FactoryLink upgrade


Thread Starter


Does anyone have any experiiences to share with me on upgrading FactoryLink V4.3.1 or similar early version (running on OS2 Warp V4.0) to FactoryLink V6.6 or 7.0 (running on NT or Win2K)?
Was it "painless" or ?


Henry Aguado

I had an ugly experience trying to perform the upgrade. First of all, our vendor told us that there was a utility to re-populate the config tables. When we got down to the conversion, we found out that it was not ready for prime time and had to do much more re-keying than we thought (or budgeted for). Some other "gotchas" include
but by no means are limited to:
-Reports will not print as they did before, necesitating re-formating the reports.
-There may be graphics glitches that will require modifying the screens.
- If you are adding web-clients to the system, you will have disastrous results (graphics distortions) unless the web-client stations are using the same video resolution that the screens were developed in.
Good Luck!
painless + factory link is an oxymoron. USData will probably hit you for a full price for the upgrade since you are spanning so many revisions. Watch out for nickel and dime add ons that really run the price up. Factory link changed significantly at rev. 6-FLECS. It actually got a tiny bit easier to use. I don't
know how well your app will transfer up to 7.0 since I've never moved an application that far, but when I moved from version 4.4.1 for Win 3.1 to ver. 6 for NT the conversion went faily well with only a few items needing do-overs. I was moving from Win3.1 to NT however. I have four licences at 6.5 and believe it or not, still two on Win 3.1. In 1999 we upgraded three systems from FL-DOS (yes that right, DOS) and ditched FL altogether and used CiTect.

If you use MPS (multi-platform-save and resore) you should do OK. Too bad OS2-warp never caught on. I'm curious, does FL support Linux? Anyone know?

Last time I attended any FL training was in Dallas about '97. USData had a fairly new CEO and he attended lunch with a group of us students. He asked us why we thought FL was loosing market share. Almost unaminously the group answered it was because the cost was way way out of line with the competition, and the advantages offered weren't significant. They didn't lower costs however, so they derserve the
declining marke share. Too bad, all in all its a decent HMI with more flexibility than most others I've seen. I mostly use RSView these days, but I wish it had the flex of FL.

Daniel Chartier

Well, to start with the obvious, nothing about upgrades is ever painless, exactly. At one point, I was in charge of supporting FactoryLink for a canadian distributor. I can tell you the safest path (not the painless, now...): You must upgrade (convert)to version 4.4 under NT, from which you can convert to version 6.0, and from there to 6.5 (the last version I worked with). After that, I must presume the there are more steps to version 7.0. Notice the number of steps you must follow; do not attempt to jump over one (at least keep a copy of the version you are upgrading from)because the databases are different (have different fields) at each of those
levels. Risks include database corruption, new versions of the drivers (which means rebuilding all your tables in the new version)and "orphan" graphics (their term for lost tag links in the graphic pages). So really, I would suggest going to a distributor who has technical support and has a reason to help you out (like a new version upgrade you would be willing to buy...) and ask him to help you out with his old licences. Or USData used to offer to do it for you for a (hefty) fee and a good delay.
Hope this helps...
Daniel Chartier

It will be easier for you to update to Factorilink V7 because it can be used as mono
domain. You will be able to save most part of your application.

You will lose recipe & alarm but for the alarm you can use xcell to don't declare all of them.

Best regards
Thanks all those who responded. I believe we are going to go to V6.6 step by step as this sounds like the way to go with least amount of headaches (and is $30K cheaper than V7 for our set-up, not including hardware upgrades, and engineering).

Will let you know how it turns out!

Hallo !
This is right. I've FL 7.1 in use and it doesnt work correctly at all. There are many bugs. FL 6.6 works correctly an it runs on all plattforms (eg. linux, windows, OS2). But you can wait until FL 7.2 is comming. I've seen a beta version, the most bugs are killed.

my E-Mail:
mailto:[email protected]
If you use an mps and restore it with FL4.40, convert, do the same to get to FL6.0 then it converts to 6.60 easily. I brought a 4.31 app up to 6.6 with no problems at all. The original ran on OS2Warp and then finished on an NT4.0 box with SPack 6.0.

We have also taken a 6.60 app and converted it to ver 7.0. We run legacy graphics and start the appliation with a batch file that calls the old run command (Flrun -w-d). The graphics run fine. We had trouble using the webclient ACtive X oject.