Mar 28 2026 APP 2.0.0-beta40 will be released in 7 days.
It did take a long time to have the work finished on this and it will have a major performance boost of 30-50% over 2.0.0-beta39 from calibration to integration. We extensively optimized many critical parts of APP. All has been tested to guarantee correct optimizations. Drizzle and image resampling is much faster for instance, those modules have been completely rewritten. Much less memory usage. LNC 2.0 will be released which works much better and faster than LNC in it's current state. And more, all will be added to the release notes in the coming weeks...
Update on the 2.0.0 release & the full manual
We are getting close to the 2.0.0 stable release and the full manual. The manual will soon become available on the website and also in PDF format. Both versions will be identical and once released, will start to follow the APP release cycle and thus will stay up-to-date to the latest APP version.
Once 2.0.0 is released, the price for APP will increase. Owner's license holders will not need to pay an upgrade fee to use 2.0.0, neither do Renter's license holders.
Hey team APP,
I just recently started some integration tests on a huge project im working on, a 250 pannel mosaic with much overlapp so it actually less than 250 actuall pannels but still ALOT. Registration works perfectly with 0,59'' RMS but now comes the part that scratches my head a bit. When doing the LNC calculation APP is using little to no RAM of my 128gb available and 1-4%max on my 16 core CPU. I believe that this calculation would be done in a few min rather then multiple hours when it use all the components to its max potential...
Im currently doing a 0.2x scale integration test with 4th degree LNC and 2 iteration and its calculating for like 6hrs now. I did this before and I know that it will take about 8hrs to complete the calculation of 1 iteration.
I dont know the insides or the limitations of the algorhythm behind LNC but it would be cool when this could be a bit faster in the future 😀
Best Regards
Jakob
Hi Jakob @jkbsahner,
Wow, that sounds like an awesome project on which you are working 🙂 !!
Let me explain about the LNC algorithm and it's implementation in it's current state:
LNC currently is a linear least squares regression. Our current implementation does not allow it to use multiple cpu cores allthough we are investigating if it would be possible. So this explains why it is only using 1 single cpu core. I think that it's performance will be better if you would disable CPU hyperthreading, but i have never really tested it. Theoretically it should do. The amount of RAM needed depends on the actual field of view size in pixels and the amount of overlap areas between panels. In addition the higher levels of LNC require more memory since more parameters need to be solved.
The good news is that I have already started work on a new LNC algorithm, I would call it LNC 2.0 🙂 I am reformulating the mathematical issue to make it smarter and I will try to make it a non-linear Least Squares algorithm which I know i can implement using all cpu cores ! The LNC 2.0 should work better, be much faster! and use less ram as well. It will not be included in 2.0.0 stable which has the highest priority now. But it migh be there for APP 2.1 or 2.2 I think.
Mabula
Hi Mabula,
thank you for the detailed answer!
Yeah this project is huge and the full resolution would be about 2 gpx. As this is way to much data to proccess in the raw form I downscaled the data and now the full fov is about 800mpx. I also rented a server with the mentioned specks above and 2x 4TB ssd's to integrate this data over the next few days/weeks. I need 3TB to integrate it in full resolution 😀
I hope this works and Im looking forward for the future of APP with more optimizations and new features!
Jakob
