Wrong symmetrical seam corners on pieces

Issue #937 resolved
Yiannis created an issue

It seems like symmerical seam corners are not correct on patterns.
The red lines show how symmetrical seam corners should be.

wrong symmetrical seam corners.jpg

Please find attached the "wrong symmetrical seam corners.val" file from which the image was created.

The "symmetrical seam corners simulation.val" file is attached to show how those corners should be calculated. Try changing the purple lines shape to understand how corners should change on different pattern shapes.

Comments (29)

  1. Roman Telezhynskyi repo owner

    Although the angle type is not working as intended. Right now the name doesn't represent what it does. It should truncate corner, the truncation line is symmetrical to edge relative to the opposite seam line.

    Also i don't like abstract nature of your examples. So, one of them i will ignore. I don't see big sense to try improve seam allowance for not real case. Not all abstract cases will produce nice looking seam allowance.

  2. Yiannis reporter

    Is this the use of angle type you are talking about?

    symcorner.jpg
    symcorner2.jpg

    This is indeed a symmetrical corner and it is exactly the same kind of seam corner used both in hems and the type of seams you are talking about.

    Following you can find some examples where the same symmetrical seam corner type has been applied (in both seams and hems)

    symcorner3.jpg
    Please let me know if you need any further explanation.

  3. Roman Telezhynskyi repo owner

    Please let me know if you need any further explanation.

    Sorry, just slow down little bit. I don't understand what you are trying to say.

  4. Yiannis reporter

    I didn't mean to force you to do something. I was just trying to explain that symmetrical corners work OK for both cases. That's the way it usually works in patternmaking programs. I am sorry if there is any misunderstanding.

  5. Roman Telezhynskyi repo owner

    Don't worry. Your question too much pattern making specific for me. Don't forget, i am not pattern maker. I just need translation. :)

    So, where is issue? Do i still need to fix something?

  6. Alexei Burmistrov

    Roman, the bug really exists. If you change seam allowance settings in "wrong symmetrical seam corners.val" to correct ones [A2: by second edge right angle] and [A3: by first edge right angle], you will not achieve the required symmetry. The bug shows up when the reference edge is a straight line. If it is a curve all works fine. Please, reopen the issue.

  7. Roman Telezhynskyi repo owner

    Did you test my changes? Can you show me screenshot how it is and how it you think should be?

  8. Alexei Burmistrov

    Роман, последние изменения ничего не изменили. Давайте посмотрим на ваш скриншот. То что хочет репортер -- это получить усечение, симметричное относительно выбранной стороны.. Зачем сплошь и рядом используется симметричное усечение? Давайте попробуем после сшивания по A2-A3 уложить припуск на деталь. В нижних скришотах припуск совместится с приходящим припуском по A-A3. В верхних же припуск вылезет за пределы приходящего припуска и получится брак. Поэтому поведение припусков данного типа для прямых нужно привести к таковому для кривых

  9. Roman Telezhynskyi repo owner

    Ничего не остается. Вам придется показать на моем скриншоте "правильную" форму. Пока что я не вижу где здесь "ошибка".

  10. Alexei Burmistrov

    А нет, я не прав. для кривых тоже неверно считается. Просто незаметно в случае угла близкого к прямому. Вот примерно поправил как должно было быть. Посмотрите что будет c совмещением, если сложить по линии A2-A3 и B2-B3
    screenshot_modified.png

  11. Roman Telezhynskyi repo owner

    As i already mentioned for original author the right approach to achieve this is to use custom seam allowance included as main path. I even included test file no one apparently tried to open. :(

  12. Yiannis reporter

    I'm glad you are reopening this issue. In fact, I did yesterday check the file you uploaded with the custom hemline but this doesn't seem to be a convenient way to do it. As you asked me to slow down and you marked the issue as resolved I did not send any further reply. However I will be really very happy to discuss it again.
    On the image you uploaded earlier the seam corners seem to be the "right angle" ones instead of "symmetrical" ones.
    Alexei's image seems to be correct for the staight line shapes, but on the curved line shapes it looks as if "right angle" corners have been applied.
    Would a "step by step procedure" image be of help to you?
    Let me know if you are interested in this.

  13. Roman Telezhynskyi repo owner

    @ynsgabr, actually i think @mistflock helped me to fix this issue. We still need to check this fix though. Big part of conversation happened in our Telegram chat group. Unfortunately for english speaking it is mostly occupied by russian speaking. :)

    As you asked me to slow down and you marked the issue as resolved I did not send any further reply.

    This should never stop you. I can be wrong. :) This issue is very tricker.

  14. Yiannis reporter

    That looks almost fine and should be OK for most cases! This is what I was trying to tell you!

    Just a little detail: I would be even "more correct" if the mirrored lines on the bottom shapes were curved as well, keeping the same curvature as the side seams of the piece.

    Thanks a lot for your effort on trying to understand what's in a patternmaker's mind and improving it at last!

  15. Roman Telezhynskyi repo owner

    Just a little detail: I would be even "more correct" if the mirrored lines on the bottom shapes were curved as well, keeping the same curvature as the side seams of the piece.

    Yes, i understand, but unfortunately this is limitation of the current implementation. I cannot do anything.

  16. Catherine Richard

    I think this is not really a 'bug', just a confusion between margin seams and hems. Valentina cannot understand which seam of the piece has to be mirrored (bottom, side seams ...). The symetric tools draft mirroring bottom hems perfectly well.

  17. Log in to comment