Surface looks a lot worse in Cura 14.03

[wallan 0]

I'm mainly printing models made in zBrush.

In the 14.03 version surface looks a lot worse than in previous versions.

At least at the same speed.

In order to get a surface quality from 14.03 that I have in 14.01 or 13.12 i can only print at about 25-50% of the speed compared to the other versions.

The two images below shows an area of about 15x15mm.

The black dot is only to get the camera something to focus on.

Layer 0.1

Shell 0.8

Speed 50

Here is what the surface looks like in 14.01

And 14.03

Here is also a picture showing a print where the lower 1/3 is printed at 50% speed compared to the upper 2/3 that was printed at 100% speed (50mm/sec).

 

[mnis 11]

I'm not sure if it helps, but:

Have you ever tried while printing the "material flow" directly on the machine in "TUNE" to adjust?

I know this is not the right way to permanently solve problems. But it would be very interested if you so will achieve a visible improvement.

And the problem is certainly not the material currently used, or other mechanical problems?

Markus

[IRobertI 434]

I'm far, far from an expert on the software/firmware so this is just me thinking out loud. I believe daid (the creator of cura) mentioned a little while back that he had increased the resolution of the output in the latest version. Could it be that the Arduino is now being taxed too hard due to the increase in moves it has to plan and execute and not able to quite keep up, thus causing "micro stuttering"?

 

[wallan 0]

I'm not sure if it helps, but:Have you ever tried while printing the "material flow" directly on the machine in "TUNE" to adjust?

 

Yeah, that's what i did to figure out that lowering speed helps.

Also tried higher temperarure in the upper 1/3, 240deg i think but that made no difference.

[peggyb 142]

I did this test with the same settings in 14.01-14.03-14.03test1

The 14.03 behaved different, not making smooth round circles, you could hear that it made more straight lines as a circle.

speed 50, temp 210, Ultimaker org.

 

[wallan 0]

Nice to see the problem confirmed by someone else as well.

 

[illuminarti 16]

I edited the title of the topic (14.01 -> 14.03) just so it's clear what we're talking about.

 

[yzorg 2]

1+

I have some similar problems with highres models.

Since the upgrade to 14.03

think the gcode calculation of round shapes has been changed.

Its regular and not that strongly visible on my machine.. compared to Wallans pics. it looks a bit like a fine layer of dragonscales.

Compared to 14.01 i notice a quite strong noise change when printing said round shapes.

Its slower, much louder and seems to studder a bit. at least i think i remember it printing faster there..

This happens only when i print mega-highres models... which i do alot.

 

[Daid 285]

How does 14.03 compare to 14.03-TEST from http://software.ultimaker.com/Cura_closed_beta/ ? I changed a few things related to this in the TEST.

 

[wallan 0]

How does 14.03 compare to 14.03-TEST from http://software.ultimaker.com/Cura_closed_beta/ ? I changed a few things related to this in the TEST.

 

I can do a testprint but witch of the testversions are you talking about. There is both a Test1 and a Test2.

One of the prints posted by PeggyB seems to be the test1 version if that's the one you want to compare.

 

[Daid 285]

I can do a testprint but witch of the testversions are you talking about. There is both a Test1 and a Test2.

One of the prints posted by PeggyB seems to be the test1 version if that's the one you want to compare.

 

More reports is always better.

Test1 and Test2 are equal for these goals (they differ on an point unrelated to Ultimakers)

 

[wallan 0]

Ok.

Here we go...

14.03

14.03 Test2

 

[tlrt2701 0]

I print the next test with Cura14.03Test2. The print run 3h. The first 5mm Looks very nice and smooth

Thorsten

 

[tlrt2701 0]

The print is ready. Perfect! Better surface than the 14.01

Thanks!

Thorsten

 

[dubtec 0]

would be very interested if anyone has done other comparisons.

so it looks like 14.03 TEST2 is the one too beat?

 

[Daid 285]

so it looks like 14.03 TEST2 is the one too beat?

 

Looks like it. That's good, as that's the latest with all fixes for quality in it.

 

[eldrick 58]

Got a TEST2 for the Mac?

 

[Dim3nsioneer 426]

@Daid: Is it possible that 14.03-test2 doesn't save a profile string in the gcode? Is that on purpose?

 

[dubtec 0]

Off topic? I can't get14.03 test2 to slice. It hangs and becomes unresponsive, when it try to slice. Also if I try to rotate a part.I had the same issue with previous test builds to 14.03 I have to force close. Win.8.1 64bit

[Daid 285]

@Daid: Is it possible that 14.03-test2 doesn't save a profile string in the gcode? Is that on purpose?

 

Yes, you have to add ;{profile_string} to the end code, or else it won't add it in that version.

(The BFB firmware crashes on that line)

 

[dubtec 0]

Will this cause it to "hang"?

[Daid 285]

Will this cause it to "hang"?

 

No. That's an unrelated issue, no idea what could be causing that...

 

[Dim3nsioneer 426]

Sorry, another bit off-topic question: Is the BfB-flavor on a level where it can and should be tested? I know someone with a BfB printer who is quite annoyed by the discontinued BfB software.

 

[fingerpuk 0]

Is there a TEST2 for Mac? I'm noticing some similar effects to the OP and would like to test an alternative version of Cura.

 

[Daid 285]

Is there a TEST2 for Mac? I'm noticing some similar effects to the OP and would like to test an alternative version of Cura.

 

TEST1 and TEST2 are the same except for the BFB profile and the Win8 drivers.