shrinking to zero doesn't actually scales the radius to zero #96238

New Issue

Diogo Valadares Reis dos Santos · 2022-03-08T15:08:44+01:00

Diogo Valadares Reis dos Santos commented

2022-03-08 15:08:44 +01:00

System Information
Operating system: Windows-10-10.0.19041-SP0 64 Bits
Graphics card: AMD Radeon RX 5700 XT ATI Technologies Inc. 4.5.14802 Core Profile Context 22.2.1 30.0.14023.7007

Blender Version
Broken: version: 3.2.0 Alpha, branch: master, commit date: 2022-03-04 16:11, hash: 295d5c6ef5

Short description of error
if you use alt+s and then type 0, the radius of the the curve goes to 0.001 instead of 0

Exact steps for others to reproduce the error

click on a curve point
use alt+s to shrink it
type 0, if you zoom in or look at the n-panel, you will not that the radius isn't 0.

2022-03-08 10-54-21.mp4
you can resize it to any number without any problem, but specifically at 0, instead of setting the radius to 0 it sets to 0.001

Extra note: This may be better reported in another thread but while reporting this bug I noticed that the status bar where you type the numbers wrongly says that the operation is shrink/flatten, but there is no such thing for curves, it probably should be something as scale point. I tried to find it in documentation and I only found that alt+s should be using "radius" but in this case the documentation would be wrong because we are not defining the radius of the point directly, we are scaling it. https://docs.blender.org/manual/en/latest/modeling/curves/editing/curve.html#radius

Summing up this extra note:
alt+s mode naming in curve edit in blender is probably wrong
the operation may also be wrong, but I'm not sure if its wrong in blender or if its wrong in the documentation

**System Information** Operating system: Windows-10-10.0.19041-SP0 64 Bits Graphics card: AMD Radeon RX 5700 XT ATI Technologies Inc. 4.5.14802 Core Profile Context 22.2.1 30.0.14023.7007 **Blender Version** Broken: version: 3.2.0 Alpha, branch: master, commit date: 2022-03-04 16:11, hash: `295d5c6ef5` **Short description of error** if you use alt+s and then type 0, the radius of the the curve goes to 0.001 instead of 0 **Exact steps for others to reproduce the error** - click on a curve point - use alt+s to shrink it - type 0, if you zoom in or look at the n-panel, you will not that the radius isn't 0. [2022-03-08 10-54-21.mp4](https://archive.blender.org/developer/F12907614/2022-03-08_10-54-21.mp4) you can resize it to any number without any problem, but specifically at 0, instead of setting the radius to 0 it sets to 0.001 Extra note: This may be better reported in another thread but while reporting this bug I noticed that the status bar where you type the numbers wrongly says that the operation is shrink/flatten, but there is no such thing for curves, it probably should be something as scale point. I tried to find it in documentation and I only found that alt+s should be using "radius" but in this case the documentation would be wrong because we are not defining the radius of the point directly, we are scaling it. https://docs.blender.org/manual/en/latest/modeling/curves/editing/curve.html#radius Summing up this extra note: alt+s mode naming in curve edit in blender is probably wrong the operation may also be wrong, but I'm not sure if its wrong in blender or if its wrong in the documentation

Diogo Valadares Reis dos Santos commented

2022-03-08 15:08:45 +01:00

Added subscriber: @Diogo_Valadares

Philipp Oeser commented

2022-03-08 15:38:41 +01:00

Added subscribers: @mano-wii, @lichtwerk

Philipp Oeser commented

2022-03-08 15:38:41 +01:00

Changed status from 'Needs Triage' to: 'Needs Developer To Reproduce'

Philipp Oeser commented

2022-03-08 15:38:41 +01:00

Hm, this was like that since forever.
@mano-wii : since there is no real limit in RNA, would you have an idea why transform code limits it to 0.001f?

Hm, this was like that since forever. @mano-wii : since there is no real limit in RNA, would you have an idea why transform code limits it to `0.001f`?

Germano Cavalcante commented

2022-03-08 17:40:46 +01:00

Added subscriber: @broken

Germano Cavalcante commented

2022-03-08 17:40:48 +01:00

In #96238#1318809, @lichtwerk wrote:
@mano-wii : since there is no real limit in RNA, would you have an idea why transform code limits it to 0.001f?

This has been the case since when the operator was implemented in d3028ec70d.
Maybe only @broken knows the answer.

My guess is that this was done to prevent the operation from getting stuck after successive Shrink Fatten calls if one of them results in a radius of 0.0.
But that doesn't prevent this radius from being set to 0.0 from the UI.
So it doesn't seem like a valid reason to limit the value.

Here a possible solution:

diff --git a/source/blender/editors/transform/transform_mode_curveshrinkfatten.c b/source/blender/editors/transform/transform_mode_curveshrinkfatten.c
index aa4d608de04..c9f0c949135 100644
--- a/source/blender/editors/transform/transform_mode_curveshrinkfatten.c
+++ b/source/blender/editors/transform/transform_mode_curveshrinkfatten.c
@@ -63,11 +63,10 @@ static void applyCurveShrinkFatten(TransInfo *t, const int UNUSED(mval[2]))
 
       if (td->val) {
         *td->val = td->ival * ratio;
+
         /* apply PET */
         *td->val = (*td->val * td->factor) + ((1.0f - td->factor) * td->ival);
-        if (*td->val <= 0.0f) {
-          *td->val = 0.001f;
-        }
+        CLAMP_MIN(*td->val, 0.0f);
       }
     }
   }
@@ -93,10 +92,6 @@ void initCurveShrinkFatten(TransInfo *t)
   t->num.unit_sys = t->scene->unit.system;
   t->num.unit_type[0] = B_UNIT_NONE;
 
-#ifdef USE_NUM_NO_ZERO
-  t->num.val_flag[0] |= NUM_NO_ZERO;
-#endif
-
   t->flag |= T_NO_CONSTRAINT;
 }

> In #96238#1318809, @lichtwerk wrote: > @mano-wii : since there is no real limit in RNA, would you have an idea why transform code limits it to `0.001f`? This has been the case since when the operator was implemented in d3028ec70d. Maybe only @broken knows the answer. My guess is that this was done to prevent the operation from getting stuck after successive Shrink Fatten calls if one of them results in a radius of 0.0. But that doesn't prevent this radius from being set to 0.0 from the UI. So it doesn't seem like a valid reason to limit the value. Here a possible solution: ``` diff --git a/source/blender/editors/transform/transform_mode_curveshrinkfatten.c b/source/blender/editors/transform/transform_mode_curveshrinkfatten.c index aa4d608de04..c9f0c949135 100644 --- a/source/blender/editors/transform/transform_mode_curveshrinkfatten.c +++ b/source/blender/editors/transform/transform_mode_curveshrinkfatten.c @@ -63,11 +63,10 @@ static void applyCurveShrinkFatten(TransInfo *t, const int UNUSED(mval[2])) if (td->val) { *td->val = td->ival * ratio; + /* apply PET */ *td->val = (*td->val * td->factor) + ((1.0f - td->factor) * td->ival); - if (*td->val <= 0.0f) { - *td->val = 0.001f; - } + CLAMP_MIN(*td->val, 0.0f); } } } @@ -93,10 +92,6 @@ void initCurveShrinkFatten(TransInfo *t) t->num.unit_sys = t->scene->unit.system; t->num.unit_type[0] = B_UNIT_NONE; -#ifdef USE_NUM_NO_ZERO - t->num.val_flag[0] |= NUM_NO_ZERO; -#endif - t->flag |= T_NO_CONSTRAINT; } ```

Diogo Valadares Reis dos Santos commented

2022-07-11 17:18:29 +02:00

I guess Matt isn't active anymore, so I don't know if we will get an answer anytime soon...
I usually like to set the radius of hair tips to 0 and the only problem I have is if I want to increase the size later, tho it can be done by manually setting the size in the N-Panel.
Having this soft limit usually gives me problems whenever I need to merge the tips of the curves.

I guess Matt isn't active anymore, so I don't know if we will get an answer anytime soon... I usually like to set the radius of hair tips to 0 and the only problem I have is if I want to increase the size later, tho it can be done by manually setting the size in the N-Panel. Having this soft limit usually gives me problems whenever I need to merge the tips of the curves.

blender-admin commented

2022-07-12 15:11:05 +02:00

This issue was referenced by 2d1fe736fa

This issue was referenced by 2d1fe736fabdcccdd46edfbdfc68720498c30e10

Germano Cavalcante commented

2022-07-12 15:20:13 +02:00

Changed status from 'Needs Developer To Reproduce' to: 'Resolved'

Germano Cavalcante closed this issue

2022-07-12 15:20:13 +02:00

Germano Cavalcante self-assigned this 2022-07-12 15:20:13 +02:00

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

shrinking to zero doesn't actually scales the radius to zero #96238