Rios and ssc32.exe both seem to be win32 programs. Are there any old dos programs for L5 or L6?
Why do you need support for DOS?
Both RIOS and SEQ are Win32 programs.
If you don’t find it here on the site, it probably doesn’t exist. LM is great at keeping older stuff supported in software, so if they don’t have something for a particular OS, then it didn’t get done.
I’d rather see the LM software ported to the GNU/Linux platform, which would make more sense if LM were going to port RIOS and SEQ.
8-Dale
Jeeze Dale the guy just asked if anything existed, not for somebody to go do a port. Maybe he would just prefer a simple non-multitasking command line based application that works on an old PC running DOS.
You seem to have a fair amount of time and a great affinity for the whole linux thing, a port or similar concept program to rios seems like it’d be a great project with some tangible and useful output… especially if you could run it from the command line and not have to have all the graphical workspace stuff running in linux.
I just asked why he needed support for DOS. I didn’t think anyone would actually be using DOS these days. He seems to want the functionality of RIOS and or SEQ, but I am not totally sure on that, which is why I mentioned porting.
I have too much time and not enough interesting things I can do to fill it right now.
Doing a port implies having access to orginal source code, which I don’t have in the case of RIOS or SEQ. I’ve been toying with the idea of wrting something similar to SEQ, but haven’t decided what language, besides Pythong, I might want to use. I’m fiddling around trying to learn QT and how to use it, but have not had a lot of luck with that so far.
Command line programs are pretty boring, but maybe I can figure out a way to do a GUI/Command Line thing with Python, which has bindings for QT and such. I’m not real interested in investing time in C/C++ right now, although I have written my share of code in C (not much C++). Pointers give me headaches real fast. 
8-Dale
I don’t know that a “port” per-se is necessary, but a subset of the functions that can be driven by command line calls passing parameters might be very interesting. If I understood what I was reading I think a lot of zoomkat’s web accessed pan and tilt code was using command line scripts to send commands out the serial port. I don’t know how information is returned from something like that, perhaps have the app write a results file and then parse that file when the command line returns? I think a flexible command line based application to allow commands or even execute text scripts would be an awesome tool, especially if it could be compiled to run from a linux, DOS, or win32 command line. 
You may need to make your own programs using qbasic. It is contained in the below olddos.exe file. A long time a go I made a program to use with my mini-ssc servo controller (bottom) for web or command line use, and it may work for your controller if it is put in the mini-ssc mode.
download.microsoft.com/download/ … olddos.exe
geocities.com/zoomkat/files/zkpt2.zip
Pythong?
Obviously, I meant Python.
8-Dale
Lol. I had toyed with the idea of making a QT/ KDE version of SEQ since when I was working with my Scout gaits, it was getting painful to manually enter a lot of us numbers in a text file. As far as Linux (propably dead easy to port this to Win32 CLI, i.e. it would probably compile with migwin32) I wrote a small C program that took a string on the command line and sent it to the SCC. With a bash shell and some simple commands, that’s all you need, or you extend the string parcing stuff in the program and save variants, i.e. can from the shell stepleft1, stepright2, stepleft2, stepright 1, or something. Of course having a pretty screen where you can enter angles and seconds is even better. 
Mainly this didn’t get off the ground as I have other pressing projects. I also disliked the idea that my QT code (if written) could be recompiled to run on Windows and might displace Lynx SEQ sales.
As far as I can tell there are no custom widgets and the position of the servos in the view has no bearing on their function (note, not having a copy of SEQ, I can’t tell for sure). This makes a QT rewrite not too big a project if anyone else feels up to it.
EDIT: I didn’t just echo out to the serial port since it needs to be configured properly first, and Debian Etch had an error where echoing didn’t work for awhile.
I’ve also been toying with this idea because there just isn’t any sort of utility like SEQ or RIOS for the GNU crowd. There are Python bindings for QT and OpenGL. I started to fiddle with creating a GUI for the servo window part of a SEQ type application based on QT, but I would really like to go all the way and create a fully integrated KDE application. I just don’t have the patience or brainpower to deal with C/C++ right now, and I didn’t want to have to figure out how to do stuff with KDevelop (although I still want to do that) but if I can use Python and still accomplish everything…
I’ve written 2500+ line applications in Python (a FreeBSD SysAdmin utility to do stuff with Ports and make other system tasks easier). My software even had its own small command language for running tasks, so tasks could be chained, or run from cron or terminal as needed. My software parsed command files and executed the tasks. I should dig up that code and see what I can use from it. So, I know Python is a capable langage, even though it’s still interpreted, for writing full fledged applications.
Why should we have to be worrying about this? I think what would happen is LM and Jim would just be pushed to make their software even better yet, which they already do now anyway. They are always improving their software, so anyone who actually wants to compete with them has a hard to hit fast moving target to try and keep up with. I don’t want to compete with Jim and Lynxmotion either. I just want good high quality tools for my favorite OSes. So what if your code (or mine) can be recompiled and run on Windows?
I’d actually just like to see a port of SEQ, RIOS, LynxTerm, etc, ported to the GNU world where it could be used under any GNU capable OS like Linux, FreeBSD, OpenBSD, NetBSD, and others. When there is already something available and as high a quality as the LM software is, it isn’t really attractive to have to reinvent the wheel just to get good tools for a different OS.
I’d rather be able to use the same (SEQ, RIOS, LynxTerm, etc) software, regardless of the OS I choose to run. That’s why I use Mozilla Firefox, and other software - I can use the same software regardless of whether I am using Windows, Linux, FreeBSD, etc. I don’t have to keep learning new software for each OS to do the same tasks. But that is not likely to happen since we are in a minority of users. It’s not cost effective to expend effort unless you can sell enough units to a new market.
Think of it this way though… At least we can use the same set of SES brackets to build robots, regardless of which OS we use. 
I just don’t like using Windows, but do enjoy using a well designed and implemented GUI environment. I am much happier in the KDE environment because it works the way I like to work and it actually makes sense to me. It’s bad enough that I have to switch to Windows to use my 3D CAD software, and I am not about to give that up anytime soon (and it won’t even be paid for until October). I won’t be switching to Windows to program my robots soon though, so that will be one less thing I need Windows for.
8-Dale
If you want to make KDE apps, trust me KDevelop is 10x faster. I have an old Computer Science project (I can give you the full code, it is a very poor software implementation of a Linear Feedback Shift Register) I did which along with the tutorial here: women.kde.org/articles/tutorials/kdevelop3/ is all you need to do basic QT C++ programming in KDevelop. The vast majority of the “programming” is putting all the UI widgets in places you like. I did my project in the most non-OOP way possible, so if you’ve done any C, it should be fairly easy to understand.
Yes, no doubt, for those who already know how to make it jump through hoops. I just don’t want to deal with figuring out how to use KDevelop right now, even though I could use it for Python also. All I need for Python is Kate though, which is a great editor and already has code highlighting for Python. KDE is not really well documented yet and I don’t care to have to put in the effort to figure it out right now.
I just want to design, build, and write software for unique robots right now.
I spent four years as a maintenance programmer having to analyze and make changes to code everyone else had written. It’s not my favorite thing to have to do, and I never could get promoted because I didn’t have any “papers” to show I could do what I could do. I had to deal with some of the worst code you could ever imagine, and then figure out how to modify it to do new things or to fix old bugs the original authors never bothered with.
8-Dale
Did somebody hijack a DOS discussion?
Zoomkat’s right, we’re way off topic and should wrap this up…
However, writing C++ KDE programs is not hard. Following the above tutorial got me started in a day. As a fellow programmer (well, I wear many hats as a grad student but the main thing they pay me to do is write/ maintain simulation code) I deeply feel your pain. I can also relate to the no “papers” issue, since while I probably knew more programing than nearly all the Computer Science majors at my school as an undergraduate, I wasn’t able to get certain interships at big firms (like google summer of code) because I was a physics major. Contributing to major open source projects (I have patches in kopete, though admitantly minor and did briefly in the kernel, my ethernet driver was an extension of one driver and while it worked, the higher ups thought it best to use another driver for that card) is a good way to fix this. I also managed to get a Certificate in Informatics and Modeling since my undergradute school had such a program. Its like a minor in Computer Science and basically states you have taken the following CS courses and did a thesis-level CS project.
People with no formal C skills (like some professors and grad students) can write the worst C code imaginable, probably worse than corporate code monkeys. Besides the pointer arthematic freak, we get people who still use K&R bracket style, with multiple nested if/ for loops and didn’t even indent properly. In short order it becomes impossible to figure out flow control at all. Personally I use Hungarian notation for all my code and try to provide verbose comments to not only explain what the code is doing, but what I was thinking when I wrote it.
I think the short answer was that there doesn’t readily appear to be any DOS based application laying around that implement a comparable RIOS function for the lynx 5 or 6 arm.
There are some old Qbasic files for the arms, but they are really not very impressive. I can say that because I wrote them. lol They do not have IK, just a way to position, save, and play back sequences.
Look here.
lynxmotion.com/ViewPage.aspx … ryID=21#l5