In recent times, the editor shared good news with numerous digital enthusiasts. NVIDIA will release the RTX 40 SUPER series graphics cards in January next year.
Currently, at least three models are known, namely the GeForce RTX 4070 SUPER, GeForce RTX 4070 Ti SUPER, and GeForce RTX 4080 SUPER.
Before this, only concept and parameter images of some RTX 40 SUPER series graphics cards were exposed, without any related real product exposure. Today, good news has arrived, and the physical picture of the GeForce RTX 4070 SUPER graphics card has been exposed.
The exposed GeForce RTX 4070 SUPER graphics card is from ASUS. According to the editor’s personal experience, ASUS has been very active in revealing new products, with many leaks originating from ASUS.
Beforehand, let’s introduce and compare the specifications of the GeForce RTX 4070 and GeForce RTX 4070 SUPER graphics cards to facilitate an understanding of the differences between the two and to ascertain the enhancements of the GeForce RTX 4070 SUPER.
The GeForce RTX 4070 was released on April 13, 2023. It features the AD104 GPU based on the Ada Lovelace architecture, utilizing the TSMC 5-nanometer process. It possesses 5888 CUDA cores, 184 texture units, 64 rasterization units, 46 ray-tracing cores, 46 multi-unit streaming processors, and 184 tensor cores, with a 36MB L2 cache.
The base operating frequency is 1920 MHz, with a maximum boost clock of 2475 MHz, and it comes with 12GB of GDDR6X VRAM. The memory interface is 192-bit, operating at a speed of 21Gbps, yielding a memory bandwidth of 504.2GB/s. The FP32 performance is 29.15 TFLOPS, power consumption stands at 200W, and the suggested retail price is $599. Presently, the actual market price hovers around CNY4600.
The GeForce RTX 4070 SUPER has not been released yet and is set to launch in January 2024. There is no confirmed information about its probable use of a derivative version of AD104 or A1D103. It’s based on the Ada Lovelace architecture, utilizing the TSMC 5-nanometer process, equipped with 7168 CUDA cores, 224 texture units, 80 rasterization units, 56 ray-tracing cores, 56 multi-unit streaming processors, 224 tensor cores, 48MB L2 cache.
The base operating frequency is 1980 MHz with a maximum boost clock of 2475 MHz. It will come with 12GB of GDDR6X VRAM, featuring a 192-bit memory interface, operating at a speed of 21Gbps, resulting in a memory bandwidth of 504.2GB/s. Its FP32 floating-point performance is 35.48 TFLOPS, power consumption is estimated at 285W, and the official suggested retail price is currently unknown.
Note: The GeForce RTX 4070 SUPER specifications are not from official NVIDIA sources but rather from a third party. The information provided may not guarantee absolute data reliability and is intended for reference only.
Apart from the physical aspects, what concerns everyone most should be the performance and pricing of the RTX 40 SUPER series graphics cards. Currently, there are no detailed or authoritative performance evaluations leaked. Only some general statements are available that can be temporarily used as reference points.
The performance of the GeForce RTX 4070 SUPER is approximately 14% higher than that of the GeForce RTX 4070. The GeForce RTX 4070 Ti SUPER boasts a 15% performance increase over the GeForce RTX 4070 Ti. As for the GeForce RTX 4080 SUPER, it exhibits a performance boost ranging from 3% to 5% compared to the GeForce RTX 4080.
The performance increase of the GeForce RTX 4080 SUPER is relatively modest. This can be easily understood because if the performance of the GeForce RTX 4080 SUPER were significantly higher, it could potentially impact the sales of the GeForce RTX 4090 to some extent. NVIDIA wouldn’t favor such a scenario, and it’s unlikely that Jensen Huang would take such a step.
As for the release date of the RTX 40 SUPER series graphics cards, it’s scheduled for the CES event in January 2024. The specific launch dates for the three RTX 40 SUPER series cards are slightly different but all set to be released within January, a piece of information that is essentially confirmed.
Currently, the biggest suspense revolves around the pricing of the RTX 40 SUPER series cards, a highly sensitive issue that needs consideration of various factors. This includes the significant competitive pressure from rivals like AMD. NVIDIA must devise a competitive pricing strategy to appeal to consumers, ideally aiming to maintain a price level similar to the original RTX 40 series, if not slightly higher in quantity but not in price.
Furthermore, the pricing of graphics cards often has a ripple effect. After the release of the RTX 40 SUPER series, it’s worth noting whether there will be a substantial drop in prices for the previous generation RTX 30 series cards.
Presently, there is no detailed performance evaluation or indicative pricing available for the RTX 40 SUPER series cards. Everyone will need to be patient and await further information. The editor will share the latest updates as soon as possible, so please stay tuned.
Related:
- Holding Unsold RTX 3060 & 3050? Foreseeing Price Plunge
- RTX 4080 SUPER vs RTX 3090 Ti: Game Performance Test
- RTX 4060 Laptop vs Desktop Gaming Tests: Comparable?
- RTX 5080 vs RTX 4080: Major Performance Differences