Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
basically refresh is re-charging the caps inside the SDRAM, you can do it using internal counter, by issuing refresh command every 64mS, which is setting up row address. Assume you do video decoding in this case yo are constantly accessing to the SDRAM sequentially, row by row, which is eventually, selecting each row, now in any kind of video processing you need to have at least two buffers, one buffer is using to download your stream, other buffer you playing , in case in MPEG you will have up to 4 buffers, which you are constatlly accessing , and you do full buffer scanning
not like if you executing code and stick in the loop and sometimes executing some subroutines
Your mean that if SDRAM got constant accessed and every video frame/line is going to be processed less than 64 ms then we do not need to issue refresh command for keeping those data.
If you scanning same rows for read(write) your data will be refreshed. If you processing RGB, you still need to do color space conversion from YUV, also if you enhancing video ie from 24 to 36 bits or from SD to HD you still have to think you have you accessing SDRAM
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.