Cast your vote Do you agree or disagree with GPUBoss?

Thanks for adding your opinion. Follow us on Facebook to stay up to date with the latest news!

Differences What are the advantages of each

Front view of Radeon HD 7670M

Reasons to consider the
AMD Radeon HD 7670M

Report a correction
Significantly higher memory bandwidth 57.6 GB/s vs 12.8 GB/s Around 4.5x higher memory bandwidth
Significantly higher effective memory clock speed 3,600 MHz vs 1,600 MHz 2.2x higher effective memory clock speed
Better floating-point performance 576 GFLOPS vs 240 GFLOPS Around 2.5x better floating-point performance
Higher pixel rate 9.6 GPixel/s vs 2.5 GPixel/s More than 3.8x higher pixel rate
More shading units 480 vs 96 384 more shading units
More render output processors 16 vs 8 Twice as many render output processors
Wider memory bus 128 bit vs 64 bit 2x wider memory bus
Slightly more texture mapping units 24 vs 16 8 more texture mapping units
Higher memory clock speed 900 MHz vs 800 MHz Around 15% higher memory clock speed
Lower TDP 20W vs 33W Around 40% lower TDP
Front view of GeForce 720M

Reasons to consider the
Nvidia GeForce 720M

Report a correction
More memory 2,048 MB vs 1,024 MB 2x more memory

Features Key features of the Radeon HD 7670M  vs GeForce 720M 

memory bandwidth Rate at which data can be read from or stored in onboard memory

Radeon HD 7670M
57.6 GB/s
GeForce 720M
12.8 GB/s
Entries of a 100W Kaveri APU called (KV Spectre Desktop 100W 1305) were spotted on Sisoft Sandra’s benchmark portal along with some specifications regarding the GPU, memory and a few bandwidth tests.
Radeon HD 7670M | by hassan-mujtaba (Sep, 2013)

pixel rate Number of pixels a graphics card can render to the screen every second

Radeon HD 7670M
9.6 GPixel/s
GeForce 720M
2.5 GPixel/s
Since Core M is the first 14nm device, it will meet the demands of next iteration of mobility devices boasting better performance and efficiency.
Radeon HD 7670M | by hassan-mujtaba (Sep, 2014)

texture rate Speed at which a graphics card can perform texture mapping

Radeon HD 7670M
14.4 GTexel/s
GeForce 720M
10 GTexel/s
Of course efficiency increases can only take you so far, so along with the above changes Intel is also making some more fundamental improvements to Broadwell’s math performance.
Radeon HD 7670M | by Ryan-Smith (Aug, 2014)

floating point performance How fast the gpu can crunch numbers

Radeon HD 7670M
GeForce 720M
Floating point multiplication is seeing a sizable reduction in instruction latency from 5 cycles to 3 cycles, and meanwhile division performance is being improved by the use of an even larger Radix-1024 (10bit) divider.
Radeon HD 7670M | by Ryan-Smith (Aug, 2014)

shading units Subcomponents of the gpu, these run in parallel to enable fast pixel shading

A recent founding by Planet3DNow reveals that the top Kaveri APU parts won’t be restricted to only 512 Stream processors and can feature upto 832 Stream processors.
Radeon HD 7670M | by hassan-mujtaba (Sep, 2013)

texture mapping units Built into each gpu, these resize and rotate bitmaps for texturing scenes

Unfortunately it's also difficult to determine how many ROPs or TMUs are powering it, but the most recent version of GPU-Z pegs it at 16 ROPs which would theoretically work out to 64 TMUs.
Radeon HD 7670M | by Dustin-Sklavos (Mar, 2012)

Specifications Full list of technical specs


Radeon HD 7670M  vs
GeForce 720M 
GPU brand AMD Nvidia
GPU name Thames GF117
Market Laptop Laptop
Clock speed 600 MHz 625 MHz
Is dual GPU No No
Reference card AMD Radeon HD 7670M 600 MHz 1 GB None

raw performance

Shading units 480 96
Texture mapping units 24 16
Render output processors 16 8
Pixel rate 9.6 GPixel/s 2.5 GPixel/s
Texture rate 14.4 GTexel/s 10 GTexel/s
Floating-point performance 576 GFLOPS 240 GFLOPS


Radeon HD 7670M  vs
GeForce 720M 
Memory clock speed 900 MHz 800 MHz
Effective memory clock speed 3,600 MHz 1,600 MHz
Memory bus 128 bit 64 bit
Memory 1,024 MB 2,048 MB
Memory type GDDR5 DDR3
Memory bandwidth 57.6 GB/s 12.8 GB/s

noise and power

TDP 20W 33W

Read more


comments powered by Disqus