I think the days of them saying 'im telling you thats the way it is now' are well and truly behind them after the 2013 fun and games. They are listening, and they are upgrading the creative tools.
The best example off the top of my head is the new gmask tracer, it plays catch up with what we've been asking or in terms of nuke style roto flexibility with group scale and rotate and temp offset axis independent of the actual point interpolation, but then ADDS to that and puts in the tracer in out keyer to actually become the best in class roto tool out there.
Its easy to wax lyrical about the 'good old days' when flame had the best of everything but that was in a time when they literally had the only tracker, the only displacement surfaces and ran on computers 10 or more times faster than anyone else. But that came with a premium that meant I had to work 18 hours days 7 days a week to buy my first flame and I only owned 1/3 of it even then having had to sell my soul to money men.
If I had any complaint/suggestion going forwards it would be to let go of the insistence on building all the tools. By which i mean just license mocha or neat video or primatte rather than investing time recreating the same tools. Its not only an easy accelerant to development but it also allows user migration as the tools are familiar.
The big mistake of the last 10 years has been a crazy insistence on 3d integration as though we want to make cg in flame when a client can have a week of a maya guy for the same price... But that baton has been passed and Will is a real world flame user, hence the renewed look at usability and process streamlining. And it has a LONG way to go but its at least going in the right direction now.
There is still a crazy insistence on overengineering things such as the new 10 contexts instead of just alt1 alt 2 to set a context and 1 2 to look at them we get a drop down and bump menus etc.. but im confident that will get straightened out soon enough. For me the greater opportunity is to streamline the GUI, the zoom in paint should be the same zoom as everywhere else, the gmask axis etc in tracer is exactly like in action so apply that to classic gmask too. The move to adding tabs to explode groups rather than messing up the node graph is the way to go and there's a lot of similar inconsistencies that need addressing.
Once thats all straightened out then they can look at features, but first and foremost useability is the go. i recently came to the UK on a project that grew so while sorting out my flame assist license i used a copy of smoke 2015 in demo. Wow, so horrible after flame 2016. If you doubt its getting better you need to try going backwards. Loading and saving batches at left, nodes at right, 2 up etc are all pretty streamlined. I still don't like the russian doll multilevel saving but reel groups and sticky reels kick ****.
The new version is still in flux and theres still a ton of things we could do faster in pre-anniversary simply because the desktop tools were so fast to access after 20 years of stabbing buttons on the screen rather than using shortcuts or drop down menus, swap shot and reel grouping come quickly to mind. that could all have been easily addressed by adding a swipe down to reveal the 'classic' menu structure but its probably too late for that now.
2016 ext 1 has addressed screen resolution independence and it now works wonderfully on my 21:9 LG UM95, hardware antialiasing, background reactor and so on all add to the fun. Will burn now be available for OSX? If not put it on the wish list.
Its barely been mentioned but Python hooks in flame will make extending shots to management packages like shotgun a reality with subscriptions to iterations letting VFX supers get approvals done without messing about with flame. This is flame growing up and finally becoming a team player rather than the flame island shots go to and come back from. Its still a huge pain in the **** to comp cg in flame compared to nuke which needs sorting out, id almost suggets a toxik tab so the cg guys can prep a tree and supply a sidecar file with their version of the 3d precomp saving lots of time for online worling out what left left and left right are and the rest.
I have a long list of gripes with the current flame, but despite that list the current flame is rapidly heading the right direction and this new free flame will change things so long as it works out of the box on day one. If theres any instability its done and people wont look again then the product dies when we retire. The saddest outcome from the 2013 experiment was that none of the young guys who got excited about smoke stuck around. They looked, the played they said no. And sadly thats part of the reality of flame operation, you can't dabble. If you wanna be a flame guy you got to learn your stuff and you got to put the hours in. In a world of Red Giant Universe, Videohive templates and Magic Bullet I dont know how many people will have the neccesary level of dedication.
Ultimately the legend that is Flame is based more than anything on the simple fact that its been tough as hell to become a flame artist, so if you booked a flame suite you got a consumate professional who lived and breathed the box. I wonder if that attitude will survive the inclusion of new more casual flame users. Its probably ultimately not even that important, there's room for people at the sharp end but there's a much larger space for people who can do most stuff in a flame. At the end of the day more flames means more flame work.
The most exciting part of all this is people are talking about flame again, but this time its a real product not the erm smoke and mirrors that 2013 was.
Mike
