The NVIDIA GeForce GTX 980 performance review
- < Prev
-
Page 1 of 3 - The NVIDIA GeForce GTX 980 performance review
Page 1 of 3
- Next >
The NVIDIA GeForce GTX 980 performance review
** Note - This review was conducted with an early set of drivers and doesn't represent the potential of the new card. We've since tested the card again on 2nd October with newer drivers that represent its capabilities better.
The NVIDIA GeForce GTX 980 performance review
The NVIDIA GeForce GTX 980 graphics card marks the first attempt by the company to make full use of the Maxwell architecture. However, it is not the first to make use of this new GPU architecture as that honor goes to the GeForce GTX 750 Ti. In fact, the GTX 980 and GTX 970 are based on the second generation Maxwell design, which has numerous improvements and new features that you can read up on here.
To date, the GeForce GTX 980's new GM204 GPU is the fastest chip from the GeForce series. For the GTX 980, it has a base clock speed of 1127MHz, and a 1216MHz boost clock. Despite its sky-high clock speeds, the rated TDP of the card is a low 165W. One of the touted features is the new card's improved power efficiency. As we know that the Maxwell architecture is based on the same 28nm manufacturing process as the previous generation Kepler GPU; such improvements come entirely from the strengths of the new microarchitecture.
The Maxwell architecture features a reworked Streaming Multiprocessor (SM) known as the Maxwell Streaming Multiprocessor (SMM). The GM204 features 16 SMMs, that make up four Graphics Processing Clusters (GPC) that are also found on Kepler and Fermi GPUs. However, the GM204 GPU has only four 64-bit memory controllers, like the GTX 680 chip, and a collective memory bus that is 256-bit wide. For the first generation Kepler GPU, the reduced memory bus was attributed to its targeted (and reduced) die size of 294mm squared; however, with larger die size of the GM204 (384mm squared), the number of memory controllers remained the same. But the GTX 980 video memory capacity stands at 4GB, with a clock speed of 7010MHz to provide a rather low memory bandwidth of 224.4GB/s - roughly a third that of the GeForce GTX 780 Ti.
The increased capacity of its video memory buffer will be handy as the new card is touted to be able to drive a pair of Utra HD 4K displays at 60Hz refresh rate. In addition, the new HDMI 2.0 standard is supported - this new standard will be able to support 4K video at up to 60fps. In terms of video connectivity options, there are three DisplayPort 1.2 connectors, one HDMI 2.0 connector, and one dual-link DVI port. The maximum resolution supported by the card is a whopping 5120 x 3200 pixels!
Here's a look at how the NVIDIA GeForce GTX 980 card compares against the competing cards:
NVIDIA GeForce GTX 980 | NVIDIA GeForce GTX 780 Ti | NVIDIA GeForce GTX Titan | NVIDIA GeForce GTX 780 | AMD Radeon R9 290X | |
Core Code |
|
|
|
|
|
---|---|---|---|---|---|
GPU Transistor Count |
|
|
|
|
|
Manufacturing Process |
|
|
|
|
|
Core Clock |
|
|
|
|
|
Stream Processors |
|
|
|
|
|
Stream Processor Clock |
|
|
|
|
|
Texture Mapping Units (TMUs) |
|
|
|
|
|
Raster Operator units (ROP) |
|
|
|
|
|
Memory Clock (DDR) |
|
|
|
|
|
Memory Bus width |
|
|
|
|
|
Memory Bandwidth |
|
|
|
|
|
PCI Express Interface |
|
|
|
|
|
Power Connectors |
|
|
|
|
|
Multi GPU Technology |
|
|
|
|
|
DVI Outputs |
|
|
|
|
|
HDMI Outputs |
|
|
|
|
|
DisplayPort Outputs |
|
|
|
|
|
HDCP Output Support |
|
|
|
|
|
Test Setup
These are the specifications of our graphics testbed:
- Intel Core i7-3960X (3.3GHz)
- ASUS P9X79 Pro (Intel X79 chipset) Motherboard
- 4 x 2GB DDR3-1600 G.Skill Ripjaws Memory
- Seagate 7200.10 200GB SATA hard drive (OS)
- Western Digital Caviar Black 7200 RPM 1TB SATA hard drive (Benchmarks + Games)
- Windows 7 Ultimate SP1 64-bit
Below is the list of cards we'll be testing. For the reference GTX 780 Ti card, we had the ASUS GeForce GTX 780 Ti 3GB GDDR5. We also threw in a typical end-user GTX 780 Ti card, which is the Gigabyte GeForce GTX 780 Ti Windforce 3x OC. The top-end AMD R9 290X "Hawaii" card representative was the ASUS Radeon R9 290X DirectCU II OC, in order to have a gauge of the GTX 980 performance against rival AMD.
- NVIDIA GeForce GTX 980 4GB GDDR5 (ForceWare 343.91)
- NVIDIA GeForce GTX 780 Ti 3GB GDDR5 (ForceWare 340.52)
- Gigabyte GeForce GTX 780 Ti Windforce 3x OC 3GB GDDR5 (ForceWare 340.52)
- ASUS Radeon R9 290X DirectCU II OC 4GB GDDR5 (AMD Catalyst 14.4)
Benchmarks
Here's the full list of benchmarks that we'll be using for our assessment:-
- Futuremark 3DMark 2013
- Crysis 3
- Call of Duty: Ghosts
- Thief
- Hitman: Absolution
For our temperature, power consumption and overclocking testing, 3DMark 2013 was used.
- < Prev
-
Page 1 of 3 - The NVIDIA GeForce GTX 980 performance review
Page 1 of 3
- Next >