Lightroom Mobile: Photomerge (Pano/HDR) with imported images

The amount of ram needed to merge a few hundred megapixel panorama is pretty absurd, and the amount of paging that would have to be done on a mobile device (remember, iPad pros JUST got 6GB of ram), plus the fact that you need to be ready to shut down the process at any time (phone call, Facebook, sms, etc all have the potential for you to switch apps, which then puts high memory requiring processes at a high risk of being killed) means this is not a trivial task. Much more likely is being able to use the server, but then how to provide a good quality experience in an asynchronous process (we don’t spool up a server instance every time someone launches the app, and setting a new server instance, allocating memory, firing up camera raw, loading the files, processing the files, etc all takes time, meaning you’d be sitting there waiting for what might feel like an eternity), not to mention that you might want to try different settings out.

Not unsolvable problems but also not super easy problems for a feature that not that many people use. It’s a typical engineering constraint issue—do we take the resources away from super important, highly used features to work on useful but less impactful tools or not? And with something like pano especially, it’s not something that can be done a little bit at a time.

We have had some prototypes of HDR merging working locally, but with the weird landscape of memory availability on devices (which Apple purposefully doesn’t market strongly since it’s not important to the majority of their customers), it makes it super hard for us to build something that works well on one device but doesn’t at all (or is suuuuuuper slow) on another. Keep in mind that people are not always rational and consider things only in their own perspective—this was the most expensive device I could afford, and it’s not the cheapest, so I expect it to work! Or, “it was top of the line 5 years ago, why can’t I merge 7 61MP shots?!?!?”

We want to build software that not only does the thing, but does it well, predictably, and with a good experience, for all the different types of customers we have.

Thank you for your extensive reply. I appreciate this a lot as it’s the first clear cut answer I have received from Adobe. Now I know that I should not keep my hopes up for ever getting panorama merging on iPad.

Though, I have to point out that panorama merging has been possible with Affinity Photo for quite some time and it does it well and fast. It easily stitches vertical panoramas I have had issues with in even Lightroom. I have used it to stitch several (8+) 30 megapixel RAW files together without problems. While I understand you need to decide how to spend your resources, I find the argument of not having enough RAM rather curious.

I understand you want to release a good experience but I personally see nothing wrong in keeping features exclusive to those iPads which are capable. After all, isn’t that the same on any Mac? No one expects you to support panomerging 61 megapixel files properly on a Macbook Air but, for that reason, you don’t restrict it on those devices that are capable, right? Maybe I’m looking at this wrong but it’s kind of a strange argument to me.

Perhaps I wasn't clear--it's not that stitching isn't possible, it's that even after doing all the work, it's not a great experience. I just tested out Affinity's pano stitching with one of my typical panos. It was a 17-shot pano from my Nikon D800E (36MP) which resulted in a 317MP pano (I tend to have a decent amount of overlap in my photos). This is a very common FoV / pano result for me (even though I've moved over to 47MP Sony A7Riiis--I just happened to find the Nikon based one faster for this test).

Here's what I found, on my iPad Pro 12.9" 4th gen, 1TB (6GB of RAM):

• Generate Preview: 4m 54s

• Create Pano: 3m 30s

• Render Pano: 7m 45s

That means it took over 15 minutes before I could do anything with my pano, all while I could do nothing else with my iPad.

Mobile devices, currently, don't do multi-tasking very well. When's the last time you wanted to hold onto a mobile device, waiting 15 minutes, and not being able to do anything else? Not to mention the battery drain induced by all this?

The point about a MacBook Air is somewhat spurious--it's a desktop system, which means you can easily run another app while processing happens in the background, not to mention that it's possible to outfit with 16GB of ram, reducing the amount of memory paging significantly.

Put another way, just because it's possible to do something doesn't mean it's a good use of time and resources. It's not like it's a simple switch that we'd flip and suddenly you'd have pano stitching support. Imagine it's 6 months of work (which, while I'm just making this number up, it's probably a lower limit), would you tie up six months of resources for a feature that won't be used that much because the experience would be poor, taking those resources away from making something that many more people would use?

Again, the point is not to say it's impossible, and if we can find a way to make it work leveraging our servers, so much the better for everyone.

Thanks again for explaining in a detailed way. I agree that waiting 15 minutes is a very long time. I just feel rather bummed out over the fact that means other use cases will not get to do this. My main use case is stitching vertical panoramas together (drone photography), which is mostly 4-5 20 megapixel photos. It’s not a massive project and it can be frustrating not being able to do this on the go.

I also understand the decision of whether or not to spend valuable time and resources on this matter. I don’t have any insight into what most people request or what most people use the panomerge feature for of course.

I do appreciate your insights and you taking time to answer customer questions. It would be great to get some very rough idea of “this is what we are working on” though. Not asking for a timeline but rather something that tells customers what is on the table. This was possible for Photoshop on iPad as well.

I have been using Lightroom CC on iPad since it was first released in its current cloud form and have loved using it.

Creating arbitrary limits is really hard, both technically as well as from a user perspective.

Imagine we found a way to do this easily, but only if we limited to 60% of ram (or some other arbitrary limit). And let’s say we find a great way to appropriately estimate the amount of available ram. Then imagine the experience in which on your iPad you can do 8 iPhone photos, your iPad Pro you can do 3 61MP files, but when you try a 6 36MP pano, you can’t on any device. It’s not the number of photos—you did more before. It’s not the size of the photos, you did larger photos. It’s not even the FoV, this one just has more overlap. Why isn’t this working? What’s wrong with Adobe? Is anyone even remotely intelligent over there?

You get the drift. We already have some arbitrary limits that I run into already (we have an arbitrary max pano limit of I think 600MP that I run into all the time now that I’ve started creating multi row panos with a paid or Syrp Genies), and it’s super frustrating—I don’t want to build new ones in.

Anyway, I totally hear you! I know it’s frustrating and I know you want all the things, me too! One of my guiding principles for Lr Mobile is that you can do everything on your phone. So now it’s just a balancing act of resources to device capabilities. Heck, the rate that android devices are adding system memory to their devices (now if only android could update their memory management system to be a fraction as efficient as Apple’s