How would you, a software engineer, spec out a new MacBook Pro?

Question

I've read through the MacBook threads and seen one question repeated again and again: which specs to go for in the new MacBooks. I figured it might be worth it to make a dedicated post to discuss this.To start this off, consider the following scenario:- 50% of my work is web or mobile app development- another 50% is training ML and deep learning models (currently on a cluster, but I'd prefer do it on my notebook of course)- I also do raw photo editing work, as a hobbyIn this case, what specs would you choose for your Mac? Specifically:- M1 Pro with [14, 16] GPU cores, [8, 10] CPU cores, and [16, 32] GB of unified memory- M1 Max with [24, 32] GPU cores, 10 CPU cores, and [32, 64] GB of unified memory

dougbarrett · Accepted Answer

I've got the M1 Macbook Pro, 16 GB RAM and 2 TB hard drive, and do 95% software development and 5% photoshop, related to web development, so I'm not dealing with massive files with a ton of layers.I'd say that this is more than enough power to hold me off from upgrading for the next few years, so fortunately for you I'd say if you can, the RAM upgrade may be more important than the actual CPU upgrade.

tmaly · Answer

Go with 64 GB of memory. All of these IDEs are memory hogs, it is better to have the extra memory. Also you might want to load a larger dataset into memory at some point.

stocknoob · Answer

Figure out your income/hour, how many hours/day you plan to use the device, and realize saving pennies per hour on a machine that is your lifeblood isn't necessary.You don't need to spring for extra storage, stick a 1TB SD card in the slot.

kevinherron · Answer

I'd probably get the M1 Max with 64GB but I'd like to know if there's any discernible difference when driving two 5K monitors between the various GPU options.I know it's not a great experience on my current 16-inch MBP with AMD Radeon Pro 5500M 8GB.

wmf · Answer

Max CPU, max RAM, min GPU.The full GPU would help for ML training only if that actually works.

avl999 · Answer

The answer is do not buy a MacBook pro with their ridiculous premiums on memory and disk. Buy a Dell or Thinkpad, install Linux on it, will support all your usecases and you won't be trapped by Apple, will also save some $$$s in the process.

vaughan · Answer

M1 Max 32 + 64GB.

How would you, a software engineer, spec out a new MacBook Pro?

Go with 64 GB of memory. All of these IDEs are memory hogs, it is better to have the extra memory. Also you might want to load a larger dataset into memory at some point.

Figure out your income/hour, how many hours/day you plan to use the device, and realize saving pennies per hour on a machine that is your lifeblood isn't necessary.
You don't need to spring for extra storage, stick a 1TB SD card in the slot.

I'd probably get the M1 Max with 64GB but I'd like to know if there's any discernible difference when driving two 5K monitors between the various GPU options.
I know it's not a great experience on my current 16-inch MBP with AMD Radeon Pro 5500M 8GB.

Max CPU, max RAM, min GPU.
The full GPU would help for ML training only if that actually works.

The answer is do not buy a MacBook pro with their ridiculous premiums on memory and disk. Buy a Dell or Thinkpad, install Linux on it, will support all your usecases and you won't be trapped by Apple, will also save some $$$s in the process.

M1 Max 32 + 64GB.