Interview: Page (1) of 1 - 06/19/07 Email this story to a friend. email article Print this page (Article printing at page facebook

Getting Intimate with CineForm Intermediate Part 2

CineForm CEO David Taylor discusses RA format in digital cinema, different flavors of HD By Mike Jones

Digital intermediate workflow and Lossless compression seem to have become the buzz words of the month post NAB07 where a number of companies announced either new lossless intermediate workflow products or software support for existing intermediate solutions. For California-based developer CineForm, the industry attention must have been bitter sweet on one hand confirmation and acknowledgement of what CineForm has been touting loudly for quite some time, and on the other hand a newly populated stiff arena of competing products.

DMN contributing writer Mike Jones spoke in detail with

CineForm CEO David Taylor
to get the nitty-gritty on the workflow pros and cons of CineForm intermediate and how he sees CineForm in the context of a flood of lossless online production formats now filing onto the market and into the NLE. Part 1 of this interview can be found here

MJ: CineForm is compatible with a number of NLE's such as Vegas and soon to be Final Cut Pro, but its Premiere Pro that seems to be able to take best advantage of CineForm's real-time performance. What's the reason behind this? And do you see CineForm being able to deliver the same performance levels in other NLE's in the future?
David Taylor: The extra performance we achieve in Premiere Pro is because Adobe published an SDK, an interface spec, that allows CineForm to literally replace the Premiere Pro arithmetic engine with our own engine. Its the same reason that hardware companies like Matrox can provide hardware support for Premiere Pro. Our software simply masquerades as if it were hardware. 

Wed LOVE for other NLE manufacturers to adopt a similar strategy because wed do the same thing for them that is, write our own RT processing engine. As a specific example of what we can do when we cant replace the native processing engine, we worked with the Vegas engineers a few years ago to enable extra performance within Vegas based upon its knowledge of CineForms multi-resolution decoding characteristic that I discussed before. It cant make these decisions dynamically however, but it does provide some good acceleration using this technique. 

MJ: You mentioned working with the Sony Vegas engineers, how is the relationship between CineForm and NLE developers? Are they open in assisting you to take advantage of the architecture of their NLE's? Do they give you a long-term perspective on their development plans? Or do you really have to work from the outside looking in, guessing where they'll go next?
DT: It depends on who were talking about. Part of Adobes strategy is to encourage third-party developers by exploiting their SDK, so Adobe is motivated to help us add features and performance to Premiere Pro that they arent doing themselves. We also have an excellent relationship with the Sony Vegas guys as weve been working with them for many years now. Im also happy to report that we have numerous discussions in progress with other NLE and post-production software vendors that are considering adding CineForm support within their products.

But in direct answer to your question, most of these applications (aside from Premiere Pro) arent written with the intention of enabling third parties in a big way. So when you have something valuable to offer you try to tee up a strategic relationship that adds value to the offerings of both companies.

MJ: For sometime there has been a number of flavours of CineForm Connect, Aspect and Prospect HD; and now you've added the NEO series Neo HDV, NEO HD and NEO 2k. What's the difference between all these editions? Is the NEO line set to replace Connect, Aspect and Prospect or does it offer something different?
DT: Think of it this way we have products to support HDV (8-bit and up to 1440x1080), HD (10-bit and up to 1920x1080), and 2K (10/12-bit and up to 2048x2048). There are additional features differences as well, but from a market standpoint those are the three levels of customers. In our NEO family we have NEO HDV, NEO HD, and NEO 2K, each offering the features demanded by the particular market segment. NEO is designed for non-Premiere Pro applications whereas our Aspect/Prospect family is designed for Premiere Pro and includes the Cineform RT video engine for Premiere Pro.  Aspect HD is designed for HDV users, Prospect HD is designed for HD users, and Prospect 2K is designed for 2K users. There is a products equivalence between the NEO products and Aspect/Prospect at the market levels of HDV, HD, and 2K. NEO HDV is a direct upgrade replacement for Connect HD. Well also carry the NEO name onto the Mac. 

MJ:  The idea of RAW formats and the digital negative is now very much entrenched in digital still photography. Many would say it is the biggest thing to happen to photography since digital images themselves. Now you have the CineForm RAW format for video; how does this work? Does it offer a similar post-production flexibility for video as RAW does for photography?
DT: YES! Raw in digital cinema applications is essentially the same as RAW in still photography except you run it at 24fps or higher. But Im glad you asked because this is an important emerging area for digital cinema workflows.  As youre obviously familiar with RAW used in still photography, you might also be aware that Photoshop is unaware of RAW, so you must first ?develop RAW images using Photoshop Lightroom or another tool which converts them into RGB prior to importing them into Photoshop. 

The problem with this is that developing RAW images is very time consuming due to the nature of the ?demosaicing algorithms that must be used to convert RAW images into RGB images. So what we have done with CineForm RAW is to offer the ability to defer the development of RAW images until much later in the post workflow.  When our codec plays CineForm RAW files, we can dynamically apply a simple demosaic filter to the output in real time which is suitable for use in an NLE or compositing application. This allows use of the CineForm RAW files directly on the timelines of Premiere Pro, After Effects, Final Cut Pro, and soon other applications. The secondary advantage is that files in their CineForm RAW form are much smaller than when they are ?developed into CineForm 444 or any other format. So file management and disk space is well utilized. 

Finally, we have added the concept of ?active metadata in CineForm RAW files similar to what is done in still camera RAW. This is a powerful concept that well also move into other CineForm files in the future. Think of it this way - in a traditional camera, you adjust white balance and saturation inside the camera before shooting, and the results of those adjustments are ?flattened into the resulting images as a YUV or RGB file. You have no ability to later redo those settings. If you want to change them, you are again re-adjusting the output of the camera on top of the original settings, but you have no ability to go back to the original data stream as output by the cameras sensor.  With CineForm RAW, when you specify a white balance or saturation control (often known as Color Matrix information) we simply add that data into the stream as metadata inside the camera, but we NEVER modify the data that was presented to us by the sensor. Then at playback time we decode the CineForm RAW stream, and as a post process we apply white balance and Color Matrix processing before presenting the output to the calling application. Think of it as applying white balance and color matrix processing as a Photoshop layer. You can always change or delete layers without affecting the original data beneath. We also apply the demosaicing algorithm dynamically upon decode. 

And theres one other thing. . . we also allow the addition of a 3D LUT to the CineForm RAW files at record time. This can be defined by the DP at production time and is inserted into all files. That means all shot review done by the DP contains the defined ?look he/she defined at the time.
In post, all of the active metadata is used as part of the decode process, so we hide all these complexities from the calling application. We also offer the ability to modify the metadata in post much as you can do with still-camera RAW, but. . . you can do this inside Premiere Pro (which you cant do in Photoshop). 

MJ: The shift to High Definition production right across the spectrum has resulted in more HD formats than you can poke a stick at HDV, AVCHD, DVCProHD, XDCamHD, uncompressed HD subsequently there are a huge range of cameras from numerous manufacturers that all have their own quirks and idiosyncrasies; has this been a problem for developing CineForm?
DT: Not at all! In fact it is a big opportunity for us. As an example, it took CineForm about three weeks to support the 24p mode from the original JVC HD100U it took Apple two years!  It is usually much more difficult for native codec guys to support features than for CineForm. The reason is that our format already supports most features, so by converting out of a camera format you not only get better visual fidelity, but you also unlock new features such as inverse telecine processing spatial resampling, speed change, etc. 

It is also an opportunity for us because newer formats like AVCHD are very difficult to edit.  The entropy coding used in this standard is very difficult for a PC to process efficiently, so youre much better off converting out of AVCHD.

MJ: Are there particular flavors of HD that you think provide better performance and post-production options than others? Are there particular types of HD cameras that CineForm works best with?
DT: Honestly not really. We support all forms of single-link and dual-link HD-SDI. We support all forms of HDV and DVCPRO HD. We even support the special modes from HDV cameras from Canon (like 24F / 25F / 30F) and Sonys new HVR-V1U (24p extraction from 60i), plus HDMI recording from various HDMI sources. We recently began support AVCHD although an external decoder is required. We will soon be adding XDCam HD support. Having said all this, one of the powerful aspects of our workflow is to mix and match sources from multiple different camera types on the same timeline. Our software allows framerate conversions and image frame size conversions to simplify mixed-camera productions.

Page: 1

Mike Jones is a digital media producer, author, educator from Sydney, Australia. He has a diverse background across all areas of media production including film, video, TV, journalism, photography, music and on-line projects. Mike is the author of three books and more than 200 published essays, articles and reviews covering all aspects of cinematic form, technology and culture. Mike is currently Head of Technological Arts at the International Film School Sydney (, has an online home at and can be found profusely blogging for DMN at

Related Keywords:DI, HD editing, NLE, CineForm

Content-type: text/html  Rss  Add to Google Reader or
Homepage    Add to My AOL  Add to Excite MIX  Subscribe in
NewsGator Online 
Real-Time - what users are saying - Right Now!

Our Privacy Policy --- @ Copyright, 2015 Digital Media Online, All Rights Reserved