iNMR news #107

16 views
Skip to first unread message

GB, author of iNMR

unread,
Jun 2, 2026, 4:28:46 PM (11 days ago) Jun 2
to iNMR
What's New in versions 7.1.7 and 7.1.8

Dear iNMR user,

Two new releases in a short time — here is a roundup of what has changed and what is coming next.


── A NEW LOOK FOR THE MANUAL ──

The content is the same, but everything else has been redesigned. Larger, more readable type. A sidebar that stays open so the table of contents is always visible. You can now read the manual from cover to cover in sequence, following the TOC page by page — something that was not really possible before.

The manual also adapts automatically to light and dark mode, so it is comfortable to read whatever your system preference. The overall design is cleaner and more in line with current standards.

If you are on a Mac, the manual is built into the app as always. It is also available online for quick reference from any device.


── FLATTEN: A NEW COMMAND FOR CHEMICAL STRUCTURES ──

If you work with MOL files containing 3D structures, you will appreciate the new Flatten command in the contextual menu attached to a chemical formula. It projects the structure onto a plane so that all atoms and bonds are clearly visible at a glance.

Flatten works together with Open Babel, a powerful free tool that has been part of the recommended iNMR workflow for some time. Until recently, Open Babel integration was Mac-only. As of this release, Windows users are fully supported as well.

A short video demonstrates Flatten in action:
https://youtu.be/cti1MUJKePs

If you are not yet familiar with what Open Babel brings to iNMR, these two pages are a good starting point:
https://inmr.net/articles/babel.html
https://inmr.net/Help3/pgs/formula.html


── IMPROVED 2D SYMMETRIZATION ──

2D symmetrization can now be applied even when the data matrix is rectangular or when the spectral widths differ along the two axes. iNMR automatically trims the longer axis to match. In practice, this means you can symmetrize a spectrum even if you forgot to square the matrix with appropriate zero-filling along f1.

The trade-off is resolution: the result will have one quarter of the points compared to a properly squared matrix. But in many situations that is perfectly adequate, and having the option is always better than not.


── WHAT'S NEXT ──

With these improvements in place, I would love to hear your suggestions. What else could be better?

My own next project is an internal chemical shift predictor. The Prediction and Comparison commands introduced last month are useful, but they rely on a web service built on technology that is roughly 20 years old. Machine learning has advanced considerably since then, and there is good reason to expect significantly better results from a modern approach.

Building a predictor of this quality takes time — at least a year of work. When it is ready, it will be included in version 7 at no extra charge for existing users.

Thank you for using iNMR. As always, replies to this email are welcome.

Giuseppe Balacco
iNMR developer
https://inmr.net
Reply all
Reply to author
Forward
0 new messages