Here Comes Windows 11....

[Mark Gulbrandsen]

The dual-boot thing is going to suck, for sure. As is the fact that PXE booting doesn't work the same way in a UEFI environment (I don't know the details...I've always just set everything to "legacy mode" and used it that way).

I'm fine with offering UEFI and Secure Boot as options, but forcing them is just going to result in more environments that stick with older operating systems.

I understand their feelings on keeping the old stuff out. But there is a plethora of industrial computers that are stuck on old systems because there are no options. The CNC machining industry is just one of them. The Cinema industry had for a number of years... the Qube Servers that ran on Windows. EEK!

[Frank Cox]

Christie projectors (at least my CP2210) run on Windows XP.

[Marcel Birgelen]

Christie projectors (at least my CP2210) run on Windows XP.

Noop, not really.

The TPC runs on WinXP Embedded. But the TPC itself doesn't run anything, it's just an interface to the stuff that does the actual work. The rest of the projector runs pretty much on a mix of embedded software. I guess there is some Linux or VxWorks in there too and the ICP is probably some version of TI-RTOS. And some embedded stuff doesn't even need an OS to work.

[Ed Gordon]

Exactly the same problem I had with it Ed. None of the standard Photo Apps like Photoshop or Capture will run. Those are the two main apps that I use... I do have a mobile (car) computer that can just run W-7, but it sure runs a lot cooler with Ubuntu on it.

I have an old Netbook that I used when traveling that I have running Win7 on. I actually ran Win10 on it for a few years, until the Win10 "improvements" finally overwhelmed the puny Atom processor in the Netbook. The "downgrade" to Win7 revived it. I did try Ubuntu on it, and it ran just fine, but I had software I needed that was not compatible with Ubuntu.

[Leo Enticknap]

Just made a disconcerting discovery. None of the five Windows 10 PCs in my home are hardware capable of W11, because they don't have the Trusted Platform Module. These are:

• My main desktop - home build, based on MSI motherboard, in 2018
• My DCP render machine - home build, based on Supermicro motherboard, in 2015
• My laptop - Asus Core i5, bought in 2015 (originallly shipped with W8, and upgraded)
• My work-issued laptop - Dell, Core i5, bought 2019
• My wife's laptop - Dell, Core i7, bought 2017

There is a TPM board available for the MSI motherboard that would make it hardware compatible, but at $80, that it a bit above being a trivial sum (especially for functionality that I don't need for any other purpose).

I suspect that when the penny drops that many if not most W10 PCs currently in use cannot be upgraded, this will become a significant news story. I would guess that around 90% of W7 PCs in use at the time W10 launched were completely upgradable right off the bat, and most of the others only needed a cheap hardware upgrade (e.g. 2GB more RAM). I also suspect, and hope, that the endgame will either be M$ relenting on a TPM being an absolute requirement (i.e. give you the option to install W11 on a computer without one), or extend W10 support significantly beyond 2025.

All five of those machines have UEFI capability (though it is currently disabled on all of them apart from my wife's computer), so that's a non-issue. But the TPM requirement is a big drawback. The chip shortage making computer replacement more expensive, and environmental concerns (all that e-waste), might also combine to apply pressure on Microsoft, IMHO.

[Marcel Birgelen]

We have invested heavily in hardware the last few years, with the exception of the last two years, the pandemic wasn't the optimal time to invest for us... Most of our hardware is actually still in active support, yet a lot of our workstations (machines with quad Xeon CPUs from 2018 with 256 GByte RAM) don't qualify for Windows 11. I'd say that borders on the idiotic.

I don't see any reason to upgrade to Windows 11 if they don't offer support for our hardware. Heck, if they don't change their mind, we start moving away from Windows to Linux and run our remaining Windows apps using VMs. We'll simply ditch our MS365 subscriptions (my personal mail has been broken for more than three weeks now, no-one at MS bothers fixing that anyway) and move to e.g. Google.

Microsoft isn't the unstoppable force they used to be. While Adobe offers a formidable lock-in into either Windows or Apple, the general move to web-based applications has made them more replaceable than they've been the last 30 or so years.

Maybe the e-waste caused by relegating otherwise good machines to the trash is something that can trigger some buttons at Microsoft. I seriously cannot understand how you can position yourself as a "green" company and otherwise expect people to throw out 2 years old machines. Also, just try to buy some new hardware right now. Waiting times of anywhere between tomorrow and 6 months...

Apparently, there is already a Windows 11 installer out there, that kills the TPM check and what-ya-know? Windows 11 runs just fine without TPM 2.0... But apparently Microsoft has already threatened to withhold updates for those "illegal" installs. I don't know what their deal is, but I guess they got billions of dollars of "incentives" from the likes of Intel and AMD...

[Bobby Henderson]

I bought a new Alienware X17 notebook to replace the aging Dell XPS 15 I used at home for most of the previous decade. I also have a new Dell XPS desktop system on order to replace to aging tower at my day job; Dell says that system will be delivered around Thanksgiving. I hope it arrives sooner.

I have zero near-term plans to install Windows 11 on either of those new systems. In the near term I have to consider what issues could crop up with Adobe's 2022 generation of Creative Cloud applications. I rely on Astute Graphics plugins for Adobe Illustrator. So I need to wait on that company to update its software for the new version of Illustrator. And then I don't know how forward-compatible the applications are with Windows 11. It may be well into 2022 before I install Win 11 on anything. And even then I would probably only install it on the new Alienware notebook since I don't connect it to a network.

Nearly all the PCs in my workplace are running Windows 10 Pro. I think there is at least one or two vintage PCs running Win 7. The only PC in the building I think is eligible for a Win 11 upgrade is a recently purchased Dell XPS desktop system we bought to control a pair of Hewlett Packard latex-based large format printers. Windows 10 is already bad enough for its own networking pains. I imagine mixing in a Win 11 system here and there may open a new can of worms. Unfortunately it's a bridge we will have to cross within the next year or two as we replace more computer systems.

Microsoft isn't the unstoppable force they used to be. While Adobe offers a formidable lock-in into either Windows or Apple, the general move to web-based applications has made them more replaceable than they've been the last 30 or so years.

What creative applications are going web-based? I can't think of any on the professional level that are doing so. Hell, I can't think of any open-source graphics applications doing that either. They all install on and operate off of the computer's local hard disc. The applications are too big to download and install temporarily every time they're run. And they're sure not going to be running remotely from some server. In terms of speed and overall performance it's no contest, locally running graphics apps are going to win, especially when using new hardware like NVMe solid state hard drives connected directly to the PCI Express bus, fast RAM, fast GPU boards, etc. Someone could get away with running a word processor in a web browser window. Legit graphics production, video editing, etc isn't happening as a web-based thing any time soon.

The only things in graphics that are running via the web right now are specialized services, such as cloud-based storage (very slow compared to local discs), value-added content (clip art, stock photos, stock video) or fonts. The creative applications themselves still install and operate very much like they have for 30 years, despite the word "cloud" being tossed around. The only differences now is we're not installing off physical CD or DVD discs and don't have huge retail boxes with printed manuals inside.

The only scenario I can see where a web-based setup might have an advantage in graphics creation is distributed rendering across many computers. That would be a very specialized scenario. No one doing regular traditional creative work will have such a need.

[Marcel Birgelen]

What creative applications are going web-based? I can't think of any on the professional level that are doing so. Hell, I can't think of any open-source graphics applications doing that either. They all install on and operate off of the computer's local hard disc. The applications are too big download and install temporarily every time they're run. And they're sure not going to be running remotely from some server. It's no contest, especially when using new hardware like NVMe solid state hard drives connected directly to the PCI Express bus, fast RAM, fast GPU boards, etc. Someone could get away with running a word processor in a web browser window. Legit graphics production, video editing, etc isn't happening as a web-based thing any time soon.

The only things in graphics that are running via the web right now are specialized services, such as cloud-based storage (very slow compared to local discs), value-added content (clip art, stock photos, stock video) or fonts. The creative applications themselves still install and operate very much like they have for 30 years, despite the word "cloud" being tossed around. The only differences now is we're not installing off physical CD or DVD discs and don't have huge retail boxes with printed manuals inside.

Even graphical applications aren't safe for "webification" anymore. While moving heavy 3D applications and professional video rendering into the cloud may be a few years down the road, most of the basic stuff provided by apps like Photoshop and Illustrator has already seen at least a partial move into the cloud. Yeah, performance is often abysmal, but right now there are quite a few things happening to improve on that one.

One of those things is Web-Assembly, that ditches slow-as-hell Javascript for a bytecode that can eventually be JIT compiled to the local architecture. Another initiative is WebGPU, an initiative to create something like an "OpenGL" or "DirectX" for the Web. It is actively being pushed by some major tech companies (Apple, Microsoft, Google...), all to allow direct access to hardware accelerated rendering and GPU features, directly from the web.

So yeah, expect Photoshop running in a browser and maybe even Autocad and Maya within a few years from now... I'm not saying you, me are going to like it, but the writing is euhm... in the sky...

[Bobby Henderson]

Even graphical applications aren't safe for "webification" anymore. While moving heavy 3D applications and professional video rendering into the cloud may be a few years down the road, most of the basic stuff provided by apps like Photoshop and Illustrator has already seen at least a partial move into the cloud. Yeah, performance is often abysmal, but right now there are quite a few things happening to improve on that one.

Quite a few things? Please be specific.

I use Adobe Creative Cloud applications pretty heavily, but very little of those applications operate via the cloud. Getting assets such as fonts, images or even some clips of audio & video from the cloud is one thing. Loading and running the application from the Internet is another. That is NOT happening any time soon, not while there is an extremely enormous speed advantage in running the applications on hardware inside the computer. The speed difference even for a mundane mainstream application like Photoshop is huge. Running Photoshop via the web would be laughably bad.

[Marcel Birgelen]

Quite a few things? Please be specific.

I use Adobe Creative Cloud applications pretty heavily, but very little of those applications operate via the cloud. Getting assets such as fonts, images or even some clips of audio & video from the cloud is one thing. Loading and running the application from the Internet is another. That is NOT happening any time soon, not while there is an extremely enormous speed advantage in running the applications on hardware inside the computer. The speed difference even for a mundane mainstream application like Photoshop is huge. Running Photoshop via the web would be laughably bad.

Adobe's software stack is still mostly based on client-based applications. Photoshop, Illustrator, etc. all still are client based apps, based around a decades old codebase. But Adobe isn't known to be a fast mover, if it comes down to moving their codebase to another platform. It has taken them years to move their Mac implementation to native Cocoa, a 64-bit Flash client only appeared once Flash was already declared dead.

But there are tons of entirely on-line Photo editors, like Photopea, Pixlr. Just google and you'll find like 10 of those, some of them are actually pretty sophisticated.
As for vector based web-based tools, there are also quite a few, the best known is probably Gravit Designer, which is now part of Corel.
Mind you, I don't use those apps, as they're not fit for professional use yet and for now, they're still slow as #$%, but you can clearly see a trend here.

Keep in mind that while a lot of those applications are "web based applications", they essentially do a lot of the heavy lifting on your machine, only now it's a bunch of Javascript running on your machine. Javascript is slow as f&ck, compared to native code, but if you replace that Javascript with WebAssembly though, stuff becomes a lot more performant as JIT compiled WebAssembly doesn't have to be much slower than pre-compiled C or C++ code, in which most of Photoshop and Illustrator have been written.

Microsoft has already indicated that they see WebAssembly as the way forward not to just deploy "web based applications", but essentially all applications, no matter if they're loaded via the web or from your local disk.

[Bobby Henderson]

Adobe's software stack is still mostly based on client-based applications. Photoshop, Illustrator, etc. all still are client based apps, based around a decades old codebase.

The age of Adobe's applications is irrelevant to the basic problem: the Internet is GLACIER-SLOW compared to locally installed and operating applications. It might be possible for a remote server to deliver modest office productivity applications to a bunch of dumb terminals in various locations across the country. But it's not currently possible to do the same thing with professional level graphics applications and yield performance that is anything better than horribly awful.

A person's Internet connection would need to run on the order of many gigabytes per second to match speeds a creative application can yield when running on local hardware. His connection to the remote server would have to be at least as good, if not many times better. Hardly anyone has access to such online speeds. But those speeds are common on new computing hardware.

Even if Internet speeds didn't absolutely stink compared to local hardware speeds running the apps via the web seems like a solution in search of a problem. Adobe gets enough criticism as it is for services like Adobe Fonts requiring an always-on Internet connection. Making the applications themselves require always-on connections is a bit much. Then there's the business case for the whole thing. Professional graphics applications running via the web is just too ridiculous to take seriously.

[Marcel Birgelen]

The age of Adobe's applications is irrelevant to the basic problem: the Internet is GLACIER-SLOW compared to locally installed and operating applications. It might be possible for a remote server to deliver modest office productivity applications to a bunch of dumb terminals in various locations across the country. But it's not currently possible to do the same thing with professional level graphics applications and yield performance that is anything better than horribly awful.

Adobe's application code base is very much relevant for whatever comes next. They have the advantage of the userbase, but also the disadvantage that their userbase expects some level of backward compatibility. While the first assures them of a steady income, the latter can be quite a challenge when there is a need to move to new platforms.

New players will never be able to deliver a suite that will be completely compatible with Adobe's proprietary file formats, they don't have any comparable userbase, but they also don't have the codebase legacy.

A person's Internet connection would need to run on the order of many gigabytes per second to match speeds a creative application can yield when running on local hardware. His connection to the remote server would have to be at least as good, if not many times better. Hardly anyone has access to such online speeds. But those speeds are common on new computing hardware.

Even if Internet speeds didn't absolutely stink compared to local hardware speeds running the apps via the web seems like a solution in search of a problem. Adobe gets enough criticism as it is for services like Adobe Fonts requiring an always-on Internet connection. Making the applications themselves require always-on connections is a bit much. Then there's the business case for the whole thing. Professional graphics applications running via the web is just too ridiculous to take seriously.

I guess you're over-estimating stuff a bit here. Yeah, stuff is slow, but that's primarily due to the fact that most if not all of those tools are still using JavaScript and not WebAssembly. They don't require gigabit connections to and from the server, because they're running most of the stuff right on your own machine, inside the VM of the browser. Maybe test a few of those links I've sent you. Yeah, their performance is pretty awful compared to native apps for complex files, their toolset isn't nearly as complete or advanced as Adobe's, but it isn't as dreadful as you describe it, because it actually works... well, mostly... And Corel is pouring millions of their dollars into it.

Again, read up on what stuff like WebAssembly and WebGPU is doing. It is leveraging your local processing power, it's essentially a way to deliver "hybrid" applications, where the developer decides on what resides server-side and what resides client side.

Right now, for static 2D stuff, performance should be a manageable thing, also the amount of bandwith required isn't all that terrible. I don't see myself editing 4K video on-line just yet though and heavy 3D stuff currently simply lacks an interface, but that's what WebGPU tries to solve.

Regarding bandwidth: Even though large ISPs and carriers are often reluctant to invest in proper upgrades, they can't simply stay behind forever, even my 4G is pushing 200 MBit/s nowadays and I haven't tried 5G yet. The likes of Microsoft and Amazon are pushing towards datacenters ever closer to the edge for low-latency, high bandwidth computing and a VDI machine with dedicated GPU will happily run any heavy graphics software over a VDI/RDP connection. Yeah, that isn't really web-based as it simply remotes a screen, keyboard and mouse, but it still runs "in the cloud". I've tried one of the new VDI offerings of Microsoft Azure lately and the results are pretty decent, even running DaVinci Resolve as an experiment. The problem here though is that even with a gigabit internet connection, uploading raw 4K files "to the cloud' still takes a while. Another problem is that integration with stuff like a local Wacom tablet doesn't work yet, at least not with Microsoft's VDI solution, VMWare has this working.

[Bobby Henderson]

Adobe's application code base is very much relevant for whatever comes next. They have the advantage of the userbase, but also the disadvantage that their userbase expects some level of backward compatibility. While the first assures them of a steady income, the latter can be quite a challenge when there is a need to move to new platforms.

Adobe has few problems at all when it comes to backward compatibility. They're doing better at this than some other companies I can name. I can open ancient Illustrator AI/EPS files or Photoshop PSD images made in the 1990's on the latest CC versions of those applications. The file formats have been compatible across multiple platforms. The Intel side has been pretty stable, but the Mac side has gone thru Motorola 040 and PowerPC CPUs, then Intel x86 and now lately the ARM/M1 architecture. Let's not forget about mobile platforms like iOS/iPad OS. I can work in Adobe Illustrator on my iPad, save the file, then open and edit it on my Windows-based notebook. A CC license doesn't tie me to a specific platform; I can use it on two computers, both could be on the same platform or one could be a PC and the other a Mac.

For all the criticism anyone tries to level at Adobe for having an old code base, I don't see any software vendor anywhere matching Adobe in the general purpose professional graphics market. Adobe Illustrator is the best vector graphics application available. Illustrator is over 30 years old, but it blows the doors off any of these free/low-cost upstarts. None of Illustrator's professional rivals are keeping up with it either. Most of the Johnny-come-lately vector applications feel old because they're only covering bases Illustrator did back in the 1990's.

Photoshop is still the best image editor on the market. InDesign is the industry leader in professional page layout. Competition gets a lot more fierce in the video production space, but After Effects is still an industry leader in that niche thanks in large part to the integration it has with Illustrator and Photoshop.

I guess you're over-estimating stuff a bit here. Yeah, stuff is slow, but that's primarily due to the fact that most if not all of those tools are still using JavaScript and not WebAssembly.

The problem with web-based applications goes farther than what code base the application is using. Hardware, bandwidth and latency are far bigger problems. The performance of local hardware is going to utterly trounce whatever through-put could come from a puny Internet pipe.

I guess you're over-estimating stuff a bit here. Yeah, stuff is slow, but that's primarily due to the fact that most if not all of those tools are still using JavaScript and not WebAssembly. They don't require gigabit connections to and from the server, because they're running most of the stuff right on your own machine, inside the VM of the browser. Maybe test a few of those links I've sent you. Yeah, their performance is pretty awful compared to native apps for complex files, their toolset isn't nearly as complete or advanced as Adobe's, but it isn't as dreadful as you describe it, because it actually works... well, mostly... And Corel is pouring millions of their dollars into it.

I wouldn't put all that much faith in Corel. I think the company is in trouble. Their previous owners (Vektor Capital) went on a buying spree, loading Corel down with debt. Then they spun-off the whole thing to KKR last year. CorelDRAW has been Corel's main tent-pole application from the beginning (1990). But they've done a terrible job maintaining the application in recent years. The v2019 release was riddled with bugs and saw only two modest updates. Some of the bugs have lived on thru the v2020 and v2021 product cycles. Worse yet, Corel adopted a licensing and pricing model that may end up helping kill the application. BTW, Gravit Designer has its own problems. Corel wants $50 per year for it. For a one time $50 purchase a person on a budget can buy Affinity Designer and get upgrades after that for free.

The higher ups are obviously not giving Corel the resources it needs to properly maintain CorelDRAW. The current 2021 version has had two updates (and the second update released last week introduced more new bugs than old ones it fixed). Meanwhile Adobe has had four major point-release updates for the current version of Illustrator along with at least a couple dozen minor updates. No other vector graphics application on the market comes close to that. Adobe has large public beta and pre-release programs for Illustrator and several other applications. It's easy for critics to paint a big company like Adobe as being slow and complacent. In actual practice they're putting a lot more work into their graphics applications than their rivals.

[Marcel Birgelen]

For all the criticism anyone tries to level at Adobe for having an old code base, I don't see any software vendor anywhere matching Adobe in the general purpose professional graphics market. Adobe Illustrator is the best vector graphics application available. Illustrator is over 30 years old, but it blows the doors off any of these free/low-cost upstarts. None of Illustrator's professional rivals are keeping up with it either. Most of the Johnny-come-lately vector applications feel old because they're only covering bases Illustrator did back in the 1990's.

An old code-base isn't necessarily a bad thing, as old code has often proven itself. It doesn't rot, but it can become a burden if the world around you is changing to other platforms. They've been pretty slow to adapt to Android and IOS and those clients still aren't entirely feature complete. Their code base is heavily platform dependent and as such it takes them a lot of efforts to port code to new platforms, as many of those ports are essentially new implementations of existing functionality.

That's the strong part of web-driven applications, as they don't need merely that much porting to be able to run on another platform, whether it's another OS (Linux, MacOS, Android) or another CPU architecture (Intel x86 / AMD64 v.s. ARM and maybe even MIPS).

Photoshop is still the best image editor on the market. InDesign is the industry leader in professional page layout. Competition gets a lot more fierce in the video production space, but After Effects is still an industry leader in that niche thanks in large part to the integration it has with Illustrator and Photoshop.

Adobe is an absolute behemoth in the 2D graphics world and their footprint will not easily dissolve. With many people having a large part of their careers invested into those applications. But they aren't entirely unbeatable and you can see how the competition is making some inroads. Adobe's biggest vulnerability are paradigm shifts that happen so fast, they can't follow.

I'd say that Adobe has a strong hold in the semi-pro sector for video. While they have made quite a few inroads into "Hollywood-scale productions", that world is still dominated by Avid (MediaComposer ProTools), DaVinci and the likes of Foundry (Nuke). Apple had a thing going with FinalCut, but they dropped the ball a few years ago when they decided to ditch their existing platform and simply introduced a completely new product. I guess this is a good lesson for the likes of Adobe what happens when you leave your userbase standing in the rain.

The problem with web-based applications goes farther than what code base the application is using. Hardware, bandwidth and latency are far bigger problems. The performance of local hardware is going to utterly trounce whatever through-put could come from a puny Internet pipe.

Those are all well-known problems and the reason why those applications haven't been moved to "the cloud" yet are primarily due to the current limitations of main-stream web technology. While it is perfectly possible to run applications on a server and stream the I/O to end-users, this places the burden on the cloud, latency becomes an issue and streaming at high resolutions doesn't come free either. The current Javascript VM of modern browsers, while highly optimized compared with a few years back, is a big burden. Code needs to be sent in bulky, clear text, JavaScript has serious performance and threading issues. That's why WebAssembly was proposed and implemented by the likes of Google, Microsoft, Apple and Mozilla. Those companies dominate the browser landscape and many of them have a vested interest in selling cloud-based services. WebAssembly, while not the end solution to it all, will solve a lot of the performance issues of JavaScript and will essentially revamp your browser to an "app delivery platform", where entire applications can be "streamed on demand" towards the end user and where the local hardware can be used to increase efficiency, reduce bandwidth burdens and latency.

I wouldn't put all that much faith in Corel. I think the company is in trouble. Their previous owners (Vektor Capital) went on a buying spree, loading Corel down with debt. Then they spun-off the whole thing to KKR last year. CorelDRAW has been Corel's main tent-pole application from the beginning (1990). But they've done a terrible job maintaining the application in recent years. The v2019 release was riddled with bugs and saw only two modest updates. Some of the bugs have lived on thru the v2020 and v2021 product cycles. Worse yet, Corel adopted a licensing and pricing model that may end up helping kill the application. BTW, Gravit Designer has its own problems. Corel wants $50 per year for it. For a one time $50 purchase a person on a budget can buy Affinity Designer and get upgrades after that for free.

Gravit Designer has always been implemented as a web-app first. Their desktop client is essentially their web application running in a browser kiosk. This has the advantage that they have all the features on all platforms, without the need to port it to those platforms. It's not a serious contender for Illustrator, primarily due to file format incompatibilities, but it is a nice example of what can be done with pure web technology at this point in time, although it also still shows its limitations.

I don't have much faith in Corel at all and I do have more faith in Serif/Affinity when it comes down to competing with Adobe. Their photo editor and vector graphics tool see a continuous stream of updates, rivaling most other offers out there and are just short of what Adobe has to offer. File compatibility with Adobe products will forever remain an issue though and a big factor in the decision for people to keep committed to Adobe software, as long as it works for them.

Yet, I'm pretty sure that all those companies are looking at models of how to bring their apps to the web. Adobe will be late to the party, but I'm pretty sure we'll see something like Photoshop Elements move to a web-based version first. It's not that I'm going to like all of this. The big Facebook outage has clearly shown the limitations of "the cloud" and what happens if it all goes tits up, but it's an evolution we won't be able to stop, even if it will probably still take a few years to actually get there.

There is one big advantage though: A web application, at least when done right, isn't tied into a particular operating system. Once everything has moved "to the web", Microsoft Windows and Apple MacOS essentially will become irrelevant, at least as client operating system...

[Bobby Henderson]

An old code-base isn't necessarily a bad thing, as old code has often proven itself. It doesn't rot, but it can become a burden if the world around you is changing to other platforms. They've been pretty slow to adapt to Android and IOS and those clients still aren't entirely feature complete.

Devices such as an iPad or an Android phone are not a substitute for a full fledged desktop computer or notebook for use in graphics production. At best, devices like an iPad or smart phone augment the work one can do on a traditional computer.

There are some really good graphics applications on the iPad platform, such as Procreate for example. But most of the graphics apps on that platform serve a specific niche. In the case of Procreate it's a digital painting app that works great in conjunction with the Apple Pencil. None of those graphics applications can do everything that Illustrator or Photoshop can do in Mac OSX or Windows. The apps that come closest to doing so are the iPad versions of Illustrator and Photoshop.

You keep characterizing Adobe as an outfit that has grown slow and weak. In terms of traditional graphics software (pixel-based image editing, vector graphics, page layout) I don't see any of Adobe's rivals doing any better on any platform. If we were talking about another market, such as video production, yeah Adobe has a much harder fight on its hands there with rivals like Avid and Blackmagic Design. Adobe doesn't have rivals that tough in the traditional graphics market.

Adobe is an absolute behemoth in the 2D graphics world and their footprint will not easily dissolve. With many people having a large part of their careers invested into those applications. But they aren't entirely unbeatable and you can see how the competition is making some inroads. Adobe's biggest vulnerability are paradigm shifts that happen so fast, they can't follow.

The only thing that is going to hurt Adobe where they dominate is graphics applications coming along that truly out-class Illustrator, Photoshop or InDesign. The rival application has to be good enough to convince professional creative users to switch. In the 1990's Quark Xpress was the dominant page layout application for mass-produced print publications. InDesign came along and surpassed it. If one of Adobe's rivals is going to market a "Photoshop-killer" the application actually has to be better than Photoshop. The really difficult thing is the rival has to develop superior alternatives to Illustrator and InDesign at the same time. Integration across applications is a pretty big deal.

Apple had a thing going with FinalCut, but they dropped the ball a few years ago when they decided to ditch their existing platform and simply introduced a completely new product. I guess this is a good lesson for the likes of Adobe what happens when you leave your userbase standing in the rain.

Adobe took advantage of the situation when it happened. They even offered ways for Final Cut 7 users to import their projects into Premiere Pro when they couldn't do so in Final Cut X. Premiere Pro has had its own technical problems. To Adobe's credit, in recent years they've really opened up their beta and pre-release programs to more users and their development teams interact with users more frequently. Significant point-release updates happen more often and minor updates are pushed to users much more often than I see from Adobe's rivals.

Gravit Designer has always been implemented as a web-app first. Their desktop client is essentially their web application running in a browser kiosk. This has the advantage that they have all the features on all platforms, without the need to port it to those platforms. It's not a serious contender for Illustrator, primarily due to file format incompatibilities, but it is a nice example of what can be done with pure web technology at this point in time, although it also still shows its limitations.

I haven't used Gravit Designer, but I've tried other web-based attempts in creative software. CorelDRAW has a web-based app that is sort of passable for basic things. But it's not an acceptable substitute for the desktop application. I've seen message creation software for LED signs that was web-based. It mostly sucks. The applications that work best are those which install and run in the traditional sense.

I don't have much faith in Corel at all and I do have more faith in Serif/Affinity when it comes down to competing with Adobe. Their photo editor and vector graphics tool see a continuous stream of updates, rivaling most other offers out there and are just short of what Adobe has to offer. File compatibility with Adobe products will forever remain an issue though and a big factor in the decision for people to keep committed to Adobe software, as long as it works for them.

I have the Windows and iPad version of Affinity Designer. I bought both last year when Serif was doing its half price sale. I think I ended up paying $25 for the Windows version and $10 for the iPad version. The price was cheap enough for me to be willing to experiment with it. It's also nice to at least have the application on hand if I get any client-provided art made in Affinity Designer. It's typically better to open any graphics files in their native host application than import the art into a non-native environment. That's one thing that makes it so difficult, if not impossible, to do with Adobe software. Most major corporate branding and their other graphics assets are created using Adobe software. Rival applications are very spotty at importing Adobe generated AI, EPS and PDF files.

Serif's creative applications are decent in terms of affordable creative software. I wouldn't have any problem recommending Affinity Designer to someone wanting to do his own DIY graphics work on a budget. The user interface in their applications is more polished than open source applications like Inkscape. On the other hand their applications are missing some critical features. With me being such a font nerd one glaring problem with Affinity is no support for OpenType Variable Fonts. Illustrator was first to support OTF VAR fonts, but then Adobe was involved in developing the standard. Illustrator was also at the forefront at supporting OpenType-SVG, aka SVG Color Fonts.

Serif has released a handful of minor updates to Affinity Designer over the past year. They've done more than Corel has with CorelDRAW. But they're not even close to matching the frequency of updates Adobe has given to Illustrator.

There is one big advantage though: A web application, at least when done right, isn't tied into a particular operating system.

That's only an advantage for developers, not users. I do not expect graphics applications to move to an entirely web-based metaphor anytime soon, if ever.