NVIDIA Quadro RTX 4000 Review: Turing Powered Pro Graphics
NVIDIA Quadro RTX 4000 - Turing For Workstations
Although the Turing-based GeForce RTX 2000 series hit store shelves first, it was about a week before their initial unveiling that NVIDIA CEO Jensen Huang took to the stage and announced the Quadro RTX family of workstation-class GPUs, targeting professional graphics applications. As of today, Quadro RTX series consists of four graphics cards, ranging from the monstrous Quadro RTX 8000 with its 48GB of GDDR6 to the Quadro RTX 4000, which targets more mainstream content creation professionals. It’s the later that we’ll be showing you here today.
Looking at the Quadro RTX 4000 in light of some of NVIDIA’s other pro-graphics cards, it would appear to be a rather tame solutions. It’s a thin, unassuming graphics card, that consumes only a single slot, with a relatively simple, boxy fan shroud that doesn’t feature any sort of outlandish design elements, save for a racing stripe and some branding. But don’t let that fool you – this card has some teeth, as you’ll see a little later.
The NVIDIA Quadro RTX 4000 is based on the Turing TU106 GPU, similar to the GeForce RTX 2060 and RTX 2070. Its GPU will boost up to 1,545MHz and it is paired to 8GB of GDDR6 over a 256-bit interface, with an effective data rate of 13Gbps. At that speed, the card offers up to 415GB/s of peak memory bandwidth.
In comparison to the Pascal-based Quadro P4000 it supplants in NVIDIA’s mainstream pro-graphics line-up, the Quadro RTX 4000 offers significantly more performance across the board and much more memory bandwidth, plus some additional capabilities thanks to the Turing-based GPUs cache structure and inherent features. The Quadro RTX 4000 offers more CUDA cores (2,304 vs. 1,792) and faster memory, in addition to those Tensor and RT cores. The Quadro RTX 4000 has 288 Tensor cores and 36 RT cores that are simply not present on Pascal (or any other GPU architecture at this time).
In terms of its design, the Quadro RTX 4000 stands at about 9.5” long in total, though the actual PCB accounts for only about 70% of its length. The single-slot cooler extends out past the end of the GPU and a single, barrel-type fan sit at the far edge. The fan pulls air in, blows it across thin heatsinks that cover the GPU, RAM, and VRM and exhausts it through venting at the other end.
Save for the stylized artwork on the front of the cooler’s fan shroud, and the shape of the venting, the Quadro RTX 4000 has an overarching design that’s similar to the previous-gen P4000. They are both single-slot designs, with 90-degree angles all around, and short PCBs.
The top edge of the card also features some branding, i.e. the NVIDIA logo and RTX 4000 badge, along with a couple of connectors for the stereo and sync ports, which work with the optional stereo display and Quadro Sync modules. What’s missing on the Quadro RTX 4000 versus its predecessor are any SLI connectors.
The Quadro RTX 4000 has a much higher TDP than the Quadro P4000 as well (160 watts vs. 105 watts), but the card still requires only a single supplemental power feed. At the far edge of the cooler, there is a single 8-pin PCI Express power connector that points out from the back – not up to the side of a chassis like most consumer GPUs.
The outputs on the Quadro RTX 4000 consist of a trio of full-sized DisplayPorts (DP1.4) and a single USB-C type VitualLink conncector, which offers a 4 lane HBR3 DisplayPort and USB 3.1 Gen 2 over a single connector.
Looking at the Quadro RTX 4000 in light of some of NVIDIA’s other pro-graphics cards, it would appear to be a rather tame solutions. It’s a thin, unassuming graphics card, that consumes only a single slot, with a relatively simple, boxy fan shroud that doesn’t feature any sort of outlandish design elements, save for a racing stripe and some branding. But don’t let that fool you – this card has some teeth, as you’ll see a little later.
 |
|
The NVIDIA Quadro RTX 4000 is based on the Turing TU106 GPU, similar to the GeForce RTX 2060 and RTX 2070. Its GPU will boost up to 1,545MHz and it is paired to 8GB of GDDR6 over a 256-bit interface, with an effective data rate of 13Gbps. At that speed, the card offers up to 415GB/s of peak memory bandwidth.
Save for the stylized artwork on the front of the cooler’s fan shroud, and the shape of the venting, the Quadro RTX 4000 has an overarching design that’s similar to the previous-gen P4000. They are both single-slot designs, with 90-degree angles all around, and short PCBs.
