Could someone explain what the slide bar in the below pic does?
SOLIDWORKS Model Property Tools – Part1 – Measure | CATI Tech Notes
Edit: Here's a screen shot excerpt:
Dan Pihlaja wrote: ...These settings do not determine the accuracy of the calculated results; rather they affect the amount of effort/precision that goes into computing the results....
Dan Pihlaja wrote:
...These settings do not determine the accuracy of the calculated results; rather they affect the amount of effort/precision that goes into computing the results....
I must admit that I can be rather dense at times. I just can't wrap my mind around this statement. " amount of effort/precision that goes into computing the results....".
We are talking about, generally, very high-end computers performing millions of calculations per minute.
We are also talking about a measurement between 2 "things".
Why is this slider even there?
Why isn't the default "Give me the most accurate distance possible period now and every time"?
It's not "measurement between 2 things". It's mass, area, etc. of potentially highly complex shapes. If all you make are nice blocky/cylindery things, the accuracy is probably going to be dead nuts, even on the lowest setting. If you make swoopy things with splines and surfaces etc, SW has to work harder to calculate things like volume, surface area, etc. This is what must be balanced with performance.
Josh, please tell me why anyone would want an "approximation" instead of dead-on accurate Every time?
That is the best question!
Well, as someone else once said (I am not sure who it was, or I would give them credit), if you take your windows calculator and type in 4, Square root, - 2, see what you get. ((Square root of 4) minus 2).
There is a certain amount of rounding that goes into the equation. Now, on the windows calculator, we can't control the "accuracy" of the calculation. That is done in the background of windows programming and is already set for us.
In Solidworks, I think that we can control it to a small extent. By either using the lower accuracy or the higher accuracy (not sure about the background programming here at all, or what the difference is between higher and lower accuracy when it comes to algorithms).
My assumption (I haven't tested this out), is that a sensor uses this same setting. So if you were trying to fall into a range of, lets say, volume on an extremely complex part (or parts if you are in an assembly), you wouldn't want the computer chugging for 5 minutes every time it tries to calculate the volume. It gives you a rounded number that is close. Then, once you are inside the range that you want, move it to the highest setting and get the super accurate number. This allows you to get close to the number that you want without wasting time with tedious, mind numbing waiting for the computer to catch up.
This is one scenario in which I could see it being used. Although I have not fallen into this scenario yet.
Granted, I rarely change it. And on those times where I HAVE changed it, I have not seen a difference between the Lower and Higher settings.
Dan Pihlaja wrote: I have not seen a difference between the Lower and Higher settings.
I have not seen a difference between the Lower and Higher settings.
So, the slider switch is really just a Fidget Slider?
I have personally and too often have a problem with the number of decimal places SW works to.
SW works to only 8 decimal places.
KeyCreator works to 14 decimal places.
When exporting some DXF files from SW and importing into KC, there are often gaps (teny tiny gaps) between the individual pieces of geometry. Try programming that and the CAM barfs.
Fortunately I have a workflow that "fixes" most of it.
I just did a quick little test. I have a fairly large assembly open and I think that the slider that is in the measure tool also affects the mass properties calculations (which you can get to using the "options..." button inside mass properties).
With slider all the way toward Lower (faster):
This calculation took about 5 seconds.
With the slider all the way toward Higher (slower):
This calculation took more than 30 seconds.
Volume and surface area are close (within 11 cubic millimeters of each other with respect to volume), but not exactly the same.
For me, most of the time, when I am dealing with volume, it has to do with not going over a certain volume because I am going to submerge the fixture. So the faster calculation would be good enough in this case, as long as I remembered to see what the slower calculation showed at the end.
Same thing with mass. I generally have a limit that I can't go over. I generally give my self a large factor of error to account to variations with regards to amount of coolant left over inside the assembly, differences in density from purchased materials vs. calculated materials, etc... so again, the faster calculation would generally be good.
Dan Pihlaja wrote: if you take your windows calculator and type in 4, Square root, - 2, see what you get. ((Square root of 4) minus 2). -8.1648465955514287168521180122928e-39
if you take your windows calculator and type in 4, Square root, - 2, see what you get. ((Square root of 4) minus 2).
Alright math nerds. Someone please explain this .
I tried ((Square root of 4) minus 2) on the built-in Windows Calculator and got this...
My HP-30b RPN calculator gave me the same answer as I calculated in my mind... Zero. (2 calculators on my Android phone also correctly gave zero)
My mind is blown...
Binary math in the computer is imprecise. Not all values can be represented using the number of bits available. Just as in decimal math, you cannot write the precise value of one divided by three, certain values cannot be fully represented in the computer. A basic explanation can be found here.
Retrieving data ...