The request for switching off the autoslicing is not new; please see http://umforum.ultimaker.com/index.php?/topic/2356-auto-slicing-on-off-please/&do=findComment&comment=16732 with Daid's answer.
I prefer not to have a clear platform button as I tend to accidentally click buttons which destroy my work.
The white indicator in the corner of the extruder sets the tension that is applied to the filament to press it against the knurled wheel. The lower the indicator mark, the higher the tension.
It is adjusted by inserting a hex screwdriver in the hole above it, and turning the screw. You turn the screw anticlockwise (as seen from above - 'unscrewing it') in order to increase the tension.
The white indicator in the corner of the extruder sets the tension that is applied to the filament to press it against the knurled wheel. The lower the indicator mark, the higher the tension.
It is adjusted by inserting a hex screwdriver in the hole above it, and turning the screw. You turn the screw anticlockwise (as seen from above - 'unscrewing it') in order to increase the tension.
Thanks kindly!
The request for switching off the autoslicing is not new; please see this post with Daid's answer.
Is there an adult reason not to let the user choose??
From what I can see in that post, the feature is not optional "because f*** you", more or less.
Just because the engine is fast enough, and all the kinks and bugs were fixed, doesn't mean you HAVE TO FORCE IT ON THE USERS...
Maybe leave it always on in the non-advanced mode of Cura?
Is there an adult reason not to let the user choose??
From what I can see in that post, the feature is not optional "because f*** you", more or less.
Just because the engine is fast enough, and all the kinks and bugs were fixed, doesn't mean you HAVE TO FORCE IT ON THE USERS...
Maybe leave it always on in the non-advanced mode of Cura?
With the new Cura GUI settings will be customizible to different degrees, i.e. you can select a quick setting for one option and an expert setting for another if I got Daid right when he was writing about 'settings being on the chop board'. But this rather applies to settings than to buttons.
Sometimes I also would like to have autoslicing off, e.g. when I loaded a complex object, rotated and scaled it and play around with settings and plugins. On the other hand the day would come I would forget manual slicing after an important change in settings.
If I may add a suggestion: Maybe it's possible to suppress autoslicing as long as a key such as shift key is pressed?
Software (Cura)
1. PLEASE stop automatically slicing,
Obvious? No. Not at all. Why do you want to do extra actions? The slicing is ran as background process, as lowest priority. So it does not impact system performance, it does not slow down the GUI (as it's run as 2nd process) and your GCode is ready before you are. The engine is started 0.5 seconds after a setting is changed, so you pretty much have to be faster then 0.5 seconds if you want to be faster then the current implementation.
I pretty much love this system, and you'll have to do a lot to convince me. A lot more then saying "It's obvious" (Alpha-Spock: "It's only obvious")
(Slow downs due to rotation/scaling of complex objects is not due to the slicing, it's actually the GUI that is calculating stuff on the GUI thread. Fundamental problem which I am addressing in the PinkUnicorn branch)
2. Please add the option to determine the amount of shells/floor/roof in numbers instead of only in mm.
Working on a new GUI, code-name "Pink Unicorn" which has a lot more options, and where you can choose which level of setting control you want. You want to set the number of top layers? It can do that. You want to set the whole outer shell as just 1 value, it can also do that.
You can configure exactly which settings you want to see, and some settings "mask out" other settings. Screenshot:
http://i.imgur.com/BQnUtHk.png
3. Please do something about thin walls being skipped during slicing, and/or disconnect this from the "shell thickness" setting (there are also bugs in this relationship, that I've mentioned in a previous forum post).
This is a bit like asking a car to fly. It's a fundamental thing.
4. In the installation menu, please add a prompt that removes the previous version.
This is requested quite often, I haven't looked into it yet.
Firmware
1. First, congrats on fixing the initial-feeding routine.
You're welcome, but it was a bit of a quick-fix. Also thanks Illuminarti for finding the big bug I made.
2. About the start and end routines: are you absolutely sure retracting the material from the head at the end of a print is best?
Yes. Leaving material hot in the hotend is bad for the material, it also drips out, if you just let it sit there, in which case you need the initial extrusion anyhow.
3. For the end routine: why not add an on/off feature for keeping the side fans on full blast to help the prints and plate to cool? (the head will be centered, and/or loose, and will should stay close to the plate by only lowering the plate slightly at this stage. Only when cooled, the plate goes all the way down.)
Problem of leaving the head on top of the print is that it will drip on your print, and you cannot take off the print as the head is in the way. You also cannot always lift the head more, if you use the full Z height of the printer.
4. In all the temperature-related menus, the starting value should not be zero, but rather the last target temp, or the top one of last x times the menu was accessed, or ever simpler, just use the current temp. (I have some pains in my hands, some days I count every key stroke and turn of the UM2 menu wheel...)
5. Please add a "quick-heatup" option for the heated plate. Reasons obvious. Let's say it could stay in this "hot and ready" state for ~15 mins before shutting off? For the shortcut, maybe even use a long press of the wheel.
I've heard a lot of "long press" suggestions already for a lot of different features. For obvious user-interaction-design reasons it's bad to have a long-press option. But together with question 4, I'll think about it.
Another thing that has been suggested, is to start heating up the bed as soon as you enter the print selection menu. This can save you some time in heating up the bed, or, if you leave that menu open, it will heat up the bed. However, this isn't really obvious to the user that it is happening, and might scare people...
6. The "time left" estimate on the LCD is beyond useless, esp with complex prints. It would be much easier and better to have Cura write its estimate into the output file and just have the machine count down from that.
Yes. It's a bit crappy. It has to account for a lot (adjusted speeds, incorrect estimates) I'm pretty sure there is a bug in there.
7. Oh, also, regarding the "change material" routine: Currently it heats up to 260 (I'm assuming) and pulls out slowly and then quicker, the ideal conditions for creating long thin strands of material that are sometimes left over in the tube. It should yank the material quite fast at first, as well as let the user set the pull-out temperature (let's say between 180-260, so as to avoid accidental damage).
I've tried this, didn't work. You can "yank" a lot faster with your hands then with the feeder. The filament change routine is actually something I spend quite a lot of time on. But I never got the strand removed 100%. (It heats up to the set material temperature)
Sorry, I have to add my 2 cents...
@cura pt.4 - remove when install new version: good idea as long you can choose - latest greatest isnt always true...
Further, the machine/profile/settings have to be considered. I hate to setup everything new because forgot to copy the ini's manually and removed the prev. version... :-(
Further, the machine/profile/settings have to be considered. I hate to setup everything new because forgot to copy the ini's manually and removed the prev. version... :sad:
What about a backup? :?: 
Even better than replacing an old version would indeed be importing the settings of a previous version. However, this gives some new problems if you have something like six Cura versions installed as I do (there is actually no serious reason for this; you can call me paranoid...)
:-) same to me... I have 3 installed in parallel...
Don't worry, someone is pretty much a fan of having multiple versions side-by-side.
I just did add an option to the installer to uninstall all other versions (disabled by default)
https://github.com/daid/Cura/commit/a443ba22c8c44fa4733d77f2bcbb0baeb7ea9346
Hey Daid and all,
Regarding Print End:
I think it would be better to keep the filament in the heating chamber, to prevent "bleeding". It should rather be a bit material replenished until the cooling process does not allow for a further material transport.
Why? This ... I try to explain below:
I quite often printed very small objects. At the end of the printing process, material is pulled back very far, so I need many Skirt-lines for the next clean start. It's a bit annoying, but it helps a little, at the end of printing immediately with: "Move material"; convey the material back to the nozzle opening. Then prepare the next start, certainly less problems. I did this successfully practiced Felt a thousand times.
But anyway, to leave material in the heating chamber, is actually quite straightforward, at least at the UM2, as the HotEnd is cooled down in a few minutes to an uncritical temperature range. During cooldown, some material will still flow out of the heating chamber, and likely remains in only a thinned line.
To draw material at the end of the printing process very far back, could be much more problematic and even be responsible for some startup problems. When you restart, cold material must overcome at least two transitions. These transitions could be edgy and especially the ColdEnd could be deformed. If the filament has to pass through this long path into the heating chamber again, then an already thinned filament strand is probably pushed together.
For me it looks like it is much less problematic in most cases to leave material in the heating chamber.
At the end of a print job, the print head should pause in the middle front area, nowhere else, exactly as long as it takes until the user has confirmed the print-end message. Only then the print head should hit the end stops.
Markus
I just did add an option to the installer to uninstall all other versions (disabled by default)
https://github.com/daid/Cura/commit/a443ba22c8c44fa4733d77f2bcbb0baeb7ea9346
Great - will test it soon when updating my office pc. Will the setting ini's be ported?
Hi Daid, and thanks for your thoughtful replies.
Software (Cura)
1. PLEASE stop automatically slicing,
Obvious? No. Not at all. Why do you want to do extra actions? The slicing is ran as background process, as lowest priority. So it does not impact system performance, it does not slow down the GUI (as it's run as 2nd process) and your GCode is ready before you are. The engine is started 0.5 seconds after a setting is changed, so you pretty much have to be faster then 0.5 seconds if you want to be faster then the current implementation.
I pretty much love this system, and you'll have to do a lot to convince me. A lot more then saying "It's obvious" (Alpha-Spock: "It's only obvious")
(Slow downs due to rotation/scaling of complex objects is not due to the slicing, it's actually the GUI that is calculating stuff on the GUI thread. Fundamental problem which I am addressing in the PinkUnicorn branch)
You know what, you've almost convinced me, but will it really never be optional, even in your new omni-configurable version?
See, I think your reason for not making it optional has become too emotional, sometimes you gotta give the users what they want, even things that seem illogical to you. Think of how peaceful your life could be if you just solve this "issue"...
2. Please add the option to determine the amount of shells/floor/roof in numbers instead of only in mm.
Working on a new GUI, code-name "Pink Unicorn" which has a lot more options, and where you can choose which level of setting control you want. You want to set the number of top layers? It can do that. You want to set the whole outer shell as just 1 value, it can also do that.
You can configure exactly which settings you want to see, and some settings "mask out" other settings. Screenshot:
http://i.imgur.com/BQnUtHk.png
Great job, I should say, screen shot looks amazing... Can't wait...
3. Please do something about thin walls being skipped during slicing, and/or disconnect this from the "shell thickness" setting (there are also bugs in this relationship, that I've mentioned in a previous forum post).
This is a bit like asking a car to fly. It's a fundamental thing.
Ok... but are you aware of the bugs? (try slicing a 0.4mm wall, with shell set to 0.4, 0.45, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1 and 1.2... at least in 14.03 it behaved in very unexpected ways)
4. In the installation menu, please add a prompt that removes the previous version.
This is requested quite often, I haven't looked into it yet.
Low priority, I'm sure.
Firmware
1. First, congrats on fixing the initial-feeding routine.
You're welcome, but it was a bit of a quick-fix. Also thanks Illuminarti for finding the big bug I made.
2. About the start and end routines: are you absolutely sure retracting the material from the head at the end of a print is best?
Yes. Leaving material hot in the hotend is bad for the material, it also drips out, if you just let it sit there, in which case you need the initial extrusion anyhow.
I assumed you had tested it thoroughly, just wondering.
3. For the end routine: why not add an on/off feature for keeping the side fans on full blast to help the prints and plate to cool? (the head will be centered, and/or loose, and will should stay close to the plate by only lowering the plate slightly at this stage. Only when cooled, the plate goes all the way down.)
Problem of leaving the head on top of the print is that it will drip on your print, and you cannot take off the print as the head is in the way. You also cannot always lift the head more, if you use the full Z height of the printer.
I see. Anyway, I haven't had any dripping problems yet, but I have only been using PLA, and maybe got lucky with its quality.
4. In all the temperature-related menus, the starting value should not be zero, but rather the last target temp, or the top one of last x times the menu was accessed, or ever simpler, just use the current temp. (I have some pains in my hands, some days I count every key stroke and turn of the UM2 menu wheel...)
5. Please add a "quick-heatup" option for the heated plate. Reasons obvious. Let's say it could stay in this "hot and ready" state for ~15 mins before shutting off? For the shortcut, maybe even use a long press of the wheel.
I've heard a lot of "long press" suggestions already for a lot of different features. For obvious user-interaction-design reasons it's bad to have a long-press option. But together with question 4, I'll think about it.
Cool.
Another thing that has been suggested, is to start heating up the bed as soon as you enter the print selection menu. This can save you some time in heating up the bed, or, if you leave that menu open, it will heat up the bed. However, this isn't really obvious to the user that it is happening, and might scare people...
Not a bad idea. Maybe it can start after ~10secs of inactivity in the print menu, and even let you know it's happening on the LCD with a temperature prompt that stays there until you touch the wheel again.
6. The "time left" estimate on the LCD is beyond useless, esp with complex prints. It would be much easier and better to have Cura write its estimate into the output file and just have the machine count down from that.
Yes. It's a bit crappy. It has to account for a lot (adjusted speeds, incorrect estimates) I'm pretty sure there is a bug in there.
Right. And is using Cura's estimate a viable option? Just curious.
7. Oh, also, regarding the "change material" routine: Currently it heats up to 260 (I'm assuming) and pulls out slowly and then quicker, the ideal conditions for creating long thin strands of material that are sometimes left over in the tube. It should yank the material quite fast at first, as well as let the user set the pull-out temperature (let's say between 180-260, so as to avoid accidental damage).
I've tried this, didn't work. You can "yank" a lot faster with your hands then with the feeder. The filament change routine is actually something I spend quite a lot of time on. But I never got the strand removed 100%. (It heats up to the set material temperature)
I'm sure it's tricky, yes. For now, what I do is heat up the nozzle to 170-190, yank a bit by hand, then use the "Change" routine to help with removing the rest of the material and inserting the new one. The only annoying parts of that process are the switch from "heat nozzle" to "change material" and the associated wait time, as well as the wait time until the new material is high enough up the tube to let it accelerate by pressing "Ready". (I found that if I give it about 6cm it reaches the head with some speed still, and has no problem passing the small internal "lip" that has given me much trouble at first... esp if the tip of the filament was cut in the wrong direction...)
6. The "time left" estimate on the LCD is beyond useless, esp with complex prints. It would be much easier and better to have Cura write its estimate into the output file and just have the machine count down from that.
Yes. It's a bit crappy. It has to account for a lot (adjusted speeds, incorrect estimates) I'm pretty sure there is a bug in there.
I had a look at the code. It uses card.getFilePos() and extrapolates the time it took to get that far, to the total file size. Seems like a reasonable, low cost way to get an estimate. But of course this kind of estimate can be massively off, if the printed part is not homogenous from top to bottom.
I assume Cura's calculations are quite CPU intensive and not feasible to be repeated on the little 8-bit microcontroller - especially not during printing. Maybe a possible solution is to store Curas calculations as comments in the gcode file - so each layer would include a comment like ;TIME:120 meaning it takes 120 seconds to reach this layer. Then the printer would simply subtract this time from the total time written at the start of the file.
Maybe a possible solution is to store Curas calculations as comments in the gcode file - so each layer would include a comment like ;TIME:120 meaning it takes 120 seconds to reach this layer. Then the printer would simply subtract this time from the total time written at the start of the file.
Yes, I thought of something similar but never mentioned it. Each layer or occasionally throughout the gcode file there should be information about how far into the print cura thinks you are (time-wise). Then Marlin can do it's calculation based on that.
Because some layers are large and slow, and some layers are small and fast. Cura can include this information throughout the gcode (it doesn't have to be per layer - it could instead be every 60 seconds worth of gcode).
Even fancier - Cura could include qty of retractions for the remainder of the gcode and how long it thought they would take and Marlin could see how long they *really* take and adjust accordingly. Similar with other settings. For example Cura could include the assumed acceleration and how much faster it would be if acceleration is 50% faster and Marlin, knowing actual acceleration, could interpolate based on those 2 values. Same with Jerk setting. Same with feedrate adjuster (Cura could say if feedrate is at 150% how much faster it thinks the part would print - note that it's not linear because a lot of the movement time is restricted by jerk and acceleration settings).
Why not just send it with an M117 code before each layer?
http://reprap.org/wiki/G-code#M117_in_Marlin:_Display_Message
I would like to get back de USB connection to my Ultimaker2 As is was on the Ultimaker.
So Printing from Cura to USB is possible again.
On the fly settings are Easyer to do and monitoring the process is easyer.( remotely by remote desktop )
Also the Time to Print will then be more accurata.
Greetings Adriaan
I would like to get back de USB connection to my Ultimaker2 As is was on the Ultimaker.
So Printing from Cura to USB is possible again.
On the fly settings are Easyer to do and monitoring the process is easyer.( remotely by remote desktop )
Also the Time to Print will then be more accurata.
Greetings Adriaan
You can do that just fine if you switch the GCode flavor to RepRap.
But, we won't guarantee stability on the USB for the UM2.
ArunC posted a topic in UltiMaker Cura,
.thumb.jpeg.0b7a05eafc09add17b8338efde5852e9.jpeg)
Dustin posted a topic in Firmware,
Recommended Posts
gothampixel 10
small additional request for Cura. It would be nice to have a "clear platform" button grouped with load and save buttons.
Link to post
Share on other sites