I'm not trolling, I have a honest question- What do I get with Zencoder, versus the free Transcoding I get through Softlayer (http://bit.ly/hmDGtu)

Even if I didn't already have a Softlayer server, it seems like it'd be cheaper to get one of those then hire Zencoder for many cases.

Fair question. What you get is scalability (Softlayer is explicitly for low-volume transcoding), an API (Softlayer might have one, though it's not documented at the above link), a lower error rate, higher visual quality, better compression, more options, and hands-on customer support. Not everyone needs these things, but many people do.

I would think along the same lines. Then recently I was doing contract work that included ability to upload and preview video files. My client had a very average vps and was not willing to spend much on having me install/test a bunch of packages to get video conversion going.

Bring in zencoder and I had video uploads working for him in a couple hours.

(Granted, I think someone who knows their stuff when it comes to ffmpeg etc. could very well do this in two hours without zencoder, I think. Nonetheless, this is one use case.)

I'm using Zencoder on a new project, and I've been deeply impressed at both the ease of getting started, and the quality of the tech support. Their API builder lets me fill out a form, and it automatically displays the HTTP headers and the necesary JSON.

There's a certain value to simplicity.

This isn't as interesting as I hoped it would be from reading the title. It looks like the only technical aspect of this is caching the input file, which makes sense but isn't very exciting. Is there any work being done on sharing some of the encoding work between different resolutions of the same codec? I don't know a lot about video codecs but it seems like you might be able to encode the higher resolution video and save some of the intermediate data to use with the lower resolution video to at least provide hints.

This is mostly about the business side of things. Many of our customers are encoding to multiple formats to deliver a better user experience and we wanted our pricing to adapt.

Technically, we encode all of the outputs for a file in parallel across multiple transcoding servers. We're looking for ways to reduce the number of steps involved in moving an input file through the system which may result in some sort of forking of the decode to multiple simultaneous encodes, but that remains to be seen.