VisualHub

So, John ‘Tell me what to think, John!’ Gruber links to VisualHub, a new video compression front-end by the author of the brilliantly simple video-to-iPod iSquint. For some reason, I start reading the manual – I think because if this NESTA project happens, I’m going to be encoding potentially hundreds of short videos, so things like batch operation are important to me. Apple’s Compressor is rather good, but I’m open to options.

And there, towards the back of the manual, is an entire section titled ‘Xgrid.’

Oh. My. Heavens. This is genius.

See… if this project happens, I’m going to end up with a nice little pile of MacBooks (yes, I could in principle do video production workshops in schools with Windows, but without iMovie and GarageBand, where’s the fun? So, MacBooks). MacBooks, you will recall, have fairly quick CPUs. Core’s SSE3 still doesn’t seem to be a match for Altivec for video compression, but there’s an ace up the sleeve – two processors per MacBook. And they also support gigabit ethernet.

‘If only,’ I find myself saying, ‘if only I could farm out batch compression jobs across the network.’

…which is where Xgrid comes in: it’s Apple’s clustering technology. It’s been part of Mac OS X for a while now, but I don’t think I’ve ever used anything that takes advantage of it. VisualHub does. It would have been (relatively) easy to bundle up ffmpeg and all the other open-source bits in a pretty Aqua wrapper, but Xgrid support takes VisualHub from ‘meh!’ to ‘woah!’ The 4Gb file limit is a bit of a git – movies originated as DV and longer than about 16 minutes need not apply – but for my project that’s not an issue.

Oh happy, happy day.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.