Version 26.2 is slower than previous versions

p_coelho97

Hi

I was using GlobalMapper 25 for Lidar Point Cloud Classification and updated to version 26.2.

This new version is slower and a lot more unstable. It freezes panning, zooming, selecting points, classifying points, etc. It basically freezes with every action with the same quantity of data I was using in v25.

Sometimes it even crashes suddenly and I have to load all my work again.

These problems never occurred in version 25. Any way to optimize or solve this issue?

My PC specs are:

Intel Xeon W-2295 3.30 GHz

224 Gb of RAM

1 TB SSD

GeForce RTX 3060 12GB

Thanks in advance

bmg_mike

Hi,

Can you try getting the latest Global Mapper v26.2 daily build from https://www.bluemarblegeo.com/products/global-mapper-daily-builds.php and see if that improves the behavior? We have fixed a number of issues since the initial 26.2.0 release and even have an official 26.2.1 hotfix release available.

If you still have issues, can you email geohelp@bluemarblegeo.com with issue about the crash and data that can reproduce the performance issues?

Thanks,

Mike

Global Mapper Guru

p_coelho97

Hi Mike,

Thank you for the answer. I was using the daily build b110725, and having the same issues.

I am now using 25.1, that for the work im doing now, is faster.

Thanks again

gcyx

The grid algorithm for vegetation classification in version 26.2 cannot classify vegetation and can only use the default Max Likelihood algorithm. This is also a new bug in version 26.2.

Andrew Tarr

Hi Mike, further to this thread, aside from 26.2 feeling a bit slower, I've also found that the vector display filters for areas, lines, and points aren't working with the S57 data I've loaded. Everything is always rendered regardless of what is checked. I've confirmed this with both the standard 26.2 and the latest daily build.

26.1 still works well. I just checked, and it loads (or at least renders on screen) a S57 file significantly faster than 26.2.

Thanks for all your work and for a great product!

bmg_mike

Hi Andrew,

I have looked into the vector display filter issue and was able to reproduce and fix it. The daily build tomorrow will have that fixed.

We have also fixed a number of speed issues in 26.2 in the last few weeks, so you might check the speed again in the fixed build tomorrow.

If you still have speed issues with S57 in the latest daily build, can you provide one that exhibits this issue? I checked some S57 files and they all rendered almost instantaneously, so I couldn't tell any difference. We did update a lot of the S57 symbology to be vector-based rather than raster images so they would scale and rotate better, so perhaps there is an issue there for some data that I'm not seeing.

Thanks,

Mike

Global Mapper Guru

bmg_mike

Hi gcyx,

I have been able to reproduce unexpectedly changed behavior in the Lidar vegetation classification between 26.1 and 26.2 and have created issue GM-21594 to look into that issue.

Thanks,

Mike

Global Mapper Guru

gcyx

Hi Mike. I suggest changing the default unit for ground point classification in V26 from point spacing back to meter. This is because using meter as the unit is faster than using point spacing. The default unit for ground point classification in V25 is meter.

Andrew Tarr

Thanks Mike. The S57s are looking good and filtering properly. Great work. :)

bmg_mike

Hi gcyx,

The Lidar vegetation classification in 26.2 should now work the same as 26.1 again. You might want to do a Restore Defaults on the classification dialog so you get back to the default settings (like 2.0m above ground for min height rather than 0.0).

Tomorrow's daily build at https://www.bluemarblegeo.com/products/global-mapper-daily-builds.php should have this fixed.

We have another issue open (GM-20956) for the meters vs. point spacing thing for ground classification.

Thanks,

Mike

Global Mapper Guru