Difference between revisions of "Nona GPU stitching reports"

From PanoTools.org Wiki
Jump to: navigation, search
(Other video cards: I fixed the hang with Intel 965, so update its entry.)
(Linux)
 
(2 intermediate revisions by 2 users not shown)
Line 4: Line 4:
  
 
* '''IMPORTANT: before reporting, make sure you were using nona-gpu. When stitching from Hugin, set it in the Preferences -> Stitching tab -> activate the checkbox "Use GPU for remapping". When stitching from the command line, add the -g switch to the nona command line.
 
* '''IMPORTANT: before reporting, make sure you were using nona-gpu. When stitching from Hugin, set it in the Preferences -> Stitching tab -> activate the checkbox "Use GPU for remapping". When stitching from the command line, add the -g switch to the nona command line.
 +
* recent version of nona will fall back to CPU stitching for projects that the GPU can't stitch (yet).  Read the output of nona to make sure it has not fallen back on CPU stitching even if told to use the GPU.
 
* identify your video card. http://www.techarp.com/showarticle.aspx?artno=88 should help. select your GPU manufacturer's full list and note the exact name. The more precise you are, the better it is for everybody. Note that this is not your video card manufacturer / brand. Today most video card manufacturers get their GPU from either nVidia or ATI.
 
* identify your video card. http://www.techarp.com/showarticle.aspx?artno=88 should help. select your GPU manufacturer's full list and note the exact name. The more precise you are, the better it is for everybody. Note that this is not your video card manufacturer / brand. Today most video card manufacturers get their GPU from either nVidia or ATI.
 
* identify the exact driver you were using. Do so '''before''' upgrading driver, collecting reports for older drivers is useful. If you do upgrade later, make a second report for the upgraded driver.
 
* identify the exact driver you were using. Do so '''before''' upgrading driver, collecting reports for older drivers is useful. If you do upgrade later, make a second report for the upgraded driver.
Line 9: Line 10:
 
* based on your video card and operating system, select the appropriate section to edit below.
 
* based on your video card and operating system, select the appropriate section to edit below.
 
* Edit the section and append your report at the '''bottom''' of the list (but above the lines terminating the table).
 
* Edit the section and append your report at the '''bottom''' of the list (but above the lines terminating the table).
 +
* While reporting speed / performance of your specific attempt is interesting data, don't draw conclusion from it.  The improvement of GPU vs. CPU depends amongst others also on the type of CPU that is in your system and the type of project.  Performance data is meaningful only if measured consistently and against the same project.  This is beyond the scope of this page.
 
* The developers may contact you to follow up on your report.
 
* The developers may contact you to follow up on your report.
  
Line 118: Line 120:
 
1min 7s for a 6x10MP images equirectangular panorama, 1min 47s without GPU
 
1min 7s for a 6x10MP images equirectangular panorama, 1min 47s without GPU
 
<!--END TEMPLATE-->
 
<!--END TEMPLATE-->
 +
 +
<!--START TEMPLATE-->
 +
|-valign="top"
 +
! style="border-top: thin dotted #333333;" |
 +
GeForce GTX 275
 +
! style="border-top: thin dotted #333333;" |
 +
nvidia 195.36.24
 +
| style="border-top: thin dotted #333333;" |
 +
Ubuntu 10.10 x64, Core2Quad Q9550 @ 2.83GHz, 16GB RAM
 +
| style="border-top: thin dotted #333333;" |
 +
2mins 36s for 29 18mpix images to rectilinear, sinc256, 14:41 without GPU
 +
<!--END TEMPLATE-->
 +
 +
|-valign="top"
 +
! style="border-top: thin dotted #333333;" |
 +
GeForce 8800 GT
 +
! style="border-top: thin dotted #333333;" |
 +
nVidia 195.36.24
 +
| style="border-top: thin dotted #333333;" |
 +
Ubuntu 10.04 x64, Core2Duo T8100 @2.16GHz, 3GB RAM
 +
| style="border-top: thin dotted #333333;" |
 +
Blazing fast!
 +
 +
 +
 +
 +
  
 
<!-- TERMINATING TABLE -->
 
<!-- TERMINATING TABLE -->

Latest revision as of 19:09, 4 March 2011

The purpose of this page is to collect individual experience reports and learn to know what hardware / driver combination work with nona-GPU.

Individual Reports

  • IMPORTANT: before reporting, make sure you were using nona-gpu. When stitching from Hugin, set it in the Preferences -> Stitching tab -> activate the checkbox "Use GPU for remapping". When stitching from the command line, add the -g switch to the nona command line.
  • recent version of nona will fall back to CPU stitching for projects that the GPU can't stitch (yet). Read the output of nona to make sure it has not fallen back on CPU stitching even if told to use the GPU.
  • identify your video card. http://www.techarp.com/showarticle.aspx?artno=88 should help. select your GPU manufacturer's full list and note the exact name. The more precise you are, the better it is for everybody. Note that this is not your video card manufacturer / brand. Today most video card manufacturers get their GPU from either nVidia or ATI.
  • identify the exact driver you were using. Do so before upgrading driver, collecting reports for older drivers is useful. If you do upgrade later, make a second report for the upgraded driver.
  • copy the template below.
  • based on your video card and operating system, select the appropriate section to edit below.
  • Edit the section and append your report at the bottom of the list (but above the lines terminating the table).
  • While reporting speed / performance of your specific attempt is interesting data, don't draw conclusion from it. The improvement of GPU vs. CPU depends amongst others also on the type of CPU that is in your system and the type of project. Performance data is meaningful only if measured consistently and against the same project. This is beyond the scope of this page.
  • The developers may contact you to follow up on your report.

Thank you!

<!--START TEMPLATE-->
|-valign="top" 
! style="border-top: thin dotted #333333;" |
[ENTER YOUR GPU DETAILS]
! style="border-top: thin dotted #333333;" |
[ENTER DRIVER VERSION]
| style="border-top: thin dotted #333333;" |
[ENTER SYSTEM DETAILS]
| style="border-top: thin dotted #333333;" |
[ENTER COMMENTS]
<!--END TEMPLATE-->

nVidia

This section is for nVidia powered video cards only.

Linux

GPU Detail

Driver Version

System Detail

Comments

Riva TNT2 Pro 32GB

71.86.11

Ubuntu 9.04 32bit AMD Athlon XP 3200+ 2GB

this system is obviously outdated for the task

GeForce 9600GT

xorg-x11-drv-nvidia-185.18.36

Fedora 11 x86_64, Phenom 9550, 8GB

This combination works OK

GeForce 9600GT

vesa

Fedora 11 x86_64, Phenom 9550, 8GB

This combination does not work

GeForce 9600GT

nv

Fedora 11 x86_64, Phenom 9550, 8GB

This combination does not work

GeForce 9600GT

nouveau

Fedora 11 x86_64, Phenom 9550, 8GB

This combination does not work

GeForce 8400 GS

nvidia 256.53-1

Archlinux x64, C2D T5450 @1.66GHz, 2Go RAM

1min 7s for a 6x10MP images equirectangular panorama, 1min 47s without GPU

GeForce GTX 275

nvidia 195.36.24

Ubuntu 10.10 x64, Core2Quad Q9550 @ 2.83GHz, 16GB RAM

2mins 36s for 29 18mpix images to rectilinear, sinc256, 14:41 without GPU

GeForce 8800 GT

nVidia 195.36.24

Ubuntu 10.04 x64, Core2Duo T8100 @2.16GHz, 3GB RAM

Blazing fast!




Windows

GPU Detail

Driver Version

System Detail

Comments

Nvidia 8800 GT (Factory overclocked)

190.62

Win 7 RTM x64, 4 GB RAM, 2.7 GHz C2D E6400 (overclocked)

Stitched 360x180 equirect. from fisheye images. No problems.

Nvidia 8800 GT (default clocks)

185.81

Win 7 RTM x64, 4 GB RAM, 2.7 GHz C2D E6400 (overclocked)

Remapped some images. No problems.

GTX 260 Core 216 (stock frequencies)

190.62

Win XP SP3, Sempron 3400+ @ 2.5 GHz (socket 754), 2 GB RAM

A regular equirectangular panorama that took Nona 257 seconds to process with the CPU only took 22 seconds using the GPU!

GTX 260 Core 216 (stock frequencies)

195.62

Win 7 x64, Core i5 750 @ 2.67 GHz (LGA 1154), 4 GB RAM

GPU is 10-50% faster then CPU, depends on kind of interpolation used. Remapping took 20 sec with GPU and 22 sec with CPU, bicubic interpolation used.

OSX

GPU Detail

Driver Version

System Detail

Comments

NVIDIA GeForce 7300 GT

Unknown

MacOSX 10.6.1 Snow Leopard

Crashes immediately. KERN_PROTECTION_FAILURE at 0x0000000000000000

NVIDIA GeForce 7300 GT

Unknown

MacOSX 10.6.3 Snow Leopard

Dies with "GL error: Framebuffer incomplete, incomplete attachment" Hugin 2010.1.0-svn-5156

other

GPU Detail

Driver Version

System Detail

Comments

AMD/ATI

This section is for AMD/ATI video cards only

Linux

GPU Detail

Driver Version

System Detail

Comments

Radeon X550

xserver-xorg-video-radeon 6.12.6

Debian lenny 64 bit on Athlon64 x2 3800+, 6GB

"nona: this graphics card lacks the necessary extensions for -g." (Hugin stitch output)

Windows

GPU Detail

Driver Version

System Detail

Comments

Radeon 4850

8.64-090714a1-085212C-ATI

WIN XP-sp3, C2D-E8400

Worked ok, small performance increase (74s CPU, 68s GPU)

Radeon 4870 1GB (RV770)

atiumdag 8.753.0.0 (Catalyst 10.7)

Win 7 (6.1, Build 7600), Intel Q6600 ~3.5GHz, 4GB RAM

Nona GPU remapping functional. No significant improvement in performance, maybe even slightly slower than without GPU acceleration. Average GPU usage close to 0%, max ~5%. Using latest stable driver for GPU as of this posting.


OSX

GPU Detail

Driver Version

System Detail

Comments

other

GPU Detail

Driver Version

System Detail

Comments

ATI Radeon X 1600

01.00.158

OS X Snow Leopard (10.6.1)

nona does quit unexpectedly, crash log is here

Other video cards

our current knowledge is that nona-gpu only works with nVidia and AMD/ATI video cards. If you find out something different, please let us know here.

GPU Detail

Driver Version

System Detail

Comments

Intel 965

unknown

Ubuntu 9.04

Does not support the OpenGL extensions nona GPU uses, so it always does CPU stitching.