NX vs Creo vs SolidWorks, et al

[Jim Elias]

I feel your pain. The first time I reached out to your IT department to spec me a new computer (I do my own now and just tell them what to order, but back then I didn't know any better) I specifically told them what I needed it for, foolishly thinking they'd research what I needed.

They sent me the specs for a generic laptop that they'd order for someone who just needed it to get on the web, and maybe use Windows Office.

LOL, the point was that there was no pain to feel... NX apparently doesn't have the same issues with "generic" computers that SW does.

I'm sure that at some point in data complexity, you will need a "real CAD computer" with NX as well. But it's clear that that bar is way higher for NX than for SW, and it would be really nice to someday understand why. The NX work I was doing on that secretarial laptop had appreciable assembly-and-geometry complexity, it wasn't just primitives.

My current computer could be considered a WINO (Workstation In Name Only). It has a low-power processor and only needs a glorified USB phone-charger to keep running all day. NX runs fast. SW bogs down. But my vertebrae are grateful and I don't use SW all that much anymore anyway.

[SPerman]

At the core of the problem, IMO, it the companies philosophy. Siemens has prioritized stability and robustness. DSS has prioritized marketing and cramming in new features that don't work and very few people wanted in the first place.

[bnemec]

I chuckle at the conversations about all the various "Fast" "slick" "efficient" ways of managing selection behavior or turning selectable items on and off in Solidworks. Going on in the SP3 thread now. Solid Edge users aren't much interested in all that; Solidworks has the workflow turned on it's head where the user selects some thing(s) in the vast sea of selectable items in the graphics space then SW will offer what actions are available. Solid Edge users select what they want to do then Solid Edge automatically applies selection filters based on that the action is. Much more solid approach to UI stability and performance. I notice it most in applying assembly mates/relations.

[Jim Elias]

At the core of the problem, IMO, it the companies philosophy. Siemens has prioritized stability and robustness. DSS has prioritized marketing and cramming in new features that don't work and very few people wanted in the first place.

To be fair, NX is pretty merciless when it comes to costs. There are basic functions like wall-thickness analysis which my NX bundle doesn't have and would cost me tons, so in the rare instances when I need those, I export to SW.

So either way there's no free lunch.

[jcapriotti]

I haven't use NX since NX4 but my experience with it wasn't as good. We had CAD workstations but assemblies in it were dog slow at times. Luckily we only had about 50 parts max in an assembly with the average being about 10 (medical company). Where as my current company runs SolidWorks with thousands of parts. Hard to compare stability as it wasn't apples to apples. I did enjoy NX equations capabilities at the time, and it's ability to select more geometry types to define features. I hope the sketcher is better than it was back then....SolidWorks sketcher blew NX out of the water.

I did find that NX and SolidWorks have more in common in modeling workflow than Solid Edge NX, they are both a bit more freeform in how you go about creating features. So transitioning to NX was not too difficult. SE seems more like Creo and wants to walk you through the steps, which is great for learning but seems slower and in the way later.

[SPerman]

To be fair, NX is pretty merciless when it comes to costs. There are basic functions like wall-thickness analysis which my NX bundle doesn't have and would cost me tons, so in the rare instances when I need those, I export to SW.

So either way there's no free lunch.

Agreed. I would dump SW tomorrow and switch to NX/TC, but I can't justify the added cost, which is significant.

[Jim Elias]

Agreed. I would dump SW tomorrow and switch to NX/TC, but I can't justify the added cost, which is significant.

Understandable, and I doubt that I will actually ever go off SW maintenance, as tempting as it is every time that yearly invoice from my SW VAR shows up. (I like my VAR fine, no beefs there.)

There are simply too many potential clients out there running SW. Many don't love it, but they're already there and it would be infeasible for them to switch. This is why I would never advise anyone in the gig world to pooh-pooh SW proficiency... that can be a quick way to learn about skipping meals.

[Jim Elias]

I haven't use NX since NX4 but my experience with it wasn't as good. We had CAD workstations but assemblies in it were dog slow at times. Luckily we only had about 50 parts max in an assembly with the average being about 10 (medical company). Where as my current company runs SolidWorks with thousands of parts. Hard to compare stability as it wasn't apples to apples. I did enjoy NX equations capabilities at the time, and it's ability to select more geometry types to define features. I hope the sketcher is better than it was back then....SolidWorks sketcher blew NX out of the water.

I did find that NX and SolidWorks have more in common in modeling workflow than Solid Edge NX, they are both a bit more freeform in how you go about creating features. So transitioning to NX was not too difficult. SE seems more like Creo and wants to walk you through the steps, which is great for learning but seems slower and in the way later.

When taken in the contexts of the relevant times, I think that SW when I started (97 / 98-Plus) was a lot stronger than it now is. UG/NX might well have been weaker than it now is. Ownerships, development teams, etc. have all changed.

[SPerman]

I haven't used NX in 6 years, and I've only dabbled with SE and F360, so my opinion may be skewed. The one place where SW is head and shoulders among the competition is the myriad of ways they've made available to minimize wasted motion and clicks. Breadcrumbs, shortcuts, (hotkeys and mouse gestures) context based menus, etc.

I was used to the order of operations in NX, but after working with SW, I find I prefer to pick the object(s) and then the action. If I started using another program regularly, I would adjust, but given both options, that is the path I take. (I'm also the guy who prefers a RPN calculator, or did last time I used one.)

[jcapriotti]

When taken in the contexts of the relevant times, I think that SW when I started (97 / 98-Plus) was a lot stronger than it now is.

Yes and no......stability is worse in some areas.....but functionality is much better. We started on 98 and I can think of so many things it was missing....no way I'd go back to it.

NX, I'd love to get my hands on to see what they improved. I had little problem picking it up and feel its workflow is similar to SolidWorks. I just hated the sketcher which I think was a newer implementation for NX at the time as there was something called "Pre-v18" sketches that really sucked. And before that was "curves" which some users insisted on continuing to use which I couldn't stand, and ugg...parametric curves were worse. Their sketcher was missing the types that SW had and same went for assembly constraints.......however that was long ago.

[SPerman]

I started transitioning from NX to SW in 2016 and I didn't notice any appreciable differences between the available sketch tools, but maybe going that direction I didn't realize what I was missing.

[SPerman]

There are demo versions of NX if you really want to play with it. This is a 30 day trial, but I thought when I looked into this last year there were other options.

https://trials.sw.siemens.com/en-US/tri ... _BwE&gad=1

[mp3-250]

SW sketcher is based on D-cubed engine licenced from siemens.
UG had its own sketcher pre v18 before D-cubed was aquired by Unigraphics and its tools made it inside NX.
curves had a lot of implementations.
the legacy ones are non parametric non associative, and you can draw them in the 3d space without a 3d sketch, then there are features like the bridge curves which are very useful with surfacing.
NX allow a fine tune modelling, you can control the tolerance of many features and switch from surface to solid, extrude lines or edges.
The approach with UG NX is extreme flexibility. the software is there to allow you to put together a working and stable 3dmodel out of pure garbage.
you can use a non lInear approach and if the things goes bad delete all parameters without re-importing the 3d data.
a fully parametric approach is also available, equation and relation are available consistently from every dialog unlike SW UI.(roll eyes )

NX does cost a lot, but while we can quantitify how much a software cost us, it is difficult to quantify how much all the bugs in SW costs us in terms of lost productivity, workarounds of workarounds, legacy feature that explode in your face when upgrading..
at least I can (and had to) quantify how much money costed us a critical bug as the time to remake 3d models is easy to count. Or pdm downtime due to a less then reliable administrative tool is also simple to count.
Don't get me wrong, some tools inside SW are nice concepts, but not fully implemented or implemented so bad they make your drawing implode as they become heavier and heavier. (I don't think we are the only one with cutlists of 100+ items with slow drawings)
In NX you will never read about "workflows" or the SW spin about the user doing it wrong (without the software complaining btw) , the machine not performing well, the bad alignment of the planets in the galaxy nearby etc
Breadcumbs, shortcuts, lean UI... very nice to save 3 clicks allowing us to speed modelling, just to have your assy blowing in your face for "reasons" or a drawing taking 10 minutes to load or it takes sometimes 10 seconds to draw a dimension.
And you wonder why, at least here in Japan, many companies that use SW use a 2D CAD do make their 3d data 2d drawings exporting some views from SW

[zxys001]

Creo and NX are on another suite/cost level... imho, this is really about SE (SW should not be here imho).

NX, (with my limited exposure is NOT that easy to use) for the skilled user it has many ways to go, it has an impressive tool set but still trying to catchup with todays UI's and out of reach from most users/companies.

Creo is still the best parametric / direct modeling power modeling tool, it's not easy to use, but in the right hands, can dominate and is dominate (Asia) because of the speed and it's versatility. (I have a lot of experience with it)

SE is really a much better/cheaper way to go. (from what I've test, it's a much easier to use as a parametic and direct modeling tool)

SE is more in line with Creo/SW.

Anyhow, maybe this should be "SE vs NX" and why does NX cost so much? (Hi Roland! )

[Jim Elias]

Creo and NX are on another suite/cost level... imho, this is really about SE (SW should not be here imho).

NX, (with my limited exposure is NOT that easy to use) for the skilled user it has many ways to go, it has an impressive tool set but still trying to catchup with todays UI's and out of reach from most users/companies.

Creo is still the best parametric / direct modeling power modeling tool, it's not easy to use, but in the right hands, can dominate and is dominate (Asia) because of the speed and it's versatility. (I have a lot of experience with it)

SE is really a much better/cheaper way to go. (from what I've test, it's a much easier to use as a parametic and direct modeling tool)

SE is more in line with Creo/SW.

Anyhow, maybe this should be "SE vs NX" and why does NX cost so much? (Hi Roland! )

"Creo and NX are on another suite/cost level...", nope, that's missing the point of why I chimed in.

I'm not addressing the functions. My NX bundle license is the bare-bones NX10101, it does a few things which my SW-Prof won't do, and there are a few things which my SW-Prof does, which my NX won't do. On the balance that's a wash.

BUT -- NX is way more stable and runs way faster, on hardware that is much less capable (and thus much less costly), than is needed with SW. That's the crux of what I'm often addressing, and I've never seen an answer as to why this is, though it is indisputably so. If it's the "outdated" UI, then I would tell SW -- you've "modernized" your UI too far if it's causing you to resort to this level of resource piggery.

Creo: I guess Creo/Pro-E is fine, I haven't used it much myself. I have collaborated often enough with Creo-based teams and my impression has been that Creo is about at the NX level, because it looks like we tackle similar issues with similar methods. I haven't noticed that the Creo folks are able to do anything more than me and/or do it faster.

[Jim Elias]

Yes and no......stability is worse in some areas.....but functionality is much better. We started on 98 and I can think of so many things it was missing....no way I'd go back to it.

NX, I'd love to get my hands on to see what they improved. I had little problem picking it up and feel its workflow is similar to SolidWorks. I just hated the sketcher which I think was a newer implementation for NX at the time as there was something called "Pre-v18" sketches that really sucked. And before that was "curves" which some users insisted on continuing to use which I couldn't stand, and ugg...parametric curves were worse. Their sketcher was missing the types that SW had and same went for assembly constraints.......however that was long ago.

yes, needs to be viewed in the context of the times. I likely couldn't go back to 98-plus now either. But my recollections tell me, there was a lot less waiting for SW to get stuff done, and there were a lot less crashes.

[Jim Elias]

I started transitioning from NX to SW in 2016 and I didn't notice any appreciable differences between the available sketch tools, but maybe going that direction I didn't realize what I was missing.

They're very similar. What makes the NX sketcher rock for me is the "Edit" function, especially as regards Offset relationships. In SW, when sketch Offsets go faboozled, it can get extremely laborious to fix and oftentimes, you're just forced to delete the Offset and re-constrain every segment manually. With NX, you can do an Edit and then have full control over the selection set from the ground up.

The ability to edit a sketch without rollback in NX is something I addressed earlier in another couple of posts, it is extremely useful.

There are a couple of SW things which I would like to have in NX sketch -- three-point angle dimensioning comes to mind. But I don't miss these to anywhere near the extent that I miss the NX Edit function while SW sketching.

[zxys001]

Hi Jim,

Ok, sure many moons ago I had the basic NX Mach version as well, with a few modules (have to get the bill but it was freeform surface and animation)
I could get it to crash no problem. Fact, NX handles crappy data really well, period.
Yes, NX crashed less then and recently. ... than SW but my SW models/assys are/were fully controlled parametric masters. (and wave is not even close to what I was managing in the past and present, mainly dumb models, which is fine)...and yes, SW demands much more hardware ( i have a core i9 40gigs ddr5 ram, 3070 gpu)...and I've always bought the best i can afford (own/used SW since 98+)
Creo, on the other hand was least likely to crash and could handle much larger data sets with less hardware.
NX is very good with large data sets on minimal hardware! Again, it crashes...and, I have lost data (had to had IT pull an old version and lost the rest, yep)
Maybe it's because they manage the graphic engine pipeline and memory better, like Creo does? Or maybe it's the older unix code or "X Windows" interface from years ago. When I used Pro/e on WIndows NX on a RISC CPU or Sparc, I rarely ever crashed. But, as PTC moved towards greater integration with Windoze, it crashed more!
With so many Windows configs and quality - cpu/gpu/memory and the graphics drivers, these are the main culprits.

so, yah, the next wave we have is 100% cloud based BROWSER's. I'm not impressed using it (onshape), I'm still more effective without it but I get it and as per above, $ drives this and industry feeds off it.

It's funny like watching WALL-E ... recliners and sippy cups.

"Creo and NX are on another suite/cost level...", nope, that's missing the point of why I chimed in.

I'm not addressing the functions. My NX bundle license is the bare-bones NX10101, it does a few things which my SW-Prof won't do, and there are a few things which my SW-Prof does, which my NX won't do. On the balance that's a wash.

BUT -- NX is way more stable and runs way faster, on hardware that is much less capable (and thus much less costly), than is needed with SW. That's the crux of what I'm often addressing, and I've never seen an answer as to why this is, though it is indisputably so. If it's the "outdated" UI, then I would tell SW -- you've "modernized" your UI too far if it's causing you to resort to this level of resource piggery.

Creo: I guess Creo/Pro-E is fine, I haven't used it much myself. I have collaborated often enough with Creo-based teams and my impression has been that Creo is about at the NX level, because it looks like we tackle similar issues with similar methods. I haven't noticed that the Creo folks are able to do anything more than me and/or do it faster.

[HercalloY]

You will find the whole industry has or is moving to subscriptions. Autodesk, PTC, Siemens, 3DS..

The "whole" industry is not "moving" towards this. The developers of the software are PUSHING customers in tyrannical way towards this goal that is theirs. Nothing to do with customers moving anywhere. Enough with the marketing b.s. You're biases are showing.

[Ryan-3DS]

The "whole" industry is not "moving" towards this. The developers of the software are PUSHING customers in tyrannical way towards this goal that is theirs. Nothing to do with customers moving anywhere. Enough with the marketing b.s. You're biases are showing.

My intentions are not to incite a riot by engineers and designers that are passionate about their tools. We all have a little fear when things in our jobs change. Especially the main tools we use in our jobs.
But I will clarify, the companies that control 98% of the CAD industry, not the entirety, are moving to a subscription model. There are several components that make up a subscription model and the "big guys" in the industry all have similar but not the same subscription models. Please do NOT equate the subscription model with cloud-based, cloud-ready, or cloud-native capabilities. That is different and as mentioned earlier each of the big guys do this differently.
CAD is not a siloed tool. It may not even be the bigging of the design process.
You have to look at how fast software technology changes. Software technology is changing faster than the software OEMs can keep up. That puts another level of stress on companies using the software. It is a rabid race for software OEMs and their customers to be first to market, capture the market, and maintain the market. The design, manufacturing, selling and product servicing process is rarely a single-person process. It requires collaboration, validation, simulation, specification, quality, and sometimes regulation.
If you know me, I am a "preacher" at heart. I understand the big picture, and most of the little details, and I understand the effect of change on people.

[Glenn Schroeder]

My intentions are not to incite a riot by engineers and designers that are passionate about their tools. We all have a little fear when things in our jobs change. Especially the main tools we use in our jobs.

. . . and especially when we're convinced that the change will make our job more difficult, not less.

[bnemec]

The "whole" industry is not "moving" towards this. The developers of the software are PUSHING customers in tyrannical way towards this goal that is theirs. Nothing to do with customers moving anywhere. Enough with the marketing b.s. You're biases are showing.

Close. It's not the developers though, rather the execs at the software companies driving the change. What we see is comparable to and organism that was once symbiotic feeling the pressure of competition and the actual evolution of the host. But instead of changing to continue to be mutually beneficial they are shifting to more of a parasite that's trying to convince the host that the host will die if the relationship is severed. Look at the behavior and notice many are transitioning from focusing on filling a market void to survival mode.

On the other hand I see several good VARs trying to take the high road and branch into tangent markets of CNC and robotic support, additive manufacturing, etc. They are working to find new ways to maintain that symbiotic relationship to the engineering/manufacturing world. They understand their place in the "food" chain. Reminder that engineers are similar relationship to manufacturing. Products can be made without engineers/designers, our job is to help manufacturing be more productive, efficient at producing better products.

[zxys001]

...they are shifting to more of a parasite that's trying to ....

[HercalloY]

My intentions are not to incite a riot by engineers and designers that are passionate about their tools. We all have a little fear when things in our jobs change. Especially the main tools we use in our jobs.

Please don't try to dismiss my comments with an attempt to make them out to be emotional and there for not relevant. I am no more "passionate" about CAD software as i am my set of verniers. Sure.. i enjoy using CAD applications more... but that does not equate anything with the negative connotation you are trying to paint. especially the "fear of change" crap you spout. Classic attempt at bringing emotional language into objective and rational thought. Please drop that marketing crap... you're biases are showing.. again.

But I will clarify, the companies that control 98% of the CAD industry, not the entirety, are moving to a subscription model.

And this proves my point. Ih ave no idea why you would even think such a statement would do anythign other then re-affirm what i said and what everyone knows.

There are several components that make up a subscription model and the "big guys" in the industry all have similar but not the same subscription models. Please do NOT equate the subscription model with cloud-based, cloud-ready, or cloud-native capabilities. That is different and as mentioned earlier each of the big guys do this differently.

I tell you what.. if you "big boys" stop conflating the them.... we will too. Sound like a deal? (one that'll never happen.. this is rhetorical obviously. hence tyrannical/dictatorial/monopolistic practices as mentioned).

CAD is not a siloed tool. It may not even be the bigging of the design process.
You have to look at how fast software technology changes. Software technology is changing faster than the software OEMs can keep up. That puts another level of stress on companies using the software. It is a rabid race for software OEMs and their customers to be first to market, capture the market, and maintain the market. The design, manufacturing, selling and product servicing process is rarely a single-person process. It requires collaboration, validation, simulation, specification, quality, and sometimes regulation.

All marketing fluff. Software is software is software. Its the flexible part of the hardware equation.. yes. But it is still the same as its always been. Its still a tool PEOPLE use to do something. We don't need, nor do we request.. to be locked into negative aspects and have the rental model, along with other poor practices, forced upon the software realm. especially not when we keep having people like you... continually pissing in our hands and tell us its raining... i mean.. gently pattering down with H2O from the cloud.

If you know me, I am a "preacher" at heart. I understand the big picture, and most of the little details, and I understand the effect of change on people.

He who tries to convince others of something that doesn't exist, usually for nefarious purposes. Preacher indeed.

[HercalloY]

Close. It's not the developers though, rather the execs at the software companies driving the change. What we see is comparable to and organism that was once symbiotic feeling the pressure of competition and the actual evolution of the host. But instead...

Of course. i use "devs" as just a simple, catch all and quick explanation without giving to go into such a detail explanation. We all know the way big corps work. of course.. i do often have conflicting thoughts about this as just with the Nazis... many went along willingly stoking the furnaces. So you can't give them to much of a free pass me thinks.

[Jim Elias]

<snip>

And you wonder why, at least here in Japan, many companies that use SW use a 2D CAD do make their 3d data 2d drawings exporting some views from SW

I've just started seeing that directly because I'm interfacing with an Asian supplier who models in SW... I was surprised to see that the prints appeared to have been done in AutoCAD.

@mp3-250, do you know the particulars of how that workflow functions? i.e. when they do a model change, how does this get propagated to the DWG? Thanks for your time.

[mp3-250]

I've just started seeing that directly because I'm interfacing with an Asian supplier who models in SW... I was surprised to see that the prints appeared to have been done in AutoCAD.

@mp3-250, do you know the particulars of how that workflow functions? i.e. when they do a model change, how does this get propagated to the DWG? Thanks for your time.

While I cannot speak for all companies, for what I can see here the answer to your question is: "there is not such workflow"

For mechanical design Japanese companies are still heavily relying on paper. And I do NOT mean strictly 2D CAD... but PAPER drawings made by HAND. While this give them a lot of flexibility to quickly adjust for modifications, well you can imagine the rest by yourself.
My personal experience is that compared to Europe there is a 10-20yrs gap.
In the past I noticed some Chinese and Korean suppliers were on par or well ahead European way of designing, but Japan is stuck in the 90s.

From time to time there is some research about CAD adoption.
UFJ reasearch did a survey back in Dec 2019 (It is not changed drastically)

(source ministry of industry here [jpn] https://www.kansai.meti.go.jp/3jisedai/ ... 3d-cad.pdf)

On the left is the CAD adoption % graph in mechanical design in surveyed companies (1527)

12.2% NO DESIGN DATA (meaning paper / nothing digital at all)
26.5% 2D Data
44.3% Both 2D & 3D Data
17.0% 3D Data

On the right is the design format requested from supplier and the majority apparently ask for drawings (again the paper one not even the 2D data)

54.3% drawings
23.8% 2D data
15.7% 3D data

My theory is that production is mainly paper driven and asking for drawing data comes at an additional cost so they simply cut it.

ADDENDUM: update of the 2D become a case by case scenario. Like a top assy drawing is unlikely to be touched for a "minor" change even if visible. Or a 3D is not modified for a "small" change on the 2D side. Obviously this lead to any sort of problems from a data management perspective, but to change this kind of things there is a lot to change in the work mentality.