• Welcome to the Lightroom Queen Forums! We're a friendly bunch, so please feel free to register and join in the conversation. If you're not familiar with forums, you'll find step by step instructions on how to post your first thread under Help at the bottom of the page. You're also welcome to download our free Lightroom Quick Start eBooks and explore our other FAQ resources.
  • Stop struggling with Lightroom! There's no need to spend hours hunting for the answers to your Lightroom Classic questions. All the information you need is in Adobe Lightroom Classic - The Missing FAQ!

    To help you get started, there's a series of easy tutorials to guide you through a simple workflow. As you grow in confidence, the book switches to a conversational FAQ format, so you can quickly find answers to advanced questions. And better still, the eBooks are updated for every release, so it's always up to date.

Develop module Generative Fill and Client Server


New Member
Dec 18, 2023
Lightroom Version Number
Operating System
  1. Windows 10
Just a bit of musing
I was professionally involved with computing from the mid seventies until I retired early in this century. One of the trends, in both directions, was how the balance of processing shifted between Client and Server, depending on where the computing power could best be sited and the capacity of networks (and a few other factors of course). I'm thinking of the moves from dumb terminal, to thin client, to fat client, distributed data bases, data base servers and so on.
As Lightroom has added more power consuming facilities my computer is starting to creak a little, and I was thinking that I may need to upgrade, or at least feed Nvidia's pockets with a new graphics card.
However, as we can see Lightroom has moved more in the client server direction, by offloading the generative fill processing to their cloud servers.

I'd though for a while that Lightroom's capabilities would be constrained by the capacity of desktop hardware, but I hadn't expected this development (probably not looking closely enough). Of course. I don't think we really know what the charging model will be when it becomes a part of the released product, but I wonder how much more Adobe will want to move more functionality into the server.
Any thoughts?
Interesting question. The Android apps have been handing off some edits like Lens Blur to the cloud when they're used on lower-spec devices, so it could potentially make it possible to do things like Merge on mobile devices. They still do as much as they can on the local device though, including Lens Blur on higher-spec devices.
My professional background mirrors yours. (My first computer job was using an IBM/360).
Personally, I do not think there is enough computing power available to let Adobe handle everyone's AI processes in the Adobe Cloud. Second, local client hardware is improving at a rapid rate. Apple's move to Silicon chipsets has been a revolution in local functionality. The third factor is network speeds, Things really slow down once you get outside of the local network. The reason the Client Server model worked at all was that you had a local Server serving local clients (and this was when there were only 100 mbit networks. )

Taking a look at Generative fill/Generative remove, AI is useful from a cloud perspective in that you can generate pixel data to replace image areas defined by the local computer and passed to the Cloud server. The Cloud computing can store and create multiple scenarios that are often reusable . But the merging of images still needs to take place locally.

Your typical tablet or phone mobile device simply does not have the computing power to perform complex AI processes. As in the Case of Lightroom, these are not available on the local mobile device and if they exist have to e handed off to a cloud server. Wifi Speeds on mobile devices might typically yield 250Mbps download and perhaps 40% of that for upload. Cellular speeds at 5G (theoretically 10Gps) are in reality much slower than WiFi.

So, I don't see cloud computing making much of an inroad over an ever increasing power in a desktop computer. "Firefly". as a service exists to support those operations that can not be managed at the local level of computing. And I think over time these will get included in the local app as local "horsepower" will allow.

The move has been away from "dumb terminals" toward independent computing not the other way around. In spite of its adoption of UNIX base OS, IBM is still selling Mainframe computers. But most users are reluctant to pay for Software as a service (SaaS) and even reluctant to subscribe to software licenses instead of a perpetual license. Right not the use of Firefly is low and most users can use its AI feature within the credit limits in terms of their subscription. Should that change, there will likely be a revolt an order of magnitude greater that the revolt experienced when Adone moved to the subscription model.
My professional background mirrors yours. (My first computer job was using an IBM/360).
You "youngsters!". My first was an IBM1620 in 1969 with a mind boggling 4k of memory and no disk drives (you booted from a card deck).
You "youngsters!". My first was an IBM1620 in 1969 with a mind boggling 4k of memory and no disk drives (you booted from a card deck).
FWIW The IBM 360 was introduced in 1965 although I did not use one until 1973.
P.S. I'm not sure how I've managed to get here as a new member! I've been here ages as @davidedric!
Your old login "davidedric" is still out there. All you need to do is logon as that user.
P.S. I'm not sure how I've managed to get here as a new member! I've been here ages as @davidedric!
Looks like you've ended up with two accounts under two different email addresses. Drop us an email and let us know which email you want to use going forward, and we'll merge the accounts.