Do you have a question?
Please contact us via email, contact form or phone
+49-7243-2167-23 (head office, Germany) or directly your local contact.
Special note
This knowledge base is updated regularly. If you need an information and don't find it here, please don't hesitate to contact us! Of course, we're also to your disposal at the phone:
+49 7243 2167-0
Floating point calculations
The floating point operations are very slow. Is there something wrong in my program?
Texas Instruments TMS320C6xxx DSPs used in VC Smart Cameras are not optimized for floating points operations. Please use integer calculation wherever possible, for example by using a scaling factor.
See PDF for more information. Please make sure to be logged in!
If floating point calculation is absolutely needed, the TMS320C62x/C64x FastRTS Library can be used. The execution times are then 2 to 4 times faster than with the "normal" library.
This library is free and can be downloaded from TI.