Hi Subin, I have tried both SciLab and Octave. Both are appreciably easy on system resources compared to MatLab. As for SciLab, it seems easy and simple, but it is completely incompatible with Matlab (if you dont take into account, the converters). So I couldn't use it much. It is a good solution though, if you dont mind learning another language. And as for octave, it is mostly compatible with Matlab, if you are careful enough to avoid a few pitfalls. I actually like the fact that it is driven from a command terminal (ofcourse, matlab style Koctave is available), since it can use some of the nice features of the terminal. But where it actually falters is graphics. Octave graphics is nowhere close to that of Matlab, whether it be gnuplot, or octaviz (both are buggy). Both need to improve considerably before they can challenge MatLab. Ssorry to admit it, but looks/graphics do matter for this. Anyone care to take this up, or help? Octaviz seems to be a good one. But can be improved a lot. What we need to replace MatLab is something that can actually copy MatLab's functionality. I dont usually hold this point of view, but in case of numerical analysis, MatLab is the de-facto industry standard- even for organisations like ISRO. It doesn't make sense for students to learn another, less widely used language- since this is a seriously practical issue. Octave can however take that position with a few improvements.
Subin, as for GEDA as a replacement for PSPICE, I was surprised a little. GEDA is a design tool and PSPICE is simulation tool. How were you able to substitute one for the other? Were you refering to the schematic capture, perhaps? Anyways, GEDA also seems to be lacking . It needs a lot of operations from command line. Octave is fit to be run from command line, but not GEDA. Half the operations are from GUI and half from command terminal. It needs to be more integrated. A new fully integrated UI, may be a cross platform one, can give GEDA a new lease of life. Something like KICAD's UI. As of now, it is still a hackers' toy - not something you would call production standard. Many may disagree with me on this, but most users wont even try a design tool that requires them to touch the command prompt. That's a shame considering how good the rest of GEDA is. However these are simple issues that can be solved if someone takes keen interest. We could arrange a GSoC project, or just outsource it. Or we could even make an attempt to solve it ourselves if there are enough takers for the idea. Anyone interested (with enough time at hand)? Regards, Gokul Das On Jan 4, 12:00 am, "Subin Mathew" <subin.a.mat...@gmail.com> wrote: > hi, this is Subin, Mar Baselios College, s7 Ec, Kerala University, I > think its not only Kerala University, but also most other universities > in Kerala, prefer MATLAB as a program for signal processing and > SIMULINK ( a sub program inside MATLAB) for system simulation. > > These programs are taught (as far as i know) to Electronics and > Electrical students in various semesters. For the EC batch, the > program is taught in s4, s7 and the EEE in other semester ( which i > don't know). I have attached a copy of the syllabus(a jpeg file) of > only MATLAB from the S4 syllabus. > > The fault lies in not having a very stable program (like MATLAB) to > run most commands. I have used octave and found it just OK, but not > anywhere near the skill of MATLAB, the commands are similar, but > Octave or even GNU-plot fails in providing all features used in > today's electronic industry.(of course you can share the commands and > functions and make octave better) > > There was once an OS that came through called FEL (Fedora Electronics > LAB) but it didnt install properly or i think the md5sum ckeck didnt > work, its available at zyxware, and the softwares need a lot of > updation, and validation, but their backup services (or the prople who > once wrote those GPL programs) have been stopped, and some of the > documentation is in languages other than English. > > In S4 we had learned about simple stuff that could be done through > Octave, but in S7, i think Python also needs to be implemented for the > set of experiments, or so. The MATLAB works on large features like > DSP, Image Processing, Mathematics, Signal Processing simulatons and > so on, SO the question becomes when we have one program, why care to > learn more? > > Still Octave seemed to work with much lesser RAM suck-up when i tried > complex Amplitude and Frequency bandlimitted wave programs on it, > where as MATLAB drained our college Win-XP's ram for 5 to 10 minutes > fully. > > Guys trying to learn matlab programming, Octave is a nice choice too, > same commands work, the matlab looks like a notepad model (like gedit) > and all (command window, editor, graph's, errors) integrated as one > unit, where as Octave looks like terminal ( or like MATLAB's Command > window) and slower, but safer on the PC, as i didn't need to upgrade > to 1GB + ram (as MATLAB will hang your PC even at 1GB) > > S4 also uses circuit programs like PSPICE, ORCAD, but i heard some > colleges ( i dont know which) use also GEDA, (which i used for my s6 > project and found as easy as PSPICE except some minor errors here and > there). > > The S7 we use some more programs for reading from DSP kits, those > programs are usually shipped from Texas Instruments, so i guess then > we cant say anything to them. > > By the way have anyone used scilab , how is it? > > matlab.jpg > 51KViewDownload --~--~---------~--~----~------------~-------~--~----~ "Freedom is the only law". "Freedom Unplugged" http://www.ilug-tvm.org You received this message because you are subscribed to the Google Groups "ilug-tvm" group. To post to this group, send email to ilug-tvm@googlegroups.com To unsubscribe from this group, send email to ilug-tvm-unsubscr...@googlegroups.com For details visit the website: www.ilug-tvm.org or the google group page: http://groups.google.com/group/ilug-tvm?hl=en -~----------~----~----~----~------~----~------~--~---
