Elemental Badaboom Overview and Benchmarks

Elemental Badaboom Overview

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 48
Main screen

Last, but certainly not least, we have Elemental’s Badaboom application. Like many of the applications we are evaluating today, Elemental Technologies’ Badaboom media converter uses CUDA technology to accelerate video transcoding. But, this program does not allow CUDA to be disabled so consumers are required to have an NVIDIA GeForce 8 series or later graphics card to use this software.

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 49
Video decoder options in settings menu

After meeting the system requirements, end users can use Badaboom to convert their videos for portable media devices. Elemental claims that anyone can painlessly convert video between formats, and we will evaluate how their user interface performs and functions in real-world benchmarks using movie clips and other videos in different formats.

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 50
Video encoder options in settings menu

By taking advantage of the multiple stream processors in NVIDIA GPUs—up to 240—the transcoding speed is greatly improved compared to traditional CPU-based software solutions. For computers with multiple CUDA-enabled GPUs in a system, users can launch one Badaboom for each GPU in the system and transcode multiple videos at the same time.

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 51
Audio options in settings menu

Badaboom can support multiple video formats including DivX, Xvid, MPEG-1, VC-1, AVI, MKV, MOV, MP4, WMV, FRAPS, and AAC Audio. One very useful feature for newcomers to video transcoding are available pre-configured user profiles to output for 12 types of devices including the iPhone, iPod Touch, YouTube, Blackberry Bold, Microsoft Zune, Apple TV, and several others.

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 52
Picture options in settings menu

Badaboom’s H.264 encoder and 2 channel stereo output are specifically targeted at portable media devices such as the Apple iPod and Sony Playstation Portable. Users can also convert video from multiple sources like DVDs, Video_TS folders, and other video files on your hard drive.

Here’s a quick rundown of other features available with Badaboom:

  • Easy quality adjustment – A simple slider mechanism controls the quality of final image

  • Efficiently use all of your PCs resources – Harness the GPU for transcoding while still being able to use the CPU for other tasks

  • Multiple simultaneous transcodes – If you have more than one GPU, you can launch multiple Badabooms to see the fast performance on each movie transcode

  • Auto-open converted files in iTunes – After transcoding is complete files can be automatically moved into iTunes for simple syncing
     
  • DVD title selection and preview – Only transcode the titles you want Automatic up and down scaling – Input videos are scaled up or down to match your current output device settings

 

Elemental Badaboom Benchmarks 

For our final benchmarks, we are once again tag teaming Handbrake against Badaboom to evaluate trancoding times and CPU usage. We didn’t add any video filters and effects. We just used the pre-configured device profiles to output video to the iPhone, YouTube, and the iPod Touch. The outputted formats include 640×480 MPEG-2, AVCHD 720×480 MPEG-4, and AVCHD 1920×1080 MPEG-4.

 

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 53
Overall transcoding times

 

NVIDIA CUDA-enabled Applications Roundup - Graphics Cards 54
Average CPU usage during transcoding

Finally, we are seeing some better CPU usage scores. Badaboom chewed through all of our benchmarks and gave us the best CPU usage scores we’ve seen so far. The most impressive CPU usage score came during the YouTube 1920×1080 MPEG-4 transcode that averaged only 9 percent! That’s absolutely outstanding considering most of the other applications we’ve reviewed were over 50 percent and higher. Handbrake was pretty much pegged out at 99 percent for every benchmark and was also much slower in transcoding every benchmark.

We’ve got to commend Elemental for paying attention to how much the CPU was being engaged during transcoding. There’s a bit debate right now between getting the best transcoding performance from utilizing both the CPU and GPU and or just using the GPU and leaving the CPU for multi-tasking functions. I can see it going either way depending on the user’s needs. 


« PreviousNext »