Introduction

GIGABYTE GeForce RTX 4070 SUPER AORUS Master is the company's top of the line custom-design based on the new GeForce RTX 4070 SUPER that goes on sale from today. GIGABYTE's coveted AORUS Gaming brand has several tiers, with the AORUS Master and AORUS Xtreme being the top ones. The company also has a more affordable AORUS Elite product among its custom RTX 4070 SUPER lineup. NVIDIA today kicked off its RTX 40 SUPER series, a mid lifecycle refresh for the RTX 40-series Ada, which aims to give gamers more performance at existing price points, by increasing shaders, or in some cases, even more memory. NVIDIA recommends the RTX 4070 SUPER for no holds barred AAA gaming at 1440p with ray tracing, although our testing with even the older RTX 4070 has shown these cards to be capable of 4K Ultra HD gameplay, if you know your way around your game's settings, or can get GeForce Experience to pick the right settings for you. You can also take advantage of DLSS, which is supported on nearly every new AAA and e-sports release these days, or even the newer DLSS 3 Frame Generation.



NVIDIA found itself with a gaping performance gap between the RTX 4070 and the RTX 4070 Ti, as the RTX 4070 barely enables 76% of the shaders available on the AD104 silicon they're based on, while the RTX 4070 Ti maxes it out; a gap AMD exploited with its Radeon RX 7800 XT. NVIDIA looks to fill this with the new RTX 4070 SUPER, which uses 93% of the SIMD resources of the AD104. The RTX 4070 Ti is being retired, and in its place, the new RTX 4070 Ti SUPER is being released next week, based on the larger AD103 chip. This way, at both the $599 and $799 price points, NVIDIA is offering double-digit percent performance increases within the generation; so gamers have more value heading into the spring-summer gaming season.

The GeForce RTX 4070 SUPER, as we mentioned, is based on the same 5 nm AD104 silicon as the RTX 4070 and RTX 4070 Ti, but leans closer to the latter in terms of specs. With 56 out of 60 streaming multiprocessors (SM) enabled, it a has 7,168 CUDA cores, 224 Tensor cores, 56 RT cores, and 224 TMUs, at its disposal—for reference, the RTX 4070 only has 46 out of those 60 enabled. NVIDIA has also unlocked all 80 ROPs present on the silicon, compared to 64 on the RTX 4070; as well as all 48 MB of L2 cache, compared to 36 MB on the RTX 4070. The memory sub-system is unchanged compared to the other two—you get 12 GB of 21 Gbps GDDR6X memory across a 192-bit memory bus, yielding 504 GB/s of memory bandwidth, cushioned by a large on-die cache.

The new Ada Lovelace graphics architecture driving the GeForce RTX 40-series, introduces generational uplifts in performance and energy efficiency, thanks to the new 5 nm EUV foundry process, and SIMD-level innovations. The new Ada CUDA core, in addition to IPC increases and support for higher clock speeds over Ampere; features support for shader execution reordering, a feature that improves ray tracing performance. The new 3rd generation RT core supports displaced micro-meshes, a feature that increases complexity of ray traced objects without a linear increase in performance cost; and the new optical flow accelerator, a component that lets the GPU draw entire alternate frames using AI, without involving the main graphics rendering pipeline, which is needed for DLSS 3 Frame Generation to work. The company has also redesigned the memory sub-system with larger caches on the GPU, so it could narrow the memory interfaces while increasing the memory sizes.

The GIGABYTE RTX 4070 SUPER AORUS Master is a massive graphics card, with the company's largest version of its WindForce 3X cooling solution. This card looks like it's an RTX 4090 when installed, with its 342 mm length and 153 mm height that nearly max out what an ATX mid-tower case can fit; and an over three slot thickness. The card weighs 50% more than the NVIDIA Founders Edition; all that added weight comes from the heavier heatsink, which should not just cool the GPU better, but also offer GIGABYTE the headroom needed for factory overclocked speeds.

The GIGABYTE AORUS Master offers factory overclocked boost frequencies of 2655 MHz, compared to 2475 MHz reference, which is supported by an increased power limit for better boost frequency residency. The cooling solution features a multi fin-stack heatsink that's ventilated by three large fans. The heatsink features as many as seven copper heat pipes making contact with the GPU and other hot components over a vapor chamber plate. Each of the three 100 mm fans has shark skin-inspired grooving that enhances airflow; and double ball bearings. The cooler is 50% longer than the PCB underneath, and so nearly half the cooler's airflow goes through the heatsink, and directly out the backplate, through a large cutout. There's plenty of RGB lighting, including illuminated AORUS logos on the backplate and top; and LED rings along the fan frames. There are other gamer-friendly features to be had, including dual-BIOS, and anti-sagging brace. GIGABYTE is pricing the RTX 4070 SUPER AORUS Master at $700, which is a $100 premium over the NVIDIA baseline, and halfway between the RTX 4070 SUPER and the upcoming RTX 4070 Ti SUPER.

Short 10-Minute Video Comparing 8x RTX 4070 Super

Our goal with the videos is to create short summaries, not go into all the details and test results, which can be found in our written reviews.