I agree. Set up a Node Bin specifically for Arnold. The purpose of integrating Arnold is not to perform complex modeling in Flame, but to better integrate with Maya, import large and complex FBX or USD 3D scenes, and then apply Arnold for compositing and rendering in Flame. Imagination is always great, but achieving it may be a huge undertaking.
should be unreal engine.
should be real time export to usd.
so there’s that..
3 Likes
@cnoellert -very meta…
Should tickle you pink my friend.
It will be impossible to get agreement here on priority since most of us only use specific subsets of what Flame offers, and someone’s must-have is another’s I didn’t know it even existed.
There are definite efficiencies by having a 3D feature set right inside the app, rather than round-tripping. No contest.
But there are two new realities that go against this feature request:
In today’s app ecosystems budgets are a lot tighter. The software is cheaper ($5K/yr rather that multiples of that), and the tech debt load and external forces are bigger. As a result dev resources are stretched thinner than ever, and getting big changes done is difficult and slow. See how long it’s been taking Foundry to update the 3D system. Flame is not more resource rich here.
The other aspect is that today’s 3D apps have way more complexity than what was common the last time 3D was added to Flame. Meaning, the task is probably an order of magnitude bigger lift.
So things are drifting apart, not getting closer.
And the more specialized apps get, the harder it gets to have the ‘everything’ app. We would be better off making round-trips more seamless by better format support and import functions.
4 Likes
Does the ADSK feature request website have the ability to downvote things? I do not want this and do not think development time should be spent on it. I smell another feature request!
2 Likes
I can’t believe they let go of Francis — again. What a great talent and guy. Are we down to two lead devs for Flame?
3 Likes
A feature request to ‘do not do a feature request’
we will be lucky if there are even just “two devs” on the Flame team.
Since the latest RTX technology includes built-in real-time ray tracing, game engines have been quick to take advantage of it. Flame operates in a very similar way, which is why it was able to integrate Stingray technology so easily.
When discussions about implementing Arnold come up, it sounds like a massive undertaking and out of scope with what Flame’s 3D system was designed for. Our current 3D system is designed to “cheat,” and it does that quite effectively. By adding the real-time ray tracing capabilities now available in modern GPUs, we could take that concept even further—something that will become increasingly important for the reasons I mentioned in the original post.
Ray tracing will give Flame artists the ability to perform far more advanced compositing, which will be necessary given the kind of content we’ll be working with.
Otherwise, Flame risks becoming just a conform and delivery tool for studios.
Autodesk makes Maya. I doubt they want to invest time and money implementing Ray Tracing in Flame when it already exists in their other product. Remember: they don’t actually give a shit about you or I.
2 Likes
Alright, let’s stick to subtitles improvements then, and Selectives in Image …
2 Likes
I’m still fighting to get Paint node corruptions and MotionVector Tracking fails to be fixed.
It is a pretty expensive tool for that compared to Resolve which handles more formats both in and out as well as DCP & IMF mastering plus Dolby Atmos support.
Hate to say it but the 3D environment in Fusion is much better than Flames now too.
1 Like
I know several studios that use Flame only in this capacity, conform/delivery. All comping is done in Nuke.
And my studio literally had this discussion today about switching to Resolve conform for our finishing dept, since we already use Resolve for Color.
I’ve been banging the drum for years with the dev team to please revamp action’s 3D workspace. Basically make it like Nuke or any other 3D application.
Remove the damn clipping planes in various views, show all the objects/boundaries and have a grid. Aligning objects for projections in large scenes (i.e. a shot of a large landscape) is sometimes impossible. Moving a gmask attached to an axis that’s far off in the distance is an exercise in futility. Even when you adjust the scene scale, there are times that the values are just so large that Flame doesn’t seem to be able to cope.
The upcoming reverse user group would be a good opportunity to discuss with ADSK. So is NAB.
2 Likes
We will be prompt comping in Weavy on our Apple Vision Pro’s before anything remotely near this happens with Flame. I’ve been saying it for at least a year now. It’s over (I mean all of it, not just Flame). AI will win. Whatever you have now is bonus. Enjoy it. Get it. Save it.
1 Like
If you give up you contribute to making a foregone conclusion. Sometimes that’s what it is. But you become part of the problem, and let the future happen to you instead of shaping it.
1 Like