NVIDIA shares official GeForce RTX 3070 performance figures

NVIDIA reveals official GeForce RTX 3070 performance figures

NVIDIA today confirmed that the launch date of the GeForce RTX 3070 graphics card has been postponed till October 29th. NVIDIA wants to avoid limited stock at launch, hence more cards are being produced to meet the demand. We are not sure if two weeks are going to change a lot in this matter, but the timing of this launch is definitely interesting as it lands right after the Radeon RX 6000 announcement.

In the same blog post on the NVIDIA website, we can find the very first graph showing RTX 3070 performance in synthetic and gaming scenarios.

The labels on the official graph are missing, so we went ahead and measured each graph and calculated how many pixels each bar has. This gave us values for reference. So yes, this definitely not perfect but it is better than nothing.

NVIDIA tested games with their RTX technology enabled whenever possible. It appears that DLSS was not enabled. Games were tested at 1440p resolution, which seems like a target for RTX 3070 SKU.

The RTX 3070 appears to offer slightly higher performance than RTX 2080 Ti (108% on average), but in some instances, the card is slower (Borderlands/Blender/TimeSpy Extreme). On the other hand, the card offers 147% to 163% performance of RTX 2070 in games and 153% to 226% in synthetic benchmarks.

NVIDIA officially claims that their RTX  3070 is on average 60% faster than RTX 2070:

Across a variety of ray-traced and rasterized DirectX and Vulkan titles, the GeForce RTX 3070 delivers similar or faster performance than the GeForce RTX 2080 Ti (which sold for twice the price) and is on average 60% faster than the original GeForce RTX 2070.

NVIDIA is now set to launch GeForce RTX 3070 on October 29th. On the same day, NVIDIA will lift the embargo on reviews, meaning that the time for purchase decisions may be limited. Hopefully, the supply will meet the demand, and customers will have time to read or watch the reviews.

Official NVIDIA GeForce RTX 3070 Performance

NVIDIA GeForce RTX 30 Series Specifications
VideoCardz.com	RTX 3090	RTX 3080	RTX 3070	RTX 3060 Ti
Picture				[To be leaked]
Reference Board	PG132 SKU 30	PG132 SKU 10	PG142 SKU 10	PG190 SKU 10
GPU	8nm GA102-300	8nm GA102-200	8nm GA104-300	8nm GA104-200
Die Size	628 mm2	628 mm2	392 mm2	392 mm2
Transistors	28.3 B	28.3 B	17.4 B	17.4 B
CUDA Cores	10496	8704	5888	4864
GPU Utilization	98%	81%	96%	79%
Tensor Cores	328	272	184	152
RT Cores	82	68	46	38
Base Clock	1395 MHz	1440 MHz	1500 MHz	TBC
Boost Clock	1695 MHz	1710 MHz	1725 MHz	TBC
Shader Perf.	35.6 TFLOPS	29.8 TFLOPS	20.3 TFLOPS	TBC
Memory	24 GB G6X	10 GB G6X	8 GB G6	8 GB G6
Memory Clock	19.5 Gbps	19 Gbps	14 Gbps	14 Gbps
Memory Bus	384-bit	320-bit	256-bit	256-bit
Max Bandwidth	936 GB/s	760 GB/s	448 GB/s	448 GB/s
Thermal Design	350W	320W	220W	~180W
MSRP	$1,499	$699	$499	TBC
Release Date	September 24th	September 17th	October 29th	TBC