Even More Synchronicity
I have to admit, I was originally kind of interested in this Synchronous Technology topic, but I”m quickly getting over any sort of interest that I had in it. I”m catching a fair amount of crap for it, and not really getting what I think are very good answers. Since I can”t rely on finding anyone with an objective opinion out there, I”m just going to have to put this on the shelf and wait until maybe someone from Siemens sends me a copy to review (hint, hint).\n\nSiemens has started selling this “pig in a poke” at a greatly reduced price (~$3000 – in a notice from a reseller who spammed some forums). The price seems to suggest that they see Spaceclaim as their main competition, even though they are asking you to “trade in” your SolidWorks license to do it. Of course they can”t actually take your SolidWorks license, so the whole thing is just a stunt.\n\nAnother thing that confuses me is you hear these marketing testimonials about how company X is seeing such great improvements, but at the same time you can”t get a straight answer out of anyone who has actually used it about functionality because the software is not going to be available until August.\n\nThe fact that no one can really descibe it succinctly in a way that makes sense says everything they need to say about it. So if I can”t get users to say anything intelligible about it, then it remains just another glossy ad with nothing behind it. The part that really irks me is that the people who say I”m just criticising because I”m devoted to SolidWorks are waving the SEwST flag without any personal experience whatsoever. I am at least looking for objectivity.\n\nOne of the fellows who posted a comment here sent me a couple of long emails which are equal parts rant, defense of MS Ribbon interface, incorrect memory, self contradictions, and underdog flag-waving. Scott Wertel it seems is a fan of Alibre and Solid Edge. He claims CAD is just a tool, but also defends zealotry. I”m not sure what to make of this contradiction. He claimed I had changed my mind about the SW 08 interface, but later recanted, because he was obviously wrong about that.\n\nAnyway, this is really long, but it explains a few things. I”m trusting that Scott”s one view of this stuff was enough to give him the correct impression. Here is what Scott had to say, with comments interspersed:
===============================
Religious zeal
One of the things you have to realize is that the SE community and SW community are made up of entirely different types of people. The SW community is very passionate about SW. I (k)new a guy who went unemployed for 6 months because he wouldn”t take a job working on any other software. That”s asinine. CAD is just a tool. The SE community, as a whole, agrees with that philosophy. CAD is only a tool. If it works, great; if it doesn”t, find another one. So far, SE works. When it doesn”t, I make a call report direct to Seimens and get an answer. The times I worked for companies that utilized Works, I was greatly dismayed by the type of support, or lack of support, that VARs provide. Perhaps that explains the difference in passion for a software. Or perhaps it is because SE users know that SE works (as well as any other mid-range MCAD software) but not many others in the world have even heard of it. It is always a struggle explaining to people that we use SolidEdge, not SolidWorks. Honestly, people always try to correct me “…you mean Works?” “NO! I don”t.” But I digress.
I will tell you, though, that this religious zeal you speak of is a community response due to the fact that SE finally has something new and unique to differentiate it from the competition. More importantly, Seimens is actually marketing. SE has never, read that NEVER, been marketed effectively and the community is ecstatic to have something to grab on to. So there is a difference between religious zeal of Synch Tech and the fact that users are just excited about having something worth showing off, before the competition has it.
Skeptical
When you have only the info that marketing feeds you, which is probably intentionally incomplete, and almost certainly exaggerated in some way, people speculate, looking for the truth behind the hype. Speculation is a guess. Guesses can be wrong. I keep guessing about stuff and asking for verification from someone who has seen it because I want to understand. I do think that this technology does have the capacity to be useful, but really its just another tool. Because it is not going to do away with history based approaches, it cannot be as revolutionary as the original history based revolution. To deny it has any importance would be wrong, I think, but some people are certainly overstating the case.
Also, if this is going to take a lot of training, doesn”t that contradict the whole “ease of use” mantra in the first place? This is why I don”t think Siemens really has a great message for the technical user. It doesn”t fit the low end user, and doesn”t eliminate the specialist knowledge needed to run history based CAD. This is part of the reason I don”t get it, and why I think it is not going to be the block buster people are assuming it will be.
============================
180ed
I could have sworn you wrote an article on how the SW08 UI was actually very efficient once the user got used to it. My apologies if I mis-spoke. But, can you blame SW or SE for going this route with the UI? More people are familiar with Office 2007 than they are any CAD package. Getting to know one means you”ll get more comfortable with others. Not to mention, do you pay a small licensing fee for the Vista-like UI, or spend a small fortune developing your own? I think we”ll see many more Windows-based programs with a similar UI. I”ll say that it makes no difference to me. I”ll adjust and be just as proficient. I managed to do it in Word and Excel, I can do it with CAD, too. Just need an open mind and willingness to climb the relearning curve.
Apology accepted. UI? Hold it, what are we talking about here? Ribbon UI is widely reviled. Some people like snakes or spiders as pets. Sure, you can find people out there who like the Ribbon. SW doesn”t license the Ribbon, they did spend the money to develop their own, as I heard the story. Interestingly, I think SW changed their interface at least in part to differentiate themselves from “the rest”, but it has backfired, you”re right about that, everybody looks the same more than ever now.
================================
CPDA White Paper
Interestingly enough, I haven”t seen that white paper yet. I”ve seen just about every other one, but not that one. I did a quick glance through it and want to point you to the last figure. In it, D-Cubed and Parasolid are on the last rung, Synch Tech connects the two and is a bridge to the main CAD package, either SE or NX. The point to make here is that this layer is a completely new technology. Whereas D-Cubed and Parasolid are open technologies, meaning Seimens will license it to anybody, ST is as proprietary as one can get. It will never be licensed, sold, or shared. As far as facts go, that is about as specific and we can get because all the employees at Seimens have been incredibly hush-hush about it. I don”t know if you are aware, but Synch Tech is one of the reasons Seimens bought UGS. ST was in the development stage during the acquisition and was one of the major selling points to the future value of NX and SE.
Solving relations
This is where the confusion lies and where speculation has ruined the experience, as does the “Synchronous” title. Synchronous does not necessarily mean at the same time. Most importantly it doesn”t solve assembly relationships and sketch relationships at the same time. There are no sketches. There is not even a sketch environment. There are some 2D geometry creation tools, but that is to only aid in defining regions on a surface or plane to create 3D geometry. Once the 3D feature is created, the 2D geometry is absorbed – it no longer exits. The dimensions placed on the 2D elements have been transferred to the 3D geometry.
I”ll try to explain with an example. Start by “sketching” a square and extruding it to create a box. Once you create the box, the sketch is gone. The horizontal and vertical dimensions used on the sketch now define the edges and/or side faces of the box. The sketch is no longer needed. If I need to change the size of the box, I can find the feature in the feature list because it does exist as a feature, and then edit its properties. Or, I can just add another 3D dimensions (like PMI dimensions) and change its size. Once I finish the command, the dimension is absorbed into the feature. It”s there, but not there. I can either find it again, or just create another dimension if I have to make a subsequent change. There is some logic behind removing duplicate dimensions, but I don”t have any specifics on that.
Now add a cylindrical boss to the box. Like before, you kind of start with a sketch but not really a sketch. If you had 3D wire frame geometry, or intersecting edges, or any other geometry you could use it to create a circular region (projected if need be) onto the surface of the box. Go ahead and dimension the circle to get the right size. Once you extrude it, the region and dimensions get absorbed into the 3D feature. Click the edge and change the circular edge”s diameter. Or, apply a new dimension to the circular edge to change its diameters. Click on the cylindrical surface and a dimension diameter appears there, too (if I remember correctly). Delete the initial box, and the cylinder remains because it is not history dependent on that base feature. Keeping the first box there, add another boss on the open size of the cylinder so the circular edge disappears, you can still change its diameter without having to roll back the new solid feature to “find” that edge again.
This is not direct editing, or explicit modeling, or using push/pull/move face commands that add yet another feature to the history tree. (Yeah, everybody has got those.) This is on-the-fly changing the parameters of the solid without having to worry about how it was created. This is because of how Live Rules determines relationships, not D-Cubed or Parasolid. Live Rules is that ST layer between D-Cubed and the CAD application. Your basis that building ST on the foundation of D-Cubed is on shaky ground is correct, if it actually were based on D-Cubed, which it is not.
===================
When I talked about solving sketch relations simultaneously, I was talking about SW with D-Cubed.
“Synchronous Technology” I will take to be a marketing buzzword, with as much meaning as “Sausage Integrator”.\n
To be fair, in the image you cite (captured from the white paper below) looks asmuch like D-Cubed is part of the foundation of ST as anything. ST is on top of D-Cubed just like a house on a foundation. I think you”re trying to pick and choose your analogies to make it look better than it is. The fact that D-Cubed is what ST is built on top of is at least cause for concern.
=============================
Live Rules
I brought it up, I might as well try to explain it technically with some examples. Dan did so, and based on your blog I think you understand, but let me elaborate. FYI: It is something to be excited about.
Take for instance you want to create sweep. The commands in SE and SW are the same. Create a sweep path (usually 2D sketch but can now be a 3D sketch) and a 2D cross section of the profile to sweep along the path. In the path sketch, you need to apply fillets to all the corners, the fillets need to be of a the proper radius to make sure the solid doesn”t buckle when wrapped, and most importantly, you need to make sure that you have tangent relationships created between the lines and arcs (via D-Cubed). What happens if you accidentally delete the tangent relationship? The line and arc and still tangent, you didn”t alter the geometry, but SW or SE (i.e. Parasolid) is too stupid to know that they are tangent. With live rules, it finds tangency based on actual physical geometry and not applied constraints. It does it on-the-fly, it does it live. In this case, synchronous may be that it finds not only the geometric constraints, but also geometry constraints (hard to differentiate).
Now take that example and apply it to solid geometry, not sketches. I have a box with the edges filleted. I don”t have a tangent relationship between the fillet and the sides. I have geometry that is tangent, physically, so live rules picks up on this and uses it. If I want to lock-it-in, so to speak, or always make something tangent (or coplanar, or parallel, or whatever), then you can add what are called “persistent” live rules between the geometry. This is more liken to assembly relationships, but on the 2D part. To do this in old SE or SW, you would have to create that geometry in the sketch just the right way to be able to apply that D-Cubed relationship. What if you can”t sketch a 2D cross section like that? No worries with SEST, because they don”t even have to be the same feature to have relationships exist between them. They don”t even have to be the same part in the case of ST for assemblies. This is a fundamental change in solid model methodology. It truly is amazing (and different) but the only way to grasp it is to experience it yourself and that won”t happen until August.
The other example I like about Live Rules is the fact that there are no parent-child relationships. What was a child, can become a parent. Have you ever wanted to keep a hole centered, and then the face of the solid a certain distance from the whole? You can plan that out with layout sketches and other “guru” methods, but you always have to have that surface first and the whole dimensioned to it because you need solid geometry to cut the hole from. If I want to keep the hole in the same place and move the surface, I would have to really dig into the feature tree and find the design intent to make sure I lock down the right dimensions and change the other ones. Usually it requires changing the size of the block, then going into the hole feature and changing its locating dimension by the appropriate amount to move it back. With ST, you just add a dimension between the hole and the surface, change the dimension (clicking the button to lock the hole”s position for this edit) and the surface will move. Very easy.
Do you have design intent where you need to keep certain things symmetric? Not a problem, because you can lock symmetry so a single dimension change to one surface will actually move the opposite side surface by the same amount to maintain symmetry. Again, live rules. But that is where a lot of rank and file SE users are scared of Live Rules. How much does it pick up? How does it find what it associates and changes automatically? How can I force it to only change what I want, locking in other geometry if necessary? Live Rules makes it very easy to make change, almost too easy. How can I be assured that some purchasing agent, or machinist isn”t accidentally altering geometry he shouldn”t (besides using PLM or file security measures) while they are viewing the file? And that is also where the big learning curve is going to come in. How to become proficient at what associations live rules makes on the fly and what it doesn”t, and when I need to add explicit relationships and when I just need to hot-key an override.
Right now, rank and file SE users know as much about SE with ST as you do. We are scared, but welcoming the change with a bit of pessimism. It is new, it is revolutionary, it not just hype. But we won”t know the total implications of this until it released in August and we actually get to use it.
===========================
If a 3D path is something new for a sweep in SE, no, SE and SW are not basically the same. None of the requirements that you cite are requirements in SW, except under special circumstances.
You”re talking about feature recognition, which is another very iffy sort of thing. The rest of this sounds reasonable.
===========================
History
Yes, history can still exist. For the first few releases, if not permanently, users have the option through an administration control panel to turn on ST or not. If the admin turns it on for the users, they still have the option of modeling the old way -with history – or the ST way.
Speculation
I don”t think any seats of any CAD package are going to be converted to SE just because of ST. I do think that a lot of SW or Inventor houses will buy one seat of SE with ST just to be able to handle imported geometry easier. I think customers debating between SpaceClaim, CoCreate, or KeyCreator will now look at SE with ST instead, because it appears to be the best of both worlds, plus a drafting package to boot. Will they buy? Who knows. I don”t think this answers all of our prayers. I do see a lot of machine shops picking it up for that reason. A lot of CAM packages handle imported geometry well, but this is a way to clean up some of it before going to CAM, if necessary.
I think SE with ST is a great buzzword. But beyond that, I think we are looking at a revolution in CAD the same way Pro/E revolutionized CAD in the 80s.
===============================
My thing with history is “can you do both in the same model?” Is it one or the other, or can you do both? The meaningless “synchronous” implies both.
I have strong doubts about equating ST to the Pro/E revolution. With Pro, it took a long time for other companies to challenge their dominance. That wouldn”t happen today I think because of the vicious competitiveness of the market.