolers and connectors.–
Ray tracingNo dedicated RT CoresThird-generation RT Cores; 121 RT TFLOPS
AI coresNo Tensor CoresFourth-generation Tensor Cores; 836 AI TOPS
Upscaling and frame generationNo NVIDIA DLSSDLSS 3.5 with Super Resolution, DLAA, Ray Reconstruction and Frame Generation
Base clock1506 MHz2.29 GHz
Cache–64 MB L2
Memory
Video memory8 GB GDDR516 GB GDDR6X (better)
Memory bus256-bit256-bit
Memory bandwidth–736 GB/s
Memory detail8008 MHz effective memory speed (MSI Founders Edition); numerical bandwidth omitted because the maker source used does not publish it.GDDR6X at 23 Gbps
Connections and power
Board power150 W (better)320 W
InterfacePCIe 3.0PCIe 4.0
Display outputs3 × DisplayPort 1.4, 1 × HDMI 2.0b and 1 × dual-link DVI-D; up to 4 displays; HDCP 2.23 × DisplayPort 1.4a, 1 × HDMI 2.1a; up to 4 displays; HDCP 2.3
Maximum resolution7680 × 43204K at 240 Hz or 8K at 60 Hz with DSC, HDR
Power connector1 × 8-pin1 × 450 W or greater PCIe Gen 5 cable, or 3 × PCIe 8-pin cables with the adapter in the box
Recommended power supply500 W750 W
Slot width–3-slot (Founders Edition)
Size (reference card)268 × 111 × 38 mm and 1096 g (MSI Founders Edition)304 × 137 mm, 3-slot (Founders Edition)
Video encode and decode–2 × eighth-generation NVENC, 1 × fifth-generation NVDEC; AV1 encode and decode

More comparisons