Forum Discussion
Altera_Forum
Honored Contributor
Honored ContributorSignaltap II
Hi!
I have two doubts. I have a proyect and I have created a *.stp file. I haven't used trigers conditions. The problem is that I want to see the captured dates continously in a 10ms time. and the windows in signaltap II only show 128 samples. I use a 40MHz clock. ¿what I have to do to capture in the same windows or same log 10ms???? and the second doubt is: ¿can I capture two signals in signaltab in a vector in order to use them later with matlab???? many thanks in advance for your help.14 Replies
Altera_ForumHi BEN
I cant use a slow clock. :S ¿Is there another option without using a slow clock???' Hi Kaz! If I do what you say it only capture the 128 samples showed in the windows and I want a vector of 10ms. ¿what can I do?? Hi the brick I cant understand. Can you explain me exactely what you mean?? or give me some reference or any link? thanks every one.- Altera_Forum
you can also use a MATLAB command to capture signals from signal tap directly into a vector\matrix. using a loop you can achieve a greater depth.
but it will probably be safer to use triggers. use the command: data= alt_signaltap_run ( 'stp_file_name.stp', 'unsigned'); - Altera_Forum
to capture data from signaltap into text file:
set all data to signed(or unsigned, not hex). Then right click mouse and click on generate text file(not sure of actual menu title). Then open text file and remove any text header leaving only the coulmns. Now it is ready to be read in matlab by typing: load filename.txt (make sure the directory is set for matlab). This generates a matrix of data called filename,then read any of its column(each couln refers to a signal or bit of a bus, so make sure which column to read): data1 = filename(:,1); %all rows of column 1 - Altera_Forum
When you created the signaltap you had an option to define the sample depth, in your case 128, which will give you 128 * 25ns or 3.2us. So in order to show 10ms you are going to need to set the depth to a much larger value which I suspect will be impractical. Could you use a slower clock to capture the data?
