Now in our store! The Zotac GeForce RTX 4080 Trinity and AMP Extreme AIRO

Inspired by aerodynamic concepts, the ZOTAC GAMING GeForce RTX 4080 16GB AMP Extreme AIRO leverages an AIR-optimized design to make the most of the world's most advanced gaming GPU, powered by NVIDIA's Ada Lovelace architecture. For gamers and enthusiasts who want the ultimate in performance, the AMP Extreme AIRO graphics card has it all - from unmatched unparalleled performance to outstanding thermal and noise performance.

The AD103 GPU is based on TSMC's 4N NVIDIA Custom Process, which is basically a customized 5nm process. Like the flagship RTX 4090, the RTX 4080 GPU is powered by the Ada Lovelace architecture. The GeForce RTX 4080 has 9,728 CUDA cores that deliver 49 shader TFLOPS performance for traditional raster graphics; 304 fourth-generation Tensor cores, which offer 780 tensor TFLOPS (with sparsity) for AI processing and DLSS; and 76 third-generation Ada-RT cores capable of 113 RT-TFLOPS for the next generation of ray-traced graphics, and 16GB of GDDR6X memory.

When the GeForce RTX 20 series GPUs were introduced and released, ray tracing and Deep Learning Super Sampling (DLSS 1.0) were the key features. The RTX 30 series GPUs are basically faster versions of the RTX 20 series. This time, with the release of the RTX 40 GPUs, NVIDIA introduces DLSS 3 along with other improvements. NVIDIA's DLSS 3 combines Super Resolution, Frame Generation and NVIDIA's Reflex technologies. DLSS 3 significantly increases frame rate while reducing latency, resulting in a smooth and responsive gaming experience. The technology sounds fantastic on paper. DLSS 2 has increased the game's refresh rate in an impressive way. This was useful, especially for graphics cards in the lower and mid-range performance classes. However, unlike DLSS 2, DLSS 3 is limited to RTX 40 series GPUs. And at the time of writing this review, few games support DLSS 3. However, expect that to change soon as the technology is adopted.

 

The most important features

 

• NVIDIA Ada Lovelace streaming multiprocessors: Up to 2x the performance and energy savings • 16GB GDDR6X graphics memory
• 4th Gen Tensor Cores: Up to 2x AI performance • Air Optimized Design
• 3rd Gen RT Cores: Up to 2x ray tracing performance • IceStorm 2.0 Advanced Cooling
• Game Ready and NVIDIA Studio Drivers • FREEZE fan stop, active fan control
• NVIDIA® GeForce Experience™ • SPECTRA 2.0 RGB lighting
(compatible with external LED strips)
• NVIDIA Broadcast, NVIDIA G-SYNC® • Dual BIOS
• NVIDIA GPU Boost™ • Die-cast metal RGB LED backplane

 

All GeForce RTX 40 series graphics cards on the market have all the features that the Ada Lovelace architecture offers. After all, they are powered and manufactured by the same GPU. What sets the RTX 4080 graphics cards apart from their add-in card partners is the design and aesthetics. Some manufacturers have their own PCBs apart from the card's cooler. But basically all RTX 4080 cards have the same performance as NVIDIA sets the limits and controls how these graphics cards perform. Zotac introduced a new graphics card design with the release of the RTX 40 series. You have the Trinity range and the AMP Extreme AIRO. The Trinity (OC) cards are cheaper and have less RGB lighting. The non-OC variant features the base clock speed, while the OC variant has factory overclock settings. The AMP Extreme AIRO has a higher or faster boost clock rate by default. But don't worry, as I mentioned earlier, they will all perform similarly regardless of whether it's an OC or non-OC variant, all thanks to NVIDIA GPU Boost technology. The Zotac GeForce RTX 4080 Trinity OC features Zotac's advanced IceStorm 2.0 cooling system. It also features a "freeze" fan stop, active fan control, and dual BIOS. As for the RGB lighting, it features Spectra 2.0 RGB lighting compatible with external LED strips. Let's take a closer look at the card itself to see how it performs.

 

 

GeForce RTX 4080 Specs

Graphics CardGeForce RTX 4090GeForce RTX 4080GeForce RTX 3080 TiRTX 3080 10GB
Fabrication Process TSMC 4N
NVIDIA Custom Process
TSMC 4N
NVIDIA Custom Process
Samsung 8nm
8N NVIDIA Custom Process
Samsung 8nm
8N NVIDIA Custom Process
Transistor Count 76.3 billion 45.9 billion 28.3 billion 28.3 billion
Architecture Ada Lovelace Ada Lovelace Ampere Ampere
GPU AD102 AD103 GA102 GA102
Graphics Processing Clusters 11 7 7 6
Texture Processing Clusters 64 38 40 34
Streaming Multiprocessors 128 76 80 68
CUDA Cores 16384 9728 10240 8704
Tensor Cores 512 (4th Gen) 304 (4th Gen) 320 (3rd Gen) 272 3rd gen
RT Cores 128 (3rd Gen) 76 (3rd Gen) 80 (2nd Gen) 68 2nd gen
Texture Units 512 304 320 272
ROPs 176 112 112 96
Boost Clock 2520 MHz 2505 MHz 1665 MHz 1.71 GHz
Memory Clock 10501 MHz 11200 MHz 9500 MHz 1.44 GHz
Memory Data Rate 21 Gbps 22.4 Gbps 19 Gbps 19 Gbps
L2 Cache Size 73728 KB 65536 KB 6144 KB 5120 KB
Total Video Memory 24 GB GDDR6X 16 GB GDDR6X 12 GB GDDR6X 10 GB GDDR6X
Memory Interface 384-bit 256-bit 384-bit 320-bit
Total Memory Bandwidth 1008 GB/s 716.8 GB/s 912 GB/s 760.3 GB/s
Texture Rate 1290.2 GigaTexels/sec 761.5 Gigatexels/sec 532.8 Gigatexels/sec 465.1 GTexel/s
Connectors 3 x DisplayPort 1 x HDMI 3 x DisplayPort 1 x HDMI 3 x DisplayPort 1 x HDMI 3 x DisplayPort 1 x HDMI
Minimum Power Supply 850 Watts 750 Watts 750 Watts 750W
Total Graphics Power (TGP) 450 Watts 320 Watts 350 Watts 320 W
PCI Express Interface Gen 4 Gen 4 Gen 4 Gen 4
Publication date October 2022 November 2022 June 2021 September 2020

 

 

Related products