Achieve much higher peak flows by using a running average on max flow calculations

[pixie-wrangler-irl]

I would like to preface this with I am not a professional coder so I have no idea how challenging this would actually be to implement.

In my mind it should be possible to make many small extrude paths much faster with significantly higher flows than a constant flow rate would allow. Think to when you freely push the filament by hand through a hotend. How much faster you can dump the pre-heated plastic in the nozzle? VS when that pre-melt is gone and the filament is actively melting during the flowing process… I bet instantaneous flow rates for the “pre-melt” could be 400% to 500% higher than the constant flow rates we tune for with the long sweeping wall tests.

In my mind the added settings look like this:
Instant max flow rate XX.X m^3/s - The actual peak flow based on a short move with high accel.
Target AVERAGE flow rate XX.X m^3/s - Just the running average max flow, used for “long” extrusions paths
Flow Averaging time X.XX seconds - How long does it average the flow rate for.

A relatively simple running average during calculating speeds for max flow should allow for massively increased speeds on interrupted features especially with travels in between. As well as blazing fast wall extrude path line starts until the melt rate equalizes.

Another version of this might utilize a variable for the volume of the melt zone. Allow Instantaneous max flow rate of XX.X mmm/s flow rate for X.XX_m^3_Melt_Zone_Volume. Time of filament spent in melt zone would still have to be a factor somehow.

Possible Testing methods: two small towers with perimeters that come out to roughly the volume of the hotend melt zone. add in a "slow travel speed", enough to let the filament cook in the melt zone for just an extra moment if its a fast printer. start cranking up max flow to see what it can do in short bursts of nozzle pre-melt print volumes.