AMD Radeon R9 M380 vs AMD Radeon R9 M385
AMD Radeon R9 M380► remove
The AMD Radeon R9 M380 is a dedicated mid-range graphics card for laptops. It is based on a 28nm GCN core with 12 compute units and a GDDR5 memory controller. Despite the similar name, the Radeon R7 M380 is a much slower as it features only 10 compute units and DDR3 memory.
Features of the R9-M380 include video decoding for MPEG-4 AVC/H.264, VC-1, MPEG-2, and Flash directly by the AMD GPU. Multi-View Codec (MVC) and MPEG-4 part 2 (DivX, xVid) HD videos are compatible as well.
The R9 series also supports automatic graphics switching between the integrated GPU and discrete GPU. Called Enduro, the technology supersedes AMD's Dynamic Switchable Graphics and is similar to Nvidia's Optimus. Furthermore, the M380 can directly support multiple monitors using Eyefinity Technology if Enduro is disabled.
Other features include ZeroCore to reduce the power consumption when the display is turned off and Power Gating to power down areas of the chip that are not used.
The integrated HD audio processor is able to transmit HD Audio (TrueHD or DTS Master Audio) over HDMI and DisplayPort (e.g., for Blu-Ray videos). Additionally, it allows audio output simultaneously and in parallel to multiple devices with the new Discrete Digital Multipoint Audio (DDMA) feature.
AMD Radeon R9 M385► remove
The AMD Radeon R9 M385 is a mid-range notebook graphics card for laptops. It features a GCN (Graphics Core Next) based architecture and is manufactured in 28nm. Up to now only a GFXBench 3.1 OpenGL benchmarks leaked.
The most likely a bit slower Radeon R9 M380 features 12 compute cores clocked at 1 GHz and a GDDR5 memory controller. According to the naming, the performance should be quite similar to it.
|AMD Radeon R9 M380||AMD Radeon R9 M385|
|Radeon R9 M300 Series|
|Codename||Strato Pro GDDR5|
|Memory||6000 MHz||6000 MHz|
|Bus||128 Bit||128 Bit|
|Max. Memory||4096 MB||4096 MB|
|DirectX||DirectX 11.1, 5.0||DirectX 11.1, 5.0|
|Transistors||1500 Million||1500 Million|
|Technology||28 nm||28 nm|
|Features||DirectCompute 5.0, OpenCL||DirectCompute 5.0, OpenCL|
* Smaller numbers mean a higher performance
1 This benchmark is not used for the average calculation