ProE Wildfire 5 full of bugs

[ClaudiuCraciun]

After 2 weeks of using ProE Wildfire 5 I can say it's a disaster.

This one is the feed back from users. I believe this program should be friendly,easy to use and if in wildfire 4 were some functions that worked, should work also in Wildfire 5.

Doing these changes seems that PTC is not able to retest all the new changes and appears problems(bugs).

Sincerly I don't know :

I'm working for PTC or for the company with that I have the contract.

Why after we pay a lot of money for the maintenance I have to call PTC and tell them that the program have some bugs that before didn't exist.

CC

[ClaudiuCraciun]

Did you know that deleting ordinate dimensions in Wildfire 5 will cause you problems in case that you want to show back this dimensions?

Try not to delete them in case that later you need to show dimensions again. Better use erase.

Then will be another problem if your colleague need to show this dimensions on another view and doesn't know that is hidden on a specific view.

CC

[ClaudiuCraciun]

For people that used to work on previous versions of proe in a offset section proe used to work like this:

Until Wildfire4 if a feature or geometry that was used as a reference in an offset section was deleted, the section continue to work but you were getting a warning message that is saying Cross Section need to be redefined

Starting with Wildfire 4 I found that if a feature or geometry that is used as a reference in an offset section is deleted the section automatically is deleted. There is no chance to redefine the section (must be recreated again.

More than this. In Wildfire 5 they introduce No Rezolve Mode Option that is supposed to keep feature that are loosing the references unsupressed or not deleted even if the references of o feature are missing, Even in 5 the section is deleted doesn;t matter of the Resolve Mode or No Resolve Mode Option

Yesterday day I call PTC and the guy told me that he discuss with his colleagues and this is the functionality of ProE now.

DO YOU BELIEVE THAT IS OK FOR A SECTION THAT IS LOSING A REFERENCE TO BE DELETED AUTOMATICALLY OR TO KEEP IT AND HAVING A MESSAGE THAT CROSS SECTION NEED TO BE REDEFINED?

Based on your feedback I want to go forward with this call. I don't want these people that are working now for PTC to destroy what other people did more than 15 years back and worked ok.

PLEASE GIVE ME YOUR FEED BACK

Thx,

CC

[Kevin]

I think I would agree with you. I haven't tried in WF5 but my guess is that the process would be similar to WF4, but you can redefine the offset section references so the section doesn't get deleted (I haven't tried with models created in earlier versions but I'm guessing there could be problems depending on how the references were created). If you delete a reference that causes the offset section to be deleted UNDO the delete. One way you may not be able to UNDO a delete is if you have the View Manager open when you delete the reference. In the View Manager RMB the section, select Redefine, and select Section. Redefine the Sketch Setup and References for the Section.

[BurtLand]

Delete and erase seem to work fine on M070.

If you look in the correct view in the Drawing under annotations, the dimensions are listed out

If you Erase them they are greyed out and you can later unerase them.

If you Delete them then you need to Show them again.

Works rather nicely actually.

[ClaudiuCraciun]

HI,

In build 60 if you have ordinate dimensions (liniar dimension from model converted to ordinate) and DELETE them you cannot bring them back.

The only way to bring them back was to select all dimensions from a specific view, delete them and show them again.

Try with oridinate dimensions. maybe they fixed this issue on build 70 but in 60 it's working like this

CC

[Kevin]

In M040 for ordinate dimensions you need to delete only the baseline and all dimensions associated with it are deleted, no need to select individual dimensions. You can then show all the dimensions again using Show/Erase. Haven't tried later datecodes.

[BurtLand]

What release are you running?

Its frustrating - bugs that were fixed in M030 are reappearing broken again in M070.

Bugs that were in WF4 are still there in WF5 !

I dont know how they test the code but its not really good enough.

[ClaudiuCraciun]

Another problem I found:

I have a part that contain a pattern table.

I created more tables for this pattern and using relations I activate a different pattern table.

I showed dimensions for a pattern that contains 4 items in drawing.

If I activate a table that contains for example 2 items I'll see dimensions just for these 2 item until i'll regenerate. After regeneration in drawing appears dimensions for the the other 2 items that existed when the patten table had 4 items but these features don't exist anymore now in part.

The problem is these dimensions appears when I regenerate the model from drawing but dissapear when I'm updatting the sheet.

If I'll check the drawing status all the time apears as regenerated and updated (even if I have dimensions for 4 items, even if I have dimensions for 2 items)

Solution maybe you believe it's easy:

Delete the dimensions.

If I delete these dimensions being ordinate dimensions I cannot show back when I'll activate the the pattern table with 4 items (this one bbeing another bug related under this discution.

In conclusion for people that don't have time to read this explanation:

In drawing I have part dimenension from features that doesn't exist in part but appears in drawing when this one is regenerated!!

[Kevin]

Just off hand it sounds like you may be having the double regeneration issue since you are using a relation to choose between tables. Without a relation I'm able to switch from a table of four items to two and the drawing updates appropriately. Going back to four the baseline didn't need to be deleted. To show the dimensions the features not the view needed to be selected. Going from two to four the dimensions could be shown by selecting the view. If a dimension was deleted the baseline needed to be deleted before the dimensions could be shown again.

[ClaudiuCraciun]

Hi Kevin,

Thank you for do some tests on this issue.

I'm using Build M060.

I tried also to do this changes without relations but the same issue. I delete the relations that change the table, activate the table manually , regenerate the model, update the drawing and the same problem.

Also for the ordinate dimensions that don't appear in drawing.

I tried selecting the feature, selecting the view but the same result. I believe the baseline must be deleted and then the dimensions will appear.

If I delete the baseline this one will delete all the other dimensions related to the baseline that Ineed them in drawing.

If I use erase again it's bad in case that I neet to show this dimension back I have to search to see on witch view was erased, select hidden dimensions and see where this dimension is located.

Usually for the components I need to automate them depending on the configuration chosed I need to show, or erase dimensions. But now with the new changes seed that I'll have some truble.

Ex:

When those dimensions that appears from features that doesn't exist PTC suggested me to delete them and will not appear.

I deleted them but when the configuration of the component required to have them I have touble to show them.

What I don't understand is why PTC is changing stuff that worked before whitout knowing the entire functionality. These are mistakes that they are doing with money from people that pay the maintenance.

I don't know who requested to do this changes that cause issue.

Seems that we pay maintenance to PTC to find ourself problems and issues in this software that were not cought but their testers.

[Kevin]

When you dimension using ordinate dimensions are you creating added (created) dimensions? Were the dimensions a different color from the others when the feature table was changed? If so, this will definitely cause the problems you're seeing and will cause you to lose dimensions if you delete the baseline. Added (created) dimensions if deleted need to be recreated (worked the same in previous versions). Added (created) dimensions will change color letting you know references have been lost but the dimensions will not be deleted. The testing I did only used shown model dimensions which didn't cause the loss of dimensions that you're seeing or leaving unwanted dimesions. I had to delete the baseline but as long as the dimensions where shown model dimensions they would all show in the dialog box. One thing that has changed is that the show erase dialog is now only used for model dimensions (use to be used to show both model and added dimensions). So if you can (just something to keep in mind) trying using only shown model dimensions. If you have to use a mixture of created and shown ordinate dimensions create a seperate baseline for the created dimensions. You can hide one of the baselines. The annoying thing is all the dimensions associated with the baseline you hide will be hidden so you need to select the dimesions again and unhide them leaving the baseline hidden.

[ClaudiuCraciun]

I don't think so that you did the test in Wildfire 5 M060.

The problem appears in Wildfire 5 M060 and I'm using just model dimensions and there is no mixture between drawing and model dimensions.

If you have this build just test it and see it.

After I installed ProE on more than 100 machines and prepare all the users with the benefit of using Wildfire 5 a basic stuff like this is not working.

I don't know how it's possible a basic stuff like this not to work and I believe in previous build is working as you tell me.

CC

[ClaudiuCraciun]

In our projects we are using most of the time UDFs. Last week I tried to do an UDF using a runner feature from Mold Module. It's something similar with a sweep but with predefined shapes. My colleagues told me that long time ago they used to have before an UDF for this feature that it worked but now Is not working anymore.

I tried all options to have an UDF that will apply this runner but with no success (feature continue to fail before to be applied)

Soon I think I'll change my opinon about PTC. It was a very good software before but now I believe is going down or the other sftwares are going very quickly up.

CC

[Kevin]

I just tried M060 and for my setup I get the same behavior as I get in M040. If an ordinate dimension is deleted it can't be shown until the baseline is deleted. Deleting the baseline deletes all other ordinate dimensions associated to it but as long as they are shown model dimensions they show again. The annoying part is they have to be cleaned up to get them to apear as they did before. The table patterns also updated in the same manner. Wonder if it could be hardware related?

[ClaudiuCraciun]

I found that deleting the baseline can bring the dimension back when I used Show annotation. In my case I brought all dimensions from the model and all of tham have the same baseline. And now deleting this baseline is like redoing again the view. I don't this so that this should be the functionality. At least was not the same in W4.

This one is a bug that cause problems on our projects.

is it there a work around? For the moment what i found is deleting the base line and show dimensions again.

Why is working like this: I have to explain to more than 100 people that believe that this software is not working properly, it's difficult to use, not friendly and when they are comming with issues like this or the other related here I cannot do anything.

CC

[Kevin]

Some notes you might want to bring up or think about:

1) You have to select the Annotation node of the desired view under the Drawing Tree (I realize you already know this). In previous versions there was an option on the Show/Erase dialog box that you could select that would determine whether erased dimensions were displayed again. As far as functionality is concerned the same things can be done as in previous versions the process is different. Maybe a request should be made to add functionality so erased dimensions can be displayed in the dialog box instead of having to look in the Drawing Tree.

2) Deleting a shown model dimension is new functionality with WF5 so it can't be directly compared to functionality in previous versions because you could only erase shown model dimensions. It certainly seems that you should be able to show an ordinate dimension once it has been deleted. My guess as to why is the baseline is treated as part of the dimension definition; notice that if you delete a baseline and try to re-show the dimensions and just select the dimensions in the dialog box the baseline is automatically selected. Since the baseline is still present the dimension is not considered deleted. Since delete is new functionality only PTC can say if it's working the way they intended. It still works the same way in M070.

[ClaudiuCraciun]

The problem is that I have to give proper explanation to some guys that don't like proe and try to find any small issue to complain about.

In this case this should be a basic functionality of proe and I have to explain them why is it like this:

- Is it the new functionality of proe?

- They need to delete the baseline in case that they want to show dimensions again?

- Is it a bug?

Anyway my colleagues gave me another homework that is interesting and i don't know how they've got this. See the attachemnt

A simple part, a curve that contains points required later to be used for ref pattern point (the new functionality in W5)

The black dimensions are comming from curve feature (points position).

Pattern of the holes is built using point reference.

If I measure the position of the hole using the datum references that I used to create the curve(points) I'm getting a small difference 52.9982 instead to be the same position as the point has 53 (even if I choose proper projection).

The same for dimensions of 37.

The same issue is for every hole. If I measure the position of the point it's correct but if I measure the position of the hole it's a small difference.

These guys are saying that are scare to use ProE in case that there are problems like this witch one is the correct value. Dimension from feature or from measurement.

[Kevin]

The problem is that I have to give proper explanation to some guys that don't like proe and try to find any small issue to complain about.

In this case this should be a basic functionality of proe and I have to explain them why is it like this:

It may be that people want the software to work a certain way but that doesn't mean it does.

- Is it the new functionality of proe?

Yes. As mentioned before erased dimensions can be re-shown but you need to select them in the drawing tree. The Show dialog box is now used to show dimensions that have never been shown or were deleted. Deleting dimensions is new in WF5.

- They need to delete the baseline in case that they want to show dimensions again?

Yes. A way around this that I came across is to to create a baseline for each dimension instead of making a chain of dimensions off of one baseline. You can create a rule layer that collects the baselines and exclude the ones you want to show. You 'll want to rename them so they can be more easily identified with an associated dimension. This way if someone deletes a dimension you can tell which baseline to delete fairly easily so the dimension can be re-shown if needed.

- Is it a bug?

Nothing I've seen in any documentation points to it being a bug but rather working they way it's intended (at least for now). I would definitely like to see changes if possible.

Anyway my colleagues gave me another homework that is interesting and i don't know how they've got this. See the attachemnt

A simple part, a curve that contains points required later to be used for ref pattern point (the new functionality in W5)

The black dimensions are comming from curve feature (points position).

Pattern of the holes is built using point reference.

If I measure the position of the hole using the datum references that I used to create the curve(points) I'm getting a small difference 52.9982 instead to be the same position as the point has 53 (even if I choose proper projection).

The same for dimensions of 37.

The same issue is for every hole. If I measure the position of the point it's correct but if I measure the position of the hole it's a small difference.

These guys are saying that are scare to use ProE in case that there are problems like this witch one is the correct value. Dimension from feature or from measurement.

Without being able to see the model I would say check to make sure the model is contrained correctly. I was able to get this to work with no problems in measurements (they reported the same values).

[ClaudiuCraciun]

Hi Kevin,

thank you for all the time spend and giving me the solutions and explanations.

I don't want to get it wrong. I'm a guy that love ProE and will love forever. I like if there is a problem in proe to find a work around for that proplem but the people with that i'm working are looking for problems to find excuses why their projects that take longer or why they have mistales etc.

Some times I understand them and I'm agree with them but most of the time i'm not agrree.

I attached also the file with differrences between dimensions uded in part and measurement.

I found the solution but is it normal for proe to work like this?

My colleagues have problems to create simple protrusion and now I have to explain what it's inside proe and why these problems are generated.

thx again.

CC

[Kevin]

Yes, this is common too. In this case ProE is having trouble creating the geometry due to intersecting geometry that creates small geometry edges (counterbore radius being the same as the round radius on the corners). Check what happens if you remove the counterbore, if you increase or decrease the counterbore diameter by 1 mm (13 and 15 respectively), and if you increase or decrease the counterbore diameter by 2 mm (12 and 16 respectively). The solution here is to look at the part accuracy. In this case increasing the accuracy (decreasing the number in the accuracy dialog) corrected the problem. Changing the value from 0.01 to 0.009 corrected the issue. You may want to set it as low as it can be set but it can affect regenration times. For this model the absolute accuracy can be changed to 0.002. Also take a look at what happens if you change to relative accuracy (the problem goes away with the default value of 0.0012).

[ClaudiuCraciun]

Hi Kevin,

soon i'll become the bad guy of this forum.

If you change the absolute acuraccy from absolute 0.01 to 0.1 the measurement is good.

If I change th absolute acuracy from absolute 0.01 to 0.02 it's working.

So is not required to increase the accuracy it's working also when you decrease this one sa I believe is not accuracy.

How come it's working for 0.1 and 0.002 but for a value between is not working

Now in this company a genius came and said we need to change the default accuracy that PTC recomend not to change (relative 0.0012) to absolute accuracy of 0.01 because all customers and the entire molding industry in Europe are using this accuracy (absolute 0.01) and if we supply projects with the default one that si comming from proe they don't accept the projects.

What i did I update more than 5000 components with this update. A lot of work when some features that didn't work during this changes and required to be redefined.

This user that came with this issue and in a way was right. Witch dimension is correct? The part dimension or the measurement. What happend if this issue will cause problem if the position required to be acurate and an error like this will cause problem in assembly. He try to blame on proe because he found this issue and in future when there will have a mistake he can blame again on proe

I found from begining that changing the accuracy will solve the problem.

He told what happend if tomorow it happend again but nobody check the dimension. He said was lucky because he've seen the issue.

What happend if I'm not cacking this error?

CC

[Kevin]

Sorry if my answer came across a little confusing but I'm not saying that accuracy is the problem. I'm saying the geometry is the problem, the sharp corners left from the material removal with the counterbore, and that changing the accuracy is the solution if you're going to keep those. Zoom in on those corners with the part highlighted and you will see what talking about. Having geomerty like that always causes problems, is considered bad geometry, and should be avoided. The problem is the geometry calculations result in a feature size that is less than the size of the accuracy value you have specified. The values of 0.02 and 0.1 work because the geometry calculations aren't resulting in feature sizes less than those values. With the geometry you have you're going to find values that do and don't work whether you increase or decrease the value. Increasing the accuracy (decreasing the value) should result in better geometry. The question I would ask is why an absolute accuracy of only 0.01 (I realize there may be reasons for it)? Based on this one part by itself rejecting it because it's accuracy isn't 0.01 seems silly because the geometry can be created without the accuracy being 0.01 but I do understand you have to work within the guidelines and standards you have. If you need the accuracy to be 0.01 the solution is to create geometry that doesn't have those corners, and based on the part, the sharp corner geomerty serves no purpose and should be removed. If anyone here modeled a part with corners like that we would never hear the end of it. One option, remove the counterbore from the hole, create a 14 mm extruded cut at the corner, group the features and point pattern them, and then add 7 mm rounds to the four inner corners then measure and see what you get. For other parts you're having problems I'd also take a look at the geometry. If you're having a lot of problems like these I would suggest that modeling guidelines for users are in order.

[Kevin]

I took another look at the point pattern definition and did find something that would be nice to know the answer to. The holes are in the right locations. What I found to be off was the reference point in the pattern definition. If you unlink from the external sketch and edit the definition of the internal sketch that replaces it you will see what I'm talking about. There is a point with a vertical and horizontal reference line. If you create dimensions from the model planes and increase the sketcher display to 4 decimal places you will find that the point is off from where it should be. If you move the first holes position to another location either on the model or off of the model the other 3 holes will follow the sketch points and the measurements to them are what they should be. It only appears to happen with rounded surfaces, whether part of your extruded sketch for the first feature or whether they are added as round features. So what would be nice to know is what causes the reference to be off or is that a bug.

[ClaudiuCraciun]

This functionality was a good one on the previous ProE versions:

In MFG Module you were able to select multiple sequences and mirror them.

Now in Wildfire 5 this one is not working anymore. I have to select one by one avery NC sequence that required to be mirrored and do the mirror function.

Before was much easier (selecting a bunch of sequences), select the mirror plan and everithing done.

If this change in wildfire 5 represent improvement something in bad. Nobody check what is going on in the new versions?

Sorry but I have to add all these issue here for:

- people to know what problmes they will have when are using this version

- maybe somebody at PTC will see these issues and will do something in future not to happend.

CC

[ClaudiuCraciun]

Guys, this one is very bad one.

I don't know is you are familiar with Pro/E MFG but having a trajectory sequence mirrored in Wildfire4 worked like this:

- First is doing the main sequence (if this sequence contains multiple trajectory is doing one by one every trajectory and at the end of each trajectory is doing a retract operation)

- Second is doing the mirrored trajectories the same after each mirror trajectory is doing a retract operation

Now how it's working in wildfire 5:

- first is doing the main sequence (if this one containd multiple trajectory is doing one by one evey trajectory and at the end of each trajectory the tool is doing a retract)

- second is doing the first trajectory mirror (up to here everything ok)

- trurd is going with no retract to the first plunged point from the main trajectory (thru the reference part destroing in this way the Ref Model and after this is going to the mirrored sequences second trajectory)

This issue is very bad specialy when you have maufacturing files already done in Wildfire 4, 3, 2 and everithing was ok but when you opend them in W5 and try to generate the ncl or G code you get this problem.

Usualy these manufacturing files have been checked before and nobody check them now in vericut to see if there is any issue and start to destroy parts and tools.

Sincerly, what don't understand is during these 3 weeks from when I switched to wildfire 5 I found several issue:

- Some of them have an easy work around but I see no reason why PTC changed the functionality and I believe the person that agreed with the changes at PTC I believe doesn't know the entire functionality that existed before

- Some of them are dezaster

I don't know how come people on those movies from PTC site can lies about ProE W 5 and say that represent Design Without Barriers, or Key to Efficient Design, no pther software like ProE

Placing all these issues here is not because I want to complain about ProE. Is because there are people that change this software without knowing how the software is working and nobody is doing enything about this.

[Kevin]

Sincerly, what don't understand is during these 3 weeks from when I switched to wildfire 5 I found several issue:

- Some of them have an easy work around but I see no reason why PTC changed the functionality and I believe the person that agreed with the changes at PTC I believe doesn't know the entire functionality that existed before

- Some of them are dezaster

Are you refering to me or just in general?

[ClaudiuCraciun]

Hi Kevin,

Sorry I didn't refer to you. I refered to PTC and all the discutions that I had with them

I appreciate everithing that you did and finding solutions for the problems that I found. You are a nice guy that like proe and seems that you have a lot of experience with ProE

Also me I like ProE but sometimes when I found that there are bugs or problems that affect my projects it's frustrating so I decided to addd all these complains here maybe somebody from PTC will take a look to these issue and will check whay there are so many issues.

CC