FPGA Intellectual Property
PCI Express*, Networking and Connectivity, Memory Interfaces, DSP IP, and Video IP
Intel Support hours are Monday-Fridays, 8am-5pm PST, except Holidays. Thanks to our community members who provide support during our down time or before we get to your questions. We appreciate you!

Need Forum Guidance? Click here
Search our FPGA Knowledge Articles here.
5953 Discussions

FFT MegaCore RAM allocation

Honored Contributor II



I was trying to implement a 512-point 16-bit FFT megacore in the Cyclone II family device. The device has 36 M4K RAM blocks available. The 512-point 16-bit FFT needs 22 M4K blocks. I have a question regarding the RAM allocation for the FFT megacore. The FFT Megacore IP Tool does not have an option to customize the memory usage. Does it mean that the FFT Megacore will randomly allocate memory for computation or it will automatically allocate a given chunk of FPGA on-chip RAM for computation? Is there any flexibility to use External Memory e.g. SRAM or SDRAM for FFT megacore?  


If I also want to use the FPGA on-chip RAM for data storage, e.g. I used a RAM 2-port Megafunction to access a chunk of memory, will the FFT megacore also use that piece of memory for computation? This really confuses me.  




0 Kudos
1 Reply
Honored Contributor II



The Quartus software will auto allocate the RAM location for FFT IP. It is not possible to use the external memory for FFT IP and user has no control on it.  

Once the RAM location is used by FFT IP, it is unable to allocate for other function. If the internal memory location is not sufficient, the Quartus software will fail compilation with relevant error message.  



Best Regards, 



(This message was posted on behalf of Intel Corporation)