# Keep underexposure in conversion

**URL:** <https://forums.negativelabpro.com/t/keep-underexposure-in-conversion/10047>\
**Category:** Digital Camera Scanning\
**Created:** [February 25, 2026, 10:07pm UTC](https://forums.negativelabpro.com/t/keep-underexposure-in-conversion/10047 "2026-02-25T22:07:42Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jacopastorius](https://avatars.discourse-cdn.com/v4/letter/j/a88e4f/32.png) [@jacopastorius](https://forums.negativelabpro.com/u/jacopastorius)\
**Post date:** [February 25, 2026, 10:07pm UTC](https://forums.negativelabpro.com/t/keep-underexposure-in-conversion/10047/1 "2026-02-25T22:07:42Z")

</div>

Hi all. i just started my process of home scanning. I scanned with much satisfaction a few bnw rolls and today i started with color. Unfortunately the scans came out not very good. I think the problem is i shot this roll with a camera that has a not very well calibrated light meter and it tend to underexpose so i got a lot of noise/grain. I think it is due to negative lab pro which in the conversion process because if i make the curves inversion i got a underexposed image and no grain. Is this correct?

Film: kodak gold 200

camera: leica SL

Lens: sigma 70 macro 2.8 art

Settings: iso 100, f5.6 1/80 +1 exposure compensation

Here is an example: the firs screenshot is from an image converted with NLB, the second one is a curves inversion

> **[1 hosted at ImgBB](https://ibb.co/gM3FwcCx)**
>
> Image 1 hosted on ImgBB

> **[2 hosted at ImgBB](https://ibb.co/21LP3VQP)**
>
> Image 2 hosted on ImgBB

---

<div class="post-metadata">

**Author:** ![winslow](https://yyz1.discourse-cdn.com/flex029/user_avatar/forums.negativelabpro.com/winslow/32/6058_2.png) [@winslow](https://forums.negativelabpro.com/u/winslow)\
**Post date:** [February 25, 2026, 10:54pm UTC](https://forums.negativelabpro.com/t/keep-underexposure-in-conversion/10047/2 "2026-02-25T22:54:01Z")

</div>

It would be helpful to see the negative, but an underexposed photo will inevitably show more grain and additional color noise in the conversion.There are ways to reduce the visible grain/noise by tweaking what information from the negative NLP is reading.

NLP cannot work magic though.
