Can Pano2vr handle 1500 panoramas project?

Q&A about the latest versions
Post Reply
User avatar
Hrvoje
Posts: 144
Joined: Sat Feb 28, 2009 7:34 pm
Location: Zagreb, Croatia
Contact:

Thu Mar 25, 2021 1:03 pm

Hello

I wonder is there a limit of how much panoramas can pano2vr pro current version handle?
Question is related from both aspects:
- as panorama creator
- as panorama user

I have a request for about 1300-1500, panoramas so I have a concerns if P2VR can handle it.

My biggest projects so far had no more than 100 12500 pix panoramas
and memory consumption was about 5-8 gigs of ram with occasional program crash.

This new project would have 5 x 15K pixels panoramas, 20 x 12K ones and rest would be 7,5K ones made by Ricoh Theta Z1

I have fairly strong machine:
CPU: Amd Ryzen 9 3900X
SSD: fast m.2
GPU: radeon RX 580
RAM 64 GB

Also, will such big project have big impact on loading times and user memory consumption?

Thanks,
Regards
User avatar
Hopki
Gnome
Posts: 11269
Joined: Thu Jan 10, 2008 3:16 pm
Location: Layer de la Haye, Essex UK
Contact:

Thu Mar 25, 2021 5:14 pm

Hi,
The number of nodes is not the issue, so far we have created a project with 100,000 nodes.
The requirements were lots of free space on the image cache location/scratch drive, this also needs to be fast, so a fast SSD.
And 16GB Ram and up would also help speed things up.
Multi-Core CPUs also speed things up, I have a 10 core iMac Pro and when processing all 10 cores are used.

That said Pano2VR has settings to aid large projects.
Under the Settings/Preferences and Advanced tab you will see:

1.png
1.png (12.35 KiB) Viewed 92 times

If you deselect Load all images immediately Pano2VR will open only loading the first node.
As you select another node then that will get loaded.
With Unload Images to save memory selected, that is what it will do, as deselect a node, after a time it will unload from memory so you can work on each node without the fear of filling up the memory.

However, the best way which does not include changing any settings is to convert your input images to TIFF's (tif) using Pano2VR.
This will convert them to Tiled TIFF's.
This process adds about a third more to the file size, that's on the understanding the input images were TIFFs to start with.
The best way to do this is to set up a Droplet with a Transformation output set to Equirectanular, size Auto and file type TIFF.

What this does is embeds all the multiresolution data into the images, this means Pano2VR does not need to load and remap the input images meaning the project will open very, very fast.
Once the first output is made, then again subsequent outputs will be fast.

You can also stagger the output, If you select to create the output, let Pano2VR run for a bit, then you can Cancel.
Then Create the output again, it will carry on from where it left off.

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/
User avatar
Hrvoje
Posts: 144
Joined: Sat Feb 28, 2009 7:34 pm
Location: Zagreb, Croatia
Contact:

Fri Mar 26, 2021 5:42 pm

Hello Hopki

I'm glad to see that P2VR can accommodate such number of nodes.
Thanks for the detailed and comprehensive answer.
Suggested settings and workflow will help me with other projects with 100 nodes projects as well.

Best Regards
Post Reply