The website doesn't say if price is for one op sys or all

So I'm wondering how much it would cost for all 4 versions? and/or for just the Windows and Linux versions? I've never had a Mac unless you count the Apple IIe; but you never know I might pick one up one day if I ever decide to publish any programs. I might find a future version worth 236ドル.00 for all four, but not right now.

Actually the only reason I'm considering the Windows version right now is because my Linux system won't accept 32 bit compatibility for some reason so I'm thinking of temporarily using my virtual Windows XP machine and getting into learning more about RunBASIC. I've always loved the concept. I probably won't use it as a web server (except possibly for some testing and to learn how to set it up) until the next version comes out. But I would like more control over the way it displays on the screen, font etc. as well as over the timeouts for programming reasons. Sometimes I'm called away from the computer and as a result I occasionally lose work because of the timeouts. Font control is important to me also as my eyes are getting old and worn out from many years of abuse.

Anyway, I thought I'd ask about pricing on multiple platforms as there's nothing about it on the website.

I purchased both Windows and Linux versions back in 2008, then in 2012 after upgrading to 64bit linux I started getting segmentation faults when trying to use runbasic.
I have asked for help from the far reaches of the internet and even up to Mr Gundel but have had no luck getting it to run.
Even a clean install on the same computer and installing 32bit libraries RB will not run.

However I have had success with a new install of Mint Linux 64bit(Cinnamon) plus 32bit libaries on a small atom computer, RB will run.
I then , via the intranet start RB on the computer that will not run RB and code from there. ie , I navigate to the IP:http://192.168.20.10:8008/seaside/go/runbasicpersonal
and start coding.

Not sure you need all OS releases of RB, just serve it up from one location.

Shoot I think one of the most prolific coders of RB serves up most of his code on a netbook, maybe 100ドル secondhand.

I have the same problem with the free version on my 64 bit Linux Lite system. Segmentation fault. But I can run the Windows version in a Windows XP virtual machine using Virtual Box. I could probably get by with that until the new version comes out, which could be awhile. But I would like those added features and I would like to try setting up the server on the web and get a feel of what it's capable of now. I'm thinking the next version will be powerful enough to do what I want to do, and I'd like to get a head start on my website idea. I absolutely love the concept of web programming in BASIC.


Metro said:


One of the reasons I was hesitant to purchase the licensed version of RB a few years ago was that all the sites I could find made with RB were extremely slow, except on the Run BASIC website, which always worked well. I tested some of the code available on the Run BASIC Wiki and most of it ran well in the free version, but I've been hung up on the fact that I couldn't find anything on the web that wasn't slow.

Someone brought up the idea of creating a BASIC to HTML converter on Carl's BASIC FB group the other day; and since I know Carl plans to upgrade Run BASIC after LB5 is completed I started thinking about it again. I have this really great idea for a local community based interactive website. But I don't have the skills to put it together using PHP or any other web language. I think I could pull it off if I could program it in BASIC.

Since RB is so different from other languages I'm also thinking it would be more hack proof than most web languages.

Hi Ho Donny,

I run a number of runbasic apps on our company server. It's an older Debian install, 16 gig, Intel I7 two core and running off an ssd. I'm able to easily run 100 or so clients at a time. Most all the apps access mysql for one purpose or another and seeing load averages of 2 or less is quite normal. It'll slow down at some point or another, but the client community has never complained. One thing I noticed. RunBasic will cleanup after itself quite nicely after a client drops out.

Your Mileage May Vary. It all depends on the size of the app, how much memory and cpu you have.

One thing I can share. Since RunBaisc has no inkling of 64bitness... Just be sure to have RunBasic on a multilib Linux box. Then move to the RunBasic directory and run "ldd rbp". Install the 32 bit verions (hope you have a similar 32bit box to your 64bit box) that it may complain about.

I was able to do this and get RunBasic to work on one particular Debian box in headless mode ONLY. No graphic presents on the desk top. It works well enough for my purposes.

Why RunBasic craps out on some 32bit Linux boxes? I really, really wish I knew.

At about this point in time. I'd pay almost anything for a RunBasic 2.0 that would run on "ANY" 64 bit Linux box!

Carl, this is so badly needed I cant say how much!

jerry

I'd pay double what I paid initially.
However It's all about hours in the day for Carl, maybe it's time to release RB to the open source community.
Carl can still be proud of the fact he created it.

Of course there is the risk of a forking of LB.

Hmmm! I guess I'm the only one running windows here..
So far I haven't had any issues with speed. Probably because of the system I have. It has about 300 users. There are only about 50 people who do updates. The updates are complex with lots I/O. Everyone is looking at information.

With the price of computers, I'll toss money at the hardware, instead of people and software.

I have a PC that I've tried several Linux with RB, but I just haven't had the nerve to go for it.

Hi Jerry..

Most all the apps access mysql for one purpose or another and seeing load averages of 2 or less is quite normal

What code do you use to access mySql. I don't remember the numbers exactly, but years ago I did benchmarks between SQLite and mySQL and mySQL was more than twice as fast as SQLite. Don't know if those numbers are still true today.

Thanks for the help..
Dan

G'day Dan
maybe Jerry would be kind enough to share...
runbasic.proboards.com/thread/935/connecting-databases-contd

Hi Dan and Meercat.

First off, there is really nothing wrong with sqlite. I used it for a long, long time and it has served me very well. However, at work we have a mysql server resource that is to die for! Also, I was asked to drop sqlite in favor of mysql anyway. So I found a way.

So how did I do it?

RunBasic has no native mysql support so I had to write my own support for it.

I simply wrote a mysql shim in FreeBasic, that slides between RunBasic and mysql. I wrote an app that accepts command line input and prints the results to the console. From RunBasic, I build a query string and submit it to the shim with rc$ = shell$(query$). If the query is successful, the data is returned in rc,ドル otherwise the shim reports "error".

Queries sent to the shim look like: "R,Eservice.ESmachines,x,x,x,x,x,x,x,x,x,x,x,x". The first comma separated variable indicates a simple select, the next is the database and table, the next string of "x"'s is what Im looking for in the database. All very fast and a boat load of fun. My shim supports: R simple selects, W updates, C make new tables, D drop a table if it exists.

I have very specific select needs for my RunBasic applications and my FreeBasic code shows it.

It works great. It allows me to clean up all the sqlite code from my RunBasic and I get to utilize the speed of mysql on a cluster.

Email me direct is you need freebasic help.

Thanks Jerry,


Just the opposite of my SQLite experience. I get locks that are forever and they cannot be released until you kill the RB server.

I've looked at a process similar to what you are doing. Problem is my queries and updates are complex and can't be done with a simple query. LB now has ODBC support so I guess it would be possible to do it with LB.

How fast is the connection and receipt of info form Just Basic.

As an example I have a Program that does Pipe Lineup connections between Tanks. There are about 3000 possible pipes, and 1000 tanks. They pick the from and to tank. It usually takes a few thousand requests as it gets pipes and based on that pipe it determines other pipes it should try. As it works it's way through the pipes it finally finds the pipes you should use.
Do you think it would be fast enough to get about 1000 request. Right now it usually takes less than a minute to compute a pipe lineup.

The big reason I like mySQL is replication. I have systems that do not have time for backups.

I'll have a look at your idea and see if it will work for me..

Thanks for your help..
Dan..

Howdy Meerkat.
I had a RunBasic app using sqlite and accessed by about 50 clients or so. It was hammered pretty regularly and never had a lock problem. But did have a couple of corrupted DB's from time to time if a lot of updates poured in. On mysql, no issues.

Whatever you can do in mysql can be done with my "shim" method. It all depends how complex you need to make it.

My shim is all FreeBasic. The latency of using a shim depends entirely on your network. My clients are all on an internal VNP, the mysql instance is on a cluster and nothing slower than 1G ethernet. All queries are a finger snap from RunBasic shell$() to evaluating the return data. The freebasic shim, runs locally to the mysql database as does the runbasic client. You have to expect, the more data you ask for, the longer you wait. It makes sense after all. A few hundred bytes are quite fast, a few thousand are slower, but not painfully so. Your "pipes" may be different.

You'll have to expirement some.

Thanks Jerry.

I'll give it a try..

I've always had lock problems. I guess it's because of my weird complex SQL statements.
When Liberty Basic 5 came out it used the same code as Run Basic to do SQLite. For some reason the lock problem on Run Basic was ignored. I couldn't wait to prove the problem on LB. I narrowed the problem down to a program doing only 2 SQL commands, and it locked the DB every time. You had to restart LB to release the lock, just like the RB problem. You can read the LB post. The program was called "Lock Test Monster". Carl said he found LB was not releasing all the controls and LB now runs the test program without locks. He said he'd make those changes to RB, but I haven't seen anything yet. If he makes those changes to RB, I wouldn't be having this issue..


Thanks again for the help..
Dan