Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - PolarNick

Pages: 1 2 [3] 4 5
General / Re: New alignment parameter for background suppression
« on: April 21, 2020, 05:59:44 PM »
Yes, it works. I scanned an object on turntable from three angles (i.e. three circles of photos) and for each angle I just additionally took photo without object (i.e. photo of clean turntable + background). Then I masked these three photos fully and applied masks to tie points.

Nice! :)

I don't know for sure what leads to deletion of /etc/X11/xorg.conf, it can be something like system distributive upgrade or packages installation or packages update.

You surely need to backup this file somewhere if it happens again. It seems that another way to prevent this deletion is to move this file from /etc/X11/xorg.conf to something like /usr/share/X11/xorg.conf.d/my_xorg.conf (you can just try it, or try to find another way from man).

Sorry, I meant not /usr/share/doc/xserver-xorg-video-intel-hwe-16.04/xorg.conf, but old xorg.conf from your old message (I also attached it).

Can you please save xorg.conf (attached to this message, it has ~64 lines) to /etc/X11/xorg.conf and try again?

Is output of lspci | egrep -h "VGA|3D controller" still includes 03:00.0 VGA compatible controller: NVIDIA Corporation GF119 [NVS 310] (rev a1) ?

If yes - you can try to use your old xorg.conf file (from your message) - just save it as /etc/X11/xorg.conf and reboot.

If it doesn't help - make sure that /etc/X11/xorg.conf wasn't automatically changed after reboot and provide Xorg.0.log

You Xorg.0.log has multiple errors (Ctrl+F for string (EE)). May be your xorg.conf file was changed? (for example because of some kind of packages installs/updates)

Can you please post your xorg.conf again?

Bug Reports / Re: spurious fork
« on: July 05, 2019, 01:29:50 PM »
This is problem from sklearn:



Code: [Select]
import os
os.environ['JOBLIB_MULTIPROCESSING'] = '0'
import sklearn

General / Re: I have an intel NUC iCore 8 using Iris Intel Video
« on: April 02, 2019, 05:58:14 PM »
What is your output of the clinfo command?

General / Re: GTX or RTX
« on: March 25, 2019, 01:57:18 PM »
2x1080ti will be faster than single 2080ti. But of course 2x2080ti or 3x2080ti will be even faster but less cost effective :)

General / Re: External GPU (eGPU) experience with AgiSoft Photoscan
« on: April 28, 2018, 08:16:34 PM »
Input data for computations on GPUs are generated on CPU, so when you have very fast eGPU - your CPU can became the bottleneck (especially when computations are in low quality - because CPU work part to GPU work part ratio will be bad).

When slow GPU added (AMD 560) - it adds pressure in input data preparation part, i.e. CPU tries to feed both GPUs and so fast eGPU will be used less.

P.S. this is just typical reason for the described behavior, and if my speculation is correct - you will see speedup with AMD used in case of higher depth maps quality. (because of better CPU work part to GPU work part ratio)

General / Re: New alignment parameter for background suppression
« on: March 14, 2018, 02:06:57 PM »
You can read about masking here.

Also note that magic wand can be used in incremental style with multiple clicking while holding Ctrl.

General / Re: Tesla P100 and virtualgl
« on: February 14, 2018, 01:39:29 PM »
M40/M60/K80/V100 driver version is: 387.34,it's FREE.

P100 requires special drivers included with NV Grid 5.1. This driver needs to pay. Free drivers do not have OpenGL features.
Interesting. I didn't know that. I was thinking that V100 is a successor of P100, and so I expected driver politics to be the same.
cloud-scripts were updated and tested on P100 PCIE. So if you will install clean Ubuntu 16.04 and follow instructions as is - everything should just work. If it will not work - it will be interesting to look at your cloud-scripts/configure.log, cloud-scripts/start_vnc_server.log, /etc/X11/xorg.conf, /var/log/Xorg.0.log and output of nvidia-smi execution.

General / Re: AWS remote cloud server setup
« on: February 13, 2018, 03:00:57 PM »
You can checkout cloud-scripts - they will configure Ubuntu 16.04 on p2/p3/g2/g3 EC2 instance with full OpenGL GUI support (via TurboVNC).

Other Languages / Re: Работа с VMware
« on: January 26, 2018, 05:17:15 PM »
Я тестировал под KVM, с VMware опыта не было.

Полезные статьи на которые я опирался (до настройки и запуска KVM все так же и под VMware):


1) Требуется поддержка IOMMU процессором и материнской платой (IOMMU должна быть включена в BIOS).

2) GPU прокидывается только вместе со всеми устройствами той же IOMMU-группы. (подробнее)

3) Прокинутая GPU должна использоваться эксклюзивно (нельзя использовать одну GPU и в host, и в guest, как и нельзя использовать в нескольких guest). Поэтому host-система обязана игнорировать прокидываемые GPU (см. тут про "vfio-pci" и "Isolation of the guest GPU").

4) Нужен дисплей подключенный к GPU напрямую. Большинство мониторов умеют переключаться между несколькими кабелями, т.ч. это небольшая проблема. Об этом так же пишут тут в комментариях по ключевому слову "monitor". Кроме того в случае если нужна поддержка полноценного OpenGL с удаленным доступом и без подключения дисплея - можно сконфигурировать гостевую систему этими скриптами (поверх TurboVNC можно поднять noVNC сервер, и подключаться с любого компьютера через браузер).

Заметки для NVidia Geforce:

5) Драйвер Geforce откажется работать если заметит виртуализацию. Он проверяет сигнатуру гипервизора в cpuid. Поэтому в случае KVM нужно выставить kvm_hidden=on (см. devtalk, askubuntu, detecting virtualization).

6) Не создавайте датацентров с Geforce, т.к. в датацентрах нельзя использовать драйвера для Geforce (см. EULA):
No Datacenter Deployment. The SOFTWARE is not licensed for datacenter deployment, except that blockchain processing in a datacenter is permitted.

Четкого термина что такое по мнению NVidia датацентр пока нет, когда появится - напишу на devtalk.

Bug Reports / Re: Function createDifferenceMask not working correctly?
« on: January 19, 2018, 02:02:30 PM »
You need to reshape numpy array to (height X width X numberOfChannels):

Code: [Select]
import cv2
import numpy as np

chunk =
camera = chunk.cameras[0]
image =

img = np.fromstring(image.tostring(), dtype=np.uint8)
assert (len(img) == image.height * image.width *
img = img.reshape(image.height, image.width,

# Please, note that OpenCV assume that image is BGR, while PhotoScan provides RGB, you can fix this with numpy:
#     img = img[:, :, ::-1]
# Or with OpenCV:
#     img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)

# You can show image in window with OpenCV:
#     cv2.imshow("window with img", img)
#     cv2.destroyWindow("window with img")

I think it isn't possible to avoid PhotoScan Python API overhead. Because even if you somehow could apply Cython (for example) - you can't get rid of Python layer in PhotoScan API.

Pages: 1 2 [3] 4 5