The $ 5,000 AMD Radeon PRO W6800X Duo graphics card is 14% faster than the NVIDIA RTX A6000 in the Octane rendering benchmark



[ad_1]

The AMD Radeon PRO W6800X Duo graphics card with dual Big Navi 21 GPUs beat NVIDIA’s GeForce RTX 3090 and RTX A6000 in the Octane Render benchmark. The benchmarks published by Bare Feats show more than Octane Render and also compare the Monster Card in other content creation apps with older Pro & RX cards.

The $ 5,000 AMD Radeon PRO W6800X Duo graphics card is 14% faster than the NVIDIA RTX A6000 in the Octane rendering benchmark

The AMD Radeon PRO W6800X Duo graphics card is the only dual-chip graphics card available in the RDNA 2 family and is exclusive to the Apple Mac Pro. The card was announced last month and costs US $ 5,000 to buy from the Apple Store. The card comes in the MPX module and is only compatible with the Mac Pro 2019 system. Here are some of its highlights:

Schenker introduces gaming-focused XMG APEX ‘AMD Ryzen 5000’ and entry-level XMG FOCUS ‘Intel 11th Gen’ laptops – with RTX 3050 Ti GPU and up to RTX 3070

  • Two W6800X GPUs, each with 60 compute units and 3,840 stream processors
  • 64 GB of GDDR6 memory (32 GB per GPU), each with a memory bandwidth of 512 GB / s
  • Up to 30.2 teraflops in single precision or 60.4 teraflops in semi-precision
  • Four Thunderbolt 3 ports and one on-board HDMI 2.0 port
  • Four DisplayPort connections routed to the system to support internal Thunderbolt 3 ports
  • Support for up to eight 4K displays, four 5K displays, or six Pro Display XDR displays
  • Full-height MPX module fills an MPX bay and uses additional power and PCIe bandwidth
  • Integrated Infinity Fabric Link connection connects two W6800X GPUs up to 84 Gb / s in each direction; the external Infinity Fabric Link connection allows two W6800X Duo modules to connect four W6800X GPUs

AMD Radeon PRO W6800X Duo GPU Specifications

The flagship variant of them all is the Big Navi 21 dual GPU Radeon Pro W6800X which offers up to 64GB of GDDR6 memory and dual 60 CU GPU arrays that offer a total of 120 compute units or 7680 cores. The clock speed of each core is maintained at 1979 MHz, providing up to 30.2 TFLOP of FP32 and 60.4 TFLOP of FP16 computing power. The card offers up to 1 TB / s of memory bandwidth with its dual 256-bit bus interface. The two reduced Navi 21 GPU arrays should bring the total power of the DUO to around 400W.

AMD Radeon PRO W6000 Series Graphics Cards

Graphic card GPU Process node Computing units / Cores Calculation rate (FP32) Calculation rate (FP16) VRAM Memory bus Memory bandwidth TDP Price
Radeon Pro W6800X Duo Navi 21 TSMC 7 nm 60/7680 30.2 0TFLOP 60.40 TFLOP 64 GB of GDDR6 256 bits x2 1 TB / s 400W To be determined
Radeon Pro W6900X Navi 21 TSMC 7 nm 80/5120 22.20 TFLOP 44.40 TFLOP 32 GB of GDDR6 256 bits 512 GB / s 300W To be determined
Radeon Pro W6800X Navi 21 TSMC 7 nm 60/3840 16.00 TFLOP 32.00 TFLOP 32 GB of GDDR6 256 bits 512 GB / s 300W To be determined
Radeon Pro W6800 Navi 21 TSMC 7 nm 60/3840 17.83 TFLOP 35.66 TFLOP 32 GB of GDDR6 256 bits 512 GB / s 250W $ 2249 US
Radeon Pro W6600 Navi 23 TSMC 7 nm 28/1792 10.40 TFLOP 20.8 TFLOP 8 GB of GDDR6 128 bits 224 GB / s 100W US $ 649
Radeon Pro W6600M Navi 23 TSMC 7 nm 28/1792 10.40 TFLOP 20.8 TFLOP 8 GB of GDDR6 128 bits 224 GB / s 90W To be determined

AMD Radeon PRO W6800X Duo GPU benchmarks

As for benchmarks, the AMD Radeon PRO W6800X Duo has been tested in several tests, and the main one we’ll be focusing on is Octane Benchmark which generally offers better performance on NVIDIA RTX cards. The Radeon PRO finished rendering in 14 seconds with a score of around 717 points. NVIDIA’s fastest workstation card, the RTX A6000, finishes the same render in 16 seconds (628 scores) while the fastest gaming graphics card, the RTX 3090, finishes rendering in 15.2 seconds (662 scores).

  • AMD Radeon PRO W6800X Duo – 717 points / 14 seconds
  • NVIDIA GeForce RTX 3090 – 662 points / 15.2 seconds
  • NVIDIA RTX A6000 – 628 points / 16 seconds

This puts the AMD Radeon PRO W6800X Duo at 14% faster than the NVIDIA RTX A6000 and 8% faster than the RTX 3090. Now, in terms of price, the Duo costs a lot more since the A6000 can be found for around $ 4,500 to $ 5,000. US while the RTX 3090 has an MSRP of US $ 1499, but even with inflated prices it can be found between $ 2000 and US $ 3000. Compared to the latest generation AMD PRO Vega II Duo, the new solution based on Navi 21 is approximately 18% faster and 37% faster than the Radeon RX 6900 XT (Single-GPU). The outlet also shared some numbers in Da Vinci Resolve which you can see below:



[ad_2]
Source link