×

Management

Dialog Box August 2006

6 Aug, 2006 By: Cadalyst Staff

---

Readers have their say.

TALK TO US To be considered for publication, you must sign your letters and include your company's name, address, and daytime telephone number. Letters are subject to editing for length and clarity. Send to: email: editors@cadalyst.com

Locked Out
The latest review of graphics cards mentions framelock and Genlock, but nowhere, including your Web site, can I find out what these are. When I searched on your Web site, I found several references to 3Dlabs. I couldn't see the date for them, but I wanted to tell you one thing that you already know: 3Dlabs has gotten out of the graphics card business for CAD workstations. What the dirty, rotten, lousy scoundrels have done is left all of the people who bought their previous high-dollar graphics cards up the proverbial creek. With my high-end Dell workstation, I purchased a Wildcat 7210 card that cost me $2,162. Less than three years later, AutoCAD and Inventor changed driver requirements totally, and my very expensive graphics card is now junk. I have Inventor Professional 11, and I cannot even look at a drawing in it. I have called and e-mailed 3Dlabs, and it basically said, "Sorry Charlie." That is not right. To just turn around and abandon us is about as lousy customer service as you can get. I know 3Dlabs is owned by Creative, who makes the Soundblaster audio cards. Needless to say, I'll never purchase one of those audio cards as long as there is a competitor with something similar. In the meantime, thank you very much for the online magazine. I do enjoy it, and I read it every month.

Editors Respond
You don't really need to worry about framelock or Genlock unless you're doing video editing or animation. One synchronizes different video streams and the other frames. I suspect you were looking at an old review (one done prior to 3dlabs going out of business). We haven't covered any of its products since we heard the company was leaving the CAD market. I'm sorry you ended up with a useless card. Unfortunately, that has happened fairly often in the CAD space.

---

Downloadable Hatch Patterns
I was wondering if you have downloadable AutoCAD hatch patterns for LT2007.

Editors respond
Yes, we do. You can go to www.cadalyst.com, click on Get the Code! and search for hatch patterns. We've had many hatch patterns submitted by readers over the years, interspersed among all the LISP code.

---

Affordable Modeling with Carrara
In Cadalyst, you asked for CAD tidbits. There is a program for modeling, animation and rendering that we use here that, to my knowledge, has never been mentioned in Cadalyst. The name is Carrara, and it is from Daz. Models come from Autodesk Inventor and are exported as SAT files. Carrara is intuitive and easy to use, as opposed to [other rendering systems that have] a big learning curve and endless icons. There is also an advanced building modeler, ArchiTools, a separate add-on to Carrara, which is used for modeling buildings. The renderer includes global illumination (fancy name for radiosity), caustics and HDRI. Also included is the full package of animation as well as a scene wizard with a ton of presets in many different categories and a texture room with a huge set of textures in every category that you could want. I used this program to make a laboratory fly-through for a bid on a project. The lab included Inventor models, an environmentally safe room and overhead location stills. Our company was awarded the contract at the end of the presentation. We were told that we had captured the vision perfectly. It is much less expensive than any other package with these types of tools and I was making renderings the first day I installed the software. There are many free tutorials available.

---

Under Pressure
I'm a regular reader of Cadalyst and enjoy your articles. I was hoping you could possibly recommend a 3D CAD program for my line of work, which involves pressure vessels. Assuming you're not familiar with what these are, they're steel tanks with flanged connections and can have a number of structural reinforcements. SolidWorks and the like may be overkill for something like this because there are no moving parts. Right now I'm using BricsCad, which is on a par with AutoCAD 2000. I would greatly appreciate your input. Thanks!

Bill Fane Responds
1. AutoCAD 2007 has greatly improved 3D capabilities: check out the Cadalyst Labs Review.

2. Just because something has no moving parts does not mean it isn't a candidate for a parametric solid modeler such as SolidWorks or Autodesk Inventor.

Parametric modelers have a number of features that would perfectly suit what you are doing.

1. You can play what if. Dimensions are the reverse of AutoCAD's associative dimensions; in a parametric modeler if you change a number in a dimension and then the geometry updates, meanwhile maintaining relationships such as tangents, parallels, equal sizes, concentrics and so on.

2. 2D working shop drawings are created semiautomatically in a fraction of the time.

3. Dimensions can include formulas and equations relating them to other dimensions and variables, and they can be linked to cells in an Excel spreadsheet. A single model thus could represent a complete range of pressure vessels. All you would need to do is to change the volume and pressure, for example, and everything would update. Wall thickness, flange sizes, bolt patterns (including size and quantity of bolts), size and quantity of connections and so on would all update in compliance with the ASME Pressure Vessel Code. This also includes the 2D working drawings

4. Item 3 also includes mating parts. Your parts may not move, but the vessel still attaches to foundations, pipes, valves and so on. Everything will stay in step.

5. You can do complete physical analyses, including weight, centre of gravity and stress analysis.

So which is better, SolidWorks or Inventor? It depends, given that they are about the same price. My personal preference leans toward Inventor; among other things it also includes a copy of AutoCAD in the same box. The bottom line is that it depends on what your customers and suppliers are using. Compatibility is usually a bigger issue than specific features in the software.

Hope this helps.

---

Acrobat Insider Newsletter to the Rescue
Just wanted to send a quick thank you to author Michael Dakan for his Acrobat Insider for AEC Professionals newsletter [about how to use Adobe PDF stamps]. The day before it arrived, I learned that I needed to add a piece of information to every title block in a 181 drawing set. And I don't have the original CAD files for most of them. After reading about stamps, I realized that this would be the easiest way to add the information. And it's going great. Thanks again for the timely information.

---

Lost Layers
I understand this -- all objects on the layer BLOCK TEXT and all objects on the layer BLOCKTEXT move to the new layer TEXT -- but how do I make TEXT go back to BLOCK TEXT and BLOCKTEXT when I want to send it back to our clients?

We work with Pizza Hut, and they create layers we don't need. When we send it back to them, they need to see the layers they created. For example, they have five A-ROOF layers, and we have only two A-ROOF layers. How do I get back the layers?

Bill Fane Responds
If I understand your problem correctly, there is no quick and easy solution.

If you move an object from one layer to another, it has no history of where it originated from, short of using Undo. This, of course, will also undo any editing you did in between, which I assume you don't want to do.

The good news is that AutoLISP programming will allow us to attach extra, private data onto objects. This function is called extended entity data.

My "Learning Curve" columns here and here explain how to attach and retrieve extended entity data.

In your case, you would attach the original layer name before moving the object, then retrieving the original layer name and returning the object to that layer when you are done.

Hope this helps.

---

Lesson Learned
Hey, here's a hot tip for you fellow newbies out there. If you have an opportunity to do some outsourcing CAD for an architect or firm, Learn Their Standards! Nothing will tweak a customer's faith in your abilities to provide this service more than submitting a drawing with your own cute ideas as to what text styles and layers should consist of! I Know, I just got a somewhat "What the heck is this?" e-mail from a customer, and it is quite a humbling experience! Basically after three semesters with straight As, a year of OTJ experience and a CAD by the B***S attitude, I have learned the best lesson yet. Do yourself a favor and get a template -- or a set of standards from your customer so you can avoid having to adjust your billable hours.

---

Get Rid of the Oldsmobile
Autodesk is suffering from what I call General Motors-itis. It owns the market in the 90th percentile; it decides what, why and when; it lets you know what its decision is and then cancels what you already have. Alfred Sloan had a word for this: planned obsolescence. (Or was it Thorsten Veblen?). With so large a market share, the only reason to innovate is to spur upgrades. The '56 Chevy will be way better than the '55! The product will never be perfect, because they do not want it to be perfect. They want it to be eternally upgradeable.

Of course, we see where GM is today. The concept of adding another brick on top of the previous one leads to a pile that ultimately will topple, as it has in Detroit. The way foreign automakers have competed, and won, is to abandon Alfred Sloan's blueprint. It worked in its time and made a lot of people wealthy, but today is no longer the time to own GM stock.

The major problem with AutoCAD is that it's a drafting system. Like GM, Autodesk needs to get over the founding principle, which has become obsolete. The best CAD systems today are first about designing buildings, then about documenting them, which is the natural order of the process. Autodesk needs to get rid of its Oldsmobile and focus on the functions the market wants and they do not provide. They have tried, by purchasing Maya and Revit. But my experience with Revit is that Autodesk is trying to bring Revit closer to the past rather that use it to take the company into the future.

Yearly upgrades to Excel? Is there a lesson here?

Excel users have learned that they don't need constant upgrades. Excel 2000 works just fine. So Microsoft launches its insulting Dinosaur campaign. Let's hope that Autodesk doesn't try this route. What we need from Autodesk is more innovation and less nostalgia. Microsoft is promising that the next version of Office will be a radical overhaul.

It's time for a radical overhaul of AutoCAD. Not a yearly dribble of new features that other systems have had for years.

---

Simply Switch
My reaction to Autodesk's greed and a real lack of concern for AutoCAD users is to switch to a new software company that is not taking advantage of us. It will hurt at first (retraining, new add-on software, compatibility with old DWG files, etc.), but we will come out better in the long run. And Autodesk will lose numerous students who would have been future users and buyers.

Shame on you, Autodesk !

---

Smaller Pills, Less Pain
I prefer to make changes as little as possible, as your list shows. But we have done both the leap frog method and the subscription service. I have decided that the subscription service has made for a shorter learning curve with each software release. I suppose that means smaller pills to swallow on an annual basis.

---

Custom Roll-Out
In response to the article "Cadalyst Readers Balk at Yearly AutoCAD Upgrades," I agree. We are going to have to create a special project to push this software to all of our users' PCs. We add many customization VBA and AutoLISP scripts to the AutoCAD native software. Every custom piece of software has to be tested and retested prior to our installing the latest version of AutoCAD to any user PC.

---

Costly Step Down
Nice article on Autodesk's upgrade/subscription policy. You mentioned offering "lower" upgrade prices to move to vertical products. That's how an IT manager got us all on AutoCAD Mechanical. It was cheaper to upgrade from AutoCAD 2000 to AutoCAD Mechanical 2004 than it was to AutoCAD 2004.

As you pointed out, future upgrades are at a higher price. But were you aware that if you wanted to downgrade from a vertical to plain AutoCAD, Autodesk charges a fee? We inquired about upgrading from AutoCAD 2004 Mechanical to plain AutoCAD 2006. In addition to the upgrade fee to get on a newer product, there was an additional fee to move off the vertical. I don't recall what the fee was, but it was enough that we stayed on the Mechanical vertical, which does offer us some nice options.

---

New Releases Incomplete
I agree strongly with many of the points made by Steve Johnson in his "Bug Watch" column, "Don't Feed the Bugs."

In my opinion, a new release of AutoCAD doesn't become a true release until it functions correctly. From that point of view, we are now paying Autodesk to be its beta testers. As you stated, portions of the 2006 release are still not fully functional (i.e., the CUI), and for my department's needs, the 2007 release is entirely unusable at this time due to several known issues or bugs that probably will not be fixed until the next release.

I am glad you touched on the issue of paying for annual subscriptions. Which is worse, paying for a year's worth of subscriptions without getting a new release or getting a release that is pushed out long before it is ready? I say the latter is worse, because now I have wasted many hours on upgrading, only to be forced back to the previous release. Because we are unable to use 2007, will we get our money back for this year's subscriptions? In what other industry can a company do shoddy work, admit that their product needs more work, not bother to fix the problems in a timely manner, but still keep the money they were paid to do the work?

I agree with your take on releasing the product with incomplete features. Ever since Autodesk went to the annual release schedule, a lot of the new and revised features have that incomplete feel to them. Changing features just for the sake of change is not good, and for many of us who have used AutoCAD the same way for 10 or 20 years, we are used to how things worked and didn't need most things to be revised.

What we truly want in a new release is for the existing features to work faster, and for the features that needed tweaking to work better (i.e. hatching still doesn't work well). Maybe the new CUI is better than the old way, maybe not. But for those of us who used the MNU and MNS files for years, we had a system that worked and weren't asking for a change, and now we are forced to use a system that is slower and more cumbersome to accomplish the same task.

It's very apparent that the push for annual releases that are incomplete is the root cause of some bugs occurring in very basic features. For example, grid lines showing up on image plots to HP plotters, and Map 2007's inability to consistently read the world file for rectified images. These issues are both known to Autodesk, and are examples of basic, day-to-day functions that no longer work correctly. Proper handling of rectified images is as simple as reading a coordinate from a text file and putting the image at that coordinate! And if a new CAD software company released a product that could not do something so simple and universally expected as plotting correctly to an HP plotter, they would be ridiculed right out of business.

I've heard the complaints of many users who say they will be dropping their subscriptions because it simply is not worth upgrading every year to a product that never works correctly out of the box. Some users have also stated they are contemplating dropping Autodesk altogether. There may not be enough users who feel this way for Autodesk to feel it financially, but hopefully it is enough to at least prompt Autodesk to rethink their current policies.

---

Consider the Customer
I hate to keep harping on a subject, but can anyone show me where these subscription models do anything for the customers?

We used to get upgrades/patches included in base pricing. Now we get to pay for it? Such a deal! Sign me up!

If Autodesk wants to get me excited about a subscription, it had better freeze upgrade pricing as long as I keep my subscription current. And that means holding pricing for more than one year, I'm talking about as long as I keep my subscription paid up! Ten years, good for me, and Autodesk doesn't make out too bad either. But with no guarantees as to holding fee structures, subscription pricing is variable with each new release and year that passes. It does not compute. Maybe Autodesk would like the sign-ons and passwords to my checking account, too? Take what you want Autodesk, I know you have my best interests at heart . . . Unrestricted subscription pricing amounts to the same thing . . . (and I'm not really exaggerating very much).

In fairness, Autodesk isn't alone in following this pricing strategy, but it is pushing customers toward alternative software. Or at least it's opening the door for competitors. My employer stayed with AutoCAD 2002 due to some customized piping/routing software we use. We stepped off the upgrade bandwagon quite a while back. Then when we looked into upgrading, there was the reseller waving the repurchase stick. So we went whole hog and looked into other 2D and 3D packages. In the end, we went with a competitor's 3D product. Yes, they have a subscription pricing model too, but we wouldn't even have looked at a competitor if the pricing from Autodesk hadn't gone crazy.

Their loss, our gain.

It's funny how much trouble Autodesk has had getting its huge installed base excited about the move to new-versions/3D. Autodesk owns the keys to the kingdom. It could price its products in a lot of different ways to entice the switch and befuddle its competitors. But it can't figure out how to have its cake and eat it, too. It can't figure out how to make a compelling case for costly annual upgrades along with subscription costs, because they only make sense as a revenue generator for the company. Customers see no benefits.

Autodesk is also unwilling to let the real world intrude into its revenue plans. Customers/companies don't want to have to implement a new software package every 12 months, it takes too long to get everything running smoothly across the corporate networks, it takes too long for custom/vertical market add-ons to get updated and it takes too long to work out all the bugs and install all the patches that are required to keep it all running smooth. We're supposed to see an advantage to doing that every spring? Please.

As another example of "if it wasn't for these darn customers, we could really make some money" mentality, look at the DWG version translator. There was no way in heck Autodesk would have ever have released that until SolidWorks developed one and released it on the Internet. Autodesk was using the "you can't open newer-release drawings with the old version" as a stick against their customers, as another way to move them onto the current release. I can't imagine that SolidWorks' release of the DWG Gateway made Autodesk very happy. And in truth I'm surprised SolidWorks didn't get sued under the DCMA laws for doing so. Perhaps Autodesk is developing a SolidWorks translator to turn the guns back onto SolidWorks??

And the wheel turns again . . .

---