Hi, I'm using an ADSP-TS201 TigerSHARC Processor on a T2V6 board (http://www.bittware.com/products/hardware/prod_desc.cfm?ProdShrtName=T2V6). I'm compiling with VisualDsp++ 5.0
I'm not getting the processing performance claimed by the comment block in some assembler files. For example, I'm using the VisualDsp++ DSP library. I'm using the function _cvecdot.asm. In the assembler file, it says that this function should take ~(2 * n) cycles to complete. (I assume that n is the number of complex elements, is it true?). I have an array of 1156 complex samples, so I would expect it to take ~2312 cycles; or ~4.63 us @500Mhz. But I'm getting ~7.7 us. The same goes for other functions (_vecdot.asm, etc) , all taking approximately almost 100% more time that the declared in the comments in the .asm files.
I'm compiling for debug, but I think this should be a problem, because this functions are written in assembler.
Does anybody know if I'm doing something wrong, or if this performance is expected?
Thank you very much for your time, and sorry for my english!