For years, I've used an old color based CAD standard in our office. I'm very familiar with it. We rather simply came up with our own layer names, more or less in plain English. All objects are typically drawn on the color "ByLayer", and the printed color and thickness of the lines are determined by the color, at our choice, via a plot style table (pen assignments), that I'm sure most of you will generally be familiar with.
So, for example, I know that a line that I see on my screen in model space as red will plot as black at a thickness of 0.030" on the printed page. For in-house production of a drawing, this system seems to work fine.
Now, when one of our files is transferred to an outside party, the system seems to break down. Usually, we don't, nor are we asked to, provide to the receiving party the pen assignment (CTB) file. And, even if we did, there arises a fundamental problem in collaborating with others, namely, that the system that we have created for ourselves will undoubtedly be different than the system that the receiving party has created for their use. And so, I suspect it is typical that a receiving party employing a similar system does not even bother to ask for CTB files, definitions of the layering system, etc., but rather takes the information provided in the DWG file, and modifies the properties of objects within that DWG file to fit their own needs. Then, if it comes back to our office, again without a CTB file, things aren't usually the same way they were given. And so on and so on, round and round we go....
So the idea of having a National CAD Standard (NCS) seems appealing, for it would seem on the surface to promote the facilitation of building and site data using a common language. However, as I begin to consider implementing the NCS, there appears one immediate problem with its implementation, namely that a user of the standard must purchase it from the publisher, and the cost is $405 last time I looked. This gets you the latest version, but when a new version comes out, one must pay again. I suspect the use of the standard is therefore not widespread. For a larger firm, perhaps the cost of acquiring the standard would be a drop in the bucket. But, for smaller firms, the cost becomes more of a concern. Parenthetically, I wonder whether NSPS would consider negotiating with NIBS on behalf of the NSPS membership, to acquire for its members the right to access the NCS without charge.....
However, even if the NCS were free, there remain additional problems with its implementation, similar to the problems described above with our own current system. For (correct me if I'm wrong) the NCS appears to provide a template that is designed to be adapted by the user to fit its own needs. Surveyors might generally agree to put all their drawing objects on a layer beginning V- (discipline designator for surveying and mapping), but whether to put bearing and distance labels on V-ANNO-BRNG or V-CTRL-BRNG or V-PROP-BRNG or V-PROP-TEXT or some other combination appears to be up to the end user to decide.
Now, in its defense, the NCS does appear to encourage some level of consistency to layer naming, and provides a framework for same. But, as far as I know, the NCS allows you to more or less freely choose a layer name, using the prescribed codes, as long as it is reasonable and follows the general format DD-MAJR-MNR1-MNR2-STAT, where DD is the discipline designator, MAJR is the Major Group Layer Name, MNR1 is the 1st (optional) Minor Group Layer Name, MNR2 is the 2nd (optional) Minor Group Layer Name, and STAT is the Status. Additional MINR group modifiers may be inserted between MINR and STAT.
Significantly, the NCS does not appear to govern the color used for a layer, the linetype used for a layer, the thickness of a line, the color printed on the page, etc. So, implementation of the NCS still leaves much choice in the hands of the end user. And this freedom of choice will invariably produce variety, and variety will produce some of the same symptoms that I now experience with the current system.
Of course, a contract arrangement will allow the super to specify the standard that must be followed by the sub, and this will tend to promote harmony between super and sub. But the contracted standard only works within projects that follow it. And, since the contracted standard is arbitrary (even if it meets NCS requirements), it will not readily transfer to another party that hasn't been following the same standard. Am I right?
Sorry for the long-winded post, but what do you all think about the NCS? Is it worth the effort to implement this?
Best,
Al
I believe this version offered free from the Corps of Engineers is very nearly -if not exactly - the same as the US National.
I'm pretty big on CAD Standards but I confess that I've never actually seen the US National in it's entirety. If you could get 100 guys to pitch in $5 each they would all soon have a copy.
Our system will not let me access that site for security reasons...
> Our system will not let me access that site for security reasons...
Perhaps try it from home. I have the files but they are a little too big to email. The site itself is benign.
Maybe try your state DOT standards. It seems that they are free. That's what we use. I like the .stb for plotting that WisDot uses. The basic plotstyles are Existing (everything is shaded), Proposed (everything is black), and Normal (WYSIWYG). For lineweight, set it up in the layer and use ByLayer.
I was able to register, log in and retrieve the files without any problem. Thanks.
I've been contemplating that approach as well, but in MA, the templates that are provided differ widely from the National CAD standard.
Standards are always dependent on what your client, or your client's client, requires. U.S. Army Corps of Engineers started requiring digital CAD (MicroStation) files around twenty years ago. The web site you were referred to above is the current version of the standards they required then. It is an expansion of the National CAD standard. If you, or your client, are working for any Department of Defense related agency it will likely be required.
If the survey work is to be used by or for a contract with a state agency they may have their own standard. I worked in two different standards at the same time twenty years ago. I would expect there are firms that work with more than that on a regular basis. Addressing different CAD standards should be a part of negotiations on contracts. Simply a part of the world we live and work in these days.
This is a fantastic post and a topic about which I tend to be very passionate. As a CADD junkie, I berate others in my office regularly for not "following the standards".
But there are, as you mentioned, pitfalls. I think this sums it up nicely:
Problem: 100 different competing CADD standards.
Solution: Let's make a single standard for everyone!
Next Problem: 101 different competing CADD standards.
For several years, I was employed by a very large (multi-national) corporation that did all kinds of engineering and surveying across the US and Canada. There were a few hundred offices and so the company said that there would be a "company-wide CADD standard" so that if offices shared drawings, there would be no issues in compatibility.
Unfortunately, it never worked out that way. Since most of the offices were actually small companies that were acquired by the parent company, these small offices would pretty much just keep using their old standard (if they had any) or simply do the minimum necessary to look like they are following the company standard. The end result was that every office had their own "version" of the company standard.
Honestly, I don't have any specific solutions. At my office, we use (mostly) a version of the NCS. I tell all of my team that they need to follow the standard template that I've set up as closely as possible. But I know that sometimes there are unique circumstances that require unique layers/styles/etc. In those cases, use some rational thought and name the layer something that is sensible. If you have a layer showing existing topography of garden statuary (which now I should make a standard layer for that), don't put that line work on a layer named "v-topo-shot". Call it something like "v-topo-grdn-stat" or whatever. And put a descriptor in the field for that. Make notes on a non-plotting layer if you need to. My goal is always to be as clear as possible so when another person opens that drawing, they can understand what is being shown.
Oh, and always, always, ALWAYS make things color and linetype "ByLayer" so that when I insert it into my drawing as an xref, I can simply do some layer color changes and it will work just fine with my ctb files.
Skwyd,
Thanks for understanding what I'm trying to get at. It's a design problem. And, it appears as if there is no one silver bullet that's going to solve this problem. I had naively thought that maybe following the NCS would solve it, but while it might improve things a bit, it isn't perfect. Different clients will have different standards, and by contract, I will either follow them or not.
As a business decision, I'm thinking that it might be wise for a small business to simply use a standard that already exists rather than trying to create one from scratch. And further, it might be wise to choose a standard that is required by your biggest client. So, if most of the work that requires following a CAD standard is done for client A (say, a multi national surveying and engineering firm) then maybe use their standard - not only for their projects, but for your own projects as well. Of course, it isn't perfect, because there will still be other clients that require following a different standard.
Again, thanks for posting.
Al
In the past I have requested CTB files from others that are providing CADD drawings to me. Especially if I'm going to want to plot plan sheets and such for a submittal. And I've provided my CTB files to others for that same reason.
I thinking having a standard within your own operation (no matter how small) is a good idea. It provides some consistency in how the plats and other drawings you send out look. I've had a county surveyor tell me before that he always knows one of my maps because they all look the same. That made me pretty happy.
I'm working for a smaller firm with about 12 employees total. And 4 of them don't do CAD too much (2 owners and my main field crew). But we still work on continuing to develop and review our CADD standards regularly. Plus with all of the things that we're starting to use in Civil3D, it is helpful to have a common starting point between my boundary drawings, my topography drawings, and the design drawings that the engineers are putting together.
I even go so far as to develop a standard for the folder structure on the server so that the various drawings and other files are stored in similar fashion on our file server for every project. It isn't perfect, but it is a time saver. And the more employees involved, the more important it seems to be to have at least a modicum of a standard in place.
> I even go so far as to develop a standard for the folder structure on the server ...
I don't consider that going far. I actually think that is the first thing that should be done.
I really do not like the NCS and I have one client who wants everything in NCS.
I have a nice layer translator through Carlson that converts my layers to theirs and then back again.
Well, that's something that I'd like to know more about. Is that a built in feature of Carlson Survey, or something special? What's it called?
Thanks.
I'm a big proponent of standards - and there are so many to choose from! I generally set up my templates using the Corp of Engineering - it's a free guide and based on NCS. I also prefer using named plot styles and using the layer settings to drive my design. Using the standards makes it much easier for me to integrate with GIS.
I have worked in offices where the CAD person would not compromise or budge on his choices for layers or styles, so the surveyors ended up doing their own thing. It really makes a mess when trying to share data.
for conversions, from AutoCAD forum: LAYTRANS
"Just make sure you have a file with their layers in it.
If you work with this company a lot, you can save the layer mapping you'll be doing and put it on the network so you'll be able to do it faster next time."
If I recall correctly, there was someone that used to work with me that was developing a LISP routine that would go through each layer in a drawing and prompt the user for a new layer name, color, and linetype. I don't know what ever became of that though as I no longer work with that person.
I've heard of third party software that does layer translations using a template. I believe you have to define the translation initially.
The NCS is cumbersome at times. And some of their suggested layer names are based upon abbreviations that are a bit of a stretch. But, I've made it work for my office so far.
One of the key principles that I use when it comes to CADD standards is that I don't care what the name of a layer is as long as someone can look at that layer name and have an idea of what is supposed to be on it.
I've received drawings from others with layers named something like "LL2AD001". Looking at that name, I've no idea what is supposed to go on that layer. Then, when I check the drawing, that layer is a design layer with the proposed interior walls of a building that I'm staking. To me, that layer name is worthless. The name "A-WALL-DSGN" is cumbersome, but it DOES indicate that it might have something to do with walls, came from the architect's plans, and is a design layer.
> I've received drawings from others with layers named something like "LL2AD001". ...
It wasn't that long ago that file and directory names where limited to 8 characters, and not much longer before that you had to type in every command character by character. A lot of users who came up in that environment continue to limit their naming out of habit, and newer user see what the old dogs are doing and try to emulate them without knowing the why. It's kind of a pet peeve with me. I'd like to buy a vowel, Pat.
When I started on C&G in '89 there was a very short list of available layers whose names were not user defineable. Microstation made a big leap to having 64 levels (64! what will we do with them all?) but, again, the names were not user defineable. Guys my age who learned in this environment are now managers.
Most of the "crazy" layer names I get are from architect drawings, actually. And I would venture to say that these layer names aren't coming from the "old-timers" that learned in the days of DOS limitations on file names and such.
In the past I put out a lot of effort to help the "old-timers" (that taught me pretty much everything I know) improve their CADD skills and show them things that can make their tasks easier.
I still recall being in College learning AutoCAD on an old XT machine running AutoCAD 6 in DOS. Those wondering moments when you realize you had to regen your drawing and it was going to take at least 20 minutes for this to happen. So you'd tell your buddy to watch your backpack while you went to the student center to get a soda for the both of you.
I'm fortunate that I learned right when new technology was just budding in this industry. I'm a big technophile, but I also can appreciate doing things "the long way" because you can catch a lot of blunders that way.
I believe this version offered free from the Corps of Engineers is very nearly -if not exactly - the same as the US National.
Link not working. Try https://www.wbdg.org/ffc/army-coe/cad-bim-technology-center.
As I just re started another new biz, I am interested in this... but I don't often get asked to share cad files, so that's not a big issue.?ÿ I prefer to KISS with the layer names just what they are for.?ÿ
?ÿ
If I ever do need to share and someone at a big firm needs it in NCS or WTF ever system, they should be able to do that in about 30 seconds with the software available.?ÿ ?ÿ
?ÿ
If I'm reading that right, they want me to pay them something like $500 for the privilege of putting a prefix of?ÿ "V-" in front of my layers then I'm free to pretty much do what I want after that????ÿ Seriously??ÿ ?ÿYeah, no.?ÿ ?ÿ?ÿ