NVIDIA GeForce RTX 4070 GPU appears in new Reflex Latency Performance benchmarks

In Counter-Strike 2, NVIDIA compared the reflex latency performance of the upcoming GeForce RTX 4070 GPU with older RTX 3060 & GTX 1060 GPUs. NVIDIA is bringing its Reflex technology, which reduces latency in a variety of games, especially eSports games, to Counter-Strike 2, as reported by The Verge. Valve’s Counter-Strike 2 recently made headlines for its innovative new features, including dynamic smoke and sub-tick responsiveness, which players of the venerable first-person shooter will experience when it launches.

NVIDIA has shared some new latency benchmarks showcasing how their technology helps improve the gaming experience in Counter-Strike 2 because the game is all about responsiveness. NVIDIA’s Reflex is one such technology that is essential because it makes the game much more responsive. With the Reflex mode on, we experience improvements ranging from 20% to 70%.

It’s intriguing how much better results are produced by the Reflex Mode on older GPUs. For instance, using a GeForce RTX 4070 results in a 20% sold reduction (8 ms vs 10 ms), but downgrading to an older GPU, like the RTX 3060, results in up to a 32% reduction (11 ms vs 16 ms). When compared, even the GTX 1060 exhibits a significant 35% reduction.

This demonstrates how crucial NVIDIA’s Reflex technology is for games like Counter-Strike 2. The comparison that has been drawn here is what matters more. NVIDIA is comparing the GeForce RTX 4070 to cards in different price ranges, including the RTX 3060 and GTX 1060, in addition to using an unreleased graphics card.

The GeForce RTX 4060 may not yet be available, and since the RTX 3060 and GTX 1060 are the most widely used GPUs on the market, NVIDIA may have needed a GPU from the 40 series to compare with these two. The most popular GPU on Steam at the moment is the RTX 3060.

The AD104-250/251 GPU core is anticipated to be included in the NVIDIA GeForce RTX 4070 GPU.

Nvidia GeForce RTX 4070 GPU has a 36 MB L2 cache, which is 9x more than the GA104 GPU in the RTX 3070. It has 5888 CUDA cores, 12 GB of GDDR6X VRAM clocked at 21 Gbps, and drives bandwidth up to 504 GB/s.

A 200W TGP is anticipated for the graphics card, which equates to an 8% power savings over the RTX 3070. The slide also shows the average gaming power, which is listed at 186 Watts and is 13.5% and 22.5% lower than the RTX 3070 and RTX 3070 Ti, respectively. While the card should perform on par with or faster than the RTX 3080 12 GB, which consumed anywhere between 300-320W, this is the same power consumption as the NVIDIA RTX 3060. That is a significant improvement in efficiency.

Also read:

• AMD Ryzen 7 7800X3D is reportedly only slightly slower than the AMD Ryzen 7 7900X3D
• Asus ROG Ally: A supposed April Fools Joke which hit us right in face with the brick of reality

Source