Maybe things changed drastically somewhere in the last five years, but when I experimented with a Hackintosh “memory efficiency” wasn’t exactly something I’d associate with OSX. My first attempts were to run that thing in a VM and the entire system was practically unusable unless you have it 4-6GB. I know unused RAM is wasted RAM, but the RAM was actually used by the system processes rather than part of some kind of I/O cache like on other operating systems. With 8GB I could do web browsing and even compile a hello world program I somewhat reasonable time.
I agree with Apple that 8GB should still be plenty of RAM today… for a $600 PC. I know they’re just doing popcorn pricing, but with how cheap RAM is (especially when you leave out the controllers on the chips) they should start their popcorn pricing at 16GB for the price they’re asking.
With their build quality, their excellent screens, their incredible speakers and their practically unbeaten battery life, it makes sense that Macbooks are out of reach for many consumers. These laptops are all top-of-the-line models and for similar specs you’ll be paying a similar price if you go Windows. However, their arbitrary limitations on both storage (and storage speed) and memory are just an insult to their customers.
Do they store 32-bit integers as 16-bit internally or how does macOS magically only use half the RAM? Hint: it doesn’t.
Even if macOS was more lightweight than Windows - which might well be true will all the bs processes running in Windows 11 especially - third party multiplatform apps will use similar amounts of memory no matter the platform they run on. Even for simple use cases, 8 GB is on the limit (though it’ll likely still be fine) as Electron apps tend to eat RAM for breakfast. Love it or hate it Apple, people often (need to) use these memory-hogging apps like Teams or even Spotify, they are not native Swift apps.
I love my M1 Max MacBook Pro, but fuck right off with that bullshit, it’s straight up lying.
Do they store 32-bit integers as 16-bit internally or how does macOS magically only use half the RAM? Hint: it doesn’t.
As a Mac programmer I can give you a real answer… there are three major differences… but before I go into those, almost all integers in native Mac apps are 64 bit. 128 bit is probably more common than 32.
First of all Mac software generally doesn’t use garbage collection. It uses “Automatic Reference Counting” which is far more efficient. Back when computers had kilobytes of RAM, reference counting was the standard with programmer painstakingly writing code to clear things from memory the moment it wasn’t needed anymore. The automatic version of that is the same, except the compiler writes the code for you… and it tends to do an even better job than a human, since it doesn’t get sloppy.
Garbage collection, the norm on modern Windows and Linux code, frankly sucks. Code that, for example, reads a bunch of files on disk might store all of those files in RAM for for ten seconds even if it only needs one of them in RAM at a time. That burn be 20GB of memory and push all of your other apps out into swap. Yuck.
Second, swap, while it’s used less (due to reference counting), still isn’t a “last resort” on Macs. Rather it’s a best practice to use swap deliberately for memory that you know doesn’t need to be super fast. A toolbar icon for example… you map the file into swap and then allow the kernel to decide if it should be copied into RAM or not. Chances are the toolbar doesn’t change for minutes at a time or it might not even be visible on the screen at all - so even if you have several gigabytes of RAM available there’s a good chance the kernel will kick that icon out of RAM.
And before you say “toolbar icons are tiny” - they’re not really. The tiny favicon for beehaw is 49kb as a compressed png… but to draw it quickly you might store it uncompressed in RAM. It’s 192px square and 32 bit color so 192 x 192 x 32 = 1.1MB of RAM for just one favicon. Multiply that by enough browser tabs and… Ouch. Which is why Mac software would commonly have the favicon as a png on disk, map the file into swap, and decompress the png every time it needs to be drawn (the window manager will keep a cache of the window in GPU memory anyway, so it won’t be redrawn often).
Third, modern Macs have really fast flash memory for swap. So fast it’s hard to actually measure it, talking single digit microseconds, which means you can read several thousand files off disk in the time it takes the LCD to refresh. If an app needs to read a hundred images off swap in order to draw to the screen… the user is not going to notice. It will be just as fast as if those images were in RAM.
Sure, we all run a few apps that are poorly written - e.g. Microsoft Teams - but that doesn’t matter if all your other software is efficient. Teams uses, what, 2GB? There will be plenty left for everything else.
Of course, some people need more than 8GB. But Apple does sell laptops with up to 128GB of RAM for those users.
My 32/16 bit integer example was just that: an example where one was half the size as the other. Take 128/64 or whatever, doesn’t matter as it doesn’t work like that (which was my point).
Software written in non-GC based languages runs on other operating systems as well.
I used MS Teams as an example, but it’s hardly an exception when it comes to Electron/WebView/CEF apps. You have Spotify running, maybe a password manager (even 1Password uses Electron for its GUI nowadays), and don’t forget about all the web apps you have open in the browser, like maybe GMail and some Google Docs spreadsheet.
And sure, Macs have fast flash memory, but so do PC notebooks in this price range. A 990 Pro also doesn’t set you back $400 per terabyte, but more like … $80, if even that. A fifth. Not sure where you got that they are so fast it’s hard to measure.
There are tests out there that clearly show why 8 GB are a complete joke on a $1600 machine.
So no, I still don’t buy it. I use a desktop Windows/Linux machine and a MacBook Pro (M1 Max) and the same workflows tend to use very similar amounts of memory (what a surprise /s).
Almost all programs use both 32bit and 64bit integers, sometimes even smaller ones, if possible. Being memory efficient is critical for performance, as L1 caches are still very small.
Garbage collection is a feature of programming languages, not an OS. Almost all native linux software is written in systems programming languages like C, Rust or C++, none of which have a garbage collector.
Swap is used the same way on both linux and windows, but kicking toolbar items out of ram is not actually a thing. It needs to be drawn to the screen every frame, so it (or a pixel buffer for the entire toolbar) will kick around in VRAM at the very least. A transfer from disk to VRAM can take hundreds of milliseconds, which would limit you to like 5 fps, no one retransfers images like that every frame.
I have a gentoo install that uses 50MB of ram for everything including its GUI. A webbrowser will still eat up gigabytes of ram, the OS has literally no say in this.
Pied Piper middle out compression for your RAM
But seriously it’s so ridiculous especially since he said it in an interview with a machine learning guy. Exactly the type of guy who needs a lot of RAM for his own processes working on his own data using his own programs. Where the OS has no control over precision, access patterns or the data streaming architecture.
Apple executives haven’t actually been computed guys for years now. They’re all sales and have no idea how computers work. They constantly saying stupid things that make very little sense, but no one ever calls them on it because Apple.
Macrumors just released an article talking about how the 8gb is a bottleneck in the new M3 models lol
I still hate that they killed the mid-range model. Your option is the lower end MacBook Air with no fan, or the higher-end MacBook Pro. There is no in between.
I absolutely love the snappiness of the m1 chip in my current 2020 MBP, and how much more efficient ARM is compared to x86, but it seems really hard to justify going an extra 300$ in the future.
I really just wish they would bring back the original MacBook (with no suffixes at the end)
I kind of want to go for the framework laptop, but I still do like ARM and given I want to do more stuff around machine learning in the future, which is already kind of difficult to run large language models with only 8 gigabytes of RAM, it at least kind of runs with ARM. On my basement PC, It will barely do anything
There are some external GPUs that can be USB-attached. Dunno about for the Mac. Latency hit, but probably not as significant for current LLM use than games, as you don’t have a lot of data being pushed over the bus once the model is up.
Even if it was like 16GB on a PC still not worth $1.6k
Especially when 16g is something like $50.
At consumer prices. There’s no way Apple doesn’t pay wholesale rates for memory.
they have the memory controllers built into their processors now. So adding memory is even cheaper, it just takes the modules themselves
Tell that to Google Chrome
But it’s $1600 Apple. Not the cheapest Mac book air.
Pairing a chip this capable with just 8GB of shared memory is also just a waste of good silicon. Which makes the price all the more insulting to me.
Like, this is the equivalent of Usain Bolt losing one of his legs
“His one leg is still more capable than regular person’s two legs”
That is exactly what Apple would say, isn’t it
The thing is even if that were true, which it isn’t, I’d still prefer him with two legs. Especially if I’m paying the amount of money I would normally pay for 50 legs.
Somewhat stretching the analogy there
It’s always nice to have as many legs as possible. I love legs
Somewhat stretching the analogy there
Your analogy is looking a bit leggy at this point.
They could just sell appropriately specced computers and make absolute bank like they do anyway, but nooo, that would be too nice of them.
I don’t necessarily even begrudge them a profit margin on RAM. Sure it’s kind of a scam but also I guess it’s just the price you pay for convenience. If you want to better price you upgrade the RAM yourself (assuming that was actually possible).
But the markup they have on RAM isn’t reasonable it’s totally insane.
If you went to McDonald’s and cheeseburger was $0.99 and then a cheeseburger with extra cheese was $2 do you think was something was up but that’s essentially what Apple are doing. Cheese does not cost $1.99, you are literally almost doubling the price for a subcomponent
Instead I feel it’s the opposite because that memory is shared with the GPU. So if you’re gaming even with some old game, it’s like having 4gb for the system and 4gb to the GPU. They might claim that their scheduler is magic and can predict memory usage with perfect accuracy but still, it would be like 6+2 GB. If a game has heavy textures they will steal memory from the system. Maybe you want to have a browser for watching a tutorial on YouTube during gaming, or a chat. That’s another 1-2 gb stolen from the CPU and GPU.
Their pricing for the ram is ridiculous, they’re charging $300 for just 8gb of additional memory! We’re not in the 2010s anymore!
Maybe you want to have a browser for watching a tutorial on YouTube during gaming, or a chat. That’s another 1-2 gb stolen from the CPU and GPU.
Or five times that amount if you’re running Chrome
The most expensive 8GB DDR5 stick I can find on Amazon is about us$35. There are 64GB sets that are under us$200!
Apple should be ashamed.
My 16GB XPS running Linux almost fills up entirely when running several docker containers, IDEA, Firefox, Teams, Postman and a few other, smaller apps, but it fits still, and I can work with it (tho I can’t wait to get my 32GB framework laptop)
Now gimme a 8GB MBP and I’ll show you that I wouldn’t get shit done on that configuration. And at 1600 it’s just crazy.
11 gigabytes of that is probably being used up by Teams, it’s a memory hog.
Teams it’s the absolute worst and it’s the only app that can sometimes crash my Linux machine
It crashes on Apple too.
It sucks that better alternatives like slack went from the big player to a small player only because of Microsofts power over businesses. If teams would win because it was actually a better product I’d be fine with that. But teams is just a pile of shit we are forced to use.
🤖 I’m a bot that provides automatic summaries for articles:
Click here to see the summary
With the launch of Apple’s M3 MacBook Pros last month, a base 14-inch $1,599 model with an M3 chip still only gets you 8GB of unified DRAM that’s shared between the CPU, GPU, and neural network accelerator.
In a show of Apple’s typical modesty, the tech giant’s veep of worldwide product marketing Bob Borchers has argued, in an interview with machine learning engineer and content creator Lin YilYi, that the Arm-compatible, Apple-designed M-series silicon and software stack is so memory efficient that 8GB on a Mac may equal to 16GB on a PC – so we therefore ought to be happy with it.
With that said, macOS does make use of several tricks to optimize memory utilization, including caching as much data as it can in free RAM to avoid running to and from slower storage for stuff (there’s no point in having unused physical RAM in a machine) and compressing information in memory, all of which other operating systems, including Windows and Linux, do too in their own ways.
Given a fast enough SSD, the degradation in performance associated with running low on RAM can be hidden to a degree, though it does come at the expense of additional wear on the NAND flash modules.
We’d hate to say that Apple has designed its computers so that they perform stunningly in the shop for a few minutes, and work differently after a few months at home or in the office.
His comment is also somewhat ironic in that much of the focus of YilYi’s interview with Borchers centered around the use of Apple Silicon in machine-learning development, which you don’t do in a store.
Saved 71% of original text.
I looked at a few Lenovo and MS laptops to see what they are charging to jumps from 8 to 16 GB.
They are very close to what Apple charges.
So, they are ALL ripping us off!I just got a laptop with 64GB of DDR5 ram for $870 or so from HP, so I wouldn’t take these specific examples you found as gospel.
It makes it not feel like a premium device
Because it’s not
In my entirely anecdotal experience, MacOS is significantly better at RAM management than Windows. But it’s still a $1,600 USD computer, and 16GB of RAM costs nearly nothing, it’s just classic Apple greed.
How did you measure this?
Sounds like “feelz” measuring to me
The main metric has been with Adobe apps. 2017 Macs with 8GB of RAM are still able to run Premiere and a few others things smoothly simultaneously. Windows machines with the same config were crashing constantly and kept going.
But I’m still not defending Apple here. It’s been 6 years, and their base level MacBook still ships with the same amount of RAM.
I’m also under the impression the M powered books are much better at thermo management and battery usage over PC versions?
ARM chips are generally better at that.
It’s not anecdotal in the least. It’s been widely tested. There’s a reason an M1 Mac mini with 8GB of RAM can load and fully support over 100 tracks in Logic Pro. The previous Intel machines would buckle with just a few.
ARM is not comparable to x86-64. The former is totally unified, the latter totally modular.
Introducing the new Apple MagicRAM©™!
