Better measurement tool

One of the most important tool in any CAD or modeling software is its measurement tool. Frankly, I find the tape measure in SketchUp very limited because it is a straight line measurement, which is the primary need for this tool. However, it would be nice for SketchUp to provide the closely related tool (that every other software has) that shows the individual measurements in all three axes.

1 Like

You don’t really state what advantage this would give you, or in what context you’d like to see x,y,z offsets. You may or may not be aware that you can specify exact xyz coordinate locations for a move destination or for the endpoint of a line.


1 Like

I have the ‘Query’ tool selected by the shortcut key combination [Shift]+[Q]. Instanly available to give me X, Y and Z for any endpoint related to the systems axes (not to changed drawing axes!)

To get x, y, z along changed axes apply the ‘Text’ tool on endpoints in current context.
If grouped, enter the group’s environment first.


Yes, I am aware of setting exact positions using [ ]. On a side note, when I have tried to use exact positions, like moving the top center of a cylinder to the origin, the first time that I do it the cylinder is obviously not at the origin but when I try it again a second time, it actually moves to the origin.

Back to the main topic, many times in my work, I am trying to integrate multiple CAD drawings from multiple sources. For example, a floor plan from an architect and framing 3D from the structural contractor. In these instances, you try to pick a reference point on both and set that reference point at the exact same 3D position. Sometimes, it is obvious that things are slightly off.

Another example: I know that one object should be directly above another object. I can use the various standard views to tell that something is off and to which direction.

In both examples, I am finding it very difficult to correct the positional relationship between two objects in SketchUp with the tape measure. With a measurement tool that shows the measurement in each individual axis, I could easily correct move the correct amount in the axis that is out.

For all your examples, I would use the move tool together with some inferences instead of the tape measure tool.

1 Like

I agree with @Cotty. It sounds like you need to review SketchUp’s inference system when using the move tool.

I am very familiar with the inference system, which is identical to AutoCAD that I have used for 20 years. I still find times, mostly in very dense and complex models, where doing a measure and move is easier because there are so many inference points to choose from that are very close together.

I think for a lo of cases the measured distance will nearly be an irrational number which will be difficult to enter for the movement?

Whatever point you are measuring to presents the exact same ease or difficulty of selection as the same point would present if used as an inferencing target, so I don’t see how measuring makes things any easier. I do, however, see that measuring then moving is twice the effort of simply moving and involves an unneccesary manual transcription of data and the attendant opportunity for input error.


I understand what you guys are saying. Let me explain in a step-by-step way. When you have a VERY large file (in number of objects, number of possible inference points, and sheer distance between points):

  1. Choose move command
  2. Choose object to move
  3. Choose reference point for the move
  4. Zoom way out, then way in to see and choose an inference point far away
  5. Zoom way out, then way in to see closer to the object being moved

OOPS. The zoom in the middle of a move command at step 5 LOSES the inference point that you just choose in step 4. So in many cases that I have experienced, it is easier to zoom in the middle of a 3D measurement, then do a move just because I cannot keep an inference point active.

It might be difficult to explain and we could also chalk it up to personal preference. I just think a 3D measurement tool is useful.

Maybe you can have a look at Fredos MoveAlong (part of FredoTools) and use a line as a move reference e.g.?

Um. how about this process:

  1. select the move tool
  2. zoom in on the object to move and click on the desired reference point
  3. zoom out until you can see the destination
  4. move the cursor over near the target and zoom back in (the object selected in 2 will have followed)
  5. zoom again as necessary until you can see the target point clearly and get an inference on it
  6. click the target point to finish the move

Hmmm. An arbitrary preference poorly explained. I suppose at this point it would be cruel to argue with you further.

However, it would seem that I sort of glossed over your initial post and failed to notice that your premise appears incorrect, namely, that the Tape Measure tool can measure only in a straight line (point-to-point might be a better way to describe it). But that’s not true; the Tape Measure tool can use inferencing like any other tool, so that by using inference locking (Shift + inference), you can constrain the measurement to the offset along any individual axis.

For example, the following picture shows a measurement between the two rectangles along the red axis, even though the true distance (along the hypoteneuse) is greater.


Thanks. Everyone has made helpful suggestions. I posted in “Feature Requests” because I do think it is useful to have a measurement tool like is available in other software packages that returns 4 values; straight point-to-point distance, X-distance, Y-distance, and Z-distance.



I am always in favor of letting the customer have it “their way”.

Having optional display of the x, y and z displacements, along with the direct-line measure, sounds very reasonable to me.

1 Like

A Measurement tool would not be complex to write as an extension. But since everyone might want slightly different options, why not invent a set of macros that can be embedded in the display string which allow people to select what gets displayed?


Just to test my understanding: if you were to first measure the inference point (from step5) then you could simply type in the coordinates (or work out the coordinates to type in). Correct?

Why not use the measurement tool to drop a guideline, then you can snap to it much easier without actually having to remember any numbers.

I can only think on one scenario that I would ever actually use a measurement tool: when someone is hovering over my shoulder and asks a dimension. Everything else is from survey, inference from specific geometry or using the “/000” or “x000” (or r-click and “divide”). Dimensions separate entities and are solely for presentation documents and normally added through layout.


I realize that there are other options as you mention (find the coordinates of the target point, drop a guide line, turn off layers to easily see target point, etc.).

My basic point is that my work involves very large, very dense models with thousands of components. As DanRathbun mentioned, I would just like the measurement tool to give 4 values because I have found that in my work, this type of measurement information is MUCH quicker than the other options when I am trying to get exact relationships between two components.

There were two main points i think you missed;

  1. to get the measurement you are looking for, you first must be looking at the target point and click it. If you are here anyway, then why not mark it rather than measure it? I initially said guideline, but it would probably be more useful to hit [L] and draw a reference line from this point (i would then hit [up arrow] and type 5000 [return]). It is then easy to find the end.

  2. just because this is the way you are used to working, it does not make it the most efficient way to work in this package. For you it might be quicker just now soley because it’s how you are used to working - i think the task could be done in less clicks/keystrokes with less mouse movement/zooming, but you would be slower because it would be different.

1 Like

… in saying that, there may be a benefit of setting a target point before selecting an object to move:

feature request have some sort of command or control key combo that would set a target - perhaps the move tool when nothing is selected and [ctrl] click, or enter a “target” mode through a r-click option, or maybe place a guideline point and be able to tag/flag it as a target.
then be able to select the object to be moved and it’s ‘source’ point, simply hit [return] (without moving or typing in any dims) and it would jump to that point.
Maybe that’s too simple? Perhaps you would need to type “t”, then you could define multiple targets: name/number each when they are created.

There must be a plugin/script to do this already?