4/24/2021 0 Comments View Vcd File
While this description looks promising - unfortunately, our single-colimn, raw ASCII data in chan1.dat and chan2.dat can not be read directly by gwave That is because gwave expects at least two columns of data in ASCII data files, the first representing the time base - and also, preferably the first row should contain data labels.So if the mouse cursor is over the graph, and the user scrolls the mouse wheel forward, the graph zooms out; if he scrolls the wheel backward the graph zooms in.
Searching simple waveform viewer - Comp.Arch.Embedded - EmbeddedRelated.com I think your own search failed because you used an overly restrictive. However, the problem is that, typically, we have to work with large datasets (millions of points) - or rather, the complete opposite of the sentiment expressed below: Why doesnt Python release the memory when I delete a large object. Do you really need a list containing 5 million integers I never do;-) Typically, this is the domain of so-called Waveform viewer software; which is, however, dominated by proprietary (and sometimes rather expencive) offerings. What this article addresses, then, is a comparison between open-source tools that could be used for this purpose. ![]() The procedure described here was performed on Ubuntu 10.04, on a MSI Wind U135 netbook (1G RAM, 1GHz typically 1.67GHz in performance mode, 20GB hard disk space). Lets assume that these were obtained from an electric circuit simulation, or an oscilloscope measurement, at a sampling rate of fs25MHz - meaning that the sampling period between each point is Ts125e64e-8 or 40ns. Thus, the entire span of 1 million points represents time series, encompassing a 40ms ( 1e64e-8 ) time period. Additionally, we will try to limit the number of decimal places in value strings (simulating, in some sense, the sampling resolution limit in typical acquisition hardware) To make this more interesting, lets say we have two channels (datasets) - each representing an FM -like signal of the kind sin(xsin(x)) ( or rather, like, so the amplitude changes as well ). Additionally, each channel has different frequencies ( fa13KHz, fb150Hz vs. Note that it is probably easier to solve these types of problems by hand or with a script; but sometimes we dont exactly know the values we are looking for - which is why a quick visual inspection of data (and a waveform viewer to do it with) is valuable. The problem above is also not completely specified; it is just mentioned, in order to show the typical type of problems that require use of waveform viewers. For instance, it can be noticed that the FM-like signal above has regions where the frequency, quite visibly, slows down - typically, we would like to identify where these regions start or end; or what is the delay between the starts of the x th such region in first vs. Given the size of the final ASCII-text data files, the easiest way to browse them as text files would be to use less directly in the terminal (or even quicker, head or tail ). One of the first things one may think of in Linux, when faced with generating numeric ASCII data, is to use the bash and a calculator, such as wcalc, to generate this data. And that is indeed possible, as demonstrated below - although quite inefficient (in this case, at least). Note that wcalc doesnt in itself have a concept of program loops, so we basically use bash to implement the loop (and call wcalc at each step). Below is a session log of bash commands; you can either: save them as a script (say gendata.sh ) and process them using.gendata.sh ( note, this requires that you made the file executable chmod x gendata.sh previously ); or you can pastetype them in the bash terminal console: clean data files first, as we append to them. Below is a session log of scilab commands; you can either: save them as a script (say gendata.sci ) and process them using scilab -nw -f gendata.sci ( note, this will still enter the console, put an exit command at end of script to exit automatically ); or you can pastetype them in the scilab command line console ( use scilab -nw to start it ): Note that scilab -nw simply starts without specialized Scilab Window, and so the console goes directly in the Terminal; however, in this mode, scilab you must press Enter to recognize key presses - important for stopping printout of large data increase stacksize - else scilab may run out of memory. Below is a session log of python commands; you can either: save them as a script (say gendata.py ) and process them using python gendata.py; or you can pastetype them in the python command line console ( use python to start it ): import numpy. At the moment, we could choose one or the other as the data set to work with - there shouldnt be (m)any differences between the two datasets: using diff will find difference in each line due to missing 0 in sci.dat. As the data files could, in principle, be obtained in a different way - the first idea would be usually to use the default GNU software for the purpose - so well start with an outline of that. The Gwave ( homepage ) software, which also goes by the name of gnuwave (not to be confused with Googles offering known as Google Wave ), is a waveform viewer, that can read raw files from spice2G6, spice3F5 or ngspice, and a tabular ASCII format suitable for use with GnuCAP or homegrown tools.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |