Forum

Author Topic: Metashape dies while building TiledModel from dense cloud  (Read 2612 times)

koriel-angelswing

  • Newbie
  • *
  • Posts: 22
    • View Profile
Metashape dies while building TiledModel from dense cloud
« on: July 01, 2020, 06:30:16 PM »
I am trying to build TiledModel from dense cloud. (FYI I am running Metashape's method in python flask and I ran flask with development mode)

But for some reasons, metashape's TiledModel build procedure is suspended and dead without any errors.

This might be caused by CPU or Memory?

My computer has

model name   : Intel(R) Core(TM) i7-8086K CPU @ 4.00GHz
stepping   : 10
microcode   : 0x6e
cpu MHz      : 4299.995
cache size   : 12288 KB

and

MemTotal:       32809936 kB

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14813
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #1 on: July 01, 2020, 06:51:32 PM »
Hello koriel-angelswing,

How many points are there in the dense cloud?

Do you have the processing log for the failed operation?
Best regards,
Alexey Pasumansky,
Agisoft LLC

koriel-angelswing

  • Newbie
  • *
  • Posts: 22
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #2 on: July 01, 2020, 07:17:13 PM »
Hello koriel-angelswing,

How many points are there in the dense cloud?

Do you have the processing log for the failed operation?

point cloud size: 738742419 points

and

There was no specific logs for failing but this was the log when it died.

Linear system solved in 286.787s
Got Iso-value in 18.1315s
Iso-Value -1.25426
99816962 faces extracted in 191.279s
decimating mesh (99685672 -> 230231)... done in 63.7897 sec
subdividing model... done in 0.049283 sec
extracting tiles... done in 0.006782 sec
saved block in 0.164512 sec
loaded tiled model data in 0.009422 sec
loaded block index in 0.000318 sec
Grid size: 4070 x 4070 x 4070
Tree depth: 12

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14813
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #3 on: July 04, 2020, 11:48:14 PM »
Hello koriel-angelswing,

Have you checked the memory consumption during this stage?

What is the tile size that you are using?
Best regards,
Alexey Pasumansky,
Agisoft LLC

koriel-angelswing

  • Newbie
  • *
  • Posts: 22
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #4 on: July 06, 2020, 01:59:06 PM »
Hello koriel-angelswing,

Have you checked the memory consumption during this stage?

What is the tile size that you are using?

I did not check the memory consumption but it should be high at maximum I guess.

I used 256 tile_size, the default value.

koriel-angelswing

  • Newbie
  • *
  • Posts: 22
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #5 on: August 05, 2020, 10:38:48 AM »
Hello koriel-angelswing,

Have you checked the memory consumption during this stage?

What is the tile size that you are using?

No pings on this? haha

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14813
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #6 on: August 09, 2020, 01:35:00 AM »
Hello koriel-angelswing,

The issue could be related to the memory issues, therefore you can check the RAM modules using Memtest86 utility in single- and multi-threaded modes, just for case there are any failures.

If the processing is using RAM at max, you can extend the disk swap.
Best regards,
Alexey Pasumansky,
Agisoft LLC

koriel-angelswing

  • Newbie
  • *
  • Posts: 22
    • View Profile
Re: Metashape dies while building TiledModel from dense cloud
« Reply #7 on: August 10, 2020, 08:03:40 AM »
Hello koriel-angelswing,

The issue could be related to the memory issues, therefore you can check the RAM modules using Memtest86 utility in single- and multi-threaded modes, just for case there are any failures.

If the processing is using RAM at max, you can extend the disk swap.

The computer has been using 32GB RAM so I extended the swap to 32GB as well.

Thank you very much.