Andrew Richards
Forum Replies Created
-
[Greg Andonian] “That’s not true. FCP7 could open projects from previous versions.”
Fair enough, and Apple has stated they aren’t going to chase that dragon. However, with the new XML capability, CatDV can convert FCP7 sequence XML into FCPXML. Not a perfect translation, but then nothing could be. Interchange between any two NLEs will lose some level of detail like effects, titles, etc.
The more salient point is that the barriers to using FCPX in any real world workflow are quickly coming down. Then it becomes a matter of preference instead of outright missing capabilities.
Best,
Andy -
[Marvin Holdman] “Over the course of the last week I’ve been thinking a lot about the incremental update that just came out and what it might mean in the broader picture of how FCPX will continue to develop. It would appear that it’s going to be a slow process to see this application get the functionality and integration capabilities that its predecessor had.”
They put out a .0.1 release with a non-trivial feature payload 91 days after they shipped 10.0. This is a slow rev cycle? Compared to what?
[Marvin Holdman] “Given the fact that Apple was and is very tightlipped regarding what they are doing with their product, combined with the their suggestion that the app store delivery of FCPX will allow more frequent upgrades and their severe limitation of releasing their API’s, I wonder if this isn’t the sign of an app that is getting ready to be a continuing heartbreaker?”
They posted an FAQ stating what they were working on, when they would release it, and they have adhered to those statements. They have reiterated they will be bringing two more critical features (broadcast monitoring and multi-cam) in the next rev. This is tight-lipped? Apple might not have an emissary on these forums, but they have been uncharacteristically candid (for them) about their plans for FCPX.
And what is the “severe limitation” in their APIs?
[Marvin Holdman] “It seems obvious to me that part of the design criteria was to create a program that was App Store friendly. It would seem that much like the hardware appliances that Apple has created, they were hoping to turn this “Pro” app into vehicle to create more App store purchases. While many Apps on the store are selling for way less that $100, this one is selling for $300 and is one of their top grossing apps, as I type this. “
Certainly, Apple is all-in on the App Store as their distribution channel for software. But that isn’t because they make their money there, it is because they think it lets them deliver a more cohesive user experience for the people who buy Macs (where the real money is). The App Store isn’t a cash cow. In fact, Autodesk said the only reason they aren’t selling the full version of AutoCAD on the App Store is that there is presently a cap on prices at $999. If the App Store is anything like the iTunes store, its profits are a drop in the ocean.
[Marvin Holdman] “When you look at what they chose to launch with as a 3rd party app, Automatic Duck, I think we’re seeing what they envision their partners to be like. Higher priced plug-ins for more specialized applications. I don’t know for sure, but I would also suspect that App Store sales by these venders put a percentage of the sale in Apples pocket. Higher prices, higher profit for Apple for being nothing more than the gatekeeper.”
Yes, they want to pass off niche functionalities to third parties (though OMF is arguably not so niche). Regardless, Automatic Duck was never for sale via the App Store. And I doubt Apple had any say in what the price was. If anything, Apple’s attitude about software pricing is to make prices disruptively low.
As I type this, I see no accessory apps (aside from tutorials) for FCPX on the App Store outside Apple’s own Motion and Compressor. I do agree there is a race-to-the-bottom effect associated with pricing FCPX at $299, but in the world of AE and Pro Tools, it is common for plug-ins to cost more than the app they are sold for. If Apple has some nefarious plan for making fistfuls of cash off the FCPX ecosystem, there isn’t even a whiff of that happening yet despite plugins and helper apps already available for sale outside the App Store.
Besides, a lot of the third party software for use with FCPX will be FREE. DaVinci Lite? AJA Control Room? BlackMagic Media Express? You can bet the forthcoming camera-native codec support will be gratis as well. The stuff they are leaving to third parties will not cost extra very often. $500 Automatic Duck OMF support will be the exception, not the rule.
[Marvin Holdman] “The more time passes by, the more it seems that FCPX was designed in the boardroom by the marketing department. The fact that it seems to advocate a less functional core with a much higher degree of 3rd party dependency seems unrealistic for an enterprise level application. “
I’ll agree that product marketing executives have a lot to do with steering product development, but the way Apple seems to be handling development for FCPX looks very shrewd to me. Think about it: what were FCP’s big development shortcomings in the past? Codec support limitations that had to wait for major revs every 18-24 months to address. Brittle legacy QuickTime frameworks that were not designed to handle the rigors of accurate video. Now? Apple is actually baking broadcast video muscle into the OS frameworks. Instead of asking the Pro Apps team to deal with low-level stuff like new codecs, they are now free to work on high-level features.
I don’t think the thing that makes rapid development practical is the App Store as we’ve heard a few times from Richard Townhill. I think they get rapid development because they have a proper foundation to build on. If they deliver on their promise for the next version, they will have achieved critical feature parity with FCP7 in less than a year. That’s impressive.
[Marvin Holdman] “I don’t remember anyone asking for further proliferation of 3rd party vendors in their editing software? It certainly wasn’t a feature that I even remotely imagined, but then again, I’m not on the board.”
What is so bad about building a platform? FCP was as successful as it was because it was so open to third party enhancement at a time when Avid only worked with Avid hardware (right down to the hard drives they “qualified” and marked up by 300%). Apple has been adamant that they think an ecosystem is better than a monolith. Maybe no one was blogging about how much they wished Apple would leave tape transport to AJA and BMD, but there were plenty of people complaining about how spotty the Edit to Tape function was in legacy FCP.
Best,
Andy -
[Chris Harlan] “I’m wondering if I transferred to a MBAir, and through the course of the day, as I sat waiting to talk to clients, at lunch, waiting in the car pool line to pick up my daughter, etc., I could make a little headway on my selects, and then, plug the MBair in via target drive to my 8 or whatever core and use it as a disk. It is SSD. TBolt DOES support target mode. I wonder if I could get enough speed.”
If you don’t mind cracking open the case, or if Apple adopts 6G SATA SSDs, you will be able to get some very impressive speeds.
Best,
Andy -
[Bret Williams] “I didn’t know anywhere was two weeks from anywhere.”
Some contents or functionalities here are not available due to your cookie preferences!This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.
https://youtu.be/Tw_YryVgLOgBest,
Andy -
[Morten Ranmar] “Does anybody know if it will work with a File Shared Volume? ( I have a shared storage solution using a Small Tree ethernet server card)”
Larry Jordan reported that NAS also works.
Best,
Andy -
Here is the developer documentation for the old XMEML DTD and the new FCPXML DTD.
Godspeed.
Best,
Andy -
Well, it seems AXEL ain’t the same as FCPXML. So I was wrong to conflate the two. AXEL is some other thing according to Philip Hodgetts.
Best,
Andy -
[James O'Connor] “Great – but does this mean I have to have Apple’s Fibre Channel system for Xsan or can I find a way to get my GraniteSTOR by Small Tree Networks (GBE and SATA) to host FCPX Events ***and now, more importantly, PROJECTS*** on the server?”
According to Larry Jordan, it is NOT Xsan exclusive.
Best,
Andy -
[Bob Zelin] “I am boring you with this, because Apple obviously is interested in EASY (for dopes like me), and FCP X defines this, as well as OS X Lion, and OS X Lion server (and iOS in general). SO, won’t a future XSAN configuration meet with this Apple mindset – or will it remain as complex as it is today ? There is a Mac Mini Server (with only one ethernet port, but Thunderbolt) – so won’t there be a “future” for XSAN on a Mac Mini ? Or will it remain the current nightmare configuration ? “
Somebody will need to make fibre channel switches easy for that to happen. It is technically possible that someone could build an “Xsan Hub” that uses Thunderbolt to carry both the FC and metadata LAN links that Xsan requires to run and requires minimal configuration to be used in an Xsan deployment, but I’d be STUNNED if one came from Apple. Maybe they will coax a partner to build one. But for now, the only way to build an Xsan still involves zoning an off-the-shelf fibre channel switch, setting up a discrete metadata LAN alongside the public LAN, and setting up the right cocktail of RAID LUNs to be made into an Xsan filesystem.
But yeah, it sure seems like they are up to something, doesn’t it?
Best,
Andy -
[Bob Zelin] “If you have a Mac Mini, and the Promise Thunderbolt to Fibre adaptor (non yet released) – can you run an XSAN enviornment – or do you have to to the WHOLE EXPENSIVE COMPLEX THING – with the 2 servers, dual switches, plus fibre cards in every MAC Pro, no iMAC’s, no MacMini clients, no MacBook Pro clients ? None of this is made clear. “
Short answer: it has to stay complicated.
Long answer: Xsan is essentially StorNext, and for a client to participate in a StorNext filesystem, the fibre channel switching is essential. Xsan probably won’t let you, but StorNext can actually make a SNFS filesystem on internal disks. But if you want more than the MDC to be able to access the filesystem as well, that’s where the switch comes in. Xsan/StorNext clients need block-level access to the disk for it to work. The only way to get multiple clients seeing disk at the block level is via switched block-level protocols like fibre channel.
[Bob Zelin] “The idea of having to buy a Promise VTrak, dual servers, Fibre cards for every MAC Pro, and a QLogic Fibre switch kind of defeats the purpose of a “simple” SAN from Apple.”
You don’t need redundant MDCs to run an Xsan filesystem, but redundant MDCs is the de facto standard install. Every client does need to have a fibre channel HBA of some kind (be it a card or a Thunderbolt adapter). The most complicated part of Xsan by far is the FC switch.
[Bob Zelin] “And as this forum demonstrates on a daily basis – there are LOTS AND LOTS of simple, easy to use, and inexpensive SAN’s already out there.”
It’s hair-splitting, but some of the “SANs” you’re referring to are really NAS. Technically, Xsan is a shared disk clustered filesystem atop a SAN.
But for video editing purposes, none of that really matters. Video editing doesn’t require block-level access to work. File-level access via TCP protocols works just fine as long as the bandwidth is there.
[Bob Zelin] “So what secret (if any) is up Apple’s sleeve ? “
I too am puzzled about Apple’s strategy for Xsan. It pretty much can’t be dead simple due to the required infrastructure for that type of SAN, but now they are giving the software away with Lion. Thunderbolt makes it possible to physically connect to Xsan with iMacs, Mac minis, and MacBooks, but that doesn’t make it easy- you still have to set up and zone a very complicated fibre switch. What’s up their sleeve? The magic eight ball says “cannot predict now”.
Best,
Andy