Pro-Engineer compared to NX7.5 1

[jeff97070]

Hi,

Let the fun begin, I have 18 years experience using Pro-Engineer. The company I work for is wanting to switch to NX7.5.
I have had one week 40 hour class room training and now about 3 months of using NX and not 100% of the time.

My initial conclusion and it could just be my years of using Pro-Engineer but Pro-Engineer is a much more flexible
easier and faster tool to use than NX. To me NX has way to much information required to get the job done like it's drawing package is very cumbersome compared to Pro-Engineer.

Please share your thoughts and I’m being opened minded about this so maybe it’s a lack of using NX 24/7 for a good year or so.

 

[JohnRBaker]

While I'll let real (and 'unbiased') users respond to your query, I will point out that NX 7.5 is two releases out-of-date.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[jerry1423]

more flexible is what ways?

 

[joycejo]

Please enlighten us on how you can begin to make a comparison after 40 hours of training and 3 months of experience after using Pro/e for 18 years?



John Joyce
N.C. Programming Supervisor
Barnes Aerospace, Windsor CT
NX6 & NX7.5

 

[looslib]

I had 15 years on Ug from UG2-V3 to NX1 when the company I used to work for decided to switch us to Pro/2000i2. I am now using Wildfire4.

1) You need more training! We trained our UG users a minimum of 13 days when they switched to Pro/E. Basic Solid Modeling-5 days, Sheet Metal Design-2days, Assembly Modeling-3 days and Drafting-3 days. We then added 2 days Pro/Intralink training and some others got Mechanica, Routed Systems and cabling.

2) There are many things in UG that are more robust than Pro/E. There are some things in Pro/E that are easier than NX. They tend to wash out over the years as they leap-frog each other with new enhancements and refinements.

3) At least Siemens doesn't release software (too often) that is not ready for production use and users can load and run an initial release. With PTC, I would NEVER turn a F00 (initial release code) loose on my users and usually it would be 4 or 5 builds before I would move to a new release of Pro/E.

4) When we switched from UG to Pro/E, the users hated the drafting module the most about Pro/E. Cumbersome, too many menus, text looks like crap, etc. Even on WF4 here, I still hear complaints about PTC drafting, and we have never used anything else besides some AutoCAd for electrical work.

What versions are you comparing? Like John said, NX7.5 is 2 versions old and 2 years behind. For reference, that would be Wildfire 5 which is 2 releases behind Creo 2.

I would move back to a NX shop in a heartbeat! For reference I am a manufacturing engineer, not mechanical and I spend most of my time these days as a system admin and working with PLM tools. However, I have done design, drafting, NC programming and GRIP programming in UG and design and drafting in Pro/Engineer.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[looslib]

I just thought of one area where 'NX has way to much information' would be very useful.

Try to find out what layer a symbol is on in Pro/E? Pro/E, to my knowledge, does not provide any information on layers by selecting an object. Even when I have the layer tree open, I cannot find what layer my drafting symbol is on. Most likely it is not assigned to any layer, so it is just in the file. Pro/E's data structure and maintenace of where in the file structure something is located is almost non-existent. In NX, every entity is on a layer, even system type stuff, although layer 257 is hard to get to at times. I know since V10 and parametric modeling the use of layers in UG/NX has not been as important as it was in pre-v10 days.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[jerry1423]

more flexible is what ways?

 

[socialcad]

NX and ProE each have their strengths and weaknesses. In general, NX is far more flexible than ProE IMO. Try turning a part model into an assembly model or an assembly model to a part model in ProE.

 

[looslib]

Part to assembly - Create a new part and import original part as a component. Our engineers do single part assemblies.
Assembly to part - It it is a single part assembly, select the model and create component in context.

I know what you mean, just teasing with the above.

I agree that NX is more flexible and provides more information about the objects in the design than PTC provides.
At the same time, there are features in Pro/E that are better than NX. Family tables of parts are better executed in Pro/E than the way NX does them. NX was way behind in developing the Template conceptfor base parts and drawings.

Like I said before, they leap frog each other with new ideas and implementation of new concepts. NX does a much better job of getting it right the first time.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[jeff97070]

I want to thank everyone for your input. I just have to much Pro-E in my blood where I can do it in sleep so the long and short of this is it's just going to take time to get NX down.

 

[mjcole]

I used ProE for 6 years before making the leap to Unigraphics v18 and was able to come up to speed with no formal training other than the CAST or Computer Assisted Self Teach tutorials which were informative but the majority of my training was accomplished by testing things out to find out how Unigraphics did things differently. At that point in time the options file for UG was a huge several hundred or thousand line .dat file which required manual editing. Most people say ProE or Creo now has a steep learning curve. So I believe it will take some time but not too much to get into a NX groove and love it.

I have not used NX in some time but plan to get a trial copy of it to see all the new powerful capabillities they have made with the Synchronous Technology features. The other resource that helped me get knowledge on how to use NX or UG back then was the site you are using right now to post this Question. looslib and I are like opposite sides of a coin. I used ProE first then got into UGNX whereas he started on UG and moved to ProE.

I applaud both companies for the robust CAD products they make but it's important to recognize that all products have flaws.

In Terms of Layers ProE has a pretty robust system for it's layers allowing certain feature types to be auto placed using layer rules which can be added in the config options. Layers in NX can be prettey complicated for new users including the Visible in view options and being able to have an item visible but not Selectable on screen. Items don't need to be on layers but most ProE users don't really care about this.

Enhancement request maybe?
As far as releasing too often that is more accurate to say for Catia who Releases functionality which takes 2 or 3 R#s to be stable V5R15 may have had new functionality that has bugs which may not be fixed till R18 if that.
Back in V18 days UG crashed for me 10+ times a week. Recent releases of Proe crash way more than in the past. For any system I'd recommend not using it the entire day. Close the program at least once a day to prevent it from loosing stability.

"It's not the size of the Forum that matters, It's the Quality of the Posts"

Michael Cole
Boston, MA
CSWP, CSWI, CSWTS
Follow me on !w¡#$%
@ TrajPar - @ mcSldWrx2008
= ProE = SolidWorks

 

[moog2]

stable....I use NX all day every day, with very large assemblies, and 4 or 5 sessions running at the same time, all with large assemblies open.
I'd say....at the most........i get one crash every 3 months....although, somethimes you'll do something stupid & it might hang up for a while, but if you leave it 10 minutes or so, it'll usually recover.
If i was using software that crashed every day i'd get very p___ed off
I've been on NX since V17, it's never crashed once a day.... before that i used catia 4.19 i don't remember that crashing that ofter either

 

[looslib]

One of the things I have always told my users of UG/NX and Wildfire is to reboot every night when you go home. Windows does not manage memory cleanly and closing a program will still leave fragments in memory. A reboot clears them. The users here are very bad about rebooting daily and even weekly. If they run into some flaky problem, I tell them to reboot before I'll even look at it. 95% of the time, they don't come back because the reboot solved their problem.

As for crashing, Wildfire is worse than NX ever was, especially with large assemblies. Before moving to 64-bit computers, we had to be very careful how we loaded some of our largest assemblies of aircraft. You could watch task monitor memory usage go up as parts were loaded by Wildfire. At 2.2GB memory used by the xtop process, Wildfire would just vanish from your screen. PTC eventually coded a memory watch into their code that would warn you when you had 5% or 10% left. Not totally helpful when loading a large assembly and you are 85% loaded.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[Kenevil]

I've been using UG/NX since about 1999. Because I didn't know better, I thought it was great until I spent a year on Catia V5. I haven't been on Pro E but I agree with your comment "NX has way to much information required to get the job done ".

It's also poke intensive and the pokes are not consistant between operators. It's like a different company designed each one independantly.

Additionally, it's not fully parameterized. This will upset a lot of people but it's true. Quick example, create a sphere with it's center point on a point. Now move that point... the sphere wont move. There are many similar situations. My company asked me to put this into a presentation to give to a "mentor". He came back with a bunch of translating options... which were not parameterized.

 

[JohnRBaker]

Additionally, it's not fully parameterized. This will upset a lot of people but it's true. Quick example, create a sphere with it's center point on a point. Now move that point... the sphere wont move.

Well you may have been using UG/NX since around 1999 but you must have stopped around 2008 before NX 6.0 was released (we're currently delivering NX 8.5) since starting with NX 6.0 all so-called basic or 'primitive' solid bodies, a Sphere being but one of these, have been "fully parameterized". For example, if you create a Sphere relative to a Point and you move the Point the Sphere would move as well.

And for the record, back in 1999 (or even earlier) you could have simply created a Sphere EXACTLY the way Pro-E (and a bunch of other systems) did; just sketch and revolve a circle whose center was located at a Point. If you had then "moved" that referenced Point, the sketched 'Sphere' would have moved as well, just like it would have done in Pro-E (and a bunch of other systems).

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[Kenevil]

Thanks for info. It seems automotive is a little behind at NX7.5. I look forward to the update if it addressed my statement but that was just one example. Helix seems to be realitivly unparameterized along with many associative line operations to points. Do you know if there were changes to those operations?

 

[JohnRBaker]

The 'Helix' has been completely reimplemented in NX 8.5 (the consensus is that it's perhaps now the MOST 'parametric' object in NX).

As for you comment about "many associative line operations to points" I'm not sure what you're referring to. Perhaps if you could be a bit more explicit I could make specific responses.

One thing that you need to keep in mind, like MOST other systems, NX is depending more and more on using Sketches for 2D curve and profile type tasks and while it is true that the older none-parameterized (i.e. 'dumb' curves) are still supported, they are being relegated to a 'legacy' class of objects meaning that they have been removed from the default User Interface layout and must be 'resurrected', as it were, by 'customizing' the UI to make them accessible once more.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[uwam2ie]

The helix in Nx 8.5 is really well done!

thank you

 

[jerry1423]

If I am not mistaken NX/Unigraphics has been able to create parametric spheres since around 1991

 

[JohnRBaker]

While that has been ostensibly true, at least in the sense that the size of a Sphere 'feature' could be changed parametrically, either by directly editing the expression controlling the Sphere's diameter or by editing the size of the circle if you used the 'Select Arc' method of creation, however, like the other so-called basic or 'primitive' objects at the time, which included along with the Sphere, the Block, Cylinder and Cone, ONLY the size parameters where editable. What was missing was the ability to control the location or orientation of this objects either by associating them with some other object, like the example of a Sphere referencing an existing Point, or by directly editing some sort of origin/orientation scheme inherent to the feature itself. That all changed starting with NX 6.0.

So while it is true that the SIZE of 'primitives' have always been 'parametric', it was incorrect to say they were 'associative', which to many people, and we agree, is an important characteristic of being 'fully parametric'. Now in our defense, UG/NX is one of the few systems which actually provided any sort of basic or 'primitive' bodies. Most other systems, particularly many of the newer 'sketcher-based' systems, expected users to create sketches and use either an extrude or revolve type of operation to create these same basic shapes, so even though the old 'primitives' were never fully associative, they were still very useful for what they were intended to be used for, as the FIRST BASIC body to which detail feature were to be added. If they were ONLY being used in that manner, the fact that they could not be associated to another object was hardly ever an issue. It was ONLY when they were being ADDED to or SUBTRACTED from an existing body that this lack of associativity became an issue and in all honesty, this was considered to be an abuse of a 'primitive' feature. These types of operations SHOULD have been performed using an extruded/revolved sketch, like all the other systems where they expected people to work this way, but BECAUSE we were kind enough to continue to even offer the ability to create these basic or 'primitive' bodies we were expected to bring them up to the latest standards of BOTH 'parametrics' AND 'associativity', which, for the record, I voted against doing since I considered it as not being relevant for the INTENDED USE of basic or 'primitive' bodies, but I was overruled.

So now you know the 'rest of the story'...

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.