What I think I’ve learned about Synchronous Technology
I’ve been guessing a lot about what functionality the term Synchronous Technology was meant to obfuscate. I’ve been both wrong and right about some of my guesses. I’m not the only one. Even Evan Almighty has been wrong and right, but mainly irrelevant. Very few people have any real “facts” about this stuff. We hear what some people want to be true, and some people just blindly repost press releases without any idea at all what it means. We’ve heard some strange emotional rants about why the world needs something like this as opposed to something or anything else, and some equally strange posts seemingly defending SolidWorks from the need to keep up with the Joneses.
I’ve been lucky enough to have Dan Staples of the Solid Edge development team and Chris Kelley, a Siemens marketing guy both stop by the blog here and leave some comments. Dan in particular has been very helpful with the only hard facts I’ve heard anywhere about Synchronous Technology. Even the ST white paper is very heavy in vague declaratives. The truth is that most of those videos posted about ST, well, I’ve been tempted to post equivalent videos of SolidWorks doing exactly the same thing. Not because I have an axe to grind, but just to show that the examples we have seen of ST have been unconvincing because the capabilities already exist in SW and other places. This is why some folks have said ST is nothing new. I think Siemens has done a wonderful job of creating hype, and a less than wonderful job of backing it up. But of course that may be the main attraction of hype – not backing it up is what makes it hype in the first place.
So what have we learned?
- Synchronous Technology means that all of the relations in a part are solved simultaneously instead of linearly.
In SolidWorks, mates in an assembly are all solved simultaneously. So are sketch relations. These are driven by D-Cubed, also owned by Siemens. What are two of the least reliable areas of the SolidWorks software? You got it, sketch relations and mates. Go back and count how many times I’ve singled those two areas out as things I would like to have fixed. I think this is a bad omen for ST. D-Cubed proves that simultaneous solutions are not an incredibly reliable source. Of course SW’s problems could be an implementation issue, and it still remains to be seen how Solid Edge’s implementation of simultaneous solutions for relations within a solid model is going to work in real every day modeling.
There is no doubt, history can be a cumbersome way to create geometry, but it is also just as often a great asset when making changes. Getting away from history is a great theory, but based on what I’ve seen of simultaneous solutions, there may be a reason why other CAD companies are not flocking to it.
- Synchronous Technology really is something new – a new way of putting together a lot of old stuff.
“Synchronous” can mean “solving relations simultaneously”. Or it can mean integrating several modeling technologies or techniques. I think the “synchronous technology” term benefits from some intentional double entendre, but at the same time is sufficiently vague that it doesn’t mean anything in specific as a stand alone phrase. All of this probably makes sloganeering marketing people wet their pants in glee. Still, clever new names or not, this is all familiar territory.
Simultaneous solutions? Check. D-Cubed. Shaky foundation. Been there, done that.
Driving dimensions and geometrical relationships on “dumb” geometry? Check. Spaceclaim.
Feature recognition? Another shaky foundation. Check. Several CAM products use this, as well as FeatureWorks.
Live Rules? Behind the marketing name, this is indistinguishable from parametric relations. In fact, it looks much like parametric relations applied directly on the solid model. Dan sort of confirmed this, I think. Parametric relations on a solid model to me is the one thing to really get excited about with ST. The screen grab below was taken from a youtube video.
- Siemens is totally changing directions in the high end and midrange markets, and is throwing this Hail Mary in the hopes that all the commotion may attract more attention than all of the Spaceclaim and CoCreate non-news.
With Synchronous Technology, Solid Edge has removed itself from the SolidWorks – Inventor fray. They have been removed for some time, they just never admitted it until now. Personally, I think the hype has been brilliant, but the failure to follow up the hype with some sort of intelligible scheme that real users can understand leaves the whole thing looking like the vacuum cleaner salesman saying “trust me!!” It’s kind of telling that I had to dig this info out of people, and that Siemens clearly does not have a clear and intelligible message for users.
Chris Kelley appears to be a one-man Solid Edge blog machine, but he is marketing, not technical. Really, are there any technical Solid Edge blogs out there? Why are they not giving us the low-down from a users point of view (or are they and I’m just not seeing it?) I’m not bashing Solid Edge as a product, I don’t know enough about it to bash it. It seems so much like SolidWorks as to be practically the same, historically, anyway. But the product hasn’t seen the broad industry acceptance that SolidWorks has seen, so Siemens had nothing to lose by changing the rules and aiming low.
What I mean by “aiming low” is that the whole direct editing movement seems to be enthralled with the ease of use idea. I think this is meant to expand the definition of a “CAD user” to mean someone lower on the specialization totem. The message here is “ease of use” rather than “power” or “control”. These ideas always seem to be diametrically opposed.
- What remains in Solid Edge is no longer a history based modeler.
Some people who don’t know (including me) have said that ST is Spaceclaim bolted on to SolidWorks. I don’t believe this to be the case. I’m now under the impression that ST is Spaceclaim bolted on to CoCreate, or there abouts. There is no history left. Siemens has abandoned history based modelers. This seems to be one of the things that Dan was saying.
Many UG users I have known have said that UG never was much for history based modeling, and with Solid Edge showing as an “also ran” in the parametric history based market space, it is no surprise that they decided to changed direction, possibly if for no other reason than to distinguish themselves from SW and IV. So now they have a chance of being on the top of the heap of non-history based modelers, and trying to capitalize on the wave of direct editing modelers really rejuvenated by the popularity and simplicity of Sketchup, and given a shot in the arm by Spaceclaim.
Summary
Solid Edge and Unigraphics/NX are both strong products, but they aren’t market leaders. Solid Edge was acquired from Intergraph, and part of NX is SDRC. How is Siemens going to grow this set of products? A bold move like completely rearchitecting the software might not have been inevitable, but it really isn’t a huge surprise.
I’m very interested to see what rank and file Solid Edge users think of these changes. For those that are happy with the history-based view of the world, Solid Edge may have just handed SolidWorks a small infusion of new customers. There are several people who are not paid by Siemens who are going around claiming that this stuff is the answer to all their prayers, even though they have never laid hands on the software or fully understand how it really works.
I’m going to guess that the new Solid Edge will have a little burst of sales just due to curiosity, but in the end, interest will wane. There is a reason why products like IronCAD, CoCreate, KeyCreator and Spaceclaim have not become overwhelmingly poplular while products like Pro/ENGINEER, SolidWorks and Inventor have. I don’t think a massive hype campaign is going to be enough to change this decades old momentum.
It may be a great idea, and it may have required some revolutionary thinking to bring it to fruition, but I think Synchronous Technology is going to become another forgettable buzzword that marked a sharp turn in the road for an “also ran” history based modeler.
Matt,
I was directed to your blog due to the Solid Edge ST news, and I’m quite amazed at the “sour grapes” attitude in your post. I can only conclude that your negative reaction is a positive thing for Synchronous Technology… “The lady doth protest too much, methinks” ! I think ST has really got you scared…
With all things new, there will always be those who will find what’s wrong with it. That may in fact be what your blog is all about. I guess it would help to put your point of view into some perspective, but I’ve never heard or know anything about you, in order to say whether your criticisms are well founded, or just school yard banter.
Bob
p.s. As for me, I am a Design Engineer using CAD for over 28 years and have been a Solid Edge user for the past 8 years with great success. Oh and the reason that there may be dearth of “technical blogs” out there for Solid Edge, might be accredited to the fact that Solid Edge is that good of a product, and their Tech Support is second to none… I guess there’s no need to be bitching on blogs if everything is working right!
****
“Sour grapes”? huh. I think I’m trying to be fair. ST has me scared? Why would software I don’t use scare me? I’m not a sales person, nor a religious partisan.
There are no tech blogs on SE because the product is perfect? Are you for real? That’s like the marketing comment to a speed comparison I heard between SW and SE 10 years ago that said that SE was slower for some operation because SE did more.
You are of course welcome to your opinion, but other SE proponents have commented here with more beneficial posts. Maybe you could have identified why you think I should be more upbeat? Just because I’m not joining in a parade of blind sycophants does not mean I have “sour grapes”. The jury is still out on this product and technology, and speculation is fair game. As more facts roll in, my point of view changes. Truly game changing developments do not come along with every press release that uses the word “revolutionary”, and I don’t think it is fair to simply assume that marketers and other non-users have any idea at all what they have on their hands. In case you haven’t noticed, marketing people are given to exaggeration from time to time, and tech folks are born cynics.
So far ST appear to be mostly a marketing strategy and less a technology (despite the name). What surprise me is the Siemens tentative to use the same marketing tools (ST) to promote both NX and Solid Edge:
1.In order to sell NX Siemens have to impress and convince high level managers (they are the decision makers for this kind of products). Obviously managers don’t have any technical background to grasp the difference and implications of parallel constraints evaluation (it was called variational at the origin) and sequential constraints evaluation (also called parametric/history based evaluation). When approaching those managers, ST is a great way to send a message that the software they are buying in not just the old stuff, it’s something new and technologically advanced.
2. On the other hand, selling Solid Edge requires a different approach. Siemens have to impress and convince end users. They may not be aware of parallel and sequential evaluation, but they can recognize when a new CAD system is really providing significant improvements. End users want practical examples, numbers, specific details. Some of those users (like you Matt) my ask some very technical questions that must be honestly and fully answered in a convincing way.
Dassault seems to know the difference between the two markets, and in fact is keeping CATIA and SolidWorks marketing separate. For those same reasons, regardless of the value and the technical details of ST, Siemens marketing strategy could benefit NX, but is not going to have any significant impact on Solid Edge sales.
Matt,
Part of the problem is that marketing natually comes before the product. People are trying to rate the movie from the previews.
I did find one SE user blog on the topic:
http://wertel.blogspot.com/2008/06/se-with-synch-tech-living-up-to-hype.html
But until the product is released and people start using Solid Edge ST in the coming months its going to be a lot of speculation.
Mark
****
Thanks for the link!
Yeah, I’m eager to get more info like Dan posted.
Matt: “For those that are happy with the history-based view of the world, Solid Edge may have just handed SolidWorks a small infusion of new customers.”
I have Solid Edge ST on my computer, but unfortunately I cannot say anything about it, thanks to my Solid Edge Voyager Partner status. However, I am going to quote from the Develop3D blog (which is on your blog roll).
Develop3D: “…when you open a new part, you have two options. You can either open the part to be built with the traditional feature and history based modelling tools. Or you can use an ST enabled template, which switches on these new tools.”
Matt, you are correct about the possibility of come customers moving from one side to another. I am not sure that you have the direction correct.
Deelip Menezes
http://www.deelip.com
****
Well, the jury is out until the facts are in. It is good news that you have an option. Dan’s comment made it sound like history was out. The flexibility will turn out to be an asset. The running list of facts about ST will be changing.
I can at least confirm what Deelip said. You can continue to work with a history based model just like in the past. The main difference customers would see is a new Microsoft Office ribbon UI.
Or, you can go the DirectModel+Synchronous way. My guess will be people with new parts, parts from other systems, poorly layed out history models, or history models that weren’t set up for what the designer wants to do today will want to go this way.
Ken Grundey talks about “Hacked and Wacked” models a bit here
http://grundey.blogspot.com/2008/04/synchronous-technology.html its an older post so lacks the details you want but may be still helpful.
****
Thanks again for the link. At the very least, we are accumulating a better idea of how this stuff actually works.
What if you want to start a parametric model and then use direct model functions? Do you have to export and import, or are the direct modeling functions always available?
I think its important to point out Direct Modeling != Synchronous.
In my mind DM is really just one way of interacting with the underlying parametric modeler. Most CAD systems have some degree of direct modeling functionality. If you have no parametric modeler, you are just editing dumb geometry and it can be pretty fast. If you have a history-based parametric modeler, it can work but you can run into performance isssues because of feature tree recomputes. So DM is maybe less useful here.
If you have Synchronous as your underlying parametric modeler, it is relatively fast to solve. So DM becomes much more useful again.
Matt,
I think it is your ability to display sarcasm through text is what makes Bob believe you are “sour grapes.” You are one of the few people able to do that. I can only recall your vocal outrage at the SW 2008 UI changes, only to 180 after you had a chance to get used to it. In the SE community, we are now facing that same change to the MS Vista UI and are approaching it with open arms instead instead of vocal opposition. I think I’m a bit more familiar with your work, so I understand the passion we share for this industry and, that sometimes, our feet get in the way of our mouths. :-O
To answer your one pressing concern about SE with Synch Tech, no users have this technology yet to be able to write a comprehensive story. Beta customers and others with NDA are the only ones who have been able to put their hands on the program. Once they do, they are legally bound not to mention anything until it is made public. So once Dan or Mark, or the usual trade rags, maybe even an analyst, leaks some information, then users can come forward with a few more specifics. Until then, even the SE community has to wait and live with the speculation of possibilities to come.
I’m in a unique position in that I didn’t beta, don’t have an NDA, but have attended an interactive sneak peak. It’s public for me and I can share to the full extent of what I know. It is limited, but I’m willing to share. How do you want to handle the Q&A?
****
I don’t remember having 180ed on the SW 08 interface. It still pisses me off. I stopped railing against it because it gets old quickly.
Well, I am obviously feeding the ST frenzy here, but I’m on a mission. People are displaying this stupid religious zeal about ST without knowing really what it is. I want to figure out what it is. I’m naturally skeptical.
You can write as much as you want, and I’ll let it go up as long as you stay on topic. If you want to just send me a long email or a Word file, I’ll post it as a regular post or you can just do comments. Your choice on that.
I learn a little bit, then speculate, get it wrong, then learn some more, speculate, etc. The latest thing I seem to have mis-speculated about is that history has disappeared. So you have History AND ST, which is adding to the
doubletriple entendre of the name. Synchronous can mean history and non-history in the same product.So, if you make a History model in SE, and then want to use non-history functionality, do you have to export and import? Or do you do the SW equivalent of an inserted part? Or can you just continue? How do history and non-history interact?
Feel free to answer other questions I haven’t asked but should have.
Not sure about being a ‘one man Solid Edge blogging machine’ I think I may have to hand that title over to Mark soon. My job in this whole thing was just to get things kicked off. Even though I am ‘just in marketing’ I do recognize the fact that there are interesting conversations that people want to have with people on our products team and I think you are starting to see that with Dan’s comments here and Mark’s posts on http://siemens.com/plm/blog that it’s not going to be just marketing people that are involved in telling our story. The great thing about the read-write web is that anyone with a good story to tell can tell it.
With respect to the specific comments on ST, I guess the best we will both be able to do is sit and wait this one out: you to see what it really can do once it ships in a product (sooner every day) and me to see if it is just another forgettable buzzword or the game changer we in marketing (and products!) think it is.
****
Fair enough. I don’t see any reason why this will become a “game changer”. It might cause SW to enhance some tools that already exist, but I don’t think you will see SW make a change as drastic as Siemens is claiming this change is.
i dont think using ST is a good modeling habit, but this certainly helps with dumb imported models.solidworks should develop instant 3d and ‘driving 3d dimensions’ further to cover imported parts. those who love playing with ST crap and are excited about it, can download a copy of google sketchup for free!
***
In all fairness, there is a bit of difference between what we assume SE w ST will turn out to be and Sketchup. Sketchup is pretty primitive.
Anyway, have you used the Move Face feature in SW? It doesn’t allow dimensions, but it is useful.
maybe google should patent the DPTM (damn powerful 3d modeling !!) technology before Siemens do the same thing with ST technology
Matt,
There is another white paper:
http://www.plm.automation.siemens.com/en_us/Images/Technicom_SolidEdge_ST_WhitePaper_tcm53-61387.pdf
As for your other question “How do history and non-history interact?”
At the part level, you have to make a choice to use history or synchronous. You can not work with both symultaneously. In theory, you could start with a history based model and switch over to synchronous later (keeping the features but loosing the history) Practially, synchronous is probably better for initial concept work so that may not happen. Synchonous doesn’t need the history so doesn’t track it. There is no way to go back and turn on the history for the work you just did. I guess you could create new history going forward from some synchronous part but I don’t know it would be practical either.
I don’t think history based modeling is dead. There is a class of design problems where it is the best solution. I think the real question on everyone’s mind is if Synchronous is better for 20% of existing design work or 80% (and how will this ratio change as the technology is further developed).
matt, i was just joking about ST comparing it to sketchup . i have used move face feature in SW , it’s also a primitive feature and the results are not as expected on filleted models and does not catch the designers intent.
Hi Matt
I did post on you blog few month ago about a topic related to the interface pointing some references on Solid Edge (SE)
I would like first to position myself has a trainer and Application specialist for and SE reseller. I am not a direct employee of Siemens so I am not in the god secret
Why I position myself like that, because I felt it is important to understand that I work backstage. It is not my main role to go in front. I have my own battles to fight to educate people on how to use SE. Dan Staples from the development; Susan and Kris from the marketing are better place than me to face public.
To better position myself you will see that I am more a process trainer than a technical one. I feel that pushing button is the easy part of modeling, understand and manage what happen in my part/assembly is more important to me.
The introduction of Solid Edge with Synchronous Technology (SEST) touch some of my inner fiber has a trainer. It will also touch me directly has I will have to educate first customer (they will be the first one to use SEST), then prospects/market (that will be attract) and new customers that will see benefit for their business.
Also we have made our launch event, some lucky customers had the chance to ask questions. Giving me more freedom to talk about what’s inside.
So let’s plunge…….
First all questions could not be answer in one single post and I whant to make sure that the base is covert.
Second I am pretty sure SE will not let down parametric modeling has it has its place.
We have to remember, since the introduction of Pro-E (that claim to be the founder of the parametric) and introduction of the middle modeler like SE and Solidworks (SW) in the mid 90’s and later Inventor (IV), parametric modeling is what has condition/drive the market.
I believe that SE in its traditional form will still exist for years and will continue to evolve. No doubt that it’s development will be influence by the Synchronous philosophy.
Third I feel that market is ready for new tools.
Has we have evolve from a straight 2D wireframe on paper to a 3D wireframe (Autocad) to a Solid modeler like SE, the market is ready for the next step or can we say a new branch (approach) in the solid modeling.
By looking at this we can say that all those evolution has one goal manage change. From the strict point of view of creation, drawing a rectangle at some point is faster in 2D than in a Solid modeling. But when we add the sum of all the rectangle and complex shape 3D get a clear advantage by 90%( I will give me some margin and give a 10% to 2D because some task are better done in real 2D).
Again if we could model in 3D without having to make change life will be good.
When a designer make a change what he looking at…. Flexibility to make sure the change could be done. Rigidity (parametric) to make sure he keeps the integrity of the model/assembly. But this is like a rectangle if we increase the base, the height need to be decrease. More flexibility is less rigidity.
So this is where Synchronous Technology came in, the key phrase we have to keep in mind
“Allow more freedom to apply unpredicted change”
From that point we can start debating.
So I mention that like a rectangle, the base and height are related. How can we have the rectangle bigger to accomplish more tasks?
Most of the people will say increase its surface. Meaning more control and more flexibility, this would be done by adding new functionality on top of existing ones.
I would say this is the easy part where most people will fall. I won’t extrapolate on this easy pitfall, because it is indirectly related to the main topic… (First place where we coul debate)
From my point of view I would say if we look at the side of the rectangle we will see a cube. This is where new paradigm takes form.
SEST give users the possibility to mage a third dimension in French we call that “Tolérance Géométrique” I believe the translation will be “geometric tolerance”. In most of the drawing we create we place those tolerances to make sure the part will be functional. But we never deal with them in 3D. I would compare, up to a certain level, to “weldment beads” in assembly. For years we simply use annotation in drawing.
Another term use in SEST is “spatial relationship” or “ Related” .
Why I mention a third dimension when we design in 3D for years? Simply because today modelers are base on a 2d wire frame sketch or 3D wire frame in case of surfacing. In order to apply change we had to go back to the sketch level.
This kind of design has two branch one call skeleton modeling the other one feature modeling. (This will bring the second place where we can debate).
Base on training session I gave over the past years and past experience with other software’s , In SW et IV I believe user are orient more to a skeleton modeling, where we need to have a sketch create in order to start adding material.
In SE we could start the process of adding material even if no sketch is predefined. In fact when we start adding/removing material we have two choices “from an existing sketch” or “create a new one”.
The difference is very subtitle but it had a huge impact for further operation and management. (Third place where we can debate about skeleton approach or feature approach)
If I came back to the main topic SEST give to user a new approach/angle to start a design and also new tools to apply change. This is also in ST where it gets more fun for me has a process trainer. We now have the ability to establish relation between face to induct a behavior. The more you manage the more you can control.
So what king of control do we have?
We have the six degree of liberty that we can control with the steering wheel.
On top of that the Synchronous Solver will induct geometric tolerance base on user entry to solve the behavior of the select face and those who surround the selection. This is the Live rule (your screen capture). Advance option in the live rule give user the ability to tweak the selection made by live rule.
For now this is my first real contribution to your blog, I will prepare another post to go a little bit depther. Hope those base explanations will but some light. I am aware that some areas are still in the dark but I will work toward placing light on those areas.
Regards
P.S. French I my main language hope the words I use give a clear image of what we try to communicate
Luc P.
****
Luc,
Thanks for the post! SEwST is limited to extruded and revolved geometry? Does it support any type of geometry more complex than that? Variable rad fillets? Anything at all that is not strictly analytical shapes?
Hi Matt
Yes everything is their loft, sweep, thin wall, variable radius
No lost of command (at 95%) was report from the previous version.
Why 95% because some command has different behaviour so it is not lost but different approach. Also for specific case user could acess the SE traditional command.
Has i said often developper type 40 words a minute so we have to give them some time to write new code.
The focus of this release was on machinery part. Analytical or prismatic depending of the point of view…..
http://www.plm.automation.siemens.com/en_us/Images/se_synch_tech_FAQ_W_7_tcm53-58911.pdf
“……….
Who are the target users for this technology?
• We see this as a pervasive technology because of the
simplicity. Our belief is this will move CAD out of a
designer specialty into an enterprise capability. So
anyone that is designing or manufacturing a product
will benefit.
• All mechanical designers across discreet design and
manufacturing will ultimately want to use synchronous
technology. Initially machine designers will see the
most benefit.”
Luc P.