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.
5982 Discussions

Problem understanding de-interlacer core.

Honored Contributor II


I'm trying to implement the De-interlacer core with the following spec, 

Algorithm - Bob scanline duplication with no buffering. 

Output frame rate = Input field rate. 

Clock frequency = 25 MHz. 


I have assumed the output to be as in the picture (pl refer attachment - output.jpg) 

With reference to that I expected one pixel output per clock but when I simulated it in Modelsim I found the output to be one pixel for 4 clock cycle (Pl refer attachement pic5, pic6, pic7). 

pic5 - zoom out of initial packet formation. 

pic6 - zoom in of packet formation. 

pic7 - after a few clock cycle from the start of packet. 


The pic also contains the generation of avalon-ST signal (SOP,EOP,VALID etc)...Kindly let me know if i'm generating proper signals. 

also regarding the packet formation for avalon-st, should i make each line in a frame as a packet or can i take the entire frame as one single packet? 


I wanted to know if there are any data sheet or documentation available on the Deinterlacer core or any timing diagram regarding the functionality of the core. I already went through VIP user guide.  


Thanks and regards, 

0 Kudos
1 Reply
Honored Contributor II

Hello Arthes, 

it's passed a while from your post so I think that now you are definitely working on something else :) 

Anyway, I'm working with something similar. I could tell you that a packet information is related to an entire frame and not a sinlgle line. In fact, if you look at control packet just previous to a video packet, you would read information about the frame that is going to come (like width, height, interlaced or progressive ecc.) 

I'd like to ask you: did the deinterlacer with bob algorithm work well? Have you tested it? 

Thank you