Scan's TekSpek

Our Aim
To provide you with an overview on New And existing technologies, hopefully helping you understand the changes in the technology. Together with the overviews we hope to bring topical issues to light from a series of independent reviewers saving you the time And hassle of fact finding over the web.

We will over time provide you with quality content which you can browse and subscribe to at your leisure.

TekSpek GPU - Pro Graphics
NVIDIA Quadro P Series

NVIDIA Quadro P Series

Date issued:

Inventor of the Graphics Processing Unit (GPU), NVIDIA has been producing cutting-edge graphics for almost 20 years. Most of those graphics end up in the GeForce line of cards that power the best, most immersive gaming experience available.

Yet the GPU's ability to work on lots of information concurrently, besting what's on offer from the latest CPUs oftentimes by an order of magnitude, means that it is an optimum solution beyond gaming. This is where professional GPUs come in, helping scientists solve problems, power self-driving cars and conduct a form of modern artificial intelligence known as deep learning.

It is inevitable that with such rich GPU heritage, NVIDIA is well positioned to meet the needs of these new, emerging markets, and its Quadro series of graphics accelerators have been widely implemented in workstation computers designed for CAD, CGI and digital content creation.

Today's Quadro GPUs use the latest Pascal architecture also present in the GeForce 10-series gaming cards and Tesla solutions optimised for servers. As Quadro GPUs need to span a large market with varying needs of performance and power, NVIDIA currently produces seven P-series cards, and we list their pertinent performance details in the following table.

Nvidia Quadro P-series Desktop Line-up

Model Quadro P6000 Quadro P5000 Quadro P4000 Quadro P2000 Quadro P1000 Quadro P600 Quadro P400
Launch Date October 2016 October 2016 February 2017 February 2017 February 2017 February 2017 February 2017
Architecture GP102GL GP104GL GP104GL GP106GL GP107GL GP107GL GP107GL
Processors 3,840 2,560 1,792 1,024 640 384 256
Peak GPU Clock (MHz) 1,417 1,607 1,544 1.470 1,455 1,455 1,170
Peak GFLOPS (SP) 10,883 8,228 5,534 3,011 1,862 1,117 599
Memory Size (MB) 24,576 16,384 8,192 5,120 4,096 2,048 2,048
Memory Bus (Bits) 384 256 256 160 128 64 64
Memory Clock (Gbps) 10,000 10,000 7,600 8,000 7,000 7,000 7,000
Memory Bandwidth (GB/s) 480 320 243.3 160.2 112.1 56.1 56.1
TDP (Watts) 250 180 105 75 47 40 30
Connectors DVI, 4x DP 4x DP DVI, 4x DP 4x DP 4x miniDP 4x miniDP 3x miniDP
Price £4,180 £1,950 £1,400 £500 £370 £250 £135


There is something for every workstation user, with the headline NVIDIA Quadro P6000 almost 20x as powerful as the entry-level P400. The cost increases with the amount of performance on offer, understandably, and eagle-eyed readers may well be able to pick out the GeForce equivalents based on the specs alone. For example, the P5000 is very similar to the GeForce GTX 1080 while the P6000 is much akin to the GeForce GTX 1080 Ti.

A few features do separate these Quadro cards from their GeForce cousins. Typically, Quadros feature more memory - there is a whopping 24GB on the P6000 - and longer warranties. Additionally, not shown, the supporting software is exhaustively tested with the applications that are intended to run on workstations, meaning regular GeForce drivers may not have the compatibility or certification to run with, say, SolidWorks, Creo, Blender, et al.

Multiple cards are often installed into a large workstations or servers, and it's possible to install up to 10 into a standard-sized server box. More typically, however, workstations, featuring up to two Xeon CPUs, will have access to one or two cards.

Quadro P-series graphics card

The Quadro cards ape their GeForce brethren with respect to cooling, too, as the P6000, pictured above, looks very similar to the GeForce GTX 1080 Ti Founders Edition card. This is no bad thing as that cooler keeps to a dual-slot form factor.

Moving on down, cooling becomes simpler on the lower-wattage parts, with single-slot heatsinks available on the Quadro P2000, for example.

NVIDIA also understands that these professional graphic cards, whilst optimally employed in a desktop computer, are increasingly needed in laptops for in-field work. This is why the Quadro P-series GPUs are also available in approved notebook workstations.

GRAPHICS FEATURES Quadro P5000 Quadro P4000 Quadro P3000
CPU/Mobile Platform Generation Kabylake Kabylake Kabylake
GPU Architecture Pascal Pascal Pascal
NVIDIA® CUDA® Parallel Processor Cores 2048 1792 1280
Memory Size 16GB 8GB 6GB
Memory Type GDDR5 GDDR5 GDDR5
Memory Interface 256-bit 256-bit 192-bit
Memory Bandwidth 192GB/sec 192GB/sec 168GB/sec
Max Power Consumption, TGP (Graphics Sub-System) 100W 100W 75W
OpenGL 4.5¹ 4.5¹ 4.5¹
Shader Model 5.1 5.1 5.1
DirectX 12 12 12
PCI-E Generation 3 3 3
Display Port 1.4 1.4 1.4
Precision FP32 FP32 FP32
NVIDIA® VR Ready Yes Yes
NVIDIA 3D Vision® Pro Yes Yes Yes
NVIDIA® Mosaic Technology Yes Yes Yes
NVIDIA® nVIEW® Display Management Technology Yes Yes Yes
NVIDIA Optimus™ Technology Yes Yes Yes

The Quadro P3000 is only available in mobile form, and the other two have similar specifications to their desktop counterparts. Importantly, the trio offers NVIDIA's multi-monitor nView technology as standard, meaning that multiple additional monitors can be attached if the laptop workspace is too limiting.

These mobile Quadros offer the same level of software support and certification, as well, and Scan Computers, as a premium NVIDIA partner, has Quadro GPU Cards in both the desktop and mobile version in stock today.