Ok, here are the pics I had promised. The exact same gcode file was used to drive the prints from one test with marlin to the next test with repetier. Only changes were to flash the firmware and co-ordinate the various and many values between the two within each firmware. Understandably some tweaks could be made to both to improve upon the print specifically for each firmware. But more important was the testing to see if the firmware could complete the print without hanging. Marlin was more successful, well actually 100% successful, where repetier was only able to finish the one print of the bear without locking up completely.
I tried many different settings and variables over the course of 2 days. I did try various different baud rates and enabling ping-pong communication, as well as manually selecting binary and manually selecting ASCI. It just did not make a difference to preventing the lockups. I am open to the possibility that a setting withing repetier firmware is not set correctly and causing some type of overflow or communications issue that is beyond my knowledge of the topic. However I just really find it odd that everything was working so well for literally months for me and suddenly many issues. I orginally thought, oh great more hardware issues with my controller. But then I think marlin would have shown the same symptoms and problems in completion of prints would it not?
This first print is of a fan duct I was printing. both are from the same gcode file and yes both quit at about the same level. However the Marlin print in the background is clearly much better and more defined. The Marlin print was accidentially cancelled by a helpful cat on the keyboard that suddenly demanded love and attnetion and I mean now sort of thing, LOL.
This print of a bear was completed by both firmware and is included simply to be complete. I believe that repetier completed this print solely due to the fact that it was not long enough a print to arrive at the lockup point. All very short quick prints complete. Only the longer and more involved print run long enough to have time to reach the lockup point. repetier is in the foreground and Marlin in the back.
This cat print shows the two repetier prints that failed next to the Marlin print in the right side. An interesting point about these prints is that the two failed prints actually took twice as long to print as much as they did in comparison to the Marlin print which is of better quality and finished in half the time. I double checked the settings and all values were matched so I cannot explain this fully. The two repetier prints on the left were printed with 2 versions of repetier, the offical current seemecnc version and the middle one was printed with the newest Alpha development version of repetier (which is packed with TONS of wizzo cool new features and fun stuff so I really hope to be running it TBH)
All of the above prints printed at a layer height of .1mm. Not all failed and successful prints are shown, various slicers were used for both firmware variants but when comparing, the same gcode generating file from the same slicer is only used for comparisons. For example kiss generated gcode file was the same file to produce the print on Marlin and Repetier, slic3r same dealio. And I also used Simplify creator 3d and the very newest Cura recently posted to our forums, labeled closed beta.
The new cura really impressed me with it's output gcode files and I will be giving it more attention than I have previously. It may be simple and have a limited feature set of adjustable user options, but they seem to have things right.
I also printed with cura and creator, but there were other variables and issues with these hosts (mostly due to my inexperience with them I fully admit) so only repetier host was used for the apples to apples comparisons of firmware.
To be honest I have always been firmly in the repetier firmware camp of believers, and although I did find marlin to be easier to flash and have it setup with less fuss (far less doming adjustments for me for some reason, it seemed to just work right from the get go) repetier seemed to have a nicer feature set and my machine audibly seemed to like it's movement routines better. Unfortunately these tests and the lockup issues with repetier make it a non-contender until I can figure out why it locks on me and why I am not getting as presentable a print out of it in any version out there. I am going to try jumping to my ramps board this evening from my rambo and see if the results are similar or if my Rambo is just being a bitch about repetier.
Any thoughts/suggestions/ideas or requests for tests or values will be welcomed.
Thanks for your time and thoughts on this weird situation guys!
Gord
