Haha. Yeah. I think some of them just get forgotten in the midst of everything else the devs have to do (hello M1 support!), so maybe this goofy thread will be a little spark to get some of them to turn up.
while more general this one is 4 years old now, no votes 
FI-01049
Now it’s got one buddy.
one more for you to make it more specific FI-02760
Honestly imho this is one of the biggest reasons flame isn’t known to handle 3d as well as the competition which is a real bummer
Yeah. Every time someone says “Nuke is better for CG” I want to put my hand up and say, “the tools to comp are the same, it’s just that flame doesn’t carry all the extra channels through the setup elegantly where Nuke does.” Which if you are dealing with a shit ton of layers in a shit ton of EXRs can be a real pain.
THAT SAID
95% of all CG layers go unused. If you’re digging into the UV and AO and Spec passes something’s gone wrong in CG town.
Lets just say the way flame handles exr multichannel really needs to be revised the list is long with what’s wrong with it but I do understand from an engineering perspective this would be a massive undertaking to basically modify each node to carry all channels, that being said if I have to press F4 20 times to view the 20th layer and miss it by one click and have to start all over again that is really wrong and needs to be retooled for the better.
Voted and wrote a little.
It blows my mind. Multichannel could be treated as a timeline with each rgba set as a version. It’s so simple. Just map that shit directly to a working paradigm that literally everyone understands. Double click on an exr source up comes the timeline and you can just move your primary up or down with a hotkey if you’re an F4 jockey or by activating with a swipe. That’s how you could handle shuffles as well… it would be sooo much more intuitive rather than the correct mechanism—multichannel pipe to a shuffle node that looks like a timeline where you simply move your primary to another layer set. Plus it could be the starting out point before a total rewrite.
UV I use often… if you’re trying to do something meaningful with the other passes in flame you’re in the wrong package imho.
I always wondered what the feature request status actually means. “Accepted” or “send to database” sound like someone reviewed it, but I can’t see if it’s like “nice idea, accepted, we’ll implement this” or “ok, makes sense, added to our never ending list in the database, but who knows if it will ever see the light of day”.
@fredwarren any insights on this? 
There is not really any difference between Accepted and Accepted - Send to Database. The former means that someone on our side deemed the request as valid; the latter means the same thing, but the person who validated the request also prefers to have an entry in the tracking database we are using in the development team.
The only status I think can be of any interest to you guys (other than Implemented of course!) are Rejected and Need More Info. Everything that is set to New, Accepted, or Accepted - Send to Database are basically equally treated in the same bucket which is much more closer to “added to our never ending list” than “we’ll implement this”. The only time I see a request and say to myself that we will implement this is in the short term is when it is a very small change that can be done right away without disrupting our currents plans but those are rare.
It is also very rare that we see a request and think that it doesn’t make sense at all or that it is something we do not want to do. Unfortunately the reality is that we don’t have unlimited resources.
Thanks a lot for the detailed answer, Fred! 
I had always wondered about some entries being still tagged as “New” even though they were already years old. Glad to hear they’re not overlooked, but in the same pool with the other requests. 
Thank you sir for all that you do. And go Bills!! I have no idea about football but I thought you might enjoy that.
I would love to have a substance textures link between Maya and Flame with the send function. Also, almost 30 years asking for a modern spline-based warper / morpher using the new g-masks tracer but with an algorithm a la elastic reality. I had to wait 27 years for the send to (TDI Explore) Maya, so there is still some hope

I would love if Flame had a soft selection feature, similar to 3DS Max for vertices when you subdivide them and want to warp them. Also, a rate/time stretch tool that is available in the timeline like Premiere or Resolve can do would be amazing. Lastly, some kind of recreation of the Adobe Warp Stabilizer that is better than the current Auto-Stabilize node would be really great since it really works like magic many times when you just need to smooth out a shot quickly and move on.
Here’s one I recently submitted:
In regards to Scopes, to have the colour sampler position remember where it was on a “per context view” basis. Especially useful when sampling the same object in two different shots for matching…
FI-02763 - Scopes: Colour Sampler Position saved per Context view.
Speaking of matching shots:
FI-02425 - More Scopes Overlay Options for Reference Grabs
This one suggests a complimentary color scheme for the default mono cyan reference grab overlay which is really hard to look at. Instead having a CMY parade overlaid on the RGB parade (or something like it). Lining images up would “cancel out” the colors, behaving like a difference.
Would love an invert button in the 3d keyer.
I’m actually surprised why the keyers don’t have matte edge, histogram and gmask built in