Adobe Photoshop is Now Native on Windows 10 on ARM

Posted on May 17, 2021 by Paul Thurrott in Music + Videos, Windows 10 with 34 Comments

Adobe hasn’t issued an announcement of this important change, but the native Windows 10 on ARM version of Photoshop is no longer in beta.

“As of May 2021, Photoshop now runs natively on 64-bit Windows 10 ARM Devices,” an Adobe support page that was first spotted by Windows Central reads. “Photoshop only runs in 64-bit operating systems, and the version of the Creative Cloud that you will need to install Photoshop is a 64-bit only application. Do not install Photoshop if you intend to continue to install 32-bit applications from Adobe.”

Adobe brought a beta version of Photoshop to Windows 10 on ARM back in November. Then, as now, there are some important missing features. The following features are not available natively on Windows 10 on ARM:

  • Import, Export, and playback of embedded video layers
  • Shake Reduction filter
  • Invite to Edit workflows are not supported.
  • Preset Syncing is not on by default
  • Windows Dial Support
  • Generator and related features
  • Opening or placing U3D files
  • Starting Photoshop from Lightroom ‘Edit In’ command
  • Oil Paint Filter
  • Spell Checking and hyphenation for Hebrew and Arabic languages
  • Plugin Marketplace panel

There are also several known issues with this release, which you can find on the Adobe support site.

As a reminder, Adobe did formally announce the Apple Silicon-native version of Photoshop for the new M1-based Macs back in March. But of course it did: Where Windows 10 on ARM is used by very few people, and certainly by a tiny percentage of creators who use Photoshop, Mac users are adopting M1-based products very quickly.

Tagged with , ,

Join the discussion!

BECOME A THURROTT MEMBER:

Don't have a login but want to join the conversation? Become a Thurrott Premium or Basic User to participate

Register
Comments (34)

34 responses to “Adobe Photoshop is Now Native on Windows 10 on ARM”

  1. johnnych

    Nice ? ! I liked the idea on the last Windows Weekly of a Teams-Surface-ARM-Book-Pro-Like device which runs all the big apps that people need to get serious work done: Outlook, Office, Teams, Adobe, etc.

    Just need the customers and user base to be receptive of it :)


    • bkkcanuck

      It is a chicken and egg scenario, and Microsoft if it was serious would identify important general 3rd party applications that they think are important based on telemetry data and set aside a fund to help encourage 3rd parties to modernize and support the platform (it is either a carrot or stick approach and Microsoft has been hesitant about the stick approach). It is funny, I do remember really early on when talking to management of the bank about hardware requirements having to be beefed up to support future versions of the application we had installed there... there only real concern was... we just need enough years leeway to have the funds budgeted to do it... I think a lot of times it is not an unwillingness to move forward but the lack of communications and making sure budgets and project plans are in place well ahead of required changes.

      • hrlngrv

        | it is either a carrot or stick approach and Microsoft has been hesitant about the stick approach


        Hesitant probably because the stick approach would run afoul of US and EU regulators and possibly leave MSFT with US$/€ billions in fines. Part of the reason regulators aren't so concerned about Windows any more is because MSFT hasn't been bullying ISVs for years. If MSFT were to begin to do so again, it's likely to certain regulators would become interested in wringing fines out of MSFT.

        • bkkcanuck

          I am not talking of stick as in punitive action, but in deprecating specific interfaces that are long time legacy and giving it a timeline til that functionality will no longer work.

          • hrlngrv

            How much ACTIVELY MAINTAINED software uses those specific interfaces? Can you name any commercial Windows software available for sale today which uses anything MSFT could deprecate by, say, 2024? If VB.Net or VB6 software developed in-house would be an example of what needs deprecating, there are LOTS of MSFT's enterprise customers who simply will not tolerate MSFT doing so.


            Outside work I figure the vast majority of software MSFT can't afford no longer to support is abandonware, no longer maintained by its IP rights holders but still in use by uncomfortably large numbers of Windows users. The moment MSFT removes support for such software from Windows, MSFT spurs some consumer PC users to try Linux and wine. FWIW, wine under Linux can run a lot of older Win32 software. Many people using Windows today probably don't need to keep using Windows. How eager is MSFT to give home PC users reasons to try Linux?


            The problem with spending decades building up an 85% user base is that there's no vast pool of other potential customers to replace half the Windows user base if MSFT pisses then off. I figure when MSFT reached 1 billion Windows users across all versions, MSFT effectively lost the ability to redefine Windows. That would have been back in the late 1990s or early 2000s.

            • bkkcanuck

              We no longer have win16 in Windows...

            • bkkcanuck

              What ends up happening is you have a number of years where it is supported, you then have a number of years where it is supported but not in the most recent Windows version.... then you have a number of years running unsupported version of Windows (which is not an option for corporations), new hardware is only supported in the newer versions... and if you are not going to maintain the 3rd party software it starts dropping in popularity and lower revenue (with the likelihood of it becoming irrelevant). At a certain point that version of Windows that was left behind becomes itself irrelevant.

  2. hrlngrv

    How many production as opposed to testing systems are running WOA?

  3. bschnatt

    Great news. Now I just need to find a way to squeeze more memory onto the motherboard of my x2 laptop. D'oh! Kidding - don't really need this app...

  4. jaredthegeek

    Mac users have no choice in adopting M1 if they continue to stay with Apple. Even the iPad pro has gone to M1. Photoshop on an M1 iPad is going to brutally beat a Surface Pro in performance. There are limitations but those keep getting fewer and fewer.



    • johnnych

      Apple users are excited for M1 because it offers lower power usage and longer battery life (aka real world improvements and impacts) -- also Apple has shown us all a clear path forward into the future of computing from their side of things.


      Microsoft needs to drive its engineers to do the same thing but on the Windows side of course, regardless of what Intel or AMD have planned - push those ARM chips forward Qualcomm!

      • lvthunder

        You would think Intel, AMD, and Qualcomm would make some chips with the RAM and everything built in like the M1.

        • Paul Thurrott

          This is kind of an interesting difference between monoculture (Mac) and the more diverse PC market. Apple builds chips for itself. AMD and Intel build chips for multiple hardware makers, who also have multiple suppliers for RAM, graphics, etc. There are pros and cons to each way.
        • bkkcanuck

          The memory on the M1 is 'on package' not 'on silicon' (similar to AMD chiplets are 'on package' but not 'on silicon (together)'). AMD actually already does that for the chips that go into the Playstation 5 and new X-Box processors (though with those they use GDDR6 rather than LDDR memory) . Understand though that with x86 processors that are not as efficient, you are adding more heat to the package than you would before (more heat generated within a smaller area) so other compromises would have to be made somewhere to accommodate it.

          • Truffles

            That's it. The fundamental constraint is heat dissipation over a given area. x86 is maxed out, so if they want to add anything to the actual silicon they've got to subtract something else.


            In contrast, Apple seems to have the headroom to add all sorts of specialised functions on to their silicon: memory, neural engine, image processing, machine learning, media encoders/decoders, crypto etc. I'm guessing they've still got a ton of thermal room remaining before they need to start making hard decisions in the name of heat dissipation.

  5. crunchyfrog

    Seeing this is a step in the right direction for WOA. I suppose the next big thing is for Microsoft to fully support WOA with their own apps and to get x64 support out the door.

    • bkkcanuck

      Yes, Microsoft should have had all their applications currently supported on the latest architecture at launch so they would be fully support WoA. Unfortunately a lot of Microsoft 'changes in direction' seem to be made on the fly in a reactive way -- rather than a coherent long-term strategy which makes these types of transitions... difficult to do within the time-frame management gives to the teams. Even if Microsoft was still susceptible to their version of agility -- the overall strategy should be in place to make sure the core pieces are in place so that the delta needed in development to achieve these last minute changes in direction is rather small.

  6. lvthunder

    Now Windows just needs a chip that performs like the M1.

    • rmlounsbury

      And a better emulation layer like Rosetta 2 on macOS. Native ARM apps run fine but when you have to run a 32-bit app the experience falls over.

      • bkkcanuck

        When you don't shed legacy in a consistent manner and insist in trying to be all things to all people for all time - you get into a situation that makes it a might more difficult to implement things like Rosetta 2. In the software development world there is a word for what Microsoft accumulates by not shedding the legacy... it is called 'technical debt' ... and Microsoft has a lot of it in Windows.

      • Greg Green

        I think Rosetta is a translator/converter that translates to arm code and saves it.

        • bkkcanuck

          Yes, Rosetta 2 does translate ahead of time the x86 code to ARM code but most of the functionality is provided through calls to the OS - which are the same on both platforms. For Microsoft to support all legacy, they would have to port all of the legacy interfaces to ARM, device drivers etc. (including a not insignificant amount of code that is written in embedded assembly language). Apple by deprecating then removing legacy interfaces before the move has a much smaller amount of work to do to implement something like Rosetta 2.

  7. Truffles

    My masochistic side feels the need to seek out the inevitable head-to-head YT reviews.

Leave a Reply