Hi!
I´ve just read a recent post, which in part answers this question and on the other hand makes me ask:
Why choose one or other image type as input?
What is the difference between equi, strips, crosses, etc?
Are there any pros and cons of one over the other and which one should be used in which cases?
Thanks, Martin
Input image type
-
Hopki
- Gnome
- Posts: 13025
- Joined: Thu Jan 10, 2008 3:16 pm
- Location: Layer de la Haye, Essex UK
- Contact:
Hi Martin,
The different projection are for software that will use them.
Example HDR Studio only uses image cross.
Regards,
Hopki
The different projection are for software that will use them.
Example HDR Studio only uses image cross.
Regards,
Hopki
Garden Gnome Support
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
Hi, Hopki!
Thanks for your reply. So that´s the main difference?
But not the only one, because you´ve also mentioned in another post that it is convenient to use cubefaces or a strip for multiresolution. Right? Are there any other reasons?
Thnaks again!
Martin
Thanks for your reply. So that´s the main difference?
But not the only one, because you´ve also mentioned in another post that it is convenient to use cubefaces or a strip for multiresolution. Right? Are there any other reasons?
Thnaks again!
Martin
-
360Texas
- Moderator
- Posts: 3684
- Joined: Sat Sep 09, 2006 6:06 pm
- Location: Fort Worth, Texas USA
- Contact:
You asked "Why choose one or other image type as input?
Well this is a bit of an educated guessing game. You tell me what input flavor you use and I will tell you mine.
So I will add my flavor regardless of the output type and use.
Canon 5d/ NN-R1, Sigma 8mm. We shoot HDR so 4 images x 3 (-1,0,+1)= 12 images around 7° tilt up [no nadir or zenith required] . RAW to TIFF 16bit color depth, Convert to 16bit blended 4 single TIF images, Stitch and output type 16bit TIF, Import to Pano2vr 16bit TIF equi-rectangular image(s). Each pano could be 180+ mb. A lot of work taking high quality images. AND we never use .jpg's.
Reduce input load time, use max INPUT of 5 pano's + pano2vr Pro .... just use a super fast 64 bit CPU computer with a lot of RAM. Your mileage will vary depending on your workflow !
Time to share, what INPUT do you use ??
Well this is a bit of an educated guessing game. You tell me what input flavor you use and I will tell you mine.
So I will add my flavor regardless of the output type and use.
Canon 5d/ NN-R1, Sigma 8mm. We shoot HDR so 4 images x 3 (-1,0,+1)= 12 images around 7° tilt up [no nadir or zenith required] . RAW to TIFF 16bit color depth, Convert to 16bit blended 4 single TIF images, Stitch and output type 16bit TIF, Import to Pano2vr 16bit TIF equi-rectangular image(s). Each pano could be 180+ mb. A lot of work taking high quality images. AND we never use .jpg's.
Reduce input load time, use max INPUT of 5 pano's + pano2vr Pro .... just use a super fast 64 bit CPU computer with a lot of RAM. Your mileage will vary depending on your workflow !
Time to share, what INPUT do you use ??
Hi, Dave!
I always use an equirectangular image as input. But in another thread Hopki mentioned why it is convenient to use Cubefaces as an input image. Therefore my post asking what input type is better to use in which cases, regardless of the process to generate that input type.
I shoot with my 10.5 on a NN4 head. Every 60º and 1 or 2 zenith shots, depending on the zenith. Usually bracketed -2/0/+2.
Pocessed in Dxo PRo to eliminate the CA. Output as DNG. Then adjustments in ACR. Converted in 16bit tifs.
Fused in Enfuse. Stitch in PtGui and then imported in Pano2VR.
Could you please explain what you meant with:
"Reduce input load time, use max INPUT of 5 pano's + pano2vr Pro .... "
Does it mean to reduce the size of the final swf?
Thanks, Martin
I always use an equirectangular image as input. But in another thread Hopki mentioned why it is convenient to use Cubefaces as an input image. Therefore my post asking what input type is better to use in which cases, regardless of the process to generate that input type.
I shoot with my 10.5 on a NN4 head. Every 60º and 1 or 2 zenith shots, depending on the zenith. Usually bracketed -2/0/+2.
Pocessed in Dxo PRo to eliminate the CA. Output as DNG. Then adjustments in ACR. Converted in 16bit tifs.
Fused in Enfuse. Stitch in PtGui and then imported in Pano2VR.
Could you please explain what you meant with:
"Reduce input load time, use max INPUT of 5 pano's + pano2vr Pro .... "
Does it mean to reduce the size of the final swf?
Thanks, Martin
-
Hopki
- Gnome
- Posts: 13025
- Joined: Thu Jan 10, 2008 3:16 pm
- Location: Layer de la Haye, Essex UK
- Contact:
Hi,
A strip is a cubic image, and requires less processing to turn to cubes as its already a cubic image.
But Pano2VR has to still work with it twice, open the strip then extract the separate cubes. For normal size images, non Giga Pixel this all happens quiet quickly but for giga pixel not having to process the strip also saves a lot of processing time.
So in conclusion:
Strip = the convenience of one image file and much faster to process then Equirec
Cube faces = six images per pano, faster to process than strip.
Regards,
Hopki
A strip is a cubic image, and requires less processing to turn to cubes as its already a cubic image.
But Pano2VR has to still work with it twice, open the strip then extract the separate cubes. For normal size images, non Giga Pixel this all happens quiet quickly but for giga pixel not having to process the strip also saves a lot of processing time.
So in conclusion:
Strip = the convenience of one image file and much faster to process then Equirec
Cube faces = six images per pano, faster to process than strip.
Regards,
Hopki
Garden Gnome Support
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
If you send an e-mail to support please send a link to the forum post for reference.
support@ggnome.com
https://ggnome.com/wiki/documentation/
-
360Texas
- Moderator
- Posts: 3684
- Joined: Sat Sep 09, 2006 6:06 pm
- Location: Fort Worth, Texas USA
- Contact:
All I am talking about is "opening" a previous multi panorama project... the Pro version has to load up ALL the pano's so you can make the minor changes.
I think there is something called 'Experimental: [ ] Smart loading' option found in the FILE | Settings| Images | where once you are satisfied that you have built your project contents with ALL the multires tiles and toolbar graphics in the /OUTPUT folder you can 'create all' again... without having to do generate the tiles again.
ALL of the above is off topic and really does not address your questions.
The different cube face configurations 'Cross, horiz and vertical' etc I think were designed for use in different external viewers. Typical tile sequences were cube faces 1 left ,2 back ,3 front, 4 right ,5 top,6 bottom order. In other different viewer need the display tile sequence to be 4,3,2,1, 6 and 5. And maybe Quicktime viewer order sequence was 1,2,3,4,6,5. Pano2vr being a converter program allowed you to preselect the cube faces to be ordered so you could use different viewers. Another javascript viewer from ARCSOFT is called FSPViwer64.exe.
http://www.fsoft.it/panorama/FSPViewer.htm
I think there is something called 'Experimental: [ ] Smart loading' option found in the FILE | Settings| Images | where once you are satisfied that you have built your project contents with ALL the multires tiles and toolbar graphics in the /OUTPUT folder you can 'create all' again... without having to do generate the tiles again.
ALL of the above is off topic and really does not address your questions.
The different cube face configurations 'Cross, horiz and vertical' etc I think were designed for use in different external viewers. Typical tile sequences were cube faces 1 left ,2 back ,3 front, 4 right ,5 top,6 bottom order. In other different viewer need the display tile sequence to be 4,3,2,1, 6 and 5. And maybe Quicktime viewer order sequence was 1,2,3,4,6,5. Pano2vr being a converter program allowed you to preselect the cube faces to be ordered so you could use different viewers. Another javascript viewer from ARCSOFT is called FSPViwer64.exe.
http://www.fsoft.it/panorama/FSPViewer.htm