1.Only do reorder if requseted.In by object mode,we only need
to collect the filaments in the object for filament grouping
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Id88f2449c6d3c5d45b9ff87513308d997fa72da1
(cherry picked from commit 7b5818fe6c538123373243ec8dfedc3c2fa68e6d)
1.Always do filament group to get correct filament map
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Ie4e61c310f6c0cfeacb2a49c4f065f9674b6a35f
(cherry picked from commit 242554557b1ecd72355841826af707580fc583b6)
1.Caused by uninitialized filament map in mapping for AMS
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: I71ce6001fc6f2e72eb9303fcaba0bac16ad70dc9
(cherry picked from commit 48023e4c154c118c9396b6065b7e2476970fd441)
1.Use max flow network to handle limit
2.Support setting master extruder id
3.Fix the issue in the KMedoids algorithm where data is overwritten
after each retry.
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Idd2bedf39f61e7a65eb4199852f60b8fbebe0a7d
(cherry picked from commit 3cfb49a1b9dc2c76066ec441f1028f99a4bf99c4)
when the ams is 0/0, should not consider the print connect status
jira:none
Change-Id: I2288ff15894d5c734a98b830df6678cab2bd12fc
(cherry picked from commit 20837717d411016251c4831586b0fa61350a08b2)
1.If the group result differs little in flush,we will choose the one
that best fits the ams filaments
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Icd147b406e3494c841ef13564ad1b1231ad798fd
(cherry picked from commit 0b95bdd9d950918ea6979da6b4d62b2d2cd25b99)
1.Use float to store the weight copmputed in the procedure.Avoid the
issue where filament weight remains at 0 after changing filaments
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Iff0cfc6e22f34affbc232dbfe196f27ee06d2d9e
(cherry picked from commit 994e4f4840a6734ab479e10d39462cc692bc4d77)
1.Use min cost max flow to solve the tool order
jira:NEW
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: I909845039b67c7fe3ddd42580ad3f1d71d52262d
(cherry picked from commit 0716b8518ef62e0eac7c45de8bafb1458d8f9a8e)
1.In auto mode,display the statistics of auto mode and single
extruder
2.In manual mode,display the statistics of manual mode and auto mode
3.Support by object mode
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: If54c4db79f33d5278c0f18d01ab0518e8660f9c7
(cherry picked from commit 2dbc5c939666e294c805ee4bf33ec09976688be3)
1.Should calculate cost if filament used in that layer is 1,because
we should consider the last filament used in previous layer
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: I5838af77f1d73bfa07c65bd6ee12ae352dc3b571
(cherry picked from commit 0756b51eb2b457a1ddfddc0ab8e445c779737597)
1.Use forcast when filament num in current layer and next layer both
smaller than limit number
jira:NEW
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Ia21bb22ce353f0c74eeaf748c6d9f9c4b315ec18
(cherry picked from commit c49c0c8f723e82e4e87cd931b7d5070142fb0781)
1.Add filament flush,filament change count reduced by multi
extruder
NTOE: cases when printing by object haven't been handled
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Iaaea5f8ffae2345df5a6f1dc605342d835974d48
(cherry picked from commit ad65cdb909b29210117f690a119ef76b70373da4)
1.When n becomes large,the original algorithm to get best filament
sequence will cost too much time and memory.Use a greedy algorithm
instead.Always select the next filament with fewest flush
jira:NONE
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Iabc924e1bdb0a07af0a6ef0bbdd62e54ce54f052
(cherry picked from commit 3167e34e609d0eecb481fd550e6550024b72bb67)
1. save to appconfig and project setting
2. use it to group
jira: none
Change-Id: Id4048fc1b47f6904b2e9c0154aaa3a2b03590437
(cherry picked from commit da32b878b3d7ca95ae5c4786102848c1e8a5ab2d)
support drag drop to modify the filament map
jira: none
Change-Id: I0ed3180a0fa8e95e7c871bb039eb844faccd1344
(cherry picked from commit 67f67d76889668fdd80ae5e496b6cbca5f771a43)
1.When filament <10, do the reorder with next layer.This can reduce
10% flush in some cases
2.Support custom filament seq
3.Use caches to speed the algorithm
jira:NEW
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Ie1af9841f7165802d350eae962efe8febafbb357
(cherry picked from commit cffb4a8564844b1892eb42cfe8e883a52e70fb6a)
1.When capacity is greater than the num of filaments, always choose the
map result that can be accommodated
2.In BestFit strategy,always try to fill up the existing capacity
3.In BestCost strategy, just try the group with fewest flush
jira:NEW
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: Ifd6d64b77774039e57ffff26cf2243a4d3f89054
(cherry picked from commit cddf8cae27f4320f830b463e92703d3a6cf344e7)
1.When n<10, calc all case cost
2.When n>10, first k-medoids algorithm first
3.Enable setting group size
jira:NEW
Signed-off-by: xun.zhang <xun.zhang@bambulab.com>
Change-Id: I625f47e0235c70e440c6d489b052a156fbffca3f
(cherry picked from commit 9ec276d3d7114fff7a33213c3b47ce88df85f2ee)
1. backend support multi_extrude data structure
2. Compatible with third-party calibration
3. fix bug when get extruder in gocde export process
Change-Id: I5dac9abdd9907a521a1ba9b480f9e05640591bc1
(cherry picked from commit 21e6271e59ea8e4924866275566617d14a4b2b6e)
1. support recommended filament map when print by object
2. placeholder_parser support function filament_change
3. extruder_id of filament_map is start from 1
Change-Id: Ide8019cd4a165a25972f22706ff685c3005aa031
(cherry picked from commit b42d94e1d05236b8b7b2e65b4a24810eecf040cb)
Compare the custom gcode z to `0.5*(layer_z[n] + layer_z[n+1])` instead of `EPSILON`
to compensate float rounding error during gcode processing (SoftFever/OrcaSlicer#7834)
