Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Chuckeee

macrumors 68000
Original poster
Aug 18, 2023
1,820
4,603
Southern California
Any Thoughts or comments on the Forbes Article A STEM Degree Requires a Real PC see: https://www.forbes.com/sites/tirias...em-degree-requires-a-real-pc/?sh=50a0923378b8

This article states “While these Apple processors and PCs may be optimized for a wide variety of entertainment, productivity, and design applications, they are not optimized for many STEM applications. As I mentioned previously, it is difficult or sometimes impossible to run some STEM applications on x86-based laptops without a discrete GPU. It is even more of a challenge on a Mac because in my experience, the vast majority of STEM-oriented applications are written on and for x86-based PCs.

I strongly disagree. A SOC Mac is more than capable and in my experience superior. What a STEM student requires is a laptop that they can write and run Python (and possibly R), write and run MatLab (and possibly Mathematica) and spreadsheet with plotting. What a student does NOT need is to be battling an operating system, fighting with configuration management and constantly worrying about viruses, malware and ransomware. And a battery that last longer than 40 minutes, is really useful too.

Seems like the article’s authors [Falsely] believes student need a gaming machine to do STEM course work.

Any other thoughts?
 

adrianlondon

macrumors 603
Nov 28, 2013
5,008
7,524
Switzerland
I assume it depends what they need it for. STEM is a huge area.

Having just completed a master's in Medical Informatics, my Mac proved much more useful than those with PCs, as having Unix as the underlying O/S made everything much easier.

I believe that with a dedicated GPU some of the machine learning stuff would have been faster, but we were mainly using Tensorflow within R (using reticulate) and I'm not sure that supports GPUs anyway.
 

leman

macrumors Core
Oct 14, 2008
19,202
19,063
Depends on the field. For doing stats and data analysis in Python or R, M-series laptops are phenomenal. If you work with deep learning, a computer with an Nvidia GPU will likely be more useful, although Apple is catching up. For many engineering disciplines however, there is no alternative to Windows simply because the software is not available on other platforms.
 

jz0309

Contributor
Sep 25, 2018
10,122
26,460
SoCal
Any Thoughts or comments on the Forbes Article A STEM Degree Requires a Real PC see: https://www.forbes.com/sites/tirias...em-degree-requires-a-real-pc/?sh=50a0923378b8

This article states “While these Apple processors and PCs may be optimized for a wide variety of entertainment, productivity, and design applications, they are not optimized for many STEM applications. As I mentioned previously, it is difficult or sometimes impossible to run some STEM applications on x86-based laptops without a discrete GPU. It is even more of a challenge on a Mac because in my experience, the vast majority of STEM-oriented applications are written on and for x86-based PCs.

I strongly disagree. A SOC Mac is more than capable and in my experience superior. What a STEM student requires is a laptop that they can write and run Python (and possibly R), write and run MatLab (and possibly Mathematica) and spreadsheet with plotting. What a student does NOT need is to be battling an operating system, fighting with configuration management and constantly worrying about viruses, malware and ransomware. And a battery that last longer than 40 minutes, is really useful too.

Seems like the article’s authors [Falsely] believes student need a gaming machine to do STEM course work.

Any other thoughts?
consider the source ...
 

headlessmike

macrumors 65816
May 16, 2017
1,234
2,517
I'm a physicist working as a researcher and teacher at major University. In my experience, macOS dominates the scene with various flavors of Linux being a distant second. The change from Intel to ARM has, if anything, only made things better in terms of performance for most tasks. Nobody I know does any heavy lifting on their local machines anyway. That's all done on dedicated computing (CPU or GPU) clusters running Linux.

Nothing that the students do requires a powerful personal computer. As other have pointed out, most of their computer time is spent using a text editor. I could probably still get by today with the iBook I used back in the day. Python, LaTeX, SSH, maybe a Fortran compiler, and bit of patience is all I'd really need. To be fair though, the iBook did have discrete graphics. So maybe they’re right?!
 
Last edited:

richmlow

macrumors 6502
Jul 17, 2002
379
273
I'm a mathematician working as a researcher and teacher at a big state university. I do all of my mission-critical work using Macintosh computers. That includes running intricate custom-designed Mathematica programs, scientific computing , creating slide presentations for conference talks, typesetting research articles with LaTeX, and web design.

I use my Windows machine to play games on.


richmlow
 

throAU

macrumors G3
Feb 13, 2012
8,832
6,998
Perth, Western Australia
I strongly disagree. A SOC Mac is more than capable and in my experience superior. What a STEM student requires is a laptop that they can write and run Python (and possibly R), write and run MatLab (and possibly Mathematica) and spreadsheet with plotting. What a student does NOT need is to be battling an operating system, fighting with configuration management and constantly worrying about viruses, malware and ransomware. And a battery that last longer than 40 minutes, is really useful too.

Whilst I love my Mac, you run the hardware that's best for the specific application. It's possible that the application in question *depending on the specific sub-field of STEM* may be Windows specific. Particularly for high end engineering, etc.

There's a bunch of CUDA specific code out there, there are cheap high end discrete GPUs out there and Windows is the dominant platform for a bunch of niche applications.

Part of your post is correct: You don't want to be fighting the OS to run these very niche applications through emulation on a platform they were not written for.
 

TechnoMonk

macrumors 68000
Oct 15, 2022
1,692
2,279
Whilst I love my Mac, you run the hardware that's best for the specific application. It's possible that the application in question *depending on the specific sub-field of STEM* may be Windows specific. Particularly for high end engineering, etc.

There's a bunch of CUDA specific code out there, there are cheap high end discrete GPUs out there and Windows is the dominant platform for a bunch of niche applications.

Part of your post is correct: You don't want to be fighting the OS to run these very niche applications through emulation on a platform they were not written for.
Most stem programs that need CUDA have dedicated labs and servers. Students run on those servers. Most of the windows laptops have gimped GPUs compared to desktops/workstation variations of same GPU.
 

throAU

macrumors G3
Feb 13, 2012
8,832
6,998
Perth, Western Australia
Most stem programs that need CUDA have dedicated labs and servers. Students run on those servers. Most of the windows laptops have gimped GPUs compared to desktops/workstation variations of same GPU.

Was just one example - if you need to run say, Solidworks, it will run a lot better on Windows than in emulation in a virtual machine.

Yes, Macs are great but sooner or later if you're running a different platform to everyone else in your course, you're going to run into edge cases.

For general stuff - go nuts, run whatever you like.
 

5580463

Cancelled
Dec 4, 2022
75
293
While I love my Macs, the apple silicon architecture has complicated some things. In my prep for a cybersec certification I had to use two VMs, security onion and a custom "cyberops workstation", none of which could work properly on my M1 mini, and only through emulating (so very slow). In data-related work, I remember getting some libraries (e.g. tensorflow) to work on apple silicon took a fair bit of extra work, compared to an intel Mac I had lying around. That was some time ago though, so maybe those issues are fixed by now.
 

splitpea

macrumors 65816
Oct 21, 2009
1,134
396
Among the starlings
The students are required to learn and use very specific purpose-built math/engineering/CAD/etc software that nobody bothered to make Mac versions of — and yes, if you’re writing high performance desktop software instead of electron apps, adapting and maintaining it for multiple platforms is a ton of extra work.

It’s not that it couldn’t possibly be made to run on Mac. It’s that doing so wouldn’t be financially worthwhile for the companies selling it. And due to Macs being rare for decades and even rarer in their fields, I’m sure a lot of professors expect students to be on Windows.

This is probably not the case at all universities, but is at enough to have inspired such an article.
 

headlessmike

macrumors 65816
May 16, 2017
1,234
2,517
The students are required to learn and use very specific purpose-built math/engineering/CAD/etc software that nobody bothered to make Mac versions of — and yes, if you’re writing high performance desktop software instead of electron apps, adapting and maintaining it for multiple platforms is a ton of extra work.

It’s not that it couldn’t possibly be made to run on Mac. It’s that doing so wouldn’t be financially worthwhile for the companies selling it. And due to Macs being rare for decades and even rarer in their fields, I’m sure a lot of professors expect students to be on Windows.

This is probably not the case at all universities, but is at enough to have inspired such an article.
Forbes aside they pretty clearly state the issue is software compatibility which..is 10000% true about basically all software when people need pcs over macs.
This is a very field-specific issue. As I already mentioned, in my and adjacent areas (physics, mathematics, astronomy), Macs outnumber PCs. This is largely because macOS strikes a nice balance between commercial support (like Windows) and open software designed for Unix-like systems (like Linux et al.).
 

philstubbington

macrumors 6502
Any Thoughts or comments on the Forbes Article A STEM Degree Requires a Real PC see: https://www.forbes.com/sites/tirias...em-degree-requires-a-real-pc/?sh=50a0923378b8

This article states “While these Apple processors and PCs may be optimized for a wide variety of entertainment, productivity, and design applications, they are not optimized for many STEM applications. As I mentioned previously, it is difficult or sometimes impossible to run some STEM applications on x86-based laptops without a discrete GPU. It is even more of a challenge on a Mac because in my experience, the vast majority of STEM-oriented applications are written on and for x86-based PCs.

I strongly disagree. A SOC Mac is more than capable and in my experience superior. What a STEM student requires is a laptop that they can write and run Python (and possibly R), write and run MatLab (and possibly Mathematica) and spreadsheet with plotting. What a student does NOT need is to be battling an operating system, fighting with configuration management and constantly worrying about viruses, malware and ransomware. And a battery that last longer than 40 minutes, is really useful too.

Seems like the article’s authors [Falsely] believes student need a gaming machine to do STEM course work.

Any other thoughts?
I seemed to get by with an AIM65 - obviously this was 1000 years ago 😉
 
  • Like
Reactions: headlessmike
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.