# Native Apple Silicon Support
**URL:** https://forums.negativelabpro.com/t/native-apple-silicon-support/6017
**Category:** Feature Requests
**Created:** [March 7, 2023, 10:52am UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017 "2023-03-07T10:52:03Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@jmac](https://forums.negativelabpro.com/u/jmac)
#### Post date: [March 7, 2023, 10:52am UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017/1 "2023-03-07T10:52:03Z")
Hello NLP Forum!
I know that this feature request was touched on several months ago but I am wondering if there is any further insight as to whether the new build of NLP will have native Apple Silicon support.
I am setting up my new MacBook Pro and would prefer to not install Rosetta 2. I don’t need install NLP right away and would be happy to wait for v3 if it will have native support for M1, M2, etc. Macs.
Many thanks and looking forward to the new features in v3.
---
### Author:  [@cvandebroek](https://forums.negativelabpro.com/u/cvandebroek)
#### Post date: [March 8, 2023, 11:15am UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017/2 "2023-03-08T11:15:45Z")
Hey jmac – Welcome to the forum!
If you’re not shy of using terminal tools, you can install homebrew and compile all needed files from source. ImageMagick and ExifTool are needed. Once the tools are compiled, you can simlink the executables, so NLP will make use of them.
If the above doesn’t ring a bell or scares you… You’ll have to wait for both ImageMagick and ExifTool to provide precomiled Apple Silicon builds. Though I’m not sure when that actually will happen – if ever at all.
@Nate If you’re reading this, happy to provide my Apple Silicon builds for you to review.
---
### Author:  [@nate](https://forums.negativelabpro.com/u/nate)
#### Post date: [March 8, 2023, 2:41pm UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017/3 "2023-03-08T14:41:14Z")
> [@cvandebroek](#):
>
> If you’re not shy of using terminal tools, you can install homebrew and compile all needed files from source. ImageMagick and ExifTool are needed. Once the tools are compiled, you can simlink the executables, so NLP will make use of them.
Interesting. Good to know that it is running natively on Apple Silicon compiling through homebrew.
I’m curious what difference you see in conversion speed when using your symlinked builds vs the included self-contained build (with Rosetta 2)… perhaps you could try timing a batch conversion of 20 or so negatives, first using your symlinked builds, and then using the included build with Rosetta… (you might need to run the test on the included imagemagick builds more than once, as I believe the first time Rosetta is used, there might be some added time for translation.)
> [@cvandebroek](#):
>
> @Nate If you’re reading this, happy to provide my Apple Silicon builds for you to review.
Please feel free to send me anything you think would be helpful! You can private message me, or email me at [nate@natephotographic.com](mailto:nate@natephotographic.com).
Just note that based on my past experience, there is was a big difference in the effort required to get imagemagick up-and-running on a single system (via homebrew or another package installer), vs creating a self-contained executable that can be pre-packaged to work on any system.
Also, I’m not on Apple Silicon myself, so that really limits my ability to test and create an Apple Silicon build, at least until I’m able to update my system. But, if anyone is able to get self-contained builds of ImageMagick and ExifTool, then it should be fairly easy for me to incorporate them!
Just digging through my notes, here were a few resources that may helpful :
> **[How to package ImageMagick binaries?](https://community.adobe.com/t5/lightroom-classic-discussions/how-to-package-imagemagick-binaries/m-p/4741910)**
>
> My plugin uses ImageMagick's convert and, rather than hassle users with downloading it themselves, I'd like to package it with the plugin. This seems straightforward for Windows since IM provides builds of convert.exe. OSX is not so clear since IM...
> **[Imagemagic OSX static relocatable build](http://blog.schdbr.de/imagemagic-osx-static-relocatable-build/)**
>
> After long frustration, let's do this. We want: A relocatable binary (just the binary, nothing else) Static plugins (delegates) such as png Static freetype support so we can annotate images Homebrew Homebrew seems to be the most straightforward...
>
---
### Author:  [@cvandebroek](https://forums.negativelabpro.com/u/cvandebroek)
#### Post date: [March 8, 2023, 6:54pm UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017/4 "2023-03-08T18:54:40Z")
I’ve timed conversions on 21 colour negatives and these are the results:
_I made three runs in total, removing the first result to compensate for Rosetta 2 conversion/translation/initialisation._
**Native:**
- Run 1: 23.58s
- Run 2: 24.11s
- Average: 2.27s
**Rosetta 2:**
- Run 1: 27.46s
- Run 2: 28.18s
- Average: 2.64s
Avg difference Native vs. Rosetta 2: 15.07%
I’ll dig the subject of creating redistributable binaries and will share further info with you. 15% speed increase in conversion time might be relevant when converting full rolls (or many of them for that matter) but hardly noticeable on single images or batch sizes \< 15 images.
---
### Author:  [@jmac](https://forums.negativelabpro.com/u/jmac)
#### Post date: [March 9, 2023, 9:35am UTC](https://forums.negativelabpro.com/t/native-apple-silicon-support/6017/5 "2023-03-09T09:35:34Z")
Thank you to both Chris and Nate for your replies.
Unfortunately, I am not familiar enough with using Terminal to complile the tools as described so I’ll have to wait for a native Apple Silicon installer or install Rosetta 2. I could probably muddle through and get it to work without installing Rosetta but there is also a non-zero chance that I would make a complete mess of things!
That said, there is clearly a lot of expertise in this community so I’m hopeful that in time we might see a version of NLP that works with Apple Silicon natively.