Well i love flame! Hopefully we can continue using it for years to come.
1 Like
Hi guys,
The way I see this subject is that there aren’t enough flame artists in the market (blame whoever you want), and while a shot can be moved from one artist to another one (different time zones or one sets the shot and another one finish it, holidays or someone gets sick… plus add as many variables as you want), there are always compers available to take the shot. This is not possible to achieve with flame artists.
We love the idea that we can manage complete projects, but that’s unrealistic. Simple projects? Maybe. But for sure not complex projects where 3D (animation, FX, lighting…) have all the control. There’s no single lighter out there without a nuke, and they love checking what they are creating. Nuke is the most popular app for comping but especially because everyone knows how to use it.
Therefore, post-production companies are not relying on flame due to the lack of artist. Every app has its own pros and cons, but we need more flame artists in the arena.
Nuke is legion.
Flame is rare avis.
In my opinion, things are like this because for some years Flame instead of concentrating on the development of new technologies has pursued the idea of being software that can do a little bit of everything and this can clearly be seen in the updates that most sometimes they involve improvements to existing modules but rarely propose something new that makes it attractive as a true competitor of Nuke. So anyone who wants to work in VFX learns to use Nuke, not Flame.
I don’t know if my impression may mean anything to you guys, I am NOT a flame artist although I am trying to dabble into it for various reasons.
What I found since the last time I look at it carefully is that Autodesk got their act together, in a pretty impressive way;
- The new EXR multichannel, color management, etc… works really well.
- The Python integration is really good and allows to put Flame into a pipeline in ways other tools like Resolve would only dream.
- The conform process is the best I have seen, by far… like really far. In this respect, only Resolve is close-ish… Nuke Studio is very very very far behind.
- In terms of rendering speed, well, no contest really.
- when talking about typography, finishing in general… no contest once again.
- When editing, Resolve I find it better, specially with the extra hardware, but Flame is close enough
And I could go on for a while, bottom line is, I feel once you go to these cheaper tools you are going to come back really quick… mark my words.
I hope it helps
5 Likes
1 Like
I think Flame’s edge over Nuke was always its capacity to comp & edit all the way to presenting in a supervised session. Being able to quickly interact with the client’s comments and show the result right away is still not great on Nuke. Now, as Flame artists, we are trained for that high pressure situation and it’s the total opposite of the Nuke experience.
Now, after the pandemic and with a remote workflow taking over almost 100% of the time, I feel Flame has lost some of that “show-piece” status. Its name still holds sway, a mystique, that is owed in large part, to the brilliant Flame artists before us that could pull these sessions off with swagger.
Long live Flame!
3 Likes
It’s simply computer progress that’s stole flames thunder.
When we first used flames on an onyx or onyx 2 with 8 cpus runnjng at 150/300 mhz the average mac was running on cpu at 16hz and no pcs came close to that so all Softimage seats were on indies or later impacts. Sgi ruled the world.
Back then running an sgi was the equivalent of traveling at light speed compared to driving a delivery van.
Once shake arrived I knew it was the beginning of the end. Running shake on a laptop I could comp HD faster than the Flame could so I did. At that time I’d bounce shots from flame to me 17jnch mac pro for all sorts of tweaks with cool plugins Ive mostly forgotten the names of but grain surgery (best plugin name ever) keylight and zbigs keyer whose name escapes me.
But more than that once I accepted leaving the walled garden of flame the world was mine. Tracking in mocha, object removal in mokey, retimes using cine stretch and more. This quickly showed the aging chinks in flames armour. No longer did it have the best keyer, no longer did it have the best time warp, no longer did it have the best tracker and so on.
So the problem for flame wasn’t the power of any single competitor it was every other software in the world.
Later I moved to smoke on Mac as at least then I wasnt trapped in a Linux island. I could bounce from avid to resolve to smoke do hard thjngs in Nuke and use after effects plugins to finish jobs without exporting and transferring files half the time. When smoke 2013 became the mess it was I conceded defeat and bought a Flame on Linux and it worked well enough but any speed advantage Linux gave me was more than eroded by export and transfer times.
That’s why flame on Mac is the only reason flame still exists.
However that’s not why flame is still such a rare beast. It’s the baselight vs resolve arguement. Yes baselight outperforms resolve but hardly anyone cares as resolve is affordable. Whilst that may historically seem to no longer be the case in practical terms it is. Flame needs fast storage and as much ram as you can throw at it. The ability of Nuke to play back over a network shames flame for performance add in the multichannel stuff and it’s clear flame can compete without a major structural rebuild. A rebuild that probably isn’t worthwhile as batch and action are already quite unwieldy to see what’s going on vs a node graph in Nuke.
I’ve used flame a long time and I love it and id love the system to continue to flourish but while they spent a decade adding Playstation quality cg features and explode nodes the node based systems were fixing comp problems.
At the end of the day flame has a fairly rigid workflow structure that struggles to compete with nuke. Look how easy the roto tools are in Nuke compared to how overly complex they quickly become in flame. We excuse it by saying well we send roto out but why?
Same with the shortcuts how ridiculous are flame shortcut keys to nukes equivalent. Most shortcuts now take almost all the fingers of both hands compared to nukes single key taps.
Flame continues to succeed at rhe high end because of the artists who are now famous in their markets. Clients know the flame guy and know he will deliver. This is entirely due to hard earned reputations and a willingness by facilities to sell their top people hard to retain clients. But post houses learned from this mistake too as they hate paying our wages 
hence Nuke artists are kept anonymous and interchangeable. It’s no accident that a facility boasts how many Nuke seats it has but has bill or ted in flame. Having a hero compositor only works on jobs small enough to go across one guys desk - commercials, promos, short form. Anything needing a huge team needs continuity across the project and interchangeable worker ants rather than queen bees.
I’m glad I got on flame early and I loved being a flame artist for the majority of my career. But then I look back fondly at my time running one inch rooms with a cmx and ampex ace, or the years in harry suites and year in Henry. But I’m not attached to the tools. I’d love flame to flourish and become the beast it once was but in film and tv critical mass wins.
Once enough people use a system they get all the plugins they can acquire outside innovation as its amortized across the installed user base. Look at how much sparks used to cost vs ofx or after effects plugins to see.how that works. Or how the siggraph white papers on warp stabilizer or Content aware fill went from research to product in one version of after effects.
Volume makes innovation possible.
For flame to win at this stage it needs to be a $100 a month subscription. That’s not trying to insult anyone at autodesk but that’s probably the wide adoption breakpoint in 2022. Even at that price it would see resistance due to the hardware requirements but I think at 1200 a year you’d see user base growth which is what its future survival demands. Depending on a mostly 50+ user base wont last very long:)
3 Likes
This pattern exists across the industry and multiple software packages.
I would say that Foundry has threaded that needle better with Nuke Indie. I know several people that adopted it once that door opened. $500/yr is a manageable cost. And they firewalled it against abuse with three limitations: you can only render <= 4K resolution, scripts are encrypted and cannot be shared with full versions (to prevent shops from outsourcing to Nuke Indie armies), and you only get forum based support. No tickets, nobody to call. That’s a fair trade-off to build the independent user base. And they can acquire the clients and over time upgrade to the full versions once they have the jobs to support that.
ADSK has the infrastructure to do something similar with Flex, but the pricing and terms are still off and they have to figure some way of avoiding the outsourcing problem without choking features. Make projects non-shareable. Just single artist. Change the token/period count, and offer more granularity. 24hrs is great for facilities, it kills you if you get a few notes form a client a few days later. The basics are in place to do all of this better. Wouldn’t be a big lift.
And I know I’m a broken record on that, but personally the lack of Windows support is an issue. No easy fix I know, but for a lot of reasons to do with Apple’s roadmap, all my other systems are Windows at the moment. Flame lives on a Linux island for me, so I do have to render/transfer. Some licenses are floating and could run on Linux, but not all of them. And for plugins like BorisFX, floating licenses are lot more expensive for a boutique shop, yet essential. So I keep running a mixed setup. It adds to the tax.
The other big app that still is a hold-out like Flame is Baselight, and they seem to be doing ok for now, but adoption friction seems similar. Both Nuke and Flame benefit from a family environment that is healthy and can cross-fund a lot of infrastructure. Smaller shops like SGO and Filmworkz who have gone affordable subscription struggle keeping up with the pace at which all this technology is evolving and not being able to have enough hands to stay competitive on features while keeping up with tech debt and ecosystems.
Does anyone have a sense of how many Indie subscribers the Foundry has vs. full licenses? Would be interesting to know.
1 Like
The bigger twist is of course the attempt to transition from high cost low volume to high volume low cost.
I can think of only one example in this industry in my entire working life and that’s DaVinci resolve.
Cmx tried to go down market jsit before the end as did ampex with the ace 25. Quantel did make some inroads with edit box after the Henry but never went high volume.
A close second to resolve would be avid. Now it’s a subscription at 40 bucks a month the rumors of its imminent demise has fallen silent despite the juggernaut that is Adobe creative cloud. But that’s mostly to do with its shared storage strength making it the only choice for team driven editorial projects. It’s harder to identify projects that can only be done in flame.
Flame is still a unique product with unique strengths and weaknesses but it’s no longer the sole workhorse and as such can’t command the price for the bulk of the market especially with its underwhelming network rendering options. I don’t have a solution but unless something happens it becomes less relevant every year. And that’s a damn shame.
2 Likes
Another small but significant plugin keyer is Zbig. It was developed at Berlin’s Center for New Visual Design (CFB) and is an advanced chroma-key compositing plugin avaliable for use with programs such as Digital Fusion and Illusion. Zbig includes tools such as intelligent spill replacement, shadow adjustment, colour correction and many others that help create convincing and ‘invisible’ composites.
Its original author Zbig Rybczynski is an extremely successful film maker. His work in film and video has earned him 3 MTv Music Video Awards, three American Video Awards, three Monitor Awards for Best Director, the 1986 Billboard Music Video Award for Most Innovative Video, and many others. After working at the CFB between 1994-2001 on compositing and motion control, in 2001 he moved back to Los Angeles and became a member of Ultimatte’s and iMatte’s creative teams.
From the old post at
I recall it being really good at transparency and shadows beating primate, ultimatte and keylight for results with little tweaking.
1 Like