TheSwamp
CAD Forums => CAD General => Topic started by: HarleyHetz on April 02, 2007, 09:34:43 AM
-
I run AutoCAD 2K6 at home, and 2K2LT at work. I was talking to a friend yesterday who draws schematics for a living, he was helping me with a "block" question, and he said something that got me to thinking, so I figured I'd get some second opinions. :-)
He told me that he saves his blocks as a drawing file instead of as a block. He says that makes them more managable. That if you want to change a "block", you pretty much have to explode it, make whatever changes you want, then resave it. Whereas if you save it as a drawing, you can open it anytime you want, change it, save it as a new version, or write over the old version. :| (Doesn't seem like that much difference to me??)
But the question I have is how does this affect the overall drawing size when you are using many blocks inside of a single drawing. Does it use more memory to put a bunch of "drawings" into your drawing as blocks as opposed to putting "blocks" into your drawing as blocks??? :ugly:
-
I'll attempt an explanation of this one :-)
By saving the blocks as separate dwg files he is probably able to manage this block library better because he can use file management to sort these blocks/dwgs into logical folders.
When you simply store them in a dwg you have no way of managing the separate blocks since they only exist as part of another drawing.
The way the autocad handles a block is:
The block definition is non-graphical information contained within the dwg and only exists once.
Block references are the graphical piece. There may be many references referring back to this block definition. When you update the block definition all references to that block are updated as well.
So whether you store a block within a file or as an external file has no real revelance to the file size unless you are storing unused blocks in a drawing in which case you are doing yourself a dis-service.
Did I confuse you? :-)
-
Once a "drawing" is inserted in a file, it becomes a block, so there shouldn't be any file size difference.
We use separate drawings because they are much easier to manage and update when necessary, and when inserted don't carry any additional baggage of unused blocks.
Blocks do not have to be exploded to edit, check out REFEDIT on blocks.
-
I used to use the "separate drawing" method, prior to Dynamic Blocks. Now, I find DB's are much easier to manage if they're already in a "container" drawings, and since we only access them through Tool Palettes, they don't bring any unwanted content with them.
When the blocks were in their own drawings, they were stored in exploded form to avoid "block references itself" errors.
-
Not to steal this thread, but....
We (almost) ALWAYS explode our blocks, once we insert them into our dwgs. - I thought everyone does this, (?).
By exploding the blocks, we remove any references within them once we purge, right?
Is this a 'normal' process to explode the purge every block?
craigr
-
Not to steal this thread, but....
We (almost) ALWAYS explode our blocks, once we insert them into our dwgs. - I thought everyone does this, (?).
By exploding the blocks, we remove any references within them once we purge, right?
Is this a 'normal' process to explode the purge every block?
craigr
This would certainly increase the file size!
If a drawing containeda block with 6 instances of a block reference inserted in the drawing. Then what is being stored is one block definition (non-graphical) and 6 (graphical) block references (pointers to block def.). When you explode these (block references) back to entities each entity is stored individually, so this would increase the drawing approx. 6 times, even after purging.
-
This would certainly increase the file size!
If a drawing contained a block with 6 instances of a block reference inserted in the drawing. Then what is being stored is one block definition (non-graphical) and 6 (graphical) block references (pointers to block def.). When you explode these (block references) back to entities each entity is stored individually, so this would increase the drawing approx. 6 times, even after purging.
That makes sense to me but we explode blocks that almost never have any blocks within blocks, so when we explode then purge them, they are no more than lines, circles, etc....
I have always figured this was a much 'cleaner' way of finishing the dwgs - nothing hidden to cause problems later.
No problems, right?
What would be the advantage of not explode / purging them?
craigr
-
We (almost) ALWAYS explode our blocks, once we insert them into our dwgs. - I thought everyone does this, (?).
AAACK!!! May a herd of Yaks infest your lower colon.
Using blocks for repetative elements reduces files size, sometimes drastically. A steel structure using unit blocks for the members may have a file size of 300K while the same file using 3Dsolids would be 2MB.
Blocks, even without attributes, can be extracted into intelligent data. The unit blocks in the structure above can be extracted, using LISP, and the length of each beam (X scale) and height of each col (Z scale) can be multiplied by the member weight (number following the "x" in the block name) for a quick material weight.
Adding attributes to blocks allows a much greater degree of intelligence extraction using tools like EATTEXT. Simple things like furniture can carry a manufacturer's part number, location, phone number, or anything else you may need. A manager can then extract the data and find any piece of furniture he likes.
Simple blocks like section marks or detail bubbles can be quickly redefined to match client standards, and updated over an entire directory in minutes.
Exploding blocks around here will get you "knee-capped"
-
AAACK!!! May a herd of Yaks infest your lower colon.
:-D :-D
Exploding blocks around here will get you "knee-capped"
Around here you will be fed to the lobsters, ... ALIVE
But hey; all kidding aside, to each their own. :-)
-
We never explode, because we generate an excell spread sheet from the block attributes....but I tried a test and a drawing with over 750 small blocks (3-10 attributes each) the drawing went from 925 KB to 1.08 MB after exploding......
-
I would think it depends somewhat on your industry. Ours is HVAC Building Automation & most of our dwgs consist of line / wiring diagrams. Blocks that we do not explode are as examples - a wiring terminal, Ground symbol, small things like that. Everything else is exploded.
Our blocks do not contain any of the info you folks mentioned, hence no info to lose.
The parts we supply for a system have the part number listed on the dwg with the devices.
For our Security projects, I have gotten floor plans from outside firms that have furniture in them that are blocks which may have some of these attributes you mention - I explode all of them & put them on a layer called 'furniture'. I have no use for this other info. I basically just want a layer for all of the furniture & such, then isolate it from the walls, doors, windows & such - which go on a layer called 'floorplan'. I often have to do quite a bit of exploding & erasing to get everything isolated & on their own layers. The floor plans are copied into our title block and retained by us. The info that I am purging would not be missed for our purposes.
craigr
-
I have gotten floor plans from outside firms that have furniture in them that are blocks which may have some of these attributes you mention - I explode all of them & put them on a layer called 'furniture'.
You've probably doubled or tripled the file size
-
For our Security projects, I have gotten floor plans from outside firms that have furniture in them that are blocks which may have some of these attributes you mention - I explode all of them & put them on a layer called 'furniture'. I have no use for this other info. I basically just want a layer for all of the furniture & such, then isolate it from the walls, doors, windows & such - which go on a layer called 'floorplan'. I often have to do quite a bit of exploding & erasing to get everything isolated & on their own layers. The floor plans are copied into our title block and retained by us. The info that I am purging would not be missed for our purposes.
I'd use layer translator instead of exploding. Just select all the offending layers and put it on furn or layer zero.
Dan
-
the beauty of blocks is you only store the information to create them once (a block def) and all graphical info you see on screen ref's this definition to draw it, all it needs is the rotation and scales to be stored in the dwg file.
for example a block definition may look like this -
line y,y,z,...(info about line)
circle ...
etc, may be many many more graphical items.
a ref would look something like this -
blockref -
insert name (used to reference the definition)
insert position
rotation
scale x,y,z.
(perhaps a little more info)
so when it comes down to it, it doesn't need to store 'all' the graphical object info in the dwg file for every block, only the bare essentials to insert and place the block whose info is stored in the one definition.
If you explode them it will create graphical entities for each block ref and they will need to all be stored as simple objects thus increasing file size.
-
Not sure if I read that right. I've never used Lt.
In Lt you can make a block but not refedit it?
-
I basically just want a layer for all of the furniture & such, then isolate it from the walls, doors, windows & such - which go on a layer called 'floorplan'. I often have to do quite a bit of exploding & erasing to get everything isolated & on their own layers.
hmmm... another thought. Make the layer called furniture, then change all the blocks to that layer. When you FREEZE layer furniture all the blocks will disappear regardless of the layers used to create them. Seems like it would not only save file size, but the time you spend fiddling with exploding and changing the blocks.
-
Believe me, I'm not trying to be arguementitive.
There are a few things I am after for our dwgs -
I want to put all 'little importance' items on a layer then make the color 254, so it is very light when printed. And be able to turn them off without searching through a bunch of layers with names that usually make little sense to me.
The remaining walls, doors & Windows are on thier own layer that gets a color of 253, so when printed is a little darker than the furniture.
Security devices, labels & notes are all on a layer called 'security', which is blue and when printed is a good contrast from the rest.
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
Believe me, I have learned much from your suggestions, but I still don't see how I can get what I want without exploding them.
I understand that I am spending alot of time doing what I call 'cleaning up' the dwgs when I get them - it sucks spending soooo much time doing this. I just don't see a better way of doing it.
craigr
-
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
That's why you use the "refedit" that Randy suggested. Since a block is defined independent of all it's insertions, you change the master definition and you update all blocks... that is unless you exploded them all, because then they aren't blocks anymore.
Say I place 80 objects randomly in a room. Those objects' value are equal to the variable "X". Whatever "X" is... is what these are, and that's all they know. The goal of these objects is to always equal "X"... they have no other goal in life.
Now... "X" is defined in this little book that keeps the definition of all Variables. We'll call it the "Blockle" In the Blockle, it says that "X" has an insertion point. It also has a 2" circle drawn with the center at the insertion point.
Therefore all these 80 objects are a circle placed at the same point as their insertion point (the blue grip every block has that places itself when you insert the block or select it)
Well the great Blockle author came around and says "Gee... I want "X" to now have a verticle line bisecting it... " and it was good. So now the definition of "X" is that there will be a 2" circle located with the center at the insertion point... and it will have a vertical line bisecting it.
Well all these objects out there know that the definition has changed as soon as it happened... so they want to be "X"... so they now have a line bisecting the circle... automagically. And it was good.
You are the Great Blockle Author. The Blockle is the list of Block Definitions. The objects are all instances of the block. The Author redefines the definition via the "refedit" command.
Does that clear it up, because it was kinda fun to write... :oops:
-
Believe me, I'm not trying to be arguementitive.
There are a few things I am after for our dwgs -
I want to put all 'little importance' items on a layer then make the color 254, so it is very light when printed. And be able to turn them off without searching through a bunch of layers with names that usually make little sense to me.
The remaining walls, doors & Windows are on thier own layer that gets a color of 253, so when printed is a little darker than the furniture.
Security devices, labels & notes are all on a layer called 'security', which is blue and when printed is a good contrast from the rest.
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
Believe me, I have learned much from your suggestions, but I still don't see how I can get what I want without exploding them.
I understand that I am spending alot of time doing what I call 'cleaning up' the dwgs when I get them - it sucks spending soooo much time doing this. I just don't see a better way of doing it.
craigr
Get lt extender (or a full version of Acad) and use MP's routine:
http://www.theswamp.org/index.php?topic=3020.msg37746#msg37746
Ron
-
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
That's why you use the "refedit" that Randy suggested. Since a block is defined independent of all it's ..........
Does that clear it up, because it was kinda fun to write... :oops:
Josh, this only works if you have multiples of identical blocks, right?, in my case, I can have upto 900 separate, unique blocks....
PS: layer translate is the BOMB!!!!
-
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
That's why you use the "refedit" that Randy suggested. Since a block is defined independent of all it's ..........
Does that clear it up, because it was kinda fun to write... :oops:
Josh, this only works if you have multiples of identical blocks, right?, in my case, I can have upto 900 separate, unique blocks....
PS: layer translate is the BOMB!!!!
well... yea.... quite so. Your case is a PITA I bet ;) although I understand the benefit.
-
I guess I stole this topic.
For that, I apologize. I didn't think that Blocks were that involved, if I had known, I would have started a new topic.
It's obvious I still have alot to learn. I will look into your many suggestions.
Thanks,
craigr
-
Believe me, I'm not trying to be arguementitive.
No problem there
I really wish there was a way to change the layers of the blocks AND their colors. As I understand it, to change the properties of the block, you have to click on EVERY single block. If I could redefine the properties in mass, I wouldn't have to explode them.
Believe me, I have learned much from your suggestions, but I still don't see how I can get what I want without exploding them.
I understand that I am spending alot of time doing what I call 'cleaning up' the dwgs when I get them - it sucks spending soooo much time doing this. I just don't see a better way of doing it.
You might want to talk a look at using the Quick Select (qselect) command in conjunction with your editing of the blocks (however you decide to edit them). It allows you to select in mass quantiles with similar criteria so you can process them. Very powerful tool. I get files from our other cad package that has nested blocks within blocks (sometimes 4 to 5 nested levels deep). I have to use combination of explode and refedit and the qselect command makes it go faster.
But then again you might be already using the qselect command, but I figured I would mention it.
-
Craig, I have a question. Are these blocks the same blocks from the same supplier each time? If so, refedit them once to layer 0, and save that file as a "base.dwg", erasing everything in the file except the block definitions. Then when you get the next file from that supplier, just open the "base.dwg" saveas a new file and insert the suppliers file into the the new file. The blocks in the supplier's file will take on the definition in the "base.dwg".
Just a thought anyway.
-
There are a few things I am after for our dwgs -
I want to put all 'little importance' items on a layer then make the color 254, so it is very light when printed. And be able to turn them off without searching through a bunch of layers with names that usually make little sense to me.
The remaining walls, doors & Windows are on thier own layer that gets a color of 253, so when printed is a little darker than the furniture.
Security devices, labels & notes are all on a layer called 'security', which is blue and when printed is a good contrast from the rest.
Lets think outside the box (block).
You are receiving cad files from us Archies and whomever are you not?
Have you thought of xref'ing these files in, changing all the colors of the layers of the xref (in the host drawing) to the colors you choice and then drawing your devices over top. My MEP consultant does this with our drawings. what is nice about this is when we have to coordinate between offices. and he uses mickystation. He Xrefs my stuff in and I xref his stuff in for back check purposes. Xref'ing keeps stuff separate in a nice containers and gives you layer control.
-
Not sure if I read that right. I've never used Lt.
In Lt you can make a block but not refedit it?
That's correct, although LT 2007 and up have the Block Editor, which gives you a lot of the same functionality.
-
Our floor plans DO come from anywhere we can beg them from. They are even sometimes bitmaps or paper copies that are hand carried in for me to scan!! - I can't do much with these.
But, 75% of the time, I do get actual CAD floor plans that I 'destroy' to get what I want. I thought that if there were standard layers, I could automate what I do, but since there are no 'standard' layers between companies, I do what I have to do.
XRefs - I looked into XRef -ing dwgs, and have a couple of sites set up like that, but our project leaders never got the hang of it, and didn't like it. 'We don't need to do all of that'.
From the sounds of it, I need to learn this 'refedit' thingy. It may solve alot of my problems.
craigr
-
Oops - I guess I Refedit won't work for me. - I missed that.
Would one of the 'add-on' programs have it?
I am still trying to justify an 'add-on' prg to management.
craigr
-
..... 'We don't need to do all of that'
I wish the CDC would put this disease at the top of their list. We have it flare up from time to time in my office too.
-
I understand that they are doing their job - watching the dollars. They are usually pretty good at getting me what I need, much better than some companies - from what I have read on this board.
I believe that if I can show that the add-ons would save enough money, they would spring for it. But, it's difficult to explain to them as they know even less about CAD than I do - in fact I run circles around them - (I know, that's sad).
We have a very long history of using CAD, but only its most basic functions. There was NO automation / macros before I started! I have come up with MANY with the help of you folks. I am constantly looking for ways to cut down on my time. Hence, my wanting to learn about blocks & every other question I have asked on here.
The plus side of this is that they have noticed the time savings from my macros, standards developement, etc.... Which may be why I am hesitant to ask for the money for the add-ons. I am not sure how much money they would save for our basic CAD needs.
On our Security dwgs - If I spend 3 hrs on a very small job, half of that is usually spent bringing the CAD from others into our title block and 'cleaning it up'. I think this is a horrible waste of time. I would LOVE to automate this.
craigr
-
Thanks Kate.
I'm not agreeing with the kneecapping anymore.
-
WOW!! :-o
That's a lot of info on blocks, it'll take me a week to digest it all!! :lmao: And even then I may not!!
I appreciate all of the great info, and it looks to me that for now, my best bet is to do as my friend said and save the blocks as their own dwg file so I can change them when I need to.
However, this brings up another thing that I noticed. If I change one of these dwg files, the blocks I've already inserted into a drawing where I'm using them are not updated, at least not until I delete one of them and replace it, or just add another, then I am asked if I want to update the definition. I say yes and it changes all of the blocks...is there a way to do this without replacing or adding one...in other words, is there a command to update the blocks somewhere? (I would just look in the help, but I'm at work not at home where I have the drawings!)
Thanks again to all who chimed in, and thanks to you Craig for taking this in a direction that helped me learn a ton!! :-D
-
You should not need to erase any of the blocks you are redefining. All you need is to begin inserting the new block and when prompted if you wish to redefine, choose yes and all blocks with that name receive the new definition. You can then just cancel the insert or if you wish, complete the insert somewhere and erase it, but that is extra and unecessary steps.
-
If I change one of these dwg files, the blocks I've already inserted into a drawing where I'm using them are not updated, at least not until I delete one of them and replace it, or just add another, then I am asked if I want to update the definition. I say yes and it changes all of the blocks...is there a way to do this without replacing or adding one...in other words, is there a command to update the blocks somewhere? (I would just look in the help, but I'm at work not at home where I have the drawings!)
I'm sure someone around here has a block updater routine (CAB??). However in lieu of that, insert it just as you said above, except after you've said "yes" to redefining the block, don't place the block, just hit the escape key. The block will have been redefined.
<dang, dino beat me to it again>
-
I'm sure someone around here has a block updater routine (CAB??). However in lieu of that, insert it just as you said above, except after you've said "yes" to redefining the block, don't place the block, just hit the escape key. The block will have been redefined.
<dang, dino beat me to it again>
Here (http://www.theswamp.org/index.php?topic=6810.msg83740#msg83740) is one.