What's your beginners tip?
-
@tig said:
See http://www.smustard.com/script/SelectionMemory for free, or the inexpensive v2 there too...
thanks a bunch. that will work.
-
I did not read all of the other tips, but I'm sure someone has said this already, and it's worth repeating.
"SAVE OFTEN" ...SketchUP has a tendency to "crash often".
-
This may have been said before but please SAVE YOUR MODEL FREQUENTLY, Every few minutes, just take the 10 seconds out to push [CTRL] + [S]. you don't want to be in the position where your model bug splats (crashes) and you lose all your work. I learned this the hard way, but i hope you won't.
Another tip: Learn by watching. Look at other users' models, especially their WIP (Work in progress. AKA an unfinished model) models. You might see how they do it. You might learn a new trick or a new way to use a certain tool. Don't be afraid to ask someone how they made a model either. -
@unknownuser said:
Look at other users' models, especially their WIP (Work in progress. AKA an unfinished model) models. You might see how they do it. You might learn a new trick or a new way to use a certain tool. Don't be afraid to ask someone how they made a model either.
Agreed. But about saving, that's why SU has autosave
Though it is always good to be on the safe side. -
Save is cool but when your file becomes some big that is less cool
-
The problem with autosave is exactly what Pilou says. When your file becomes rather large and it "hits" you the worst time, it can take forever. Imagine you are in an operation with a heavy ruby plugin (which can take a while in itself, too) and then in the meanwhile comes the autosave.
I basically turned it off and got used to saving the model at every possible and "logical" time (mainly after some major steps). However it would be a risky thing to suggest this...
-
Ah, well, being a rather lazy person, I never make models so large tht it takes a long time to save 'em, but I get your point.
-
@gaieus said:
The problem with autosave is exactly what Pilou says. When your file becomes rather large and it "hits" you the worst time, it can take forever. Imagine you are in an operation with a heavy ruby plugin (which can take a while in itself, too) and then in the meanwhile comes the autosave.
I basically turned it off and got used to saving the model at every possible and "logical" time (mainly after some major steps). However it would be a risky thing to suggest this...
ditto
although when you forget to manual save.....................
happens more than i will admit lol -
I did the same. And now I am hard wired to hit save before any plugin operation
-
Yes indeed, I also forget to save sometimes and once it did happen to me that I splatted and lost some hours of work - exactly the worst timing as it was a tight deadline.
That's why I also wrote it would be risky to advise this...
-
Model somthing that is accesable to you, your house, letterbox, a piece of furniture etc model and delete/build until you get it right. Emulating the real world will force you to uncover new aspects of SU you may not otherwise stumble upon.
-
As a beginner myself (more aptly called a newb) I can say with confidence that the first
thing someone should do is RTFM (read the 'manual'). It's long, boring, etc etc et al ad nauseum but more than worth it.Just the toolbar and tools sections are worth the time and effort. There are so many context and tool altering keyboard
key combination's for the various tools that it'll really be worth your while. -
It's a mistake I made by the bushel when I started using SU.
I made multiple copies of grouped geometry instead of making them components.
My rule now is if ANYTHING is more than 1 occurrence, it is a component.So when I did convert one of the group copies to an identical component (select, explode, make comp) I was faced with the task of replacing the same groups with that same component, in the exact same position. When I moved the comp into the groups position, it would snap into position and I was not able to select only the group to delete it.
I tried several things with variable success, until I stumbled on this simple solution.
Before you move the comp onto the group, edit the group and add a single line from any point to outside its existing bounding box, then close it.
Move the comp into place. It will still snap in place.Now simply click on the extended group line and you can easily select and delete only the group.
There is also a TT tool that replaces identical groups with an identical comp, but it does the whole drawing, which is not what you may want, and if even the slightest difference is present, or the group is nested, it will not replace the group.
-
If you want to completely get rid of Sketchups face shading while keeping the colour, then turn on the shadows and set the light slider to 0 and the dark slider to 80. I found this useful for making the RBG value of the faces to match the one in the material editor.
-
Hey
thanks for your beginning tips. They are really helpful for me to develop my interest and expertise and most importantly understanding the tool.
-
Regarding non-coplanar problems. If the attempt to face keeps failing, even after you have drawn and redrawn all the lines multiple times, here are a few ways to help find the culprit. Not always successful, but usually eliminates the co-linear line problem.
If your triangular, rectangular or multi-line entity fails to face, you need to revert it to triangles. Draw lines from vertex to vertex, being careful to let SU tell you that you are using endpoints only, before you click. Zooming in helps, but if you encounter the clipping plane getting in close, then turn OFF perspective. (Camera; Perspective) Then you can get in real close.
Another problem with drawing lines is where you are trying to draw a line (from A to B) that is not quite on-axis, and SU will tell you it is "constrained by line at point". Simply ESC and draw the line from B to A. Usually works. This is probably the biggest culprit creating co-linear lines.
Now let's say you do have a suspected co-linear line. You delete both of them, and redraw a single line, but the face still won't form. The culprit here is the tiny line segment formed at the vertex after the 2 co-linear lines were drawn. It can be nearly invisible unless you zoom in real close, and that takes time. So rather than go looking at all 3 corners for it, try this.
Turn ON the Entity Info dialog box. You also need to turn on hidden lines. With the object viewed such that there are no other endpoints behind the suspect vertex, use the left to right select in a very small box just covering the vertex, and not any whole lines. If Entity Info says no selection, the problem is elsewhere. If the Entity Info says there is a very tiny line (or more than 1 line) then there is the problem. Just delete all of them. Do all 3 vertexes. This may deface other adjacent faces, but now you can delete a line and redraw it endpoint to endpoint to reform the faces.
Note that this will NOT find a vertex gap, but deleting and redrawing lines at endpoints cleared of tiny fragments will cure the gaps.
Another culprit is trying to face on a curve or arc line. Since there is no real endpoint within the length of the line, placing other lines to form a face may not be right on, and this can create a tiny fragment or gap to throw off the face. Just explode the curve/arc and that will help finding the true endpoints on the curve.
When you have the whole object faced with triangles, you can start deleting the internal lines making all those triangles to get back to your multi-line face. If a face disappears doing this, then that triangle was not co-planar with the rest of the face, and you have to redraw that section in a co-planar manor. Or, just UNDO the deleted line, and make it soft and smooth (Entity Info box again).
And with all that, I still spent 2 1/2 hours yesterday trying to get a triangle to face, and ending up deleting the whole section around it and starting over.
-
@jgb said:
I made multiple copies of grouped geometry instead of making them components.
My rule now is if ANYTHING is more than 1 occurrence, it is a component.So when I did convert one of the group copies to an identical component (select, explode, make comp) I was faced with the task of replacing the same groups with that same component, in the exact same position. When I moved the comp into the groups position, it would snap into position and I was not able to select only the group to delete it.
I tried several things with variable success, until I stumbled on this simple solution.
Before you move the comp onto the group, edit the group and add a single line from any point to outside its existing bounding box, then close it.
Move the comp into place. It will still snap in place.Now simply click on the extended group line and you can easily select and delete only the group.
I only make components, not groups. Then I never need to waste time making the conversion from group to component later. If I want to replace one component with another, it's automatic so I don't have to add any extra geometry to be able to get hold of one later. I've never found a case where a group is preferable to a component. Even when there's only one occurrence of something, I make it a component.
-
@dave r said:
I only make components, not groups.
No argument with that. I almost always end up converting a group to a comp when finished with its design.
The main reason I keep groups is to cut down on the extensive list of comps (many of which would be sets of comps) I would create during the design. I use the groupings to segregate sections of geometry to avoid unwanted merging, then explode the group once done, but within a comp. I also usually group several comps into 1 coherent set, where I may have a slight difference in the set makeup in various places in the model. Groups and comps both have their advantages and should be used accordingly. -
"Groups and comps both have their advantages and should be used accordingly."
Yup. Just never have found a need for groups even when nesting components. Depends upon ones workflow and style and it is good the SketchUp is so flexible.
-
@Dave R:
One of your best beginner's tips only recently mentioned elsewhere, is the use of an upscaled component instance to do intricate modeling which would be very difficult (empty faces, unresponsive inferencing, zoom clipping)with the object at its intended scale. Since scaling an instance does not affect the other instances, this allows great freedom. All editing done on the larger instance is immediately reflected in the smaller. When finished, simply delete the larger instance.
This one really stuck with me. I have pretty much stopped whining about the scale problem.
Salute
Advertisement