NVIDIA GeForce GTX 1650 vs NVIDIA GeForce RTX 5070

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

Only show differences

Better figure, where one is clearly better

Share comparison

GeForce GTX 1650 compared with GeForce RTX 5070

	NVIDIAGeForce GTX 1650	NVIDIAGeForce RTX 5070
Overview
Released
Released	April 2019	March 2025
Performance
Chip maker
Chip maker	NVIDIA	NVIDIA
Architecture
Architecture	Turing	Blackwell
Shader cores
Shader cores	896	6,144
Boost clock
Boost clock	1,665 MHz or 1,590 MHz	2,512 MHz
Architecture
Architecture	NVIDIA GPU specification baseline; the photo shows PNY's original stock-clock GDDR5 Single Fan board. The first-sale date is for the original GDDR5 model; a maker first-sale date for GDDR6 was not established.	GB205 GPU; 48 SMs
Manufacturing process
Manufacturing process	–	TSMC 4nm (4N NVIDIA custom process); 31.1 billion transistors; 263 mm² die
Ray tracing
Ray tracing	No hardware RT Cores	48 fourth-generation RT Cores; 94 RT TFLOPS
AI cores
AI cores	No Tensor Cores	192 fifth-generation Tensor Cores; 988 AI TOPS
Upscaling and frame generation
Upscaling and frame generation	No NVIDIA DLSS support	DLSS 4.5 with Super Resolution, DLAA, Ray Reconstruction, Frame Generation, Multi Frame Generation and Dynamic Multi Frame Generation
Base clock
Base clock	Original GDDR5 is 1,485 MHz; GDDR6 is 1,410 MHz	2.33 GHz
Cache
Cache	–	48 MB L2
Memory
Video memory
Video memory	4 GB GDDR5 or 4 GB GDDR6	12 GB GDDR7 (better)
Memory bus
Memory bus	128-bit	192-bit (better)
Memory bandwidth
Memory bandwidth	128 GB/s or –	672 GB/s
Memory detail
Memory detail	PNY publishes 8 Gbps and 128 GB/sec for GDDR5; its GDDR6 PDF gives 12 Gbps but labels its bandwidth figure as Gbps, so numeric GDDR6 bandwidth is omitted.	GDDR7 at 28 Gbps
Connections and power
Board power
Board power	75 W (better)	250 W
Interface
Interface	PCIe 3.0	PCIe 5.0
Display outputs
Display outputs	PNY's pictured GDDR5 Single Fan specification lists DisplayPort 1.4 and DVI-D. Its GDDR6 Dual Fan PDF conflicts internally, listing HDMI 2.0b in the features and HDMI 2.1 in the table; check the selected board.	3 × DisplayPort 2.1b (UHBR20), 1 × HDMI 2.1b; up to 4 displays; HDCP 2.3
Maximum resolution
Maximum resolution	7680 × 4320 on PNY's specifications	4K at 480 Hz or 8K at 165 Hz with DSC
Power connector
Power connector	NVIDIA's current comparison table lists one 6-pin for both revisions; PNY's pictured stock GDDR5 card explicitly needs no auxiliary connector, while its GDDR6 Dual Fan needs one 6-pin.	1 × 300 W or greater PCIe Gen 5 cable, or 2 × PCIe 8-pin cables with the adapter in the box
Recommended power supply
Recommended power supply	NVIDIA specifies 300 W; requirements depend on the system and partner card	650 W
Slot width
Slot width	PNY's pictured GDDR5 Single Fan and GDDR6 Dual Fan are dual-slot designs	2-slot (Founders Edition)
Size (reference card)
Size (reference card)	No retail Founders Edition; pictured PNY GDDR5 Single Fan is 1.61 × 6.22 × 4.92 inches. PNY GDDR6 Dual Fan is 188.3 × 115.1 × 39.3 mm.	242 × 112 mm, 2-slot (Founders Edition)
Video encode and decode
Video encode and decode	NVIDIA identifies the original encoder as Volta; GDDR6 cards can have Volta or Turing encoders depending on whether the GPU is TU117 or TU106/TU116. PNY specifies PCIe 3.0 x16 for both pictured/reference partner designs.	1 × ninth-generation NVENC, 1 × sixth-generation NVDEC; AV1 encode and decode