Dedicated graphics cards RTX 40 SUPER heading for a close start. A few weeks ago we covered the first leaks that hit the Internet about this new product line, and today a special NVIDIA event was announced before the gates of CES 2024, which would confirm the previous information.

The Technology and Electronics Show in Las Vegas is not only the first to be held every year, but it is also the most important in the annual calendar in terms of the number of exhibitors, visitors, journalists and importance to the industry. Simply put, it’s the ideal term for presenting products that cover the entire first part of the year. A good example is Intel, which announces its new generations of processors every year. Even the green giant participates in the fair every year and there are always notices of interest. In 2023, it introduced the RTX 40 Mobile and Ti variants of the desktop models. There will be news next year.

You already know that NVIDIA usually sells special lines of their main graphics card models. We are talking about those that bear the designation ‘SUPER’ or the same ‘Ti’. Architecturally they’re not new cards and those in question will use the familiar ‘Ada Lovelace’ like the rest of the 40 series. But they’re interesting because they increase the performance of the standard models, plus more supply usually leads to a price adjustment across the range. And we continue to insist: prices, official from manufacturers and real from sellers, must moderate.

RTX 40 SUPER version

From what we know so far there will be three new models this would increase the lineup of 40 series cards: RTX 4080 SUPER, RTX 4070 Ti SUPER and RTX 4070 SUPER.

We remind you of its specifications (always unofficial):

GeForce RTX 4080 Super

• Manufactured by TSMC in 5nm.
• Graphics core AD103-400.
• 10,240 shaders at an unspecified frequency.
• 320 texturing units.
• 112 raster units.
• 320 tensor cores of the fourth generation.
• 80 cores of the third generation RT.
• 320-bit memory bus.
• 20 GB of dedicated GDDR6X memory.
• TGP 320W.

GeForce RTX 4070 Ti Super

• Manufactured by TSMC in 5nm.
• Graphics core AD103-275 or AD102-175.
• 8,192 shaders at an unspecified frequency.
• 256 texturing units.
• 96 raster units.
• 256 tensor cores of the fourth generation.
• 64 cores of the third generation RT.
• 256-bit memory bus.
• 16 GB of dedicated GDDR6X memory.
• 285W TGP.

GeForce RTX 4070 Super

• Manufactured by TSMC in 5 nm.
• Graphics core AD103-175 or AD104-350.
• 7,168 shaders at an unspecified frequency.
• 224 texturing units.
• 96 raster units.
• 224 tensor kernels of the fourth generation.
• 56 cores of the third generation RT.
• 192-bit memory bus.
• 12 GB of dedicated GDDR6X memory.
• TGP 225W.

As you will see, a significant increase in the number of CuDA cores (Shaders) and – although it is not confirmed – also the operating frequencies and memory bandwidth. The TDP consumption also increases in the case of the RTX 4070 SUPER, while the size of the dedicated memory remains the same for all three models.

And the prizes…oh the prizes! It is said that they will not be far from those that were established for the launch of standard models. Which would at least force them to cut back. And it’s missing. We will tell you.