# New source of error

**URL:** <https://forums.maslowcnc.com/t/new-source-of-error/22404>\
**Category:** Uncategorized\
**Created:** [September 27, 2024, 9:31pm UTC](https://forums.maslowcnc.com/t/new-source-of-error/22404 "2024-09-27T21:31:30Z")\
**Posts on this page:** 1\
**Showing post:** 56

<div class="post-metadata">

**Author:** ![dlang](https://yyz1.discourse-cdn.com/flex031/user_avatar/forums.maslowcnc.com/dlang/32/859_2.png) [@dlang](https://forums.maslowcnc.com/u/dlang)\
**Post date:** [October 30, 2024, 5:19pm UTC](https://forums.maslowcnc.com/t/new-source-of-error/22404/56 "2024-10-30T17:19:52Z")

</div>

Bar wrote:

> > [@dlang](#):
> >
> > I¢m not placing the term, what is the gradient decent algorithm?
> 
> The one discussed here: [Loose belts during calibration - #67 by bar](https://forums.maslowcnc.com/t/loose-belts-during-calibration/22248/67)
> 
> It seems to be somewhat unstable though which is giving me pause

can it be added as an option for people to try?

But I think we want to fix the projected length → belt length and belt length  
 → projected length calculations before we do anything else.

David Lang

---

_[View the full topic](https://forums.maslowcnc.com/t/new-source-of-error/22404)._