Blackmagic Design Teranex Processors

df2616_teranex_sm

In recent years, Blackmagic Design has thrived on a business model of acquiring the assets of older industry icons, modernizing their products, and then re-introducing these cornerstone brands to an entirely new customer base. Top-of-the-line products that were formerly out of reach to most users are now attainable, thanks to significant price reductions as part of the Blackmagic Design product family.

Teranex is just such a case. It’s a company with whom I am well acquainted, since we are both Orlando-based. I remember their first NAB off-site, whisper suite. I’ve used their conversion and restoration products on a number of projects. At one point they were moving into the consumer TV space under the then ownership of Silicon Optix and later IDT. In that period, I produced the popular HQV Benchmark DVD and Blu-ray for them as an image test vehicle for consumers. As with many companies in the pro video space, they’ve had a past filled with ups and downs, so it’s great to see Blackmagic breathe new life into the technology.

Teranex Processors

Blackmagic Design offers three rack-mounted Teranex products. These are separate from the Teranex Mini line, which does not offer the full range of Teranex processing, but is comprised of more targeted units for specific conversion applications. The rack-mounted standards converters include the Teranex Express, Teranex 2D Processor and Teranex 3D Processor. All three offer more or less the same processing options, with the exception that the Express can work with 4K Ultra HD (3840×2160). The 2D and 3D Processors only go as high as 2K (2048×1080). Outside of that difference, they all handle up/down/cross-conversions between SD (NTSC and PAL), HD (720 and 1080), 2K and UHD (Express only). This includes frame sizes, as well the whole range of progressive and interlaced frame rates. There’s also aspect ratio correction (anamorphic, 16:9, 14:9, zoom, letterpox/pillarbox) and colorspace conversion. Add to this de-interlacing and 3:2 pulldown cadence correction. The key point, and why these units are must-haves for large post operations, is that they do it all and the processing is in real-time.

The Teranex 2D and 3D Processors can also function as i/o devices when connected via Thunderbolt. By purchasing one of these two models, you can skip the need for an additional Blackmagic Design capture device, assuming you have a Thunderbolt-equipped Mac Pro, iMac or MacBook Pro. With the purchase, you also get Blackmagic Design Ultrascope waveform monitor software that runs on your computer. When you run one of these units with Apple Final Cut Pro X, Adobe Premiere Pro CC or their own Media Express application, the response is the same as with a standalone i/o device. This is an optional use, however, as these two units can operate perfectly well in a standalone installation, such as part of a machine room environment. They do tend to have loud fans, so either way, you might want to keep them in a rack.

The biggest difference between the 2D and 3D Processor is that the 3D unit can also deal with stereoscopic video. In addition to the normal processing functions, the 3D model has adjustments for stereo images. There are also physical differences, so even if you don’t work with stereo images, you might still opt for the Teranex 3D Processor. For instance, while both units can handle analog or SDI video connections, the 3D Processor only allows for two channels of analog audio i/o to be plugged into the device. The 2D Processor uses a separate DB-25 break-out cable for all analog audio connectors. Like all Blackmagic rack products, no power plug is included. You need to provide your own three-prong electric cord. The 3D Processor features dual-redundant power supplies, which also means it requires two separate power cords. Not a big deal given the extra safety factor in mission-critical situations, but an extra consideration nonetheless. (Note: the 3D processor still works with only a single power cord plugged in.)

The Teranex Express, is more streamlined, with only digital SDI connectors. It is designed for straightforward, real-time processing and cannot also be employed as an i/o device. If you don’t need analog connectors, stereoscopic capabilities, nor Thunderbolt i/o, then the Express model is the right one for you. Plus, it’s currently the only one of the three that works with 4K Ultra HD content. The Teranex units also pass captioning, Dolby data and timecode.

In actual use

I tested both the Teranex Express and the 3D Processor for this review. I happen to have some challenging video to test. I’m working on a documentary made up of a lot of standard definition interviews shot with a Panasonic DVX-100, plus a lot of WWII archival clips. My goal is to get these up to HD for the eventual final product. As a standards converter and image processor both units work the same (excluding stereoscopic video). SDI in and SDI out with a conversion in-between.

The front panel is very straightforward, with buttons for the input standard and the desired output standard. The left side features settings for format size, frame rate, scan and aspect. The middle includes a multi-use LCD display, which is used to show menus, test patterns and video. To the right of the display are buttons for video levels and sharpening, since these models also include a built-in proc amp. Finally on the far right, you can see audio channel status, system status and presets. Last, but not least, there’s a “lock panel” button if you don’t want anyone to inadvertently change a setting in the middle of a job, as these controls are always active. When you pass any SDI signal through one of these units, the input is auto-detecting and the button layout easily guides the operator through the logical steps to set a desired target format for conversion.

As with all Blackmagic Design products, installation of the software needed for i/o was quick and easy. When I connected the Teranex 3D Processor to my MacBook Pro via Thunderbolt, all of the apps saw the device and for all intents and purposes it worked just the same as if I’d had a Blackmagic Design UltraStudio device connected. However, here the conversion side of the Teranex device is at odds with how it works as an i/o device. For example, the output settings typically followed the sequence settings of the NLE that was driving it. If I had an NTSC D1 timeline in Final Cut Pro X, the Teranex 3D Processor could not be set to up-convert this signal on output. It only output a matching SD signal. Up-conversion only happened if I placed the SD content into a 1080 timeline, which unfortunately means the software is doing the conversion and not the Teranex processor. As best as I could tell, you could not set the processor to override the signal on either input or output when connected via Thunderbolt.

Processing power

One of the hallmarks of Teranex processing is cadence correction. 24fps content that is recorded as a 30fps signal is said to have “3:2 pulldown”. It was originally developed to facilitate transferring film material to videotape. Pulldown is a method of repeating whole film frames across a pattern of interlaced video frames so that four film frames can fit into five video frames (ten fields). This pattern is called 24PN (“normal” pulldown) and the cadence of film frames to video fields is 2:3:2:3. Digital camera manufacturers adopted this technique to mimic the look of film when recording in a 24fps mode. To complicate matters, Panasonic introduced a different cadence called 24PA, or “advanced” pulldown. The cadence is 2:3:3:2 and was targeted at Final Cut Pro users. FCP featured a built-in routine for the software to drop the extra frame in the middle and restore the clips to a true 24fps during a FireWire capture. Another form of cadence is 2:2:2:4, which is common in DVD players when playing back a true 24fps DVD.

In the case of Teranex processing, it is designed to detect and correct the more common 24PN, i.e. 3:2 pulldown (2:3:2:3), but not the other two cadences (2:3:3:2, 2:2:2:4). Teranex is supposed to be able to fix “broken” 3:2 pulldown cadences in mixed timelines, meaning the pattern changes at every cut. However, when I checked this on my test project, I didn’t get perfect results. That’s most likely due to the fact that I was dealing with DV (not proper D1) content, which had gone through a lot of hands before it came to me. The best results would be if I treated every source clip individually. When I test that, the results were more what I expected to see.

Teranex technology was developed for real-time processing at a time when linear, videotape post ruled. Today, there are plenty of high-quality, non-real-time, software processing options, which yield results that are very close to what Teranex can deliver. In the case of my test project, I actually found that dealing with interlace was best handled by Blackmagic’s own DaVinci Resolve. I don’t necessarily need to get back to 24fps, but only to get the cleanest possible 30fps image. So my first target was to convert the 29.97i clips into a good 29.97p sequence. This was possible through Resolve’s built-in de-interlacing. Progressive frames always up-convert with fewer artifacts than interlaced clips. Once I had a good 29.97p file, then I could test the Teranex conversion capabilities.

I tested conversions with several NLEs, Resolve, After Effects and the Teranex hardware. While each of the options gave me useable HD copies, the best overall was using the Teranex unit – passing through it in real-time via SDI in and out. Teranex not only gave me cleaner results, as evidenced by fine edges (less “jaggies”), but I could also dial in noise reduction and sharpening to taste.

All processing is GIGO (garbage in, garbage out). You can never make awful DV look stunning in HD, much less 4K. It’s simply not possible. However, Blackmagic Design’s Teranex products give you powerful tools to make it look the best that it can. Software processing can get you close, but if fast turnaround is important, then there’s no replacement for real-time processing power. That’s where these Teranex processors continue to shine.

Originally written for Digital Video magazine / Creative Planet Network.

©2016 Oliver Peters

Adobe’s Summer 2016 Refresh

df2516_adobe_sm

Adobe is on track for the yearly refresh of its Creative Cloud applications. They have been on a roll with their professional video solutions – especially Premiere Pro CC – and this update is no exception. Since this is not a new, across-the-board Creative Cloud version update, the applications keep the CC 2015 moniker, except with a point increase. For example, Premiere Pro CC becomes version 2015.3, not CC 2016. Let me dive into what’s new in Premiere Pro, Audition, Adobe Media Encoder and After Effects.

Premiere Pro CC 2015.3

Adobe has captured the attention of the professional editing community with Premiere Pro and has held it with each new update. CC 2015.3 adds numerous new features in direct response to the needs of editors, including secondary color correction, a proxy workflow, a 360VR viewer and more.

New Lumetri features

df2516_lumetriThe Lumetri color panel brought over the dominant color correction tools from SpeedGrade CC configured into a Lightroom-style panel. For editors, Lumetri provides nearly everything they need for standard color correction, so there’s rarely any need to step outside of Premiere Pro. Three key features were added to Lumetri in this update.

First is a new white balance eyedropper. Lumetri has had temperature and tint sliders, but the eyedropper makes white balance correction a one-click affair. However, the new marquee feature is the addition of SpeedGrade’s HSL Secondary color correction. Use an eyedropper to select the starting color that you want to affect. Then use the “add” or “remove color” eyedroppers to adjust the selection. To further refine the isolated color, which is essentially a key, use the HSL, denoise and blur sliders. The selected color range can be viewed against black, white or gray to check the accuracy of the adjustment. You can then change the color using either the single or three-wheel color control. Finally, the secondary control also includes its own sliders for temperature, tint, contrast, sharpening and saturation.

In the rest of the Lumetri panel, Adobe changed the LUT (color look-up table) options. You can pick a LUT from either the input and/or creative tab. The new arrangement is more straightforward than when first introduced. Now only camera gamma correction LUTs (like ARRI Log-C to Rec 709) appear in the input tab and color style LUTs show up in the creative tab. Adobe LUTs plus SpeedLooks LUTs from LookLabs are included as creative choices. Previously you had to use a SpeedLooks camera LUT in tandem with one of the SpeedLooks creative LUTs to get the right correction . With this update, the SpeedLooks creative LUTs are all designed to be added to Rec 709 gamma, which makes these choices far more functional than before. You can now properly use one of these LUTs by itself without first needing to add a camera LUT.

New Proxy workflow

df2516_proxyApple Final Cut Pro X users have enjoyed a proxy workflow since its launch, whereas Adobe always touted Premiere Pro’s native media prowess. Nevertheless, as media files get larger and more taxing on computing systems, proxy files enable a more fluid editing experience. A new ingest tool has been added to the Media Browser. So now from within Premiere Pro, you can copy media, transcode to high-res file formats and create low-res proxies. You can also select clips in a bin and right-clip to create proxies, attach proxies and/or relink full-resolution files. There is a new toggle button that you can add to the toolbar, which lets you seamlessly flip between proxy and full-resolution media files. According to Adobe, even if you have proxy selected, any export always draws from the full-resolution media for the best quality.

Be careful with the proxy settings. For example, one of the default sizes is 1024×540, which would be the quarter-frame match for 2K media. But, if you use that for HD clips in a 1920×1080 timeline, then your proxies will be incorrectly pillar-boxed. If you create 720p proxies for 1080p clips, you’ll need to use “scale to frame size” in order to get the right size on the timeline. It’s a powerful new workflow, but take a bit of time to figure out the best option for your needs.

Adobe Media Encoder also gains the Media Browser tool, as well as a new ingest function, which has been brought over from Adobe Prelude. Now you can use Media Encoder to copy camera files and/or transcode them to primary and secondary locations. If you need to copy camera cards, transcode a full-res master file and also transcode a low-res proxy file, then this complete workflow can be handled through Media Encoder.

New 360VR viewer

df2516_360Premiere Pro CC now sports a new VR-capable viewer mode. Start with monoscopic or stereoscopic, stitched 360-degree video clips and edit them as you normally would. The viewer allows you to pan around inside the clip or view the timeline from a point of view. You can see what someone viewing with goggles sees when looking in a given direction. Note that this is not a pan-and-scan plug-in. You cannot drop one of these 360-degree clips into an otherwise 2D 16×9 (“flat”) timeline and use Premiere Pro’s VR function to keyframe a digital move within that clip.

There are other new Premiere Pro CC features that I haven’t yet tested thoroughly. These include new support for Apple Metal (an API that combines the functionality of OpenGL and OpenCL) and for grading control surfaces. Open Caption support has been improved – adding more languages and their native alphabets, including Arabic and Hebrew.

Adobe Audition CC 2015.2

df2516_auditionWant better audio mixing control than what’s available inside of Premiere Pro CC? Then Audition CC is the best tool for the job. Premiere Pro timelines translate perfectly and in the last update a powerful retime feature was added. Audition “automagically” edits the duration of a music cue for you in order to fit a prescribed length.

The Essential Sound panel is new in this update. The layout of this panel is the audio equivalent to the Lumetri color panel and also owes its design origins to Lightroom. Select a clip and choose from the Dialogue, Music, SFX or Ambience group. Each group presents you with a different, task-appropriate set of effects presets. For example, when you pick Dialogue, the panel will display tabbed controls for loudness, repair sound, improve clarity and a creative tab. Click on a section of the vertical stack within this panel to reveal the contents and controls for that section.

In the past, the workflow would have been a roundtrip from Premiere Pro to Audition and back. Now you can go directly to Adobe Media Encoder from Audition, which changes the workflow into these steps: cut in Premiere Pro CC, mix in Audition CC, and master/export directly through Adobe Media Encoder. Thus roundtrips are eliminated, because picture is carried through the Audition phase. This export path supports multichannel mix files, especially for mastering containers like MXF. Audition plus Media Encoder now enable you to export a multichannel file that includes a stereo mix plus stereo submix “stems” for dialogue, SFX and music.

After Effects CC 2015.3 and more

df2516_aeAfter Effects CC has been undergoing an overhaul through successive versions, including this one. Some users complained that the most recent version was a bit of a step backwards, but this is all in an effort to improve performance, as well as to modernize and streamline the product. From my vantage as an editor who uses After Effects as much as a utility as for occasional motion graphics and visual effects, I really like what Adobe has been doing. Changes in this update include enhanced performance, GPU-accelerated Gaussian blur and Lumetri color correction, better playback of cached frames, and a new a/v preview engine. In the test projects that I ran through it, including the demo projects sent by Adobe, performance was fast and rather impressive. That’s on a 2009 Mac Pro tower.

If you are an animator, then Maxon Cinema 4D is likely a tool that you use in conjunction with After Effects. Animated text and shape layers can now be saved directly into the Cinema 4D file format from After Effects. When you customize your text and shapes in Cinema 4D, the changes are automatically updated in After Effects for a roundtrip 3D motion graphics workflow.

Thanks to the live The Simpsons event, in which Homer was animated live using Character Animator, this tool is gaining visibility. Character Animator moves to version 4, even though the application is still technically in prerelease. Some of the enhancements include improved puppet tagging. You can record multiple takes of a character’s movement and then enable your puppet to respond to motion and trigger animation accordingly.

To wrap up, remember that Adobe is promoting Creative Cloud as more than simply a collection of applications. The subscription includes access to over 50 million royalty-free photos, illustrations, vector graphics and video (including 4K clips). According to Adobe, licensed Adobe Stock assets in your library are now badged for easy identification. Videos in your library are displayed with duration and format information and have links to video previews. You can access your Libraries whenever you need them, both when you are connected to the internet and working offline. I personally have yet to use Adobe Stock, but it’s definitely a resource that you should remember is there if you need it.

Click here for Dave Helmly’s excellent overview of the new features in Premiere Pro CC.

Originally written for Digital Video magazine and Creative Planet Network.

©2016 Oliver Peters