Elgato’s Turbo.264 HD (Turbo.264 HD Software Edition, which performs all of the same functions as the combo but without the accelerated speeds and ability to provide adaptive streaming from Elgato’s EyeTV sofware.) is hardware/software combo for accelerated video transcoding. Recently, Elgato released the sofware portion separately as the $50
The software accepts AVCHD, QuickTime, AVI, DV, WMV, MPEG-1, MPEG-2 Program/Transport Stream, MPEG-4, MP4, M4V, .MTS, H.263, H.264 AVC, Xvid, and Video_TS input. As the name suggests, it converts to H.264 video, and includes presets for encoding for iPad, iPhone, iPod, Apple TV, YouTube, YouTube HD, 720p, 1080p, and Sony PSP (and you can customize presets to your liking).
To test out the software-only version, I encoded the same three items using the software only, the software/hardware combo, and the 32-bit version of HandBrake 0.9.4 (The Party ) from an unencrypted Video_TS folder ripped using RipIt ( ) to iPad format, all using the presets for each target (HandBrake doesn’t have an iPad preset, so I used the Apple TV preset for the VideoTS folder). I also tested the direct camcorder conversion with 2 minutes and 26 seconds of video from a Canon Vixia HF100 1080p AVCHD camcorder.). My tests were converting a 21:08 AVI episode of How I Met Your Mother to iPhone format, a 43:19 720p MKV episode of Glee to Apple TV format, and a 1:38:53 movie (
All testing was performed on the latest 13-inch MacBook Pro/2.4GHz () running OS X 10.6.4.
It should come as no surprise that the software alone was much slower than the software aided by the USB stick hardware. The AVI file and Video_TS folder took almost twice as long to encode with the software, and the MKV file took more than three times as long. Interestingly, the data rates were higher with the software-only encodes, resulting in larger file sizes. (Elgato says the encoder chip “tends to be too conservative” and the company is working on new firmware to make the hardware act more like the software in that regard.)
Converting the same files with HandBrake was all over the map. The AVI file took about 30 seconds less to convert than the Turbo.264 HD software, while the MKV file took 15 minutes longer and the movie took 30 minutes longer. HandBrake also uses lower data rates, so the files it created were significantly smaller than those made by the Turbo.264 HD is most cases. (It’s also worth noting that the Elgato software kept the Glee episode at its native 1280 by 720 resolution, while the Apple TV preset in HandBrake converted it to 960 by 544.)
To try out the camcorder conversion, I launched the software, popped the SD card from my camcorder into the built-in card reader in my MacBook Pro, and clicked the Add Camcorder button that appears when it recognizes a camcorder to access the contents of my SD card. (Elgato says you can connect a USB camcorder to your Mac, but I tried the direct SD card route and was happy to see that it worked. And in fact, if you copy the contents of the SD card onto your hard drive, you can drag in the clips and work with them in the exact same way.) You can edit clips before you convert them, using the same interface as in Elgato’s EyeTV software.
You can choose which clips to convert (removing those you don’t want) as well as combine multiple clips into a single movie. And new in the latest software version is Events, a feature that looks for clips shot around the same time and groups them into a single clip. I turned Events on and it grouped the six clips I shot of my twins having lunch into one, which I then converted using the HD 720p preset. The software-only version took nearly 78 percent longer to encode than the hardware-assisted version—9 minutes and 48 seconds versus 5 minutes and 31 seconds. (I ran the conversion both directly from the SD card and from the local files copied to my hard drive—the results of the two were almost identical).
Unlike iMovie, Turbo.264 HD can access AVCHD content directly, without first converting it. As anyone who’s “imported” AVCHD video in iMovie knows—I use the quotation marks because what iMovie is actually doing is transcoding the video to the Apple Intermediate Codec, which makes the file size balloon in the process—it can take a long time and a lot of processing power to do so. This direct access and conversion is a great feature as long as you don’t care about taking advantage of iMovie’s transition and titling niceties.
One note: as the software encodes any video, frames of the video scroll by, and there’s currently no way to turn it off. This can be distracting—not to mention spoiler-inducing if you haven’t seen the video you’re encoding yet.
The results of encoding using the Turbo.264 HD software and with the hardware version were virtually identical—I couldn’t see any difference between the two. However, between the Elgato and HandBrake encodes, the Turbo.264 HD (both software and hardware) showed blocky pixelization in the exact same places during certain scene transitions and motion in the MKV episode. The HandBrake version was perfectly smooth in those same places.
Elgato says the problem lies with a bug in the CoreAVC H.264 codec the company licensed from CodeCodec (and only with a certain type of MKV file). The company says it will be replacing the codec with the H.264 decoder from MainConcept in a free update soon. I tested the same MKV video file with a pre-release version of the update and it converted perfectly without any of the blocky pixelization I’d observed earlier.
Unlike with earlier versions of the software, the audio and video were in sync in all of my tests.
Macworld’s buying advice
The Turbo.264 HD Software Edition performs the same functions as the standard, hardware-enabled version, but at half the price. But that price difference also buys you much faster encoding and the ability to use the Turbo.264 HD hardware with other video apps that support its acceleration. If you’ll use the Turbo.264 HD on a regular basis, the $100 hardware version might be a better deal in the long run. (If you want to see if you like the software, you can download a trial version that’s limited to three-minute encoding and includes a watermark.)
The larger question is whether $50 is worth it for an application whose feature set can largely be replicated by other options that cost less or, in some cases, nothing at all. If you deal with lots of AVDHD footage and don’t plan to do lots of complex editing and such, the answer may be yes. And the ability to trim videos before encoding is a nice plus. Otherwise, you may get by just fine with software such as the free Video Monkey, the 20 euro iFlicks, or others.