NVIDIA Launches The GeForce MX130 and MX110 “Maxwell” GPUs For Efficient Notebooks – Up To 2.5x Faster Than Intel’s UHD 620 Graphics

NVIDIA has silently launched their latest GeForce MX series mobility graphics chips which include the MX130 and MX110. Both of these chips are still based on the Maxwell GPU architecture and positioned at entry level mobility products.

NVIDIA Silently Launches The Maxwell Based GeForce MX130 and GeForce MX110 Mobility GPUs

As stated, both chips use the Maxwell architecture and are based on a 28nm process technology. The Maxwell architecture gave NVIDIA graphics chips a big boost over the competition in terms of perf per watt which still makes a lot of sense for mobility product designs. Both chips use the Maxwell GM108 GPU which is the most entry level SKU of the entire Maxwell GPU stack. These chips come in 25W variants and specifications are detailed below.

RelatedNVIDIA CEO Praises Nintendo Switch; Believes People Will “Be Blown Away”

NVIDIA GeForce MX130 – 384 CUDA Cores at 25W For 2.5X Performance Boost Over Intel UHD 620 Graphics

The NVIDIA GeForce MX130 includes the GM108 GPU with a fully enabled core. This chip packs a total of 384 CUDA cores that are clocked at 1122 MHz base and 1242 MHz boost clocks. The chip has been equipped with 2 GB of either GDDR5 or DDR3 memory which will be featured across a 64-bit bus interface. NVIDIA is claiming a 2.5x performance boost over the Intel UHD 620 graphics chip that ships within the Intel 8th generation processor package.

NVIDIA GeForce MX110 – 256 CUDA Cores at 25W For 1.5X Performance Boost Over Intel UHD 620 Graphics

The NVIDIA GeForce MX110 includes the GM108 GPU with a partially disabled core. This chip packs a total of 256 CUDA cores that are clocked at 965 MHz base and 993 MHz boost clocks. The chip has been equipped with 2 GB of DDR3 memory which will be featured across a 64-bit bus interface. NVIDIA is claiming a 1.5x performance boost over the Intel UHD 620 graphics chip that ships within the Intel 8th generation processor package.

NVIDIA GeForce MX Series

GPU NameNVIDIA GeForce MX 150NVIDIA GeForce MX 130NVIDIA GeForce MX 110
GPU FamilyPascalMaxwellMaxwell
GPU Process16nm28nm28nm
GPU CoreGP108GM108GM108
CUDA Cores384384256
Base Clock1469 MHz1122 MHz965 MHz
Boost Clock1532 MHz1242 MHz993 MHz
Memory (VRAM)2 GB2 GB2 GB
Memory TypeGDDR5GDDR5 / DDR3DDR3
Memory Bus64-bit64-bit64-bit
TDP25W25W25W
Performance Improvement Over Intel UHD 620 Graphics4x2.5x1.5x

Both chips support the most basic GeForce features such as NVIDIA Optimus, GPU Boost 2.0 and GeForce Experience. The chips are compliant with DirectX 12 and OpenGL 4.5 APIs while featuring full NVIDIA GameWorks support. There’s a big market for these chips if priced and positioned in the right way and obviously NVIDIA knows that, hence the launch.

The Intel Kaby Lake-G poses a serious threat to NVIDIA in the mobility market.

The main thing here is that while these deliver good performance per watt, the real challenge will be to see them going against AMD’s Ryzen mobility lineup which is equipped with Vega graphics on board. Furthermore, Intel’s and AMD’s co-venture in developing Kaby Lake-G (Intel 8th Gen CPU + AMD Vega GPU + HBM2 VRAM) is going to be a serious challenger in the industry when it arrives next year. NVIDIA currently is the top vendor for high-end and great mainstream GPUs in the mobile sector. It will be interesting to if NVIDIA maintains their mobility strength in the entry level sector which will soon be filled by a strong entry level AMD Vega core that will be featured on both AMD and Intel powered machines.

Tweet Share

Related

NVIDIA Slashes Prices on GeForce 900 Series Maxwell Cards - GTX 980 Ti Now Selling Around $450 US, GTX 970 Under $300 US

NVIDIA Launches Tesla M10 Quad Maxwell Graphics Card - 32 GB Memory, 2560 Cores Aimed at Virtual Computing

NVIDIA Readies Three Pascal GP104 GPUs To Replace GeForce GTX 980 Ti, GTX 980 and GTX 970 in June - GP104 SKUs Leaked

NVIDIA 16nm Pascal Based Tesla P100 With GP100 GPU Unveiled - Worlds First GPU With HBM2 and 10.6 TFLOPs of Compute On A Single Chip

NVIDIA's 5th Gen Flagship Pascal GPU is 70% Faster Than Maxwell in CUDA Deep Neural Network Workloads