Nvidia introduced a refreshed 2000 Series lineup back in 2019 with models having a SUPER designation at the end and here we are 5 years later and this year at CES they did that once again with their 4000 Series of cards, skipping the 3000 Series. Their CES press conference announced the new RTX 4080 SUPER, RTX 4070 Ti SUPER, and the RTX 4070 SUPER. The last of which goes on sale tomorrow January 17^th but ahead of that we have the chance to take a look at the Nvidia Founders Edition of the RTX 4070 SUPER to see what it is all about so let’s dive in.

Product Name: Nvidia RTX 4070 SUPER Founders Edition

Review Sample Provided by: Nvidia

Written by: Wes Compton

Amazon Affiliate Links: HERE

 

What’s so SUPER

Nvidia’s CES announcements were all covered in their video but to hit a few highlights before getting into what the three new SUPER cards have to offer. They touched on a few new upcoming RTX supported titles. One of those is Half-Life 2 RTX which was done using the RTX Remix tool. That tool is going into open beta offering an easy way for modders and developers to upgrade past and current games to utilize RTX technology. Anything that makes it easier to do will only help get more support and in this case, we can potentially see upgraded games that we would have never seen otherwise which is exciting.

Nvidia and Twitch announced a new way for Twitch streamers to now offer a better range of resolutions. In the past, for anyone who isn’t one of their main partnered streamers you are stuck with just one resolution. This way you will be able to stream at a range of resolutions and frame rates. Viewers who have the bandwidth can enjoy better quality 4k 60 FPS streams but users on mobile or lower-performing internet connections will still be able to watch as well.

They announced improvements in G-Sync variable refresh rate. Along with that, they covered G-Sync Pulsar which is part of their VRR or variable refresh rate technology. Pulsar bridges the gap between using variable refresh rate and ultra low motion blur to get a smooth experience but cut out the blur that you can get with it.

They then of course announced the three new SUPER cards and showed where all three will fit in their updated lineup. The RTX 4080 SUPER will replace the RTX 4080 and sit behind the RTX 4090 still. That will have an MSRP of $999, $200 less than the original MSRP of the RTX 4080. The new card has a faster base clock of 2295 MHz vs 2205 MHz on the 4080 and 40 MHz more on the boost clock with it now at 2550 MHz. The real improvement comes from more cores going from 9728 CUDA cores up to 10240. The RTX 4070 Ti SUPER replaces the RTX 4070 and has an MSRP of $799, the same as the original launch price of the RTX 4070. It gets a memory clock speed improvement but like with the 4080 SUPER, it's big improvements with the CUDA core count jumping up from 7680 up to 8448 which is a big improvement. With all of these cards, the tensor cores and RT cores will also scale up at the same rate as the CUDA cores as well. Lastly the RTX 4070 SUPER doesn’t replace the original RTX 4070, the original is staying in the lineup. The 4070 will now be $549, $50 less than its launch price and the RTX 4070 SUPER takes up that $599 price point and fits in between the 4070 and 4070 Ti SUPER. The CUDA core count goes from 5888 up to 7168, a 21.7% improvement and the same boost and memory clock speeds. Nvidia did bump up the power usage slightly with the 4070 SUPER now having a TGP of 220 Watts to the 200 Watts on the 4070, the 4080 SUPER and 4070 Ti SUPER both stay the same at 285 Watts and 320 watts respectively. Overall though I like that as a refresh the focus is on adding more cores, not cranking up clock speeds which we have seen with things like the RX 480 to RX 580.

Specifications	RTX 2070 SUPER	RTX 4070	RTX 4070 SUPER	RTX 4070 Ti
Graphics Processing Clusters	5 or 6	4	5	5
Texture Processing Clusters	20	23	28	30
Streaming Multiprocessors	40	46	56	60
CUDA Cores	2560	5888	7168	7680
Tensor Cores	320	184 (4th Gen)	224 (4th Gen)	240 (4th Gen)
RT Cores	40	46 (3rd Gen)	56 (3rd Gen)	60 (3rd Gen)
Texture Units	160	184	224	240
ROPs	64	64	80	80
Boost Clock	1770 MHz	2475 MHz	2475 MHz	2610 MHz
Memory Clock	1750 MHz	1313 MHz	1313 MHz	1313 MHz
Memory Data Rate	14 Gbps	21 Gbps	21 Gbps	21 Gbps
L2 Cache Size	4 MB	36 MB	48MB	49 MB
Total Video Memory	8 GB GDDR6	12 GB GDDR6X	12 GB GDDR6X	12 GB GDDR6X
Memory Interface	256-bit	192-bit	162-bit	192-bit
Total Memory Bandwidth	448.0 GB/s	504 GB/sec	504 GB/s	504 GB/sec
Texture Rate (Bilinear)	283.2 Gigatexels/sec	455.4 Gigatexels/sec	561.1 GigaTexels/second	626 Gigatexels/sec
Fabrication Process	TSMC 12 nm	TSMC 4N NVIDIA Custom Process	4N NVIDIA Custom Process	TSMC 4N NVIDIA Custom Process
Transistor Count	13.6 Million	35.8 billion	35.8 Billion	35.8 billion
Connectors	1x HDMI 2.0 3x DisplayPort 1.4a 1x USB Type-C	3 x DisplayPort 1 x HDMI	3 x DisplayPort 1 x HDMI	3 x DisplayPort 1 x HDMI
Power Connectors	1x 6-pin 1x 8-pin	2x PCIe 8-pin cables OR 300w or greater PCIe Gen 5 cable	2x PCIe 8-pin cables (adapter in box) OR 300 W or greater PCIe Gen 5 cable	2x PCIe 8-pin cables OR 300w or greater PCIe Gen 5 cable
Minimum Power Supply	550 W	650 Watts	650 Watts	700 Watts
Total Graphics Power (TGP)	215 W	200 Watts	220 Watts	285 Watts
Maximum GPU Temperature	83° C	90° C	90° C	90° C
PCI Express Interface	Gen 3	Gen 4	Gen 4	Gen 4
Launch MSRP	$499	$599	$599	$799