c++ - Same Direct2D application performs better on a "slower" machine -
i wrote direct2d application displays number of graphics.
when run application takes 4 seconds display 700,000 graphic elements on notebook:
intel core i7 cpu q 720 1.6 ghz nvidia quadro fx 880m
according direct2d msdn page:
direct2d user-mode library built using direct3d 10.1 api. means direct2d applications benefit hardware-accelerated rendering on modern mainstream gpus.
i expecting same application (without modification) should perform better on different machine better specs. tried on desktop computer:
intel xeon(r) cpu 2.27 ghz nvidia geforce gtx 960
but took 5 seconds (1 second more) display same graphics (same number , type of elements).
i know how can possible , causes.
it's impossible sure without measuring. however, gut tells me melak47 correct. there no lack of gpu acceleration, it's lack of bandwidth. integrated gpus have access same memory cpu. can skip step of having transfer bitmaps , drawing commands across bus dedicated graphics memory gpu.
with 2d workload, gpu spending of time waiting on memory. in case, integrated gpu has advantage. suspect second feel, geforce waiting on graphics coming across motherboard bus.
but, profile , enlighten us.
Comments
Post a Comment