Blackmagic Design eGPU

Power users have grown to rely on graphics processing units from AMD, Intel and Nvidia to accelerate a wide range of computational functions – from visual effect filters to gaming and 360VR, and even to bitcoin mining. Apple finally supports external GPUs, which can easily be added as plug-and-play devices without any hack. Blackmagic Design just released its own eGPU product for the Mac, which is sold exclusively through Apple ($699 USD). It requires macOS 10.13.6 or later, and a Thunderbolt 3 connection. (Thunderbolt 2, even with adapters, will not work.)

The Blackmagic eGPU features a sleek, aluminum enclosure that makes a fine piece of desk art. It’s of similar size and weight to a 2013 Mac Pro and is optimized for both cooling and low noise. The unit is built around the AMD Radeon Pro 580 GPU with 8GB of video memory. It delivers 5.5 teraflops of processing power and is the same GPU used in Apple’s top-end, 27” Retina 5K iMac.

Leveraging Thunderbolt 3

Thunderbolt 3 technology supports 40Gb/s of bandwidth, as well as power. The Blackmagic eGPU includes a beefy power supply that can also power and/or charge a connected MacBook Pro. There are two Thunderbolt 3 ports, four USB3.1 ports, and HDMI. Therefore, you can connect a Mac, two displays, plus various USB peripherals. It’s easy to think of it as an accelerator, but it is also an appliance that can be useful in other ways to extend the connectivity and performance of MacBook Pros. Competing products with the same Radeon 580 GPU may be a bit less expensive, but they don’t offer this level of connectivity.

Apple and Blackmagic both promote eGPUs as an add-on for laptops, but any Thunderbolt 3 Mac qualifies. I tested the Blackmagic eGPU with both a high-end iMac Pro and the base model 13” 2018 MacBook Pro with touch bar. This model of iMac Pro is configured with the more advanced Vega Pro 64 GPU (16GB VRAM). My main interest in including the iMac Pro was simply to see whether there would be enough performance boost to justify adding an eGPU to a Mac that is already Apple’s most powerful. Installation of the eGPU was simply a matter of plugging it in. A top menu icon appears on the Mac screen to let you know it’s there and so you can disconnect the unit while the Mac is powered up.

Pushing the boundaries through testing

My focus is editing and color correction and not gaming or VR. Therefore, I ran tests with and without the eGPU, using Final Cut Pro X, Premiere Pro, and DaVinci Resolve (Resolve Studio 15 beta). Anamorphic ARRI Alexa ProRes 4444 camera files (2880×2160, native / 5760×2160 pixels, unsqueezed) were cut into 2K DCI (Resolve) and/or 4K DCI (FCPX, Premiere Pro) sequences. This meant that every clip got a Log-C LUT and color correction, as well as aspect ratio correction and scaling. In order to really stress the system, I added several GPU-accelerated effect filters, like glow, film grain, and so on. Finally, timed exports went back to ProRes 4444 – using the internal SSD for media and render files to avoid storage bottlenecks.

Not many applications take advantage of this newfound power, yet. Neither FCPX nor Premiere utilize the eGPU correctly or even at all. Premiere exports were actually slower using the eGPU. In my tests, only DaVinci Resolve gained measurable acceleration from the eGPU, which also held true for a competing eGPU that I compared.

If editing, grading or possibly location DIT work is your main interest, then consider the Blackmagic eGPU a good accessory for DaVinci Resolve running on a MacBook Pro. As a general rule, lesser-powered machines benefit more from eGPU acceleration than powerful ones, like the iMac Pro, with its already-powerful, built-in Vega Pro 64 GPU.

Performance by the numbers (iMac Pro only)

To provide some context, here are the results I got with the iMac Pro:

Resolve on iMac Pro (internal V64 chip) – NO eGPU – Auto GPU config

Playback of timeline at real-time 23.976 without frames dropping

Render at source resolution – average 11fps (slower than real-time)

Render at timeline resolution – average 33fps (faster than real-time)

Resolve on iMac Pro – with BMD eGPU (580 chip) – OpenCL

Playback of timeline at real-time 23.976 without frames dropping

Render at source resolution – average 11fps (slower than real-time)

Render at timeline resolution – average 37fps (faster than real-time)

Metal

Apple’s ability to work with eGPUs is enabled by Metal. This is their framework for addressing hardware components, like graphics and central processors. The industry has relied on other frameworks, including OpenGL, OpenCL and CUDA. The first two are open standards written for a wide range of hardware platforms, while CUDA is specific to Nvidia GPUs. Apple is deprecating all of these in favor of Metal (now Metal 2). With each coming OS update, these will become more and more “legacy” until presumably, at some point in the future, macOS may only support Metal.

Apple’s intention is to gain performance improvements by optimizing the code at a lower level “closer to the metal”. It is possible to do this when you only address a limited number of hardware options, which may explain why Apple has focused on using only AMD and Intel GPUs. The downside is that developers must write code that is proprietary to Apple computers. Metal is in part what gives Final Cut Pro X it’s smooth media handling and real-time performance. Both Premiere Pro and Resolve give you the option to select Metal, when installed on Macs.

In the tests that I ran, I presume FCPX only used Metal, since there is no option to select anything else. I did, however, test both Premiere Pro/Adobe Media Encoder and Resolve with both Metal and again with OpenCL specifically selected. I didn’t see much difference in render times with either setting in Premiere/AME. Resolve showed definite differences, with OpenCL the clear winner. For now, Resolve is still optimized for OpenCL over Metal.

Power for the on-the-go editor and colorist

The MacBook Pro is where the Blackmagic eGPU makes the most sense. It gives you better performance with faster exports, and adds badly-needed connectivity. My test Resolve sequence is a lot more stressful than I would normally create. It’s the sort of sequence I would never work with in the real world on a lower-end machine, like this 13” model. But, of course, I’m purposefully pushing it through a demanding task.

When I ran the test on the laptop without the eGPU connected, it would barely play at all. Exports at source resolution rendered at around 1fps. Once I added the Blackmagic eGPU, this sequence played in real-time, although the viewer would start to drop frames towards the end of each shot. Exports at the source resolution averaged 5.5fps. At timeline resolution (2K DCI) it rendered at up to 17fps, as opposed to 4fps without it. That’s over 4X improvement.

Everyone’s set of formats and use of color correction and filters are different. Nevertheless, once you add the Blackmagic eGPU to this MacBook Pro model, functionality in Resolve goes from insanely slow to definitely useable. If you intend to do reliable color correction using Resolve, then a Thunderbolt 3 UltraStudio HD Mini or 4K Extreme 3 is also required for proper video monitoring. Resolve doesn’t send video signals over HDMI, like Premiere Pro and Final Cut Pro X can.

It will be interesting to see if Blackmagic also offers a second eGPU model with the higher-end chip in the future. That would likely double the price of the unit. In the testing I’ve done with other eGPUs that used a version of the Vega 64 GPU, I’m not convinced that such a product would consistently deliver 2X more performance to justify the cost. This Blackmagic eGPU adds a healthy does of power and connectivity for current MacBook Pro users and that will only get better in the future.

I think it’s clear that Apple is looking towards eGPUs are a way to enhance the performance of its MacBook Pro line, without compromising design, battery life, and cooling. Cable up to an external device and you’ve gained back horsepower that wouldn’t be there in the standard machine. After all, you mainly need this power when you are in a fixed, rather than mobile, location. The Blackmagic eGPU is portable enough, so that as long as you have electrical power, you are good to go.

In his review of the 2018 MacBook Pro, Ars Technica writer Samuel Axon stated, “Apple is trying to push its own envelope with the CPU options it has included in the 2018 MacBook Pro, but it’s business as usual in terms of GPU performance. I believe that’s because Apple wants to wean pro users with serious graphics needs onto external GPUs. Those users need more power than a laptop can ever reasonably provide – especially one with a commitment to portability.”

I think that neatly sums it up, so it’s nice to see Blackmagic Design fill in the gaps.

Originally written for RedShark News.

©2018 Oliver Peters

Advertisements

Hawaiki AutoGrade

The color correction tools in Final Cut Pro X are nice. Adobe’s Lumetri controls make grading intuitive. But sometimes you just want to click a few buttons and be happy with the results. That’s where AutoGrade from Hawaiki comes in. AutoGrade is a full-featured color correction plug-in that runs within Final Cut Pro X, Motion, Premiere Pro and After Effects. It is available from FxFactory and installs through the FxFactory plug-in manager.

As the name implies, AutoGrade is an automatic color correction tool designed to simplify and speed-up color correction. When you install AutoGrade, you get two plug-ins: AutoGrade and AutoGrade One. The latter is a simple, one-button version, based on global white balance. Simply use the color-picker (eye dropper) and sample an area that should be white. Select enable and the overall color balance is corrected. You can then tweak further, by boosting the correction, adjusting the RGB balance sliders, and/or fine-tuning luma level and saturation. Nearly all parameters are keyframeable, and looks can be saved as presets.

AutoGrade One is just a starter, though, for simple fixes. The real fun is with the full version of AutoGrade, which is a more comprehensive color correction tool. Its interface is divided into three main sections: Auto Balance, Quick Fix, and Fine-Tune. Instead of a single global balance tool, the Auto Balance section permits global, as well as, any combination of white, black, and/or skin correction. Simply turn on one or more desired parameters, sample the appropriate color(s) and enable Auto Balance. This tool will also raise or lower luma levels for the selected tonal range.

Sometimes you might have to repeat the process if you don’t like the first results. For example, when you sample the skin on someone’s face, sampling rosy cheeks will yield different results than if you sample the yellowish highlights on a forehead. To try again, just uncheck Auto Balance, sample a different area, and then enable Auto Balance again. In addition to an amount slider for each correction range, you can also adjust the RGB balance for each. Skin tones may be balanced towards warm or neutral, and the entire image can be legalized, which clamps video levels to 0-100.

Quick Fix is a set of supplied presets that work independently of the color balance controls. These include some standards, like cooling down or warming up the image, the orange and teal look, adding an s-curve, and so on. They are applied at 100% and to my eye felt a bit harsh at this default. To tone down the effect, simply adjust the amount slider downwards to get less intensity from the effect.

Fine-Tune rounds it out when you need to take a deeper dive. This section is built as a full-blown, 3-way color corrector. Each range includes a luma and three color offset controls. Instead of wheels, these controls are sliders, but the results are the same as with wheels. In addition, you can adjust exposure, saturation, vibrance, temperature/tint, and even two different contrast controls. One innovation is a log expander, designed to make it easy to correct log-encoded camera footage, in the absence of a specific log-to-Rec709 camera LUT.

Naturally, any plug-in could always offer more, so I have a minor wish list. I would love to see five additional features: film grain, vignette, sharpening, blurring/soft focus, and a highlights-only expander. There are certainly other individual filters that cover these needs, but having it all within a single plug-in would make sense. This would round out AutoGrade as a complete, creative grading module, servicing user needs beyond just color correction looks.

AutoGrade is a deceptively powerful color corrector, hidden under a simple interface. User-created looks can be saved as presets, so you can quickly apply complex settings to similar shots and set-ups. There are already many color correction tools on the market, including Hawaiki’s own Hawaiki Color. The price is very attractive, so AutoGrade is a superb tool to have in your kit. It’s a fast way to color-grade that’s ideal for both users who are new or experienced when it comes to color correction.

(Click any image to see an enlarged view.)

©2018 Oliver Peters

More about ProRes RAW

A few weeks ago I wrote a two-part post – HDR and RAW Demystified. In the second part, I covered Apple’s new ProRes RAW codec. I still see a lot of misinformation on the web about what exactly this is, so I felt it was worth an additional post. Think of this post as an addendum to Part 2. My apologies up front, if there is some overlap between this and the previous post.

_____________________________

Camera raw codecs have been around since before RED Digital Camera brought out their REDCODE RAW codec. At NAB, Apple decided to step into the game. RED brought the innovation of recording the raw signal as a compressed movie file, making on-board recording and simplified post-production possible. Apple has now upped the game with a codec that is optimized for multi-stream playback within Final Cut Pro X, thus taking advantage of how FCPX leverages Apple hardware. At present, ProRes RAW is incompatible with all other applications. The exception is Motion, which will read and play the files, but with incorrect default – albeit correctable – video levels.

ProRes RAW is only an acquisition codec and, for now, can only be recorded externally using an Atomos Inferno or Sumo 19 monitor/recorder, or in-camera with DJI’s Inspire 2 or Zenmuse X7. Like all things Apple, the complexity is hidden under the surface. You don’t get the type of specific raw controls made available for image tweaking, as you do with RED. But, ProRes RAW will cover the needs of most camera raw users, making this the raw codec “for the rest of us”. At least that’s what Apple is banking on.

Capturing in ProRes RAW

The current implementation requires a camera that exports a camera raw signal over SDI, which in turn is connected to the Atomos, where the conversion to ProRes RAW occurs. Although no one is very specific about the exact process, I would presume that Atomos’ firmware is taking in the camera’s form of raw signal and rewrapping or transforming the data into ProRes RAW. This means that the Atomos firmware would require a conversion table for each camera, which would explain why only a few Sony, Panasonic, and Canon models qualify right now. Others, like ARRI Alexa or RED cameras, cannot yet be recorded as ProRes RAW. The ProRes RAW codec supports 12-bit color depth, but it depends on the camera. If the SDI output to the Atomos recorder is only 10-bit, then that’s the bit-depth recorded.

Until more users buy or update these specific Atomos products – or more manufacturers become licensed to record ProRes RAW onboard the camera – any real-word comparisons and conclusions come from a handful of ProRes RAW source files floating around the internet. That, along with the Apple and Atomos documentation, provides a pretty solid picture of the quality and performance of this codec group.

Understanding camera raw

All current raw methods depend on single-sensor cameras that capture a Bayer-pattern image. The sensor uses a monochrome mosaic of photosites, which are filtered to register the data for light in the red, green, or blue wavelengths. Nearly all of these sensors have twice as many green receptors as red or blue. At this point, the sensor is capturing linear light at the maximum dynamic range capable for the exposure range of the camera and that sensor. It’s just an electrical signal being turned into data, but without compression (within the sensor). The signal can be recorded as a camera raw file, with or without compression. Alternatively, it can also be converted directly into a full-color video signal and then recorded – again, with or without compression.

If the RGGB photosite data (camera raw) is converted into RGB pixels, then sensor color information is said to be “baked” into the file. However, if the raw conversion is stored in that form and then later converted to RGB in post, sensor data is preserved intact until much later into the post process. Basically, the choice boils down to whether that conversion is best performed within the camera’s electronics or later via post-production software.

The effect of compression may also be less destructive (fewer visible artifacts) with a raw image, because data, rather than video is being compressed. However, converting the file to RGB, does not mean that a wider dynamic range is being lost. That’s because most camera manufacturers have adopted logarithmic encoding schemes, which allow a wide color space and a high dynamic range (big exposure latitude) to be carried through into post. HDR standards are still in development and have been in testing for several years, completely independent of whether or not the source files are raw.

ProRes RAW compression

ProRes RAW and ProRes RAW HQ are both compressed codecs with roughly the same data footprint as ProRes and ProRes HQ. Both raw and standard versions use a variable bitrate form of compression, but in different ways. Apple explains it this way in their white paper: 

“As is the case with existing ProRes codecs, the data rates of ProRes RAW are proportional to frame rate and resolution. ProRes RAW data rates also vary according to image content, but to a greater degree than ProRes data rates. 

With most video codecs, including the existing ProRes family, a technique known as rate control is used to dynamically adjust compression to meet a target data rate. This means that, in practice, the amount of compression – hence quality – varies from frame to frame depending on the image content. In contrast, ProRes RAW is designed to maintain constant quality and pristine image fidelity for all frames. As a result, images with greater detail or sensor noise are encoded at higher data rates and produce larger file sizes.”

ProRes RAW and HDR do not depend on each other

One of my gripes, when watching some of the ProRes RAW demos on the web and related comments on forums, is that ProRes RAW is being conflated with HDR. This is simply inaccurate. Raw applies to both SDR and HDR workflows. HDR workflows do not depend on raw source material. One of the online demos I saw recently immediately started with an HDR FCPX Library. The demo ProRes RAW clips were imported and looked blown out. This made for a dramatic example of recovering highlight information. But, it was wrong!

If you start with an SDR FCPX Library and import these same files, the default image looks great. The hitch here, is that these ProRes RAW files were shot with a Sony camera and a default LUT is applied in post. That’s part of the file’s metadata. To my knowledge, all current, common camera LUTs are based on conversion to the Rec709 color space, not HDR or wide gamut. If you set the inspector’s LUT tab to “none” in either SDR or HDR, you get a relatively flat, log image that’s easily graded in whatever direction you want.

What about raw-specific settings?

Are there any advantages to camera raw in the first place? Most people will point to the ability to change ISO values and color temperature. But these aren’t actually something inherently “baked” into the raw file. Instead, this is metadata, dialed in by the DP on the camera, which optimizes the images for the sensor. ISO is a sensitivity concept based on the older ASA film standard for exposing film. In modern digital cameras, it is actually an exposure index (EI), which is how some refer to it. (RedShark’s Phil Rhodes goes into depth in this linked article.)

The bottom line is that EI is a cross-reference to that camera sensor’s “sweet spot”. 800 on one camera might be ideal, while 320 is best on another. Changing ISO/EI has the same effect as changing gain in audio. Raising or lowering ISO/EI values means that you can either see better into the darker areas (with a trade-off of added noise) – or you see better highlight detail, but with denser dark areas. By changing the ISO/EI value in post, you are simply changing that reference point.

In the case of ProRes RAW and FCPX, there are no specific raw controls for any of this. So it’s anyone’s guess whether changing the master level wheel or the color temp/tint sliders within the color wheels panel is doing anything different for a ProRes RAW file than doing the same adjustment for any other RGB-encoded video file. My guess is that it’s not.

In the case of RED camera files, you have to install a camera raw plug-in module in order to work with the REDCODE raw codec inside of Final Cut Pro X. There is a lot of control of the image, prior to tweaking with FCPX’s controls. However, the amount of image control for the raw file is significantly more for a REDCODE file in Premiere Pro, than inside of FCPX. Again, my suspicion is that most of these controls take effect after the conversion to RGB, regardless of whether or not the slider lives in a specific camera raw module or in the app’s own color correction controls. For instance, changing color temperature within the camera raw module has no correlation to the color temperature control within the app’s color correction tools. It is my belief that few of these actually adjust file data at the raw level, regardless of whether this is REDCODE or ProRes RAW. The conversion from raw to RGB is proprietary with every manufacturer.

What is missing in the ProRes RAW implementation is any control over the color science used to process the image, along with de-Bayering options. Over the years, RED has reworked/improved its color science, which theoretically means that a file recorded a few years ago can look better today (using newer color science math) than it originally did. You can select among several color science models, when you work with the REDCODE format. 

You can also opt to lower the de-Bayering resolution to 1/2, 1/4, 1/8, etc. for a RED file.  When working in a 1080p timeline, this speeds up playback performance with minimal impact on the visible resolution displayed in the viewer. For full-quality conversion, software de-Bayering also yields different results than hardware acceleration, as with the RED Rocket-X card. While this level of control is nice to have, I suspect that’s the sort of professional complication that Apple seeks to avoid.

The main benefit of ProRes RAW may be a somewhat better-quality image carried into post at a lower file size. To get the comparable RGB image quality you’d need to go up to uncompressed, ProRes 4444, or ProRes 4444 XQ – all of which become very taxing in post. Yet, for many standard productions, I doubt you’ll see that great of a difference. Nevertheless, more quality with a lower footprint will definitely be welcomed.

People will want to know whether this is a game-changer or not. On that count, probably not. At least not until there are a number of in-camera options. If you don’t edit – and finish – with FCPX, then it’s a non-starter. If you shoot with a camera that records in a high-quality log format, like an ARRI Alexa, then you won’t see much difference in quality or workflow. If you shoot with any RED camera, you have less control over your image. On the other hand, it’s a definite improvement over all raw workflows that capture in image sequences. And it breathes some life into an older camera, like the Sony FS700. So, on balance, ProRes RAW is an advancement, but just not one that will affect as large a part of the industry as the rest of the ProRes family has.

(Note – click any image for an enlarged view. Images courtesy of Apple, FilmPlusGear, and OffHollywood.)

©2018 Oliver Peters

Luca Visual FX builds Mystery & Suspense

For most editors, creating custom music scores tends to fall into the “above my pay grade” category. If you are a whizz with GarageBand or Logic Pro X, then you might dip into Apple’s loop resources. But most commercials and corporate videos are easily serviced by the myriad of stock music sites, like Premium Beat and Music Bed. Some music customization is also possible with tracks from companies like SmartSound.

Yet, none of the go-to music library sites offer curated, genre-based, packages of tracks and elements that make it easy to build up a functional score for longer dramatic productions. Such projects are usually the work of composers or a specific music supervisor, sound designer, or music editor doing a lot of searching and piecing together from a wide range of resources.

Enter Luca Visual FX – a developer best known for visual effects plug-ins, such as Light Kit 2.0. It turns out that Luca Bonomo is also a composer. The first offering is the Mystery & Suspense Music and Sound Library, which is a collection of 500 clips, comprising music themes, atmospheres, drones, loops, and sound effects. This is a complete toolkit designed to make it easy to combine elements, in order to create a custom score for dramatic productions in the mystery or suspense genre.

These tracks are available for purchase as a single library through the LucaVFX website. They are downloaded as uncompressed, stereo AIF files in a 24-bit/96kHz resolution. This means they are of top quality and compatible with any Mac or PC NLE or DAW application. Best yet, is the awesome price of $79. The package is licensed for a single user and may be used for any audio or video production, including for commercial purposes.

Thanks to LucaVFX, I was able to download and test out the Library on a recent short film. The story is a suspense drama in the style of a Twilight Zone episode, so creating a non-specific, ethereal score fits perfectly. Drones, dissonance, and other suspenseful sounds are completely in line, which is where this collection shines.

Although I could have used any application to build this, I opted for Apple’s Final Cut Pro X. Because of its unique keyword structure, it made sense to first set up a separate FCPX library for only the Mystery & Suspense package. During import, I let FCPX create keyword collections based on the Finder folders. This keeps the Mystery & Suspense FCPX library organized in the same way as they are originally grouped. Doing so, facilitates fast and easy sorting and previewing of any of the 500 clips within the music library. Then I created a separate FCPX library for the production itself. With both FCPX libraries open, I could quickly preview and place clips from my music library to the edit sequence for the film, located within the other FCPX library.

Final Cut uses Connected Clips instead of tracks. This means that you can quickly build up and align overlapping atmospheres, transitions, loops, and themes for a densely layered music score in a very freeform manner. I was able to build up a convincing score for a half-hour-long piece in less that an afternoon. Granted, this isn’t mixed yet, but at least I now have the musical elements that I want and where I want them. I feel that style of working is definitely faster in Final Cut Pro X – and more conducive to creative experimentation – but it would certain work just as well in other applications.

The Mystery & Suspense Library is definitely a winner, although I do have a few minor quibbles. First, the music and effects are in keeping with the genre, but don’t go beyond it. When creating a score for this kind of production, you also need some “normal” or “lighter” moods for certain scenes or transitions. I felt that was missing and I would still have to step outside of this package to complete the score. Secondly, many of the clips have a synthesized or electronic tone to them, thanks to the instruments used to create the music. That’s not out of character with the genre, but I still would have liked some of these to include more natural instruments than they do. In fairness to LucaVFX, if the Mystery & Suspense Library is successful, then the company will create more libraries in other genres, including lighter fare.

In conclusion, this is a high quality library perfectly in keeping with its intended genre. Using it is fast and flexible, making it possible for even the most musically-challenged editor to develop a convincing, custom score without breaking the bank.

©2018 Oliver Peters

Premiere Pro Multicam Editing

Over the years, a lot of the projects that I’ve edited have been based on real-person interviews. This includes documentaries, commercials, and corporate video. As the cost of camera gear has come down and DSLRs became capable of delivering quality video, interview-based production now almost always utilizes multiple cameras. Directors will typically record these sections with two or more cameras at various tangents to the subject, which makes it easy to edit for content without visible jump-cuts (hopefully). In addition, if they also shoot in 4K for an HD delivery, then you have the additional ability to cleanly punch-in for even more framing options.

While having a specific multicam feature in your NLE isn’t required for cutting these types of productions, it sure speeds up the process. Under the best of circumstances, you can play the sequence in real-time and cut between camera angles in the multicam viewer, much like a director calls camera switches in a live telecast. Since you are working within an NLE, you can also make these camera angle cuts at a slower or faster pace and, of course, trim the cuts for greater timing precision. Premiere Pro is my primary NLE these days and its multi-camera editing routines are a joy to use.

Prepping for multi-camera

Synchronization is the main requirement for productive multicam. That starts at the time of the original recording. You can either sync by common timecode, common audio, or a marked in-point.

Ideally, your production crew should use a Lockit Sync Box to generate timecode and sync to all cameras and any external sound recorder. That will only work with professional products, not DSLRs. Lacking that, the next best thing is old school – a common slate with a clap-stick or even just your subject clapping hands at the start, while in view on all cameras. This will allow the editor to mark a common in-point.

The last sync method is to match the common audio across all sources. Of course, that only works if the production crew has supplied quality audio to all cameras and external recorders. It has to be at least good enough so that the human editor and/or the audio analysis of the software can discern a match. Sometimes this method will suffer from a minor amount of delay – either, because of the inherent offset of the audio recording circuitry within the camera electronics – or, because an onboard camera mic was used and the distance to the subject results in a slight delay, compared to a lav mic on the subject.

In addition to synchronization, you obviously need to record high-quality audio. This can be a mixer feed or direct mic input to one or all of the camera tracks, or to a separate external audio recorder. A typical set-up is to feed a lav and a boom mic signal to audio input channels 1 and 2 of the camera. When a mixer and an external recorder are used, the sound recordist will often also record a mix. Another option, though not as desirable, is to record individual microphone signals onto different cameras. The reason this isn’t preferred, is that sometimes when these two sources are mixed in post (rather than only one source used at a time), audio phasing can occur.

Synching in Premiere Pro

To synchronize multicam clips in Premiere Pro, simply select the matching sources in the browser/bin, right-click, and choose “Create New Multi-Camera Source Sequence”. You will be presented with several options for sync, based on timecode, audio, or marked points. You may also opt to have the clips moved to a “Processed Clips” bin. If synchronization is successful, you’ll then end up with a multicam source clip that you can now cut to a standard sequence.

A multicam source clip is actually a modified, nested sequence. You can open the clip – same as a nested sequence – and make adjustments or apply filters to the clips within.

You can also create multicam clips without going through the aforementioned process. For example, let’s say that none of the three sync methods exist. You have a freewheeling interview with two or more cameras, but only one has any audio. There’s no clap and no common timecode. In fact, if all the cameras were DSLRs, then every clip arbitrarily starts at 00:00:00:00. The way to tackle this is to edit these cameras to separate video tracks of a new sequence. Sync the video by slipping the clips’ positions on the tracks. Select those clips on the timeline and create a nest. Once the nest is created, this can then be turned into a multicam source clip, which enables you to work with the multicam viewer.

One step I follow is to place the multicam source clip onto a sequence and replace the audio with the best original source. The standard multicam routine means that audio is also nested, which is something I dislike. I don’t want all of the camera audio tracks there, even if they are muted. So I will typically match-frame the source until I get back to the original audio that I intend to use, and then overwrite the multicam clip’s audio with the original on this working timeline. On the other hand, if the manual multicam creation method is used, then I would only nest the video tracks, which automatically leaves me with the clean audio that I desire.

Autosequence

One simple approach is to use an additional utility to create multicam sequences, such as Autosequence from software developer VideoToolShed. To use Autosequence, your clips must have matching timecode. First separate all of your clips into separate folders on your media hard drive – A-CAM, B-CAM, SOUND, and so on. Launch Autosequence and set the matching frame rate for your media. Then import each folder of clips separately. If you are using double-system sound you can choose whether or not to include the camera sound. Then generate an XML file.

Now, import the XML file into Premiere Pro. This will import the source media into bins, along with a sequence of clips where each camera is on a separate track. If your clips are broken into consecutive recordings with stops and starts in-between, then each recorded set will appear further down on the same timeline. To turn this sequence into one with multicam clips, just follow my explanation for working with a manual process, described above.

Multicam cutting

At this point, I dupe the sequence(s) and start a reductive process of shaping the interview. I usually don’t worry too much about changing camera angles, until I have the story fleshed out. When you are ready for that, right-click into the viewer, and change the display mode to multicam.

As you play, cut between cameras in the viewer by clicking on the corresponding section of the viewer. The timeline will update to show these on-the-fly edits when you stop playback. Or you can simply “blade” the clip and then right-click that portion of the clip to select the camera to be shown. Remember than any effects or color corrections you apply in the timeline are applicable to that visible angle, but do not follow it. So, if you change your mind and switch to a different angle, the effects and corrections do not change with it. Therefore, adjustments will be required to the effect or correction for that new camera angle.

Once I’m happy with the cutting, I will then go through and make a color correction pass. If the lighting has stayed consistent, I can usually grade each angle for one clip only and then copy that correction and paste it to each instance of that same angle on the timeline. Then repeat the procedure for the other camera angles.

When I’m ready to deliver the final product, I will dupe the sequence and clean it up. This means flattening all multicam clips, cleaning up unused clips on my timeline, deleting empty tracks, and usually, collapsing the clips down to the fewest number of tracks.

©2018 Oliver Peters

Audio Mixing with Premiere Pro

When budgets permit and project needs dictate, I will send my mixes out-of-house to one of a few regular mixers. Typically that means sending them an OMF or AAF to mix in Pro Tools. Then I get the mix and split-tracks back, drop them into my Premiere Pro timeline, and generate master files.

On the other hand, a lot of my work is cutting simple commercials and corporate presentations for in-house use or the web, and these are often less demanding  – 2 to 8 tracks of dialogue, limited sound effects, and music. It’s easy to do the mix inside of the NLE. Bear in mind that I can – and often have – done such a mix in Apple Logic Pro X or Adobe Audition, but the tools inside Premiere Pro are solid enough that I often just keep everything – mix included – inside my editing application. Let’s walk though that process.

Dealing with multiple channels on source clips

Start with your camera files or double-system audio recordings. Depending on the camera model, Premiere Pro will see these source clips as having either stereo (e.g. a Canon C100) or multi-channel mono (e.g. ARRI Alexa) channels. If you recorded a boom mic on channel 1 and a lavaliere mic on channel 2, then these will drop onto your stereo timeline either as two separate mono tracks (Alexa) – or as a single stereo track (C100), with the boom coming out of the left speaker and the lav out of the right. Which one it is will strictly depend on the device used to generate the original recordings.

First, when dual-mic recordings appear as stereo, you have to understand how Premiere Pro deals with stereo sources. Panning in Premiere Pro doesn’t “shift” the audio left, right, or center. Instead, it increases or decreases the relative volume of the left or right half of this stereo field. In our dual-mic scenario, panning the clip or track full left means that we only hear the boom coming out of the left speaker, but nothing out of the right. There are two ways to fix this – either by changing the channel configuration of the source in the browser – or by changing it after the fact in the timeline. Browser changes will not alter the configuration of clips already edited to the timeline. You can change one or more source clips from stereo to dual-mono in the browser, but you can’t make that same type of change to a clip already in your sequence.

Let’s assume that you aren’t going to make any browser changes and instead just want to work in your sequence. If your source clip is treated as dual-mono, then the boom and lav will cut over to track 1 and 2 of your sequence – and the sound will be summed in mono on the output to your speaks. However, if the clip is treated as stereo, then it will only cut over to track 1 of your sequence – and the sound will stay left and right on the output to your speakers. When it’s dual-mono, you can listen to one track versus the other, determine which mic sounds the best, and disable the clip with the other mic. Or you can blend the two using clip volume levels.

If the source clip ends up in the sequence as a stereo clip, then you will want to determine which one of the two mics you want to use for the best sound. To pick only one mic, you will need to change the clip’s audio configuration. When you do that, it’s still a stereo clip, however, both “sides” can be supplied by either one of the two source channels. So, both left and right output will either be the boom or the lav, but not both. If you want to blend both mics together, then you will need to duplicate (option-drag) the audio clip onto an adjacent timeline track, and change the audio channel configuration for both clips. One would be set to the boom for both channels and the other set to only the lav for its two channels. Then adjust clip volume for the two timeline clips.

Configuring your timeline

Like most editors, while I’m working through the stages of rough cutting on the way to an approved final copy, I will have a somewhat messy timeline. I may have multiple music cues on several tracks with only one enabled – just so I can preview alternates for the client. I will have multiple dialogue clips on a few tracks with some disabled, depending on microphone or take options. But when I’m ready to move to the finishing stage, I will duplicate that sequence to create a “final version” and clean that one up. This means getting rid of any disabled clips, collapsing my audio and video clips to the fewest number of tracks, and using Premiere’s track creation/deletion feature to delete all empty tracks – all so I can have the least amount of visual clutter. 

In other blog posts, I’ve discussed working with additional submix buses to create split-track exports; but, for most of these smaller jobs, I will only add one submix bus. (I will explain its purpose in a moment.) Once created, you will need to open the track mixer panel and route the timeline channels from the master to the submix bus and then the output of the submix bus back to the master.

Plug-ins

Premiere Pro CC comes with a nice set of audio plug-ins, which can be augmented with plenty of third-party audio effects filters. I am partial to Waves and iZotope, but these aren’t essential. However, there are several that I do use quite frequently. These three third-party filters will help improve any vocal-heavy piece.

The first two are Vocal Rider and MV2 from Waves and are designed specifically for vocal performances, like voice-overs and interviews. These can be pricey, but Waves has frequent sales, so I was able to pick these up for a fraction of their retail price. Vocal Rider is a real-time, automatic volume adjustment tool. Set the bottom and top parameters and let Vocal Rider do the rest, by automatically pushing the volume up or down on-the-fly. MV2 is similar, but it achieves this through compression on the top and bottom ends of the range. While they operate in a similar fashion, they do produce a different sound. I tend to pick MV2 for voice-overs and Vocal Rider for interviews.

We all know location audio isn’t perfect, which is where my third filter comes in. FxFactory is knows primarily for video plug-ins, but their partnership with Crumplepop has added a nice set of audio filters to their catalog. I find AudioDenoise to be quite helpful and fast in fixing annoying location sounds, like background air conditioning noise. It’s real-time and good-sounding, but like all audio noise reduction, you have to be careful not to overdo it, or everything will sound like it’s underwater.

For my other mix needs, I’ll stick to Premiere’s built-in effects, like EQ, compressors, etc. One that’s useful for music is the stereo imager. If you have a music cue that sounds too monaural, this will let you “expand” the track’s stereo signal so that it is spread more left and right. This often helps when you want the voice-over to cut through the mix a bit better. 

My last plug-in is a broadcast limiter that is placed onto the master bus. I will adjust this tight with a hard limit for broadcast delivery, but much higher (louder allowed) for web files. Be aware that Premiere’s plug-in architecture allows you to have the filter take affect either pre or post-fader. In the case of the master bus, this will also affect the VU display. In other words, if you place a limiter post-fader, then the result will be heard, but not visible through the levels displayed on the VU meters.

Mixing

I have used different mixing strategies over the years with Premiere Pro. I like using the write function of the track mixer to write fader automation. However, I have lately stopped using it – instead going back to manual keyframes within the clips. The reason is probably that my projects tend to get revised often in ways that change timing. Since track automation is based on absolute timeline position, keyframes don’t move when a clip is shifted, like they would when clip-based volume keyframes are used.

Likewise, Adobe has recently added Audition’s ducking for music to Premiere Pro. This uses Adobe’s Sensei artificial intelligence. Unfortunately I don’t find to be “intelligent” enough. Although sometimes it can provide a starting point. For me, it’s simply too coarse and doesn’t intelligently adjust for areas within a music clip that swell or change volume internally. Therefore, I stick with minor manual adjustments to compensate for music changes and to make the vocal parts easy to understand in the mix. Then I will use the track mixer to set overall levels for each track to get the right balance of voice, sound effects, and music.

Once I have a decent balance to my ears, I will temporarily drop the TC Electronic (included with Premiere Pro) Radar loudness plug-in to make sure my mix is CALM-compliant. This is where the submix bus comes in. If I like the overall balance, but I need to bring everything down, it’s an easy matter to simply lower the submix level and remeasure.

Likewise, it’s customary to deliver web versions with louder volume levels than the broadcast mix. Again the submix bus will help, because you cannot raise the volume on the master – only lower it. If you simply want to raise the overall volume of the broadcast mix for web delivery, simply raise the submix fader. Note that when I say louder, I’m NOT talking about slamming the VUs all the way to the top. Typically, a mix that hits -6 is plenty loud for the web. So, for web delivery, I will set a hard limit at -6, but adjust the mix for an average of about -10.

Hopefully this short explanation has provided some insight into mixing within Premiere Pro and will help you make sure that your next project sounds great.

©2018 Oliver Peters

FCPX Color Wheels Take 2

Prior to version 10.4, the color correction tools within Final Cut Pro X were very basic. You could get a lot of work done with the color board, but it just didn’t offer tools competitive with other NLEs – not to mention color plug-ins or a dedicated grading app like DaVinci Resolve. With the release of 10.4, Apple upped the game by adding color wheels and a very nice curves implementation. However, for those of us who have been doing color correction for some time, it quickly became apparent that something wasn’t quite right in the math or color science behind these new FCPX color wheels. I described those anomalies in this January post.

To summarize that post, the color wheels tool seems to have been designed according to the lift/gamma/gain (LGG) correction model. The standard behavior for LGG is evident with a black-to-white gradient image. On a waveform display, this appears as a diagonal line from 0 to 100. If you adjust the highlight control (gain), the line appears to be pinned at the bottom with the higher end pivoting up or down as you shift the slider. Likewise, the shadow control (lift) leaves the line pinned at the top with the bottom half pivoting. The midrange control (gamma) bends the middle section of the line inward or outward, with no affect on the two ends, which stay pinned at 0 and 100, respectively. In addition to luminance value, when you shift the hue offset to an extreme edge – like moving the midrange puck completely to yellow – you should still see some remaining black and white at the two ends of the gradient.

That’s how LGG is supposed to work. In FCPX version 10.4, each color wheel control also altered the levels of everything else. When you adjusted midrange, it also elevated the shadow and highlight ranges. In the hue offset example, shifting the midrange control to full-on yellow tinted the entire image to yellow, leaving no hint of black or white. As a result, the color wheels correction tool was unpredictable and difficult to use, unless you were doing only very minor adjustments. You ended up chasing your tail, because when one correction was made, you’d have to go back and re-adjust one of the other wheels to compensate for the unwanted changes made by the first adjustment.

With the release of FCPX 10.4.1 this April, Apple engineers have changed the way the color wheels tool behaves. Corrections now correspond to the behavior that everyone accepts as standard LGG functionality. In other words, the controls mostly only affect their part of the image without also adjusting all other levels. This means that the shadows (lift) control adjusts the bottom, highlights (gain) will adjust the top end, and midrange (gamma) will lighten or darken the middle portion of the image. Likewise, hue offsets don’t completely contaminate the entire image.

One important thing to note is that existing FCPX Libraries created or promoted under 10.4 will now be promoted again when opened in 10.4.1. In order that your color wheel corrections don’t change to something unexpected when promoted, Projects in these Libraries will behave according to the previous FCPX 10.4 color model. This means that the look of clips where color wheels were used – and their color wheel values – haven’t changed. More importantly, the behavior of the wheels when inside those Libraries will also be according to the “old” way, should you make any further corrections. The new color wheels behavior will only begin within new Libraries created under 10.4.1.

These images clarify how the 10.4.1 adjustments now work (click to see enlarged and expanded views).

©2018 Oliver Peters