Simulation speed

[[Je...]]

In offline mode is there any way to speed up the simulation mode when executing the program? Our programs can sometimes take over 2 hours to run in real time.

[[Ca...]]

Your simulation isn't faster than normal run time? Mine runs 60-80 minute programs in seconds.

[[De...]]

On my computer if I slide the slider all the way to the right it goes too fast to watch, a program with a 2 hour runtime in the real world runs in a few minutes at full speed on simulation.

Left = Slower
Right = Faster

[[Pa...]]

Jeff,
I don't know the answer to your question, but I was wondering something similar myself.
Does offline simulation mode work well; does it actually catch collisions, problems etc?
And it works more or less in real time then? Whew, some of our programs are 6 hours!
Any info by anyone is appreciated.
Thx, Paul S

[[De...]]

It works pretty well. I know there are some times where it doesn't give a perfect representation of how the part will run on the machine, but for the most part its pretty solid.

I find it sometimes misrepresents how things will run when using patterns or on the rotary, when programming without the rotary I don't think I have seen a problem.

As for collisions, yes it is very good for finding problems before putting a program on the machine, that said, as I indicated above there are times when its not 100 percent accurate, so we still always test our programs at slow speed.

And no it does not work in 'real time' it goes much faster than it does on the machine if you turn the speed up as I indicated above.

[[Je...]]

I rarely run simulation mode (or other programmer is the primary offline programmer) so this was basically a rookie question. I was running under probe simulation mode in Planner rather than just executing the program. Under simulation mode the program runs at about 2x. Without stylus sim activated it just calculates and only takes about 5% of the time.

Stylus simulation mode in Planner does catch obvious collisions but it is definitely not perfect. My main programming tasks consist of "proving" the program and troubleshooting errors from a program created in an offline seat by our dedicated programmer. It is still tedious and requires 100% attention. It is very good for visualizing probing strategies and scan speeds.

[[To...]]

Though you can control how fast your program runs through simulation, it is important to remember that collision detection works by detecting when the stylus/probe model occupies the same graphic voxel as the part and/or fixture model at the same time. If you have the speed set too high, this condition may not be detected during the screen refresh cycles. So, if you're seriously trying to find collisions in a complex program, run at slower speeds, at least in the areas where you are concerned about collision risks.

[[No...]]

I'm 100% with Tom!
Simulation was NOT, I repeat, was NOT made to accurately depict scanning speeds and run times. The one and only focus here is collision detection and the higher speed settings mean less accuracy!
The running speed of a simulation also depends heavily on the complexity of the model geometry around the probe. You can observe this when the probe dives into a complex part on the lowest speed setting: simulation becomes much slower as Calypso has to check a lot of faces for collisions.
So if you want maximum accuracy, run it at the lowest speed. It still won't catch each and every collision (for several simulation-related reasons), but most of the standard ones. Sometimes it will take ridiculous paths to the part (e.g. through the granite 🤣 ), but with some experience you learn to ignore that...

[[Ri...]]

Please sign in to view this quote.

Then why isn't the software running at true speeds to depict the BEST accuracy?
There are features that do not run at the same time frame regardless of speed settings. Cones in particular.

[[No...]]

I'm not sure I understand what you mean. When I say accuracy I mean the accuracy of collision detection. I don't know if the lowest slider setting has the best possible accuracy, but anything higher is less accurate.
I don't see how running at true speed (as fast as a real machine?) would help to improve collision detection in any way.

Collisions are detected by calculating intersections of CAD faces (probe and machine parts vs. workpiece) in the current position of the virtual CMM. The running speed you observe on the screen is just an indirect consequence of how many intermediate movement steps are calculated between two machine positions (the number of steps is what the slider sets) and how fast your computer can calculate the collision intersections in each step. You could say the slider sets the "resolution" of the simulation, not the speed.

Btw, the speed setting in the CNC start window has no effect on simulation speed.

[[Ri...]]

Please sign in to view this quote.

Never mind.... 🤠

[[Ma...]]

Please sign in to view this quote.

How do you get to the simulation screen?

[[SH...]]

Simulation is a licenced option, to use planner you should have the licence.
first select Planner then measurement plan simulation.

[[Mi...]]

I pretty much run everything in sim. Most of my job is on an off line station so I run sim often while programming a job to check operations and then at the end to check the whole program. I'll run it on the faster setting to get through already checked operations or large groups of similar operations like multiple threaded holes. I don't use collision detect much, but I'll slow to about 3/4 or so to watch new operations and I'll slow it way down to watch a tricky operation to make sure probes clear and do what I want.

[[Am...]]

Does anyone know if its possible to set the simulation runtime close to what the actual runtime is? I was hoping to use the simulation for a project to work on the efficiency of some programs and see how much faster the programs an run with a little bit of different paths and order but the simulation was useless for this as it just goes so fast.