Makerware - Comparing Mac vs PC Time to prints....Vastly different

[James Hannon]

I am using both a Mac and a PC with my 5th Gen Replicator....   I like to work on creating stuff with the Mac, throw it in a Dropbox, pick it up on the PC (Asus) laptop and print from there using a USB cable (the logic behind this is that I don't want to tie up my Mac laptop while I'm printing a long project)

So I create 2 shapes using thinkcad and duplicate it via Makerware on the platform (see below)...    This duplication was done on the mac.  When I created a preview file, it said it would take about 7 hours and 45 minutes...  (standard (.2)  printing with raft/support)...  I then save the thing file on my Dropbox and go to the PC laptop

The PC laptop loads it in, and when I create a preview file there (using the same settings), it says it will take 6 hours and 15 minutes...(filament was diff by .02 grams which isn't much)   Woohoo, right?   No... :(

So I print it (from the PC)...  at the 3 hour mark, it says 50%....  So its looking like 6 hours & 15 mins is a possibility....   But then it slows down...At the 5 hour mark, it was only 75% done...  Finally at the 7 hour and 40-something mark it was done

I had to print it twice - once in white filament and once in black and they were both within 2 minutes of each other (and the Mac Laptop estimated time)  (7:41:41 and 7:43:40 - see picture below).  Both prints are now done and I'm happy with the final results --  So, this post was mostly to report the difference in estimates for the same file on two machines...   But I am curious why the printer had it at 50% at 3 hours and then took another 4 hours and 41 minutes to print the other 50%?

[Ryan Carlyle]

Main problem is, the slicer is not strictly aware of the acceleration settings on your printer. Acceleration (despite the name) slows down the print speed around corners to make sure the machine doesn't shake itself apart. This allows higher top speeds than non-accelerated printing, so slowing down corners actually improves overall print times. But "actual" is always slower than "theoretical" based on slicer commands alone.

Anyway, the acceleration code is executed by the firmware on the printer itself. The slicer (Makerbot Desktop) mostly just sets target speeds (or speed limits). Since the printer doesn't spend all its time at the known "speed limit", a direct time calculation from the slicer would be wrong. So Makerbot assumes a certain discount on the print speed when calculating build times in the print preview. That assumption isn't always accurate.

Given knowledge about the acceleration settings -- which you would expect Makerbot to have for their own printer, connected via USB -- it's possible to precisely calculate the true print time. There are simulators that do this. But Makerbot is kind of lazy, and it's a complex simulation, so accurate time estimates simply are not a feature they chose to include.

The "percent complete" figures displayed during the build are another layer of estimation on top of all that. The methodology used to calculate build progress is too simple to take into account certain aspects of the print geometry and acceleration settings. Finer details and smaller features cause more slowdowns.

So that explains why the various estimates would be wrong. But I have no idea why two different versions of Makerbot Desktop would give you different estimates. Presumably, the one connected to USB is using a different "speed discount factor" since it thinks it knows how the connected printer will perform. But this is all just sloppy, half-assed estimation code. Don't expect high precision.

[Robert Nee]

I only have access to a pc for printing but I notice that with the makerbot mini the print time estimates are about 2/3 of the actual print times I get.

[James Hannon]

If both the PC and Mac Software estimates were drastically off, or the PC was the accurate one (being attached and thus more "in-tune" with the specific attached printer), it would make sense...  But the unattached Mac Version was within 5 minutes of the actual time...

I printed some more stuff last night - It was quick, so I used the Mac - And I tested whether I can disconnect the USB cable from the Mac laptop, and still have it continue printing....    And it worked...  Disconnected and the job finished fine...

The night before, I couldn't even minimize the printing window on the PC version without it asking me if I wanted to Cancel the Print...

I'm thinking the decision-making folks over at Makerbot are using Macs to create the software and then just playing catchup with the PC stuff...

[Tiffany Diorio]

Hi James, 

 

 Saw your post and wanted to let you know that you don't have to keep MakerBot Desktop open while your Replicator Fifth Generation printer is printing. Once the print starts it's fine to disconnect because the entire file was already transferred over to the printer.

 If you have any other questions let me know.

-Tiffany

NOTICE: This email may contain information that is confidential or attorney-client privileged and may constitute inside information or trade secrets. The contents of this email are intended only for the recipient(s) listed above. If you are not the intended recipient, you are directed not to read, disclose, distribute or otherwise use this transmission. If you have received this email in error, please notify the sender immediately and delete the transmission. Delivery of this message is not intended to waive any applicable privileges.

[James Hannon]

Hi Tiffany, on the Mac that's true and I am able to disconnect... On the PC version however, once I sent it to print, the screen wouldn't go away unless I hit "Cancel Print"  (which I didnt want to do)