Maintain, optimize and troubleshoot your NLE
Professional cloud workflow platform
Simplified media management

Thoughts on the 2019 Mac Pro and Pro Display XDR

Today Apple unveiled the new Mac Pro, which is an attempt to woo understandably wary and skeptical creative pros. Here are my initial thoughts.

Mac Pro

From the outside it looks a lot like the old “cheesegrater” Mac Pros. It’s interesting to me that Apple chose to go back to an old design rather than attempting to push the envelope like they tried with the “trashcan” Mac Pro. But as many people preferred the older design this is certainly no bad thing.

While I wouldn’t describe the design as sexy, I doubt many creative professionals care too much about the look of a box that goes under their desk or in a server room. The trashcan Mac Pro was very much form over function and I would certainly trade form for increased function.

Apple products have a reputation for getting thinner and lighter with each iteration but it’s worth noting that the new case is slightly taller and wider than the cheesegrater case, while the depth is about an inch less. The weight is similar to the cheesegrater version.



The similarities to the old model continue on the inside. The internals should look very familiar to anyone who opened up an old cheesegrater Mac Pro. Many of us suspected Apple might offer minimal or no expansion, instead requiring users to add devices externally via Thunderbolt. Luckily this was not the case. Bringing back PCI slots got a round of applause during the keynote for some reason (no-one should applaud the fact they were ever removed), and there are eight of them this time compared to the cheesegrater’s four.

The Mac Pro supports the latest AMD graphics cards. NVIDIA cards were conspicuous by their absence, so that will be an issue for anyone relying on CUDA. Apple and NVIDIA had a falling out a while ago and it’s unclear at this point whether NVIDIA will be able to make their cards Mac Pro-compatible independently of Apple.

One of the most interesting aspects of the announcement to me were the available accessories. Not only is Apple offering rackmount options, they are even offering wheels to make it easy to transport (those cheesegraters were heavy). Apple’s phone/tablet hardware business has peaked, so in addition to services, Apple may be putting emphasis on squeezing as much money out of professionals / businesses as they can, and pro accessories would be a way to do that.

On the topic of rackmounting, it’s worth noting that Apple implied it’s a specific rackmount case you buy at the point of purchase, not simply a rackmount kit fitted to the existing case. Based on the width of the desktop case, it would be 5U in height when placed in a rack. So this isn’t really a replacement for the discontinued Xserve which was a 1U server. However, some racks have 26” in depth available which means Apple could potentially offer a deeper case that takes up less vertical space.

Another accessory is the Afterburner ProRes accelerator. While it can accelerate regular ProRes, it seems like it is designed more for Apple’s new ProRes RAW format. Apple claims it can accelerate up to three streams of 8K RAW or 12 streams of 4K RAW. Whenever we’re talking about acceleration, it’s always worth mentioning that the card isn’t all you need to achieve this - you also need storage that can handle that bandwidth too.


Basic I/O rear ports are handled via a card, implying you could add additional ports or change these to better fit your needs if necessary. But the card seems decent - it even includes USB-A ports, as well as two 10Gb-E ports (this elicited no reaction from the crowd but it definitely should have).

Apple is including a monster 1.4 kilowatt power supply in the system, capable of delivering 1280 watts of power continuously, however in reality you would only get this kind of power draw if you max out the specs, fill it up with graphics cards and set it churning away on a complex render.

In the US, most household circuits are 15 amps at 120 volts, meaning the maximum power draw for a single circuit is 1800W, with a continuous draw of around 1440W. This means that at 1280W there is a narrow overhead before the breaker trips, so you would need to be mindful of what else is plugged into the same circuit at the same time (note that a circuit may consist of multiple outlets). Most people probably won’t have to worry about this but it’s an important thing to think about if you’re planning to max out the specs.

So what about the price? The entry-level trashcan Mac Pro started at $2999 and the new model starts at $5999. In other words, without directly comparing specs, the base cost doubled from one generation to the other.

Ever since the Intel switch, people have debated whether Mac Pros actually needed server-class CPUs and ECC RAM. Apple could have included i9s and non-ECC memory in entry-level models to lower the price, or they could have simply adopted a Xeon configuration that could be achieved at a $2999 price-point.

Remember, $5999 is the base configuration so you may be looking at $8000 or more once you add accessories and tax. To max out the specs will likely cost well into five figures. So if you find yourself balking at that price, it’s likely you’re not the target market. These prices, along with options like rackmounting, suggest to me that Apple’s intending to sell these primarily to companies rather than individuals.

So while some may be disappointed in the pricing, I’m just glad Apple is making pro-grade hardware again. Let’s hope they keep it fast with regular spec bumps and not allow it to stagnate like the trashcan model.

Pro Display XDR

The specs of this display are impressive and as Apple pointed out, no display currently exists with 6K resolution, 1000 nits of brightness, P3 10-bit color and 1000000:1 contrast ratio, all in a single display. They even have a whole nano-texture glass option to reduce glare.

Just like the Mac Pro, it’s not sexy, which is perhaps more of a problem for a monitor that sits on a desk than a computer that’s hidden away. I’m also concerned about the level of heat that even necessitates a large grille on the back, which is not something you commonly see on a monitor.



There’s certainly no doubting the specs of this display, which are truly impressive. Apple is intending this to be a replacement for $42,000 reference monitors. While it offers reference modes like Rec. 709 and P3-DCI, the colorists I follow on Twitter who might be expected to purchase such a device seemed skeptical that it would truly replace professional reference monitors.

The price starts at $4999 for standard glass and $5999 for anti-glare nano-glass. If you listened carefully during the keynote, you could hear the audience groaning and muttering  when the prices were announced.

But if it truly can replace a $42,000 reference monitor then this pricing is a bargain. What is definitely not a bargain is the $999 stand. There is no way the stand costs anywhere near that to produce, so it’s massively overpriced. The VESA mount is a more-reasonable $199 so it seems as if Apple is either blatantly profiteering or trying to actively encourage people to mount their displays.

Just like the Mac Pro, it seems like it’s again geared at companies rather than individuals - there’s no way Apple would sell individuals monitors without a stand. They'd just include it in the box and increase the price.

Conclusion

The specs are great and I’m glad Apple’s making pro-grade hardware again. Companies will probably pay the Mac Pro surcharge without complaint, while individuals who don’t want an iMac will still be left wanting. Apple will make a lot of money from bulk purchases and accessories.

I’m really not sure how successful the Pro Display XDR will be. Reference monitors are niche, $42k ones especially so. It’s too expensive for people to buy as just a regular desktop display, while people who can afford $42k reference monitors may not be swayed by it and probably aren't price-sensitive anyway. For people who just want the basics as cheaply as possible, there are reference monitors available at much lower prices. So it’s for the people in the middle: those who want a reference monitor and need features like HDR for whom $3k is too low and $8k is too high  - a niche within a niche. It seems a strangely small target market for a company as large as Apple. The Apple brand may well shift some units but it's hard to see it being a big success.

Posted by Jon Chappell on Jun 4 2019 to Analysis, Apple, Hardware
Permalink

Thoughts on 32-bit codecs being phased out in macOS

You may have read Apple's article about "legacy media" being phased out in the next version of macOS or even been alerted by an error message in the new version of FCPX. The article is fairly sparse and doesn't go into detail about the change or explain why it is occurring.

To be clear, it's not that 32-bit codecs are going away, it's that all 32-bit software is going away. But it affects video software the most because the only reason for an app to still be 32-bit in 2018 is because it uses QuickTime.

Why would an app still use QuickTime over the newer and non-deprecated AV Foundation? Because AV Foundation does not support third-party codecs. I filed an enhancement request to support third-party codecs several years ago and Apple closed it as "won't fix" so their position on that seems pretty clear. (Sure, you can go to a lower level and write your own parser but if you've reached that point, why bother using Apple's media frameworks at all?)

There is a Reddit thread with quite a lot of misunderstandings about what the change means. So to clear up some of the misunderstandings in that thread: there isn't really such a thing as a 32-bit file. You can have a file stored on a 32-bit file system or in a container with a 32-bit size header but the result of this is that these files will be limited to 2 or 4 GB in size. It has no bearing on which apps can read them, so they can still be read by both 32- and 64-bit apps or codecs. Also, do not confuse the QuickTime container (.mov) with the QuickTime API - the container is alive and well and still supported by the new frameworks. (The MP4 standard's container is about 99% identical to a regular MOV container after all.)

This isn't really about 32- vs 64-bit, it's about Apple no longer allowing extensibility. Next year's version of macOS won't support third-party codecs at all, so converting 32-bit codecs to 64-bit isn't a solution. A codec is a central piece of code that handles reading and writing to a particular format. Now instead of using that central code for free with no extra effort, every app needs to create its own version of that code.

But remember: this only affects macOS's native video frameworks. Developers don't have to use them and there's nothing to stop someone writing their own DNxHD parser. You don't need to worry about your files becoming unreadable because ffmpeg will always be there. (It still supports ancient long-dead codecs like Indeo!)

This is unlikely to change much with Adobe or Blackmagic. They use their own parsers for a lot of formats so will be immune to this change. The people it affects the most are small developers, who now have to come up with solutions to problems that Apple was taking care of for them in the past. By licensing certain codecs or allowing users to install their own codecs, Apple was providing an easy, free and legal way for these developers to read these files. They may not have the manpower to write their own parsers or the legal or financial ability to license the codecs in question. ffmpeg is a solution to the technical problems, but legal ones still remain - the code has to be compiled directly on the user's computer in order to be legally compliant, which can be complicated and user-unfriendly.

It may also affect FCPX negatively too. If the cross-platform NLEs are immune to this change, it means that FCPX is the only major NLE that will be affected by it. Apple's advice in the article is essentially "buy a new camera" but will people really shelve the cameras they paid thousands of dollars for to continue using FCPX? Or will they go for the cheaper option of switching to an NLE that can open them? I think the FCPX team are great, but they are sometimes hamstrung by corporate Apple which may not have the same interests as them or may be unable to see the bigger picture of how large changes like this affect Apple's smaller divisions.

Some people have speculated that the reasons for this are because Apple is planning to merge iOS and macOS or create ARM chips for desktop computers. Apple may well do these things in future, but I think this change was made for two more mundane reasons: security and efficiency. AV Foundation is fast because it is efficient so inefficiently-coded third-party codecs may cause performance or battery life problems. Supporting third-party codecs opens up the operating system to maliciously-crafted movie files meant to exploit bugs in third-party codecs that Apple doesn't have the ability to fix themselves. (I would argue that the benefits are still worth it despite these two potential costs.)

You may be wondering what solution we are planning for our products. The majority of our apps don't use QuickTime and the only one that relies on it heavily is QT Edit. For the apps that use QuickTime, we will convert them to have some level of base functionality with the default OS frameworks. When you launch the app for the first time, it will ask if you want to enable additional codecs and if you answer yes, it will install ffmpeg for you (the exact technical way this will work and the exact codecs that will be supported is still being figured out). Advanced users will also have the ability to compile ffmpeg themselves and link the executable to our apps to ensure they can definitely read the files they need to or to have greater control or better support  for their hardware. We feel that this is the best we can make of both worlds: trying to make it as user-friendly as possible while still giving power users control.

If this change affects you, here are the feedback links for macOS and FCPX. But Apple shouldn't postpone or cancel the death of QuickTime (it absolutely should die) - the solution is to allow third-party extensibility in the new frameworks. I'm not holding out hope that this will happen, and our ultimate goal is to be cross-platform so we'd probably go down the ffmpeg route regardless, however if enough people complain they may at least continue to support some additional codecs in FCPX.

Posted by Jon Chappell on Nov 20 2018 to Analysis, Apple, QuickTime
Permalink

Apple is over-thinking the Mac Pro

Today Apple announced they are rethinking the entire design of the Mac Pro after three years without an update.

The current Mac Pro is an extremely impressive piece of design and engineering. Unfortunately the design was too clever and  over-engineered and it neither met customers' needs nor allowed Apple to easily upgrade it.

Apple issued an apology, which is extremely rare, and it's great to see they have recognized and taken ownership of the issue. It's almost unprecedented for Apple to announce unfinished products without a ship date and they're timing this right before NAB, so I'm assuming this move is to do with damage control.

The only problem is that I'm spotting a pattern here. Apple abandoned the old (pre-2013) Mac Pro and barely updated it, then after a customer outcry they released a completely redesigned model, which they then abandoned again. Now they are completely redesigning it again but it won't ship for at least a year and we don't know much about it except that it will be modular to some degree.

It sounds like Apple may be about to fall into the same trap again: over-thinking the problem in panic mode with an over-engineered solution. Apple behaves like this is a really difficult problem to solve but it's actually extremely simple. We want to use Macs, so all Apple has to do is not get in the way.

What Apple fails to grasp is that the design factors that made the iPhone successful have little relevance to the pro desktop market. Sure, design is important to creative people, but it's not #1 on the priority list - it's maybe #7 or #8. Our #1 priority is being able to do our jobs quickly with as few restrictions or workarounds as possible. I'm sure a sizable number of pro users would be happy enough with 2017 specs in the 2012 case.

If Apple wants to impress their professional customers they need to do two things: design their pro products backwards by thinking about what their capabilities should be first and then thinking about their physical design, and do a better job of maintaining a dialogue with pro customers (hint: dialogues are in both directions). There are a lot more options now and Apple should be competing for our business instead of assuming we'll just stick around like we have so far.

Posted by Jon Chappell on Apr 4 2017 to Analysis, Apple, Hardware
Permalink

WWDC - What we'd like to see in OS X 10.12

WWDC is right around the corner and Apple will most likely be unveiling a new version of OS X. Here's what we'd like to see. 

HEVC / H.265 support

H.265 (also known as HEVC), promises to halve the bitrate of its predecessor, H.264. This means that you can deliver the same video at half the bandwidth, or alternatively double the quality within the same bandwidth. This could greatly increase the popularity of 4K video on the web. 

The problem so far has been a legal one - instead of having everything tidily packaged within a single patent pool, there are two competing pools in competition with each other. 

In particular, the HEVC Advance pool initially had onerous terms that would have cost hardware and software providers considerably more money than they paid to license H.264. Most of these provisions have now been scaled back fortunately. 

Whenever a new codec comes out there is always a chicken-and-egg problem - there's no reason to make content because people can't play it and there's no reason to enable people to play it because there's no content. So now is a great time for Apple to take the plunge and kickstart the H.265 / HEVC format. 

System-wide Dark Mode

In OS X 10.11, Apple added a Dark Mode option which dims the menu bar to fit in better with dark-UI pro apps like DaVinci Resolve. 

We'd like to see this taken one step further and offer the ability to toggle the entire UI of an app light or dark. For many apps this will require no changes so it could probably be an opt-out option for developers.

Finder tags

When Apple introduced OS X 10.9 they included Finder tags which allowed you to tag a file with additional metadata.

However, this is not particularly useful because the tags appear as color dots instead of text and they are very small and overlap each other. It also means you can no longer color code the entire filename, making it easy to miss color-coded files.

So my proposal is for two types of tags: color tags and text tags. Color tags affect the background color of the file in the list and can be easily spotted when scrolling. You can only assign one color to a file.

Text tags do not affect the background color of the file, even when colored, but appear in a bubble next to the filename. So you can tag it “VFX” and “Delivery" for example and anyone can see at a glance that it is a VFX deliverable. Any tags that can’t fit in can be expanded by clicking an ellipsis icon next to the filename.

Airplay video streaming

It should be possible to stream video from a Mac video player like QuickTime or CinePlay to an Apple TV. This is something that can already be done in iOS but still hasn’t made it to the Mac.

There are some solutions to this but they are very hacky and developers would benefit from an officially-sanctioned method.

AV Foundation: Don’t treat remote files as second-class citizens

There are some very powerful tools in Apple’s AV Foundation media framework. Unfortunately many of the functions for scanning media files and extracting information are limited only to files that exist on the user’s local hard drive. This can limit the options you have for dealing with files on a remote web or FTP server.

There is no real reason for this because even though the data originates on a remote server, it still ends up in memory and/or cached to disk, depending on the situation. Because AV Foundation is built to be asynchronous, there is no reason why this data cannot be provided to an app as and when it is downloaded.

Better SVG image support

Apple needs better support for SVGs throughout the OS such as more accurate rendering in QuickLook and the ability to see thumbnails. 

We'd especially like to see an easy way to use SVGs within a user interface, which could make the aforementioned Dark Mode a little more painless to implement.

Third-party codecs in AV Foundation

Now that AV Foundation has matured and inherited many of the more advanced features of QuickTime, it's time for Apple to fix the biggest omission: third-party codecs. It would be even better if there were ways for developers to also override functionality such as streaming and caching with custom behaviors.

Posted by Jon Chappell on Jun 11 2016 to Apple, Analysis
Permalink

5 Ways Apple Could Improve iOS 9

Next week is WWDC, where Apple will likely unveil a new version of iOS. Here’s what we’d like to see.

Improved text selection

I deliberately avoid using iOS devices in situations where I need to frequently copy and paste text because it’s extremely tedious.

Here’s an exercise to try: take a paragraph of text and try to select only one or two sentences inside it. After you select a certain quantity of text, iOS tries to help you by automatically selecting the entire paragraph. Selecting exactly the desired amount of text then becomes a battle between you and the operating system. A lot of times I just select the entire paragraph, paste it and then cut it down to what I need because it’s easier and quicker, but it’s still a step I wouldn’t have to do on the desktop.

It is also especially difficult to select the text of a link because if you tap on it, iOS assumes you either want to visit the link or copy its URL.

Proper multitasking

One big problem with iOS is that it does not have true multitasking. Some functions, such as audio playback, can take place in the background but most of the time when you switch to another app the previous app will either pause or terminate.

What this means in practice is that if an app takes a long time to do something, you cannot check email or do something else while it is processing or it will probably pause and then restart when you switch back. Consequently I spend a lot more time staring at progress bars on iOS than I do on OS X.

Dark mode

Dark mode would show system UI elements with a darker skin to stop the display appearing too bright or lighting up a dark room.

This could cause complications for third-party apps so I think it should be possible for apps to see which mode the user has selected and decide in what circumstances to honor it.

Don’t autocorrect proper grammar

I have no problem with iOS autocorrecting misspelled words but it sometimes corrects valid grammar and spelling into something that is incorrect, such as automatically correcting “were” to “we’re”.

Polish

There are rumors that iOS 9 will not have significant new features and will instead be a Snow Leopard-style polish and bug fix release. I’m less in favor of this for iOS than OS X because I find iOS 8 significantly more stable than iOS 7 and the aforementioned feature requests are problems that genuinely slow me down and turn me off the platform.

However, anything Apple can do to improve the speed of the OS is appreciated, especially on older devices.

Posted by Jon Chappell on Jun 4 2015 to Analysis, Apple, Software
Permalink

6 Ways Apple Could Improve OS X 10.11

With WWDC around the corner and a likely unveiling of a new version of OS X, here’s six improvements we’d like to see.

Finder tags

When Apple introduced OS X 10.9 they included Finder tags which allowed you to tag a file with additional metadata.

However, this is not particularly useful because the tags appear as color dots instead of text and they are very small and overlap each other. It also means you can no longer color code the entire filename, making it easy to miss color-coded files.

So my proposal is for two types of tags: color tags and text tags. Color tags affect the background color of the file in the list and can be easily spotted when scrolling. You can only assign one color to a file.

Text tags do not affect the background color of the file, even when colored, but appear in a bubble next to the filename. So you can tag it “VFX” and “Delivery" for example and anyone can see at a glance that it is a VFX deliverable. Any tags that can’t fit in can be expanded by clicking an ellipsis icon next to the filename.

Airplay video streaming

It should be possible to stream video from a Mac video player like QuickTime or CinePlay to an Apple TV. This is something that can already be done in iOS but still hasn’t made it to the Mac.

There are some solutions to this but they are very hacky and developers would benefit from an officially-sanctioned method.

Files in use

One of the most annoying things about OS X is how it frequently refuses to perform an action like moving a file or emptying the trash because the file is in use, even when you can’t figure out what’s using it and haven’t opened that file in months. This bug has existed for a very long time.

I have a theory that it may be related to QuickLook scanning the file to update its icon but I haven’t been able to exactly pinpoint it.

There are some workarounds such as copying a file instead of moving it or securely emptying the trash.

AV Foundation: Don’t treat remote files as second-class citizens

There are some very powerful tools in Apple’s AV Foundation media framework. Unfortunately many of the functions for scanning media files and extracting information are limited only to files that exist on the user’s local hard drive. This can limit the options you have for dealing with files on a remote web or FTP server.

There is no real reason for this because even though the data originates on a remote server, it still ends up in memory and/or cached to disk, depending on the situation. Because AV Foundation is built to be asynchronous, there is no reason why this data cannot be provided to an app as and when it is downloaded.

OpenGL ES / Metal

OpenGL ES is a cut-down version of OpenGL that is intended for mobile devices. Metal is a low-level iOS API by Apple intended to improve performance of graphical apps.

Because neither of these are supported on OS X, it means porting graphical apps between platforms requires a lot of work. With some parts of OpenGL ES it’s easy because all you have to do is slightly modify the name of a function, but in other cases it’s trickier because there are no direct equivalents for certain commands.

Polish

One rumor making the rounds is that Apple will forego major new features in favor of polishing and optimizing its codebase.

I feel that Apple’s software quality has slipped in the past five years and the annual release cycle means that Apple is often introducing new features (and therefore new bugs) before it has finished fixing the bugs introduced in the previous version.

The last time Apple did this was for OS X 10.6 Snow Leopard and I have spoken to many people who consider it to be the best release of OS X. In fact, more of our customers are on 10.6 than 10.7 or 10.8.

I would therefore be willing to sacrifice all of the above feature requests in favor of a massive codebase polish from top to bottom.

Posted by Jon Chappell on Jun 3 2015 to DR News, Analysis, Apple
Permalink

2014 - Year in Review

2015 is now upon us so here's a look back at the past twelve months.

We introduced several new products:

CinePlay for Mac, a Mac video player with professional features like timecode, markers, safe areas and overlays. It also integrates with our Kollaborate cloud service to provide cloud video playback, commenting and synced sessions with colleagues.

Kollaborate Plugin Pack - Upload files to the cloud directly from Final Cut Pro X or Adobe Premiere Pro for Mac.

Kollaborate Folder Watcher (beta) - Automatically upload files to the cloud when copied to a particular folder on your hard drive.

Marker Import for Windows - Free Windows tool for importing markers created by our apps into Adobe Premiere Pro. There is also a Mac version available here.

We also released the following major updates free of charge:

Pro Media Tools 1.3 - Complete overhaul of QT Edit's batch processing mode, offering significantly improved performance and stability.

CinePlay 1.1 for Mac - Audio playback, gang multiple files together for simultaneous playback (useful for comparing two or more shots) and playlist support.

CinePlay 1.4 for iOS - Audio playback support.

Cut Notes 2.3 for iPad - Added an Autosave Vault, changed hold behavior to be in line with Cut Notes 1.x and greatly improved Kollaborate cloud integration.

Preference Manager 4.2 - Compatibility with Adobe Creative Cloud 2014 and Lightworks beta for Mac.

In total we released over 100 updates for our apps.

Kollaborate subscribers also received the following features at no extra cost:

  • Annotations (drawing over video)
  • Color-coded comments
  • Favorites
  • Batch uploader
  • Shared link expiration
  • Email notification filtering
  • Move files between projects
  • Metadata
  • Approve / reject files
  • Drag and drop file management
  • Dark UI option
  • Reminders
  • Free storage bump and new pricing packages
  • Server-side encoding
  • QuickLook previews
  • Alias overhaul
  • Custom departments and positions
  • Kollaborate Transfer overhaul with faster encoding and improved queue management

Our most-downloaded app was FCS Remover but our most-used app was Post Haste. Our most popular paid tool was Pro Maintenance Tools. Our most popular iOS app was Video Space Calculator.

We also improved our website in a lot of ways, but most significantly was the creation of an Account section that allows customers to login to manage their licenses and print invoices.

The most popular blog posts in 2014 were:
10 Useful Avid Console Commands (on track to be our most popular of all time)
Fixing missing database errors in DaVinci Resolve
How to protect your video content
Managing frame sequences in Batch Renamer

You may have noticed we've been a little quiet for the past couple of weeks and the reason for this is that we're working on some big updates, so stay tuned for new products and major updates in early 2015. You can keep up to date with new products, features and tips by subscribing to this blog, following us on Twitter or Facebook or by subscribing to our mailing list.

Posted by Jon Chappell on Jan 1 2015 to DR News, Front Page News, Analysis
Permalink

How to protect your video content

Sometimes it is necessary to protect video content against unauthorized redistribution. Here's an overview of the main ways of doing this.

Visible Watermarking

This involves placing text or a logo over the video to show the name of the company owning the content.

Pros

  • Simple and free
  • Doesn't require specialist software to create or view
  • Its obvious visibility provides a deterrent
  • Difficult to remove if placed in an area that cannot be cropped out

Cons

  • Disrupts the movie
  • Does not provide any physical protection mechanisms
  • No feasible way to tie a movie to a particular person
  • May increase movie encoding times

Invisible Watermarking

Some companies offer invisible watermarking by tagging content with a fingerprint. One such example would be Digimarc Guardian.

You would tag the movie and give it to the end user, then if it appears on a file-sharing website you can download it and read the tag to find out which user shared it.

Pros

  • Quick to apply to an existing movie without re-encoding
  • Can be tied to a specific user
  • Does not disrupt the movie

Cons

  • Requires specialist software to create and view
  • Does not provide any physical protection mechanisms

Content Protection Plugins

Browser plugins like Flash and Silverlight offer their own Digital Rights Management (DRM) protection systems.

Pros

  • A lot of control over exactly what the user can and cannot do. For example, you can make a movie unplayable after a certain time period.
  • Offers a physical protection mechanism

Cons

  • Requires a plugin, which may cause problems with less tech-savvy users
  • May not be playable on mobile devices
  • May be too expensive for smaller publishers (Adobe does not publish Adobe Access pricing)

HTML5 Content Protection (Encrypted Media Extensions)

This uses web technologies to allow content protection without the need for a plugin. It is frequently misunderstood as a protection mechanism in itself, but in actuality it is just a means of linking HTML5 web video with existing third-party protection systems, so a lot of the pros and cons of plugins still apply.

Pros

  • No plugins needed
  • A lot of control over exactly what the user can and cannot do. For example, you can make a movie unplayable after a certain time period.
  • Offers a physical protection mechanism

Cons

  • Currently very little browser support
  • Needs to be integrated with an existing DRM system
  • Expensive

Streaming

Streaming a file can make it difficult for users to download it, and for someone people this is all that is needed. This can be achieved with proprietary technologies like Adobe Media Server or HTML5 technologies like HTTP Live Streaming.

Pros

  • Difficult to download files
  • Fast playback times
  • Additional features can be added like automatically changing playback quality to match the user's connection

Cons

  • Requires preparation of the files in advance
  • Can still be downloaded by tech-savvy users with specific software
  • Some technologies require a streaming server

So which is best? That entirely depends on the content you are trying to protect.

  • If you are trying to prevent a paying viewer from sharing a movie, DRM is the best way to go.
  • If you are trying to prevent clients from using work in progress media without paying you, visible watermarking and/or streaming may be best.
  • If you are trying to prevent a movie being leaked before its release, a combination of physical preventions, user tracking and deterrents may be required.
Posted by Jon Chappell on Nov 24 2014 to DR News, Tutorials, Analysis
Permalink

6 Things Apple Should Improve in OS X 10.10

WWDC is tomorrow, where Apple is likely to announce a new version of OS X. Here's what we'd like to see.

1. AV Foundation improvements

There are so many improvements needed for AV Foundation that it merits an article of its own, but a few would be expanded codec support, easier extensibility, LOTS of bug fixes and the ability to stream Airplay video without ugly hacks.

2. Mac App Store improvements

Apple should allow app developers to request exemptions from sandboxing rules. We feel that the strict rules hurt the user experience and even removed one of our apps as a result of this.

Other areas for improvement would be the ability to hide third-party updates and not requiring a password for free updates (this can get tedious if you have multiple accounts).

3. Files in use

OS X frequently marks files in use when they aren't, which can cause issues copying or deleting files and also when emptying the Trash.

I have a theory that this may be caused by QuickLook trying to generate icon previews for these files but it's difficult to pinpoint.

This bug is especially frustrating when using network drives as OS X frequently creates temp files that can only be deleted after unmounting and remounting the drive.

4. Make full screen work on the current display only - don't let it blank out every other monitor

Full screen functionality was improved significantly in Mavericks with the "Displays have separate Spaces" option in the Mission Control section of System Preferences, but many people had to switch it off and revert to the old system due to compatibility issues (mine were with FCP 7).

When switching back to the old way of doing things on a shared Space, OS X blanks out all monitors even if there is nothing displaying on them.

I think the solution to this problem is a blacklist where users can specify apps that should only use the current Space on another monitor without disabling it for compatible apps.

5. Network drive / VPN improvements

It is clear that Apple is assuming most users are saving files to HFS+ formatted SSDs. This is apparent when mounting a drive over a VPN connection because Finder and many other Apple apps experience very poor performance.

In fact, I completely eschew Apple apps over a VPN and use Path Finder as a Finder replacement and AppCode as an Xcode substitute because those apps do not read or write to disk unnecessarily and handle these issues much better.

6. Bug fixes

Apple is announcing a new OS when arguably Mavericks still needs a lot of bug fixes. Some of the issues I am still experiencing in 10.9.3 are:

  • Hang when copying certain files to a network volume
  • Audio output randomly disappearing (speaker icon becomes greyed out), requiring a restart
  • Exposé sometimes fails, meaning it's difficult to exit it and switch back to an application, requiring a restart
  • Plugging in a monitor and closing the laptop lid sometimes makes it sleep randomly in the middle of me working. It also sleeps if I disconnect the charger.
  • Periodic freezes where the mouse moves but everything else is frozen.
  • Occasional apps that cannot be closed, which requires a hard reset because they stop OS X from shutting down
  • Other people have also reported issues in OpenCL apps with 10.9.3

With these issues in mind, I would be very happy if Apple institutes a Snow Leopard-style overhaul. Snow Leopard was so stable it took me a long time to upgrade and many of our customers are still using it.

I'm also hoping that Apple recently opening up the beta program to non-developers helps to address this, but if it doesn't I feel that Apple should not commit to a yearly upgrade schedule if they cannot guarantee quality.

Posted by Jon Chappell on Jun 1 2014 to Analysis, Apple, Software
Permalink

5 Features Apple Should Add in iOS 8

With WWDC and iOS 8 around the corner, we thought we'd offer up our wish list for future iOS features and bug fixes.

1. Documents app

The biggest problem with iOS is the limited file system, which can make working on files in different apps difficult.

Currently you can store movies and photos in a central place that is accessible by all apps with permission to access it. While this implementation has its flaws (e.g. apps can't delete files they've created), creating a similar storage area for documents would go some way toward improving the limitations when sharing files.

Much like Photos and Videos, these documents could be accessed and managed in a Documents (or Files) app.

2. Utility apps

Another problem with iOS is that it can be difficult to know what to do when someone emails you an unusual file format.

For example, customers often email me .log files when I'm on the move. These are exactly the same as text files, however iOS refuses to open them because they do not have a .txt extension.

Unlike OS X, you cannot force a file to open in another app, nor can you rename files, so I was stuck scouring the App Store for an app that would open them. The only app I found that would open them was a hex editor app but this was not a perfect solution because it was not designed for formatting and displaying large portions of text.

OS X has a lot of utility apps like TextEdit, Preview, Font Book, etc, but Apple seems to have delegated Mail as the only (limited) general-purpose file viewer. If Apple hopes to one day replace OS X with iOS they will need to address issues like this.

3. Better Copy & Paste

I have a lot of difficulty with copying and pasting. The primary issue I experience is that iOS tries to second-guess what I'm doing and it always gets it wrong.

For example, after you have selected a certain amount of text iOS assumes you want to select the entire paragraph, which may not be exactly what you want, but there's no way to select less.

So I find it easier to select more text than I need, paste it and then delete the parts I don't want.

4. Improved multitasking

While switching between apps is simple, the apps don't always retain their status when you switch back to them.

For example, an app that requires a network connection may have disconnected or you may have to wait for a web browser to reload the page again.

This doesn't happen every time but it occurs when iOS comes under pressure to free up memory. Now that iDevice CPUs are pretty fast I hope Apple can implement an OS X-style memory compression system to reduce occurrences.

I also hope the rumors of split-screen multitasking are true because it will help significantly with this issue.

5. Better Gmail support in Mail

I use the stock Mail app with Gmail because I have not found a third-party mail app that I like (suggestions are appreciated).

However, the Mail app seems to be frequently confused by conversations, with some messages correctly joined together and others shown as distinct messages. This can be annoying if I've deleted a conversation, only for all the deleted messages to reappear in my inbox when someone else replies. In some cases this requires me to re-delete 15 or so messages.

Another issue is that it seems to be impossible to actually delete messages. Deleted messages are archived with a special tag rather than actually being deleted. Additionally, sent messages don't show up in other apps or the Gmail web interface.

It's for reasons like the five listed above that I prefer to delegate particular tasks until I get back to a desktop computer. If the rumors are correct and Apple is creating an iPad Pro, simply adding a bigger screen won't be enough. In my opinion the changes most likely to endear it to professional usage would come from iOS itself.

Posted by Jon Chappell on May 31 2014 to Analysis, Apple, Software
Permalink