Because of the impossibility of actually being able to revise ONLY the dwg and the top assembly referenced by that dwg without HAL 9000 (i.e.: Windchill)deciding to revise everything under those 2 files (without my direction, permission, or even knowledge until later), I'm going to propose for my company (and anyone else stuck with HAL) to NOT use the W/C revision system at ALL. Instead, I'm going to propose creating a string parameter in Pro/E, and matching "attribute" in W/C that will be a "revision" parameter that will show up in our searches and formats. This way the user will have full control of revisions, as he should, instead of HAL. The rev level can be bumped back as needed if someone makes a mistake, and HAL can never take it upon itself to revise hundreds of unwanted files as has happened recently in my case. Let's not forget that as it stands now, the initiation of advancing a revision is done manually anyways, not automatically upon check-out unless you make the fatal mistake of doing a "Revise and Check Out" in Pro/E. So, in reality, what I'm suggesting is no different than currently done as far as the user ititiating the advance of the revision. It is only in the execution that it is different, and far superior because of the flexibility afforded by this technique. As it stands, now I have to try and specifically delete ONLY the last revision created by me on a huge number of files......and hope it works without deleting the entire file, or leave it alone and realize that a revision was wasted and revision traceability is now lost, perhaps forever. Tons of man-hours have already and are going to be wasted.
Pretty pathetic that it's even gotten to this point, but if the software doesn't work, I'm going to find a Pro/WORKAROUND for it's deficiencies. Sad, because Intralink worked so well in this regard, I wouldn't have to do this. So much for this being an "upgrade".......
To say that I'm furious with "HAL" and the massive performance downgrade is the underestatement of the century.....
If some consultant(s) helped you get started and migrated, you should ask for your $ back from them if they did not address this type of subject in detail.
Best bet for now may be to set all the collector preferences related to Revise to "None" at least for now.
This way, the system will not automatically collect anything, leaving it up to the users to select via collection tools.
There are a very large number of preferences (see Site, Utilties, Preference Manager). Search for Revise, then see the collector under this heading.
Yes, they did. And, I must say, in my professional opinion they did an absolutely piss-poor job. If I had the authority I'd ask for double my money back. We have an overly-invasive internet/network security system here, and for the test runs, the company had it turned off. And, we didn't have any real parts or assemblies to try out in the test runs, so none of these large assembly issues cropped up. Then, when we went live, the security system was turned back on and BAM! Instant problems.
I will tell you that in 3 different companies I've worked for that went to Windchill from Intralink, almost totally univesally, the users all HATE it, for all the reasons I mentioned. I certainly do. In fact, in the almost 3 years I've "used" it, I can tell you that not only has the system not provided any sort of performance gain over Intralink, but it literally takes me at least twice the time to do the same simple tasks, and usually far, far more than that if it even works at all. This is the only software in over 25 years of using CAD software that the more I use it, the more I hate it, instead of eventually liking it or simply forming a truce with it.
I probably don't have the admin preferences needed to set these collector preferences, but I'll try first, then pass this info along if not.
I'm still going to discuss the idea with management of creating our own revision system of parameters/attributes though and completely scrapping the W/C one in toto.
Thanks for your suggestions!
Hi Frank...
I feel your pain. Although I still think you can form that same kind of "truce" with Windchill. Remember how everyone UNIVERSALLY hated Pro/INTRALINK when it first came along? Everyone longed for the simplicity of Pro/PDM which even had some nice scripting abilities. Over time, Pro/INTRALINK got better until it worked really well... except for a few idiosyncrasies. I've used Windchill since its inception and it's gotten much, much better over time. Windchill 10 is getting us much closer to where we need to be... but I understand the path to get there has been painful.
In my opinion, most of the migration consultants and VARs selling migration services are pretty bad. At my company we were extremely lucky to work with an amazing consultant who went far above and beyond to make sure our migration went perfectly. Even with training and the support of a fantastic consultant, we still faced problems like the ones you're describing.
Mike's suggestion about looking through your preferences is definitely a good one. You should have the ability to manage some of the preferences yourself... others will be reserved for your system admins. Your best tool in working with Windchill is EDUCATION. While you can circumvent the system with parameters, I'd advice against it.
I've worked at many companies where they forbade use of certain tools in the name of keeping things simple for the less experienced users. For example, instead of using one complex feature, we were ordered to use 15 "simple" features to do the same job. My feeling is... why slow down your entire workforce? Why not train and mentor everyone to get better and make the most of the tools? It's the same with Windchill.
If you skip the revision functions in Windchill, you're missing a huge benefit of having a PLM system in the first place. Manual parameters are always subject to human error. Human error and faking the rev means you can also fake the release giving you no confidence in the final copies you're creating.
There's a fairly straightforward way to solve the problem you have:
I've been where you are and users at my company routinely hit these snags. Over the last few months we've really put a premium on education and mentoring. Support calls are down and users are getting comfortable with Windchill. At worst, we've made that truce with the software. We're moving forward and learning to use the tool for what it is while we voice our concerns to PTC. I really believe the software is getting better but it's a tough slog to make it through the initiat transition and get back to full productivity again.
If you need help with any of the suggestions... or would like me to send you some sample documents explaining the revise or delete commands, send me an email. I can't publish these for the community but I can send them to you offline.
Best regards,
-Brian
Hey Brian.
I've set the preferences I could for me, and it made no difference. I don't have admin priveledges yet so I can't set any system preferences, but I'm going to look into it.
Actually, I don't remember anyone having issues with Intralink. with the exception of a few little quirks, it always worked very well. Simple, few mouse clicks, powerful, easy to learn and use, and I 100% trusted it. Unfurtunately, in my experience with a total of perhaps 100 users in those 3 companies I mentioned, it's been nothing but a total nightmare and all the users have hated it. In fact, on eof the major reasons the DOE site I worked at went to Inventor and Vault from Pro/E and W/C is because of the PITA that is W/C.
Training is great, but I still have serious issues with the performance. For one thing, instead of being tied straight to the server, it's bogged down with all the unstable internet and Java issues. I'm constantly getting script and other errors. Where there used to be only one software to have issues with, now there are at least 3, all dependent on each other. I have huge issues with the tedious screens and picks. The simplest commands have far too many picks compared to Intralink. As I said, the worst thing is I totally do not trust the system. We've seen lost data and corrupted cache issues. In Intralink, I only had a couple different views to define. Now, there has to be close to a dozen different views I have to define. Running ANY software thru the internet is a supremely bad idea. The idea of logging in anywhere only if it was internet-based is a lie, you could set up the network to log in remotely, we did it long before W/C.
Since I started in mid-'96 on V.15, I used to eagerly look forward to every new version of Pro/E. Now, unfortunately with them hosing up the GUI every time, I dread it. When I buy a new car, I don't expect the steering wheel to be in the back seat and the brake pedal in the trunk because someone somewhere
"assumes" that it'll make me more efficient. Give me the new features, make sure they work, and change only what needs to be changed to add those features to the menu. But, as I was told by the guy who did our training (off the record), it seems the most money to be made is not in selling software but in selling training. Go figure.....Thanks for your help though!
P.S. I disagree with the human error thing. In W/C, if you make an error and bump a rev, you can't back-rev things without deleting the files and losing work. With a manual rev (which, face it, they ALL ultimately are) based on a parameter, you can easily fix your mistakes.
Ah Frank...
I don't know what to say. I've been working on the software since '91 on V.09 and I've seen it all, too. I don't really agree that PTC is concerned with selling training. I think they get caught up in trying to mimic the competition. I also think they don't have enough real world users advising them on interface changes.
To that end... more real-world engineers and long time users need to join the PTC User Technical Committees. And PTC needs to do a much, much better job of outreach and communication with it's user base. Sharing Twitter stories of how great the new software is doesn't soothe the souls of long time PTC users who feel alienated when the company switches to some new whizbang interface that does little to increase productivity.
I always love the marketing lines that tout how many mouse clicks we're supposed to save... "65% fewer mouse clicks!" Yes... once you can find the commands you need buried under new layers of useless gloss and marked with different icons...you might save a few clicks. But subtract points for the mapkeys that have been broken by the switch. Subtract points for the confusion created when the interface must be re-customized (again). Subtract points for losing the ability to put the dashboard where you want it. Subtract points for losing screen space for your model and losing it to more annoying menus. Subtract points for SLOWNESS caused by a bloated interface. Etc. By the time you subtract all the negatives and factor in user confusion... you've ended up far worse than you were in the first place.\
I need to sit down and buy you a beer Frank. I totally feel your pain on this. Maybe one day real world users... and long time Pro/E designers will get a seat at the table when these interface changes are discussed. One can only hope.
Thanks!
-Brian
Hah! I'll buy YOU a beer! Or maybe a glass of some fine whiskey. I'm in the Denver CO area. Yeah, I feel they've done a good job of aleinating their core, long-time users as well. They should look at Harley's model of keeping their core customers happy, yet add, read, useful features without forcing us to relearn everything every new (and progressively LESS useful) GUI. The WF intrface was/is bad enough with it's incomprehensible "lead vs. follow" workflow, but this Creo is a whole new level of Microsoft-copy-stupid. Ugh! I've resisted going to it so far because I have work to do, not try and figure out the "ribbon". But, I digress......
Frank, we use these settings for Revise where the important settings are not to include any dependent objects which will result in including all objects in an assembly.
I have those set and so far still have issues. Right now, W/C is running so slowly I can't even get my workspace to come up. Thanks for the info though.
My main gripe is and will be it's base in the internet. With all the problems inherent in that system, I can't imagine anyone wanting to use it for anything but leisure use. If there's an issue with Mozilla, Firefox, Internet Explorer, or any of that, BOOM!
I'm not a happy camper. What would it cost to migrate back?
Have you guys looked at the Performance Tuning Guide? It won't address all of your concerns but it could help
I can try that, and I just got a couple books on the business admin side, so hopefully I'll find something there......
Frank,
it really looks like that you got bad consultancy which is unfortunately not uncommon in the US. I couldn´t figure out which version you are using but I assume Rev 10 for now.
Please make sure that at least the WCA (Windchill Configuration Assistant) was executed and provided with correct value. In the Windchill Help Center just search for WCA and you will find a detailed description.
Further you talked about a migration and in this case it is very common that Oracle is not configured correctly. Check the SGA and PGA settings if they are sufficiently sized. Further the Optimizer settings should be set accordingly and the statistics up to date. Keep in mind that Windchill needs much more RAM on the db side compared to Intralink - up to 4 times is a good start.
If any specific functions are slow it might require an additional index in the db and you can ask tech support for help in that case.
Regarding the revise you already got the right answers. Going forward with a manual approach will cause much more costly errors in the future and prevents you to grow to additional capabilites with Windchill over time. I feel with you but there is light at the tunnel - I had several customers complaining in the past but in 99% the cases these where solved. But I agree that sometimes it takes deep investigations into the issues to find a solution since every customer environment is a bit unique.
Hope you succeed.
What the Windchill Configuration Assistant Does
WARN : wt.server.manager.startup – WINDCHILL CONFIGURATION ASSISTANT <WCA> HAS NOT BEEN USED! WARN : wt.server.manager.startup – PERFORAMANCE ISSUES MAY RESULT FROM FAILING TO USE WCA! WARN : wt.server.manager.startup – FOR OPTIMAL PERFORMANCE, WCA SHOULD BE USED PRIOR TO PRODUCT USAGE.
This initial run establishes values for an initial set of properties based on the available memory and CPU resources of the server. These settings establish things like the following:
To view the actual property values setare stored in a separate XCONF file located at <Windchill>/utilities/wca/conf/windchill/windchillconfigurator.xconf, where <Windchill> is the Windchill installation directory.
The Windchill Configuration Assistant installation adds the ConfigurationRef element to the site.xconf file to establish the link between site-managed property values and those values set as a result running the Windchill Configuration Assistant. In the initial run, these settings are propagated in your system through the use of the site.xconf file.
Thanks all for your help, and I'll try these things as I can get away from the modeling that needs to be done. That, at least, I can still accomplish.
Wow, with the difficulties that you have had with how your system was implemented, it makes our WC problems seem miniscule! Revising is simple, easy, and works very well for us. If that didn't work well I am sure I would be as frustrated as you sound! I hope it working a lot better, since this your post is over 2.5 years ago!