Very slow integrati...
 
Share:
Notifications
Clear all

2022-05-29: APP 2.0.0-beta2 has been released !

Release notes

Download links per platform:

windows 2.0.0-beta2

macOS x86_64 2.0.0-beta2

macOS arm64 M1 2.0.0-beta2

Linux DEB 2.0.0-beta2

Linux RPM 2.0.0-beta2

Very slow integration speed


(@vbocc1)
Hydrogen Atom Customer
Joined: 2 years ago
Posts: 1
Topic starter  

Hi all. I am very new to astrophotography and am not very tech savvy so I apologize in advance for any ignorance. I have previously been using DSS to stack images on my PC and then move the TIF file to APP on my Mac to do some editing. I've been reading how DSS is outdated and APP can give better stacking results so I decided to try and use APP exclusively for both stacking and the initial processing of my images. 

I first tried stacking with APP on my 2020 MacBook Pro and analyze stars was taking hours to complete. It got to around 30% when the Mac started to get extremely hot and something inside it popped and the computer broke. The Mac was less than a month old and when Apple was done repairing it they said something about having to replace the heat sink and that whatever broke on the computer usually never happens. Regardless, the computer is still very slow when I process images in photoshop so I decided to just try and stack the images on my PC instead to be safe. 

I'm now using the same data on the PC that I tried on the Mac (142 light frames and 30 of each calibration frames) and I hit analyze stars at approximately 2am. Now almost 7 hours later the analyze stars is only 66% completed. The specs of my Mac and PC are similar so I am wondering if this is a normal amount of time to process images with APP and both of my computers just don't have the power to stack more quickly. 

When trying to stack on the Mac I was allowing APP to use 7 of the 8GB of RAM that I had and using 7 threads with #CPU 4. That computer also has a 251GB SSD. 

On the PC that I am currently using I am also allowing APP to use 7 of the 8GB of RAM that I have, but it has less cores so is using 3 threads with #CPU 4. It also has a 1TB HHD. 

I'm thinking that both computers just don't have enough RAM and processing ability to integrate and stack quickly, but when I used DSS on my PC in the past it stacked relatively quickly so I am a bit confused as to what is causing it to run slow. 


ReplyQuote
(@vincent-mod)
Quasar Admin
Joined: 5 years ago
Posts: 4850
 

Hi Vincent and thanks for trying APP!

Sorry to hear your Mac developed an issue. This won't be because of APP directly, but it's likely that (since it needs a lot of processing power) another issue came to light in the machine during the processing.

Regarding DSS, yes that is lacking quite as bit in how advanced the calibration is etc, preventing more complex data to be processed nicely. Since it is relatively simple, DSS will use less processing and RAM. But there are better free alternatives as well. APP is built around some of the most advanced algorithms known in scientific literature, making it well suited for a lot of the more complex data tasks, like creating seamless mosaics for example. But also just basic calibration is better in many ways. Downside of this is, the processing and memory needs. Mabula is working on making that more efficient (or as efficient as possible) in each release.

The times you're seeing will depend on the settings in APP that you're using (LNC for instance will increase this as well significantly). But having a good computer spec will help a lot in making this faster. The more cores and ram as possible the better, besides processing on a SSD. If you want to process quicker on lower spec machines, the best is to divide the data in smaller parts. I'd advice to take it down to processing roughly 40 light frames in 1 session. This will create about 4 final integrations. You can then reload APP and load in those 4 integrations again as lights and (without calibration data this time as they are already calibrated) you integrate those 4 integrations to 1 final one.


ReplyQuote
Share: