Speaker 1: Yo what's up everyone, welcome back to another video. This video is going to be responding to some questions that were placed on the how to edit a documentary video I released about a month ago. And there were some great comments on there about creative editing and I'm going to form a part two and talk about the creative side of editing long form documentaries. There were also some technical questions that I'd love to address in this video around how to manage 4k raw workflows for long form projects. So without further ado, here are some of your questions that you placed on that video. So Von Thinking has commented, would love to see how you manage proxies and keep such a huge 4k workflow going without turning the computer into a hellfire machine. So I think this project for Oroking the documentary in question was around 26, 27 terabytes of footage. And then obviously a lot of those clips in that documentary were 4k or higher. And they were shot in Blackmagic RAW with the URSA G2 and there was a bunch of Pocket 6k RAW footage. There was also a RED camera that shot a lot of the underwater footage. And then we had a lot of Apple ProRes RAW footage for the DJI drone. And so lots of RAW footage, tons of data. And that is where the server of our studio comes into play. We've got hundreds of terabytes of data capacity and that's wired up over a 10 gigabit ethernet connection. And it's serving essentially every computer in the office with 10 gigabit ethernet. And so the read and write speeds to the server with everyone working off it, anywhere between 600 to 800 megabytes a second read and write. So it's plenty fast enough to edit RAW footage directly off, but even then it can still get a little bit laggy and bottlenecky. So what I did was I transcoded a lot of the DJI Apple ProRes RAW footage and the RED RAW footage into a mezzanine codec, which is essentially an intermediary codec that is still robust enough to deliver from. Now that's different to optimized media or proxies because proxies always need to be relinked back to the original for export. So a lot of the RED and a lot of the drone shots were transcoded into an Apple ProRes 4444 and I basically just edited straight off them. It really minimized the bottleneck in terms of reading that data off the server. So the short answer to that question is I didn't actually use any proxies in this workflow. So we're blessed enough to have the ability to read and write directly onto the server and host the 27, 26 terabytes worth of footage directly. And so a lot of the bottlenecks are alleviated when it comes to data storage. And then the bottlenecks turn to more of the computer side and actually having a computer fast enough to process that RAW footage. Blackmagic RAW footage is beautiful to work with and a lot of the technology has been built to actually be processed by the camera. And so the computer is not having to work as hard with Blackmagic RAW footage as it would be with other types and other flavors of RAW footage such as the RED R3D files. Now that's not to say that is the best way to approach a documentary of this sort. I would probably in the future transcode all of the original files into proxies and link them into DaVinci and hopefully be able to store all those proxies on a single two terabyte SSD external drive. And this would just allow me to throw it into a laptop and edit anywhere I want. Unfortunately I was stuck in the office with the edit machines that we have there because I was needing to source all of the footage directly off the server. So that is an issue but if portability isn't a problem to you then potentially you could look into a robust enough server to edit directly off. On the computer side there is definitely the need to have a lot of RAM in your computer. I've got 128 gigabytes of RAM within my 27 inch iMac and this really helps when it comes to loading large timelines. Obviously when you're loading large timelines it's trying to pull all of those clips and reference those clips at the same time in the same timeline and the RAM helps to just handle those really long timelines. So the other thing that helped a huge amount was creating a PostgreSQL database instead of a disk database. Now DaVinci has two types of databases that you can create and the PostgreSQL database gives you the ability to have a collaborative or shareable database hosted on a separate machine from your edit machine. Now I've put up a video about our server configuration here which you can check out but essentially it shows you that we're hosting the DaVinci database on a Mac Mini and that is serving via the 10 GB network the database to all of the edit machines and so the edit machines job is essentially to edit and it's not to consolidate the database and that processing is done on the Mac Mini and all these little additions really just help to alleviate the bottleneck from the edit machine and give the Mac Mini and the server responsibilities in terms of managing all that processing. If your computer itself is trying to manage all of the data and manage the database and manage the project file then that could also bog it down and cause a few bottlenecks. I will say in my experience the PostgreSQL databases are far more robust than disk databases. Now I'm not a computer technician and I don't really know what's happening underneath the surface with the two differences but in my experience PostgreSQL databases are far more robust, less crashes and more resilient to those longer form timelines. Alright, second question from Adam is how much of your organisation is done before you get into DaVinci as far as folder structures go or do you build your folder structure within DaVinci? Also how do you go about saving projects of this nature, DRP or DRA, delete source etc. So organisation for a project like this is very much done within DaVinci itself. Because this was a documentary that was shot over two years the footage on our server was coming from all sorts of different directions. I didn't feel comfortable pulling all that source footage into a centralised folder. So essentially we've got a 100TB server, there's footage all over it and I was just pulling footage from all those individual sources into the project and then all of the organisation was done within DaVinci Resolve itself. And the second part of that question is all about saving projects, DRP so DaVinci Resolve Project or DaVinci Resolve Archive, files, do we delete source material, all that sort of stuff. Now this was a huge documentary for the client and we've actually built into our pricing structure a fee to host their data on our server, 26TB is a huge amount of data and it would obviously cost us a lot of money to host that off our own back. And so as part of the documentary we are hosting the project for about a year and then there'll be an ongoing fee to continue to host that which gives us the finances to expand the server for future projects if needed. So in terms of all the raw footage we are holding onto the raw footage, in the future all of this footage might be removed from our server and put more into archive storage but for now it's just active on our server, our server is duplicated in multiple locations so the data is just sitting there for now. When it comes to DaVinci Resolve Archive projects, I haven't archived this project currently because all the footage is still active on the server. If we were ever to move the footage off the server into archive, you would then create a DaVinci Resolve Archive project for the final timeline. So essentially the final documentary, it would pull references of all those final clips and copy them into a new location with the DaVinci Resolve project. This would obviously cull the huge amount of data that is 26 terabytes and I imagine would bring it down to anywhere between three to five terabytes based on the final shots that we used in the final timeline. For smaller projects we definitely do a DaVinci Resolve Archive render and essentially render the final timeline with handles because a lot of the time if you're using long form interviews, the interviews might be anywhere from 30 minutes to 60 minutes to even 90 minutes then even if you only use a second of that interview in the final timeline, when you do a DRA file without re-encoding the handles of that clip, you're essentially having to copy that whole 90 minute interview only for the two seconds it's referenced in the final timeline. So for a massive project with a DRA, DaVinci Resolve Archive render, I would definitely re-encode everything with probably two seconds of handles either side. Thanks Adam for that question. Next question is from MJS, do you publish any advice on purging unused data once a project is finished? I think this is a similar question. I find it personally very difficult to delete any data for a client due to the nature of this documentary. There's a lot of footage in there that will be repurposed and used for years to come through a whole range of other video productions. So currently we're not purging unused data and we're not really getting rid of anything. We're just charging a fee to host the data for now and then in the future we'll be able to address what we do with that. All right and final question, it says great video, would love to know more about your technical workflow, proxies, memory management, so many timelines. I think we've addressed a lot of that, but the second part of this is I would like to make the jump from Avid for my next doc, but the long form, I've always been a bit scared to try it within DaVinci Resolve. I was definitely in the same boat. DaVinci Resolve traditionally has not been a very good NLE or non-linear editor. DaVinci Resolve started as a color grading piece of software and it's still used around the world for that today and over the years they've acquired a bunch of other companies such as Fusion and Fairlight and essentially just created this really robust piece of software that is just awesome at almost everything. With that comes a few quirks and I think there's a lot of discussion within the industry as to how robust DaVinci will be going forward because of the frequency of updates that they come out with. They seem to be releasing features every single month and I just hope that as a piece of software they prioritize the user experience, fixing bugs and don't progress too fast in terms of feature set. Having said all of that, the NLE, the non-linear editor, so the edit page and the cut page within DaVinci Resolve is really proving to be an awesome piece of software. Avid is obviously still the industry standard and it's used all around the world for a reason, but I think there's perhaps just a little bit of a lag in the industry because the industry is always slow to adopt new systems and processes. If something works every single time and is reliable in the industry then it takes a long time for that to be superseded by a new piece of tech and I think that's what's happening. DaVinci is robust enough to handle a lot of those long form projects, even feature films could be edited on DaVinci right now, but just because the entire industry hasn't taken the step to DaVinci, I think it'll be a while until you see big facilities adopting it. However, because I've had some experience in it, I would recommend people jump ship as soon as possible because I do feel like, this is just my opinion, in the next 3-5 years it will become industry standard to use DaVinci for the NLE, for the color, for the visual effects and for sound mastering. That's just my thoughts, in my opinion, using the PostgreSQL database hosted on an external machine with a robust enough machine, like a modern computer with at least 64 gigabytes of RAM and some decent graphics memory, you'll be off to the races with that. That's all I have for now, as always please comment below in the comment box if you have any questions about film making or the post production workflow side of things, I really want this channel to develop into something that gives real value to you guys and so I'm largely guided by your questions and by your comments. Really excited to get more videos out there, thank you for your support if you have subscribed, if you haven't please consider subscribing, hitting that bell notification and giving this video a thumbs up, it would really help me and I'd really appreciate it. Alright guys, peace out, we'll see you in the next video.
Generate a brief summary highlighting the main points of the transcript.
GenerateGenerate a concise and relevant title for the transcript based on the main themes and content discussed.
GenerateIdentify and highlight the key words or phrases most relevant to the content of the transcript.
GenerateAnalyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.
GenerateCreate interactive quizzes based on the content of the transcript to test comprehension or engage users.
GenerateWe’re Ready to Help
Call or Book a Meeting Now