Mirror Mode

[wildcard]

When ever I do Copy, Flip and Attach to a model, depending on how complex the model is, I'll end up making minor changes to fix things.

Correct some splines re do some patches, nothing major.

 

Pretty much everything that wasn't modified is still nicely working with Mirror Mode, but those control points I modified are no longer mirrored.

Once my model has been made whole with copy flip attach, I've to worry about making adjustments to both half's Mirror Mode takes care of that.

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

[robcat2075]

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

 

Presuming you made all your changes on one side you just do a CFA again.

[wildcard]

How I hate finding the solution shortly after I posted on the forums for guidance.

 

Selecting both points, scaling down to 0% along the y and z axis, wile both still selected inverse the x axis Pivot value in the Translate X axis.

Or at least that`s how I got both points to be mirrored again. If any one has a quicker way, please let me know.

 

*edit*

Or at least it seems to work most of the time. Any way I can Manually entering x, y and z coordinates for a control point?

Working with the Transform properties, by adding/subtracting small changes to move the CP around is a royal pain in the

[markw]

When ever I do Copy, Flip and Attach to a model, depending on how complex the model is, I'll end up making minor changes to fix things.

Correct some splines re do some patches, nothing major.

 

Pretty much everything that wasn't modified is still nicely working with Mirror Mode, but those control points I modified are no longer mirrored.

Once my model has been made whole with copy flip attach, I've to worry about making adjustments to both half's Mirror Mode takes care of that.

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

 

Try this. If you have CPs in a symmetrical model that have lost contact with their mirrored counterparts in "Mirror Mode" after "Copy/FlipAttach", you can reconnect them as follows;

First turn Mirror Mode OFF. Then left click the troublesome CP to select it. Now right click on the selected CP and a contextual menu will appear. From it select "Snap to Mirrored point(s)". Turn "Mirror Mode back on and they should be reconnected now.

[Fuchur]

When ever I do Copy, Flip and Attach to a model, depending on how complex the model is, I'll end up making minor changes to fix things.

Correct some splines re do some patches, nothing major.

 

Pretty much everything that wasn't modified is still nicely working with Mirror Mode, but those control points I modified are no longer mirrored.

Once my model has been made whole with copy flip attach, I've to worry about making adjustments to both half's Mirror Mode takes care of that.

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

 

Try this. If you have CPs in a symmetrical model that have lost contact with their mirrored counterparts in "Mirror Mode" after "Copy/FlipAttach", you can reconnect them as follows;

First turn Mirror Mode OFF. Then left click the troublesome CP to select it. Now right click on the selected CP and a contextual menu will appear. From it select "Snap to Mirrored point(s)". Turn "Mirror Mode back on and they should be reconnected now.

 

Wow, thanks Mark, I didn't know that one!

 

See you

*Fuchur*

[wildcard]

When ever I do Copy, Flip and Attach to a model, depending on how complex the model is, I'll end up making minor changes to fix things.

Correct some splines re do some patches, nothing major.

 

Pretty much everything that wasn't modified is still nicely working with Mirror Mode, but those control points I modified are no longer mirrored.

Once my model has been made whole with copy flip attach, I've to worry about making adjustments to both half's Mirror Mode takes care of that.

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

 

Try this. If you have CPs in a symmetrical model that have lost contact with their mirrored counterparts in "Mirror Mode" after "Copy/FlipAttach", you can reconnect them as follows;

First turn Mirror Mode OFF. Then left click the troublesome CP to select it. Now right click on the selected CP and a contextual menu will appear. From it select "Snap to Mirrored point(s)". Turn "Mirror Mode back on and they should be reconnected now.

 

Thanks for the guidance. I recall using it before but it didn't work for me, must have been because mirror mode was still on.

I'll toy around with it some more, and make sure mirror mode is toggled off.

[robcat2075]

I didn't know that one either!

[zandoriastudios]

great tip!

[Fuchur]

When ever I do Copy, Flip and Attach to a model, depending on how complex the model is, I'll end up making minor changes to fix things.

Correct some splines re do some patches, nothing major.

 

Pretty much everything that wasn't modified is still nicely working with Mirror Mode, but those control points I modified are no longer mirrored.

Once my model has been made whole with copy flip attach, I've to worry about making adjustments to both half's Mirror Mode takes care of that.

So, is there any way I can get control points back in sync again with there mirror control point, so they work again with Mirror mode?

 

Try this. If you have CPs in a symmetrical model that have lost contact with their mirrored counterparts in "Mirror Mode" after "Copy/FlipAttach", you can reconnect them as follows;

First turn Mirror Mode OFF. Then left click the troublesome CP to select it. Now right click on the selected CP and a contextual menu will appear. From it select "Snap to Mirrored point(s)". Turn "Mirror Mode back on and they should be reconnected now.

 

Thanks for the guidance. I recall using it before but it didn't work for me, must have been because mirror mode was still on.

I'll toy around with it some more, and make sure mirror mode is toggled off.

 

I tested it and it seems to have to do with the Mirror-Mode-Tolerance (and / or the group-snap-tolerance) - value.

If the difference of the mirrored CP is larger than the tolerance-values, it won't work.

Have a look at "Tools > Options > Modelling" and see if you can get it to work.

For my test it worked when I increased the tolerance-values there.

 

See you

*Fuchur*

[wildcard]

Got the "Snap to Mirrored point(s)" thingy to work or at least it works most of the time.

 

The problem, or I should say the my frustration, is more that when it doesn't work, I've to manually align one point to match to mirror position of the other. I don't know what the conditions are or what allows AM to find the mirror CP(s), but there are times I've to manually align them and it's the manual positioning of one CP that I find a real pain to do.

 

AM doesn't allow me to simply enter the (mirror) XYZ coordinates to move the CP.

Would there be an option to move a CP to a designated location by entering XYZ values, please share how...

 

*Quick Edit*

Thanks Fuchur,

Changing those values made my live easier, my model is of large scale so the default 0.05cm is way to small.

Will have to toy around with it a little bit to find the sweet spot that works well for me and my model scale.

[HomeSlice]

Would there be an option to move a CP to a designated location by entering XYZ values, please share how...

Hit the "Show Manipulator Properties" button in the top toolbar. Then select a CP. You should then be able to enter specific XYZ coordinates in the onscreen widget.

 

When I have to realign mirrored CPs and "Snap to Mirrored Points" doesn't work, do this:

(Assuming the CPs are mirrored *across* the Y axis)

Select both CPs.

For Scale, enter 0 for scale Y and scale Z.

Make sure you have a marker at 0 donw the center of your model.

Use the arrow keys to nudge the yellow selection rectangle right or left until the center of the selection box is at 0.

You can see how close to 0 you are by looking at the X Pivot property in the Properties panel.

In v16, whenever you nudge the selection box, you have to click once inside it in order to update the value in the Pivot property.

If in Tools > Options > Modeling, you have "Mirror Mode Tolerance" set to .02cm, then the pivot can be anywhere from -.02cm - .02cm along the X axis.

Then, if you want to do an addition step (not always necessary), right-click on one of the CPs, so only one is selected, and choose "Snap to Mirrored".

[John Bigboote]

New tip to me!

 

THAT would be a good new tip to put in the startup tips! 'Snap to Mirrored Points'... never even knew it was there!

[wildcard]

Would there be an option to move a CP to a designated location by entering XYZ values, please share how...

Hit the "Show Manipulator Properties" button in the top toolbar. Then select a CP. You should then be able to enter specific XYZ coordinates in the onscreen widget.

 

Thanks HomeSlice,

 

Like you explained, I've already been selecting both points and scale them down to 0 along the Y and Z axisnow Now that I know of the "Show Manipulator Properties" I find it less hassle to get CP moved to the right spot.

[jirard]

I would like to add something. I assume that after you copy flip attach that there are only a few control points out of whack. Try this when all else fails. After you cfa delete the unaligned points on both sides and select those splines and press the y key and reconnect all splines. This should fix it.

[wildcard]

This is really giving me a headache.

 

I know how to align two CP's through scaling the Y and Z axis or by manually entering the the XYZ coordinates through the use of the show manipulator properties and than snap to mirror points.

 

Once Copy/Flip/Attach has been applied, any modifications made to the CFA model is a time consuming task to get realigned again.

 

Cutting my model back in half and than doing CFA again often feels faster and less work, but this isn't always and option when two sides are not identical.

 

 

Don't know if there is a place to make future request or can get some plugin crazy code monkey to have a shot at creating an option to simplify this task.

A key combination similar to copy/past would really be nice.

Select the CP and Copy the data from that CP through the use of a hotkey

Select the CP that needs to be mirrored and Past the data in to the PC through the use of a hotkey.

[robcat2075]

Cutting my model back in half and than doing CFA again often feels faster and less work, but this isn't always and option when two sides are not identical.

 

that's where you lose me. Don't you want them identical?

[wildcard]

Cutting my model back in half and than doing CFA again often feels faster and less work, but this isn't always and option when two sides are not identical.

 

that's where you lose me. Don't you want them identical?

 

In most situations yes you keep both sides identical.

But I'm pretty sure there are cases where you don't have identical half`s but the majority is still identical.

 

I'm not really pleased with cutting my model up in half again to have it shortly after put back to gather again.

CFA for me pairs with bias handles and extreme magnitude values that need to be corrected, specially on larger models.

I find it a frustration to re-align CP`s that are not mirrored, fixing CP`s with bias handles/magnitude values that are out of whack after a CFA is equal frustrating

 

I'm unable to fully shape my model from one half only. I sometimes simply need to see the model as a whole to get a better feel for the shape and curves and make adjustments form there.

[robcat2075]

Implementing what you want in software sounds daunting. Some sort ofnew data would have to be stored to selectively associate CPs that were once true mirrors but are no longer.

 

From the screen shots you're showing, I don't understand what you are trying to model that can't be achieved. however...

 

My suggestion for a work flow would be...

 

1) model a mirrored form as long as possible using mirror mode or CFA

2) then work from that basic model to make the unique non-mirrored alterations

[wildcard]

From the screen shots you're showing, I don't understand what you are trying to model that can't be achieved. however...

 

My suggestion for a work flow would be...

 

1) model a mirrored form as long as possible using mirror mode or CFA

2) then work from that basic model to make the unique non-mirrored alterations

 

The screen shot only shows my typical bias handles/magnitude values going to extreme after a CFA.

 

Your work flow is correct and that is how I should be working, however I find it hard to work on only one half of a model and not seeing it a whole model. This also results in my frustration when adding and modifying CP`s and getting them correctly aligned to get them mirrored.

[robcat2075]

"not seeing a whole mode?"

 

1) have the whole model visible

2) edit on one side with mirror mode on until that doesn't keep up

3) copy your edit side

4) delete everything

5) paste back what you just copied

6) do a CFA on it

7) resume editing on one side with mirror mode on as in step 2

[wildcard]

"not seeing a whole mode?"

 

1) have the whole model visible

2) edit on one side with mirror mode on until that doesn't keep up

3) copy your edit side

4) delete everything

5) paste back what you just copied

6) do a CFA on it

7) resume editing on one side with mirror mode on as in step 2

 

Tried it, no luck with a flawless CPA

 

So for me there is step

6,5) Fix CP's out of whack

[robcat2075]

Can you post an example model that won't mirror after you CFA?

[wildcard]

Can you post an example model that won't mirror after you CFA?

 

Project file with model embedded and a quick recording of what I'm seeing on my end.

Female_Model1.prj

CFA.mov

[mtpeak2]

Once you delete half the model, select the spline down the center of the model and scale it to 0 on the X axis with an X-pivot of 0. Hooks and five pointers have issues next to the center spline, so you will need to fix them after the CFA.

[robcat2075]

This is not a problem with mirror mode not working. This is a problem with CFA not working.

 

On the left is the model CFA'd in v15 and on the right in v16. Obviously something is way out of whack there.

 

 

 

I'll AM report that.

[robcat2075]

Now an AMReport...

 

http://www.hash.com/reports/bug_view_advan...php?bug_id=5960

[mtpeak2]

If the center spline is scaled to 0 on the X axis, it CFA's fine, as I said in my previous post.

[robcat2075]

In my sample PRJ I put in the AMReport it's as zeroed as zeroed can be and it still doesn't CFA properly. Works in v15, doesn't work in v16

[mtpeak2]

As I said, it CFA'ed fine after scaling the center spline to 0. This is from the project from A:M reports (I changed the color of the model). I used the scale manipulator and the show manipulator properties to scale it.

[mtpeak2]

As you can see, the X-pivot is not 0, even though the group pivot says it's 0.

[robcat2075]

that does make it work!

 

However, that discrepancy from the 0 axis is within the mirror mode tolerance (0.05 on mine) and v15 is able to deal with it while v16 is not.

 

 

It looks as if anything with a hook is a problem.

[mtpeak2]

Does the mirror mode tolerances even have anything to do with CFA?

[robcat2075]

Does the mirror mode tolerances even have anything to do with CFA?

 

 

Probably not, but I just thought I'd through that in there.

[markw]

Thought I might just add that it's not just using C/F/A which messes with some of the CP bias settings in A:Mv16. As you see copying, pasting then flipping the whole model will also do it but only when flipping on the X axis it seems.

[robcat2075]

Thought I might just add that it's not just using C/F/A which messes with some of the CP bias settings in A:Mv16. As you see copying, pasting then flipping the whole model will also do it but only when flipping on the X axis it seems.

 

Add that as a "note" to the AMReport. I think it's the Hooks that are getting messed up.

[markw]

Thought I might just add that it's not just using C/F/A which messes with some of the CP bias settings in A:Mv16. As you see copying, pasting then flipping the whole model will also do it but only when flipping on the X axis it seems.

 

Add that as a "note" to the AMReport. I think it's the Hooks that are getting messed up.

Done. Note added.

[wildcard]

Thought I might just add that it's not just using C/F/A which messes with some of the CP bias settings in A:Mv16. As you see copying, pasting then flipping the whole model will also do it but only when flipping on the X axis it seems.

 

Add that as a "note" to the AMReport. I think it's the Hooks that are getting messed up.

 

I believe your spot on with your conclusion it might be the hooks that are causing the problems.

Removed all the hooks I could find, didn't make any other kind adjustments to the model (scale x to 0 as was talked about) and with the exception of all the holes I created by removing the hooks, the CFA worked flawlessly.

 

Created a sphere with 8 lathe steps, created two hooks on one side, cut it in half and did a CFA bam, CP's gone out of whack.

Repeated the sphere creation, but now with 64 lathe steps and on one side create two hooks, cut in half CFA and nothing, worked perfectly.

[Walter Baker]

Well thanks to everyone for the above info which will help me but I always did it the hard way.

Use the manipulator properties button on the top of the window

Click on the correct one, right down the X,Y,Z cord. the click on the wrong one and type in the numbers but be sure to do a negative when needed. You will see quickly when you do it.

Ok back to cooking supper.