[gimp-web] contents: improve GIMP 2.99.10 news.

[Jehan <jehanp src gnome org>]

commit 4e0f7b6a466ef3198abb818544c48a62aa23e0ac
Author: Jehan <jehan girinstud io>
Date:   Fri Feb 25 00:21:02 2022 +0100

    contents: improve GIMP 2.99.10 news.
    
    - Improve a bit the introduction.
    - Improve the LUT section after pippin's review.

 .../news/2022/2022-02_GIMP-2.99.10_Released/index.md   | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)
---
diff --git a/content/news/2022/2022-02_GIMP-2.99.10_Released/index.md 
b/content/news/2022/2022-02_GIMP-2.99.10_Released/index.md
index e7eb07ec..bc0834d8 100644
--- a/content/news/2022/2022-02_GIMP-2.99.10_Released/index.md
+++ b/content/news/2022/2022-02_GIMP-2.99.10_Released/index.md
@@ -8,7 +8,13 @@ Image: gimp-2-99-10-20220221-experiments.jpg
 Status: draft
 
 GIMP 2.99.10 is once again a pretty massive step in our development
-plans toward GIMP 3.0.
+plans toward GIMP 3.0. We redesigned some core concepts of GIMP ("linked
+layers", item locks' GUI, dynamics switch), substantially improve the
+new API further into releasable state, babl and GEGL gets crazy
+optimizations going on, macOS and Wayland get more love, all the while
+improving file formats… And more!
+
+The below report will go through the most interesting changes.
 
 <figure>
 <a href="{attach}gimp-2-99-10-20220221-experiments.jpg">
@@ -676,12 +682,12 @@ Citing Øyvind in a recent [report](https://www.patreon.com/posts/babl-lutefisk-
> calls fishes. The fishes are raced against each other to find the best
> conversion babl is able to create.
 
-Yet in some situations, a [LUT (Look-Up
+In some situations, a [LUT (Look-Up
 Table)](https://en.wikipedia.org/wiki/Lookup_table#Lookup_tables_in_image_processing)
-could perform better.
-This is why a new automatic step of LUT creation will be triggered in
-some cases, and its conversion speed compared with the usual universal
-conversion paths.
+will perform better, we now know what the performance will be of using a
+LUT for various combinations of input and output bits per pixel; and
+mark fishes that are better as LUTs on creation time. If such a fish is
+used more than a minimum, LUT population is triggered.
 
 Right now one drawback is that it might actually slow down slightly the
 first conversion happening in a given `input > output` conversion couple.