Messages

1

General / Re: Disappointing performance results on new MacBook Pro M1 Max

« on: April 10, 2022, 08:24:11 PM »

thxs Jenkinsm!  good to know - is it possible to disable / enable it  in a python script?

2

General / Re: Disappointing performance results on new MacBook Pro M1 Max

« on: April 07, 2022, 03:17:25 PM »

ja totally agree. I am always looking at full systems so the GPU performance might be throttled down by other components in some of the systems. But the M1Max comparison is valid never-less since the design (and speed) of the components should be (other that the additional GPU cores) the same. I get also exactly the same time stamps on both M1Max systems for tasks that use CPU only. Best, Sören

3

General / Re: Disappointing performance results on new MacBook Pro M1 Max

« on: April 05, 2022, 05:53:54 PM »

Just a tiny update to the M1Max discussion. My M1Max 10C32C arrived and I compared it directly to the M1Max 10C24C version. I found the 32C version to be another  5-10% faster depending on the Metashape process that is used. The 32C makes no difference for processes that do not use the GPU support.  Overall this was a bit disappointing. I was expecting a bit more in the range of 20% improvement but thats not what you get. A nice find however is that initial match&align is much faster when you activate "Use CPU when performing GPU acc. processing" in the Metashape GPU preferences. This speed improvement is likely a M1Max design side effect and likely also applies to the Mi1Max 24C version. With this pref setup the M1Max is even faster than a dedicated GPU like the RTX2080S or RTX3080Ti! I wonder what this means for the M1Ultra performance. However "dense point cloud processing" is slowed when this CPU/GPU pref is set.

Updated article: http://jenacopterlabs.de/?p=1572
best wishes! Sören
 

4

General / Re: Disappointing performance results on new MacBook Pro M1 Max

« on: March 18, 2022, 11:25:11 PM »

just to reactivate this thread: I am pretty much blown away by the speed of Agisoft Metashape (1.8.2) on a M1Max - it seems to be as fast as a fully specced workstation with a full RTX2080Super GPU configuration.  My findings are here:
http://jenacopterlabs.de/?p=1572&preview_id=1572&preview_nonce=bebd602a10&preview=true&_thumbnail_id=1579

kind regards! Sören Hese

5

General / Re: Phantom 4 multispectral sunlight sensor

« on: June 02, 2020, 05:40:48 PM »

Ok thxs - yes will do. We collected a a full 5 channel tif dataset for 200 ha but I will send a ZIP with 100 images in multi cam system layout at import and the micasense panel tifs. Kind regards! Sören

6

General / Re: Phantom 4 Multispectral

« on: May 31, 2020, 07:55:50 PM »

We would also like to know if support for the P4M reflectance calculation with a reference panel photo is possible soon. So far the results are totally screwed when a Micasense panel (50%) is used.  Uncalibrated processing to a dense point cloud and orthomosaic works however fine. But without the calibration option the idea of using the P4M for research purpose investigating plant properties is useless. Manual raw correction with the 50% panel might work. But Pix4D supports radiometric correction of P4M data as far as I can see!

Kind regards! Sören Hese (Friedrich-Schiller-University Jena/Germany)

7

General / Re: Phantom 4 multispectral sunlight sensor

« on: May 31, 2020, 07:42:50 PM »

interesting topic.

I am also using a P4M and so far its doesnt seem to be possible to calibrate the P4M data with a Micasense grey reference target. The panel calibration values seem to be totally wrong. I guess that the sensor data isnt fully supported by Agisoft. Any news from Agisoft? Is there a date when he P4M will be supported for reflectance calibration. We are running a 1.6.2 license cluster. Kind regards! Sören Hese (Friedrich-Schiller-University Jena)

8

Python and Java API / Re: Python Scripts for Network Processing

« on: April 12, 2019, 09:08:02 AM »

excellent Alexey! thats really helpful!

9

Python and Java API / Python Scripts for Network Processing

« on: April 09, 2019, 01:52:38 PM »

Dear all,

we are setting up a cluster for network processing large UAV (mostly POI and Nadir) datasets in a cluster of Dell linux workstations. I can hardly find any code snippets for a python script that starts the processing on the job coordination server. I have three Nodes and one coordination server running already and sending jobs from the GUI is no problem but processing from a python routine would be much more efficient. Is there anywhere a FAQ available? the Python Metashape Manual is hardly showing the full "task" syntax here. Needed is the full processing from Alignment, Matching, DepthMaps, Dense Cloud, DSM , Ortho and las export.

This snippet seams to be a start:

import Metashape
client = Metashape.NetworkClient()
chunk = Metashape.app.document.chunk
task1 = Metashape.NetworkTask()
task1.name = ‚MatchPhotos‘
task1.chunks.append(chunk.key)
task1.downscale = 2
task1.params[‚keypoint_limit‘] = 40000
client = Metashape.NetworkClient()
path = „project.psx“
client.connect(‚XXX.XX.XX.XX‘)
batch_id = client.createBatch(path, [task1])
client.resumeBatch(batch_id)

Any ideas are welcome.  Kind regards! Sören