I am not interested in a comparison between a Mac PC and Windows PC, too many variables. The comparison needs to be between an M1 Mac and an x86 Mac running the same version of macOS and the same versions of the applications. Furthermore the comparison needs to demonstrate the M1 memory usage is sufficiently better on the M1 system whereas an 8GB M1 is comparable to a 16GB x64 system. I am not interested in M1 being faster than x64 in compute performance.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I've read somewhere that apples m1 8 gigs a ram = 16 gigs of intel?
- Thread starter Ratsaremyfreinds
- Start date
- Sort by reaction score
Perhaps you missed where I said:
"I am not interested in M1 being faster than x64 in compute performance."
Or perhaps I am missing how this statement from some unknown tweet supports the claim that M1 Macs use half as much memory as x64 Macs. In which case I'd ask you to explain it.
Well, I can only give you the info that I know of. The topic of M1 “memory efficiency” is a complex one. In the end it boils down to users noticing that they seem to get more mileage out of the 8GB RAM config that expected. How much of it is real and how much of it is fantasy is unclear.
And I dont think that anyone can reasonable claim that M1 uses less RAM. One can however make a case that M1 is more agile with making memory available to applications that need it.
To clarify would imply that you provided something to clarify. I am still waiting for an explanation which might need clarification.
The information you've given is irrelevant. What you provided would not pass scientific scrutiny. What you provided might be good for a Windows versus macOS comparison. But that's not what is at issue here.
You are not wrong. I doubt that any of this “evidence” would pass scientific scrutiny. As I write before, the perception of M1 memory efficiency is born from subjective observation. These things are not easy to test scientifically. Frankly, I wouldn’t even know how to approach such an endeavor.
Then I think people should dispense with the statement that 8GB on M1 equals 16GB on x64, don't you?
As for an approach I think a reasonable approach would be to do as I said in post #57:
"The comparison needs to be between an M1 Mac and an x86 Mac running the same version of macOS and the same versions of the applications."
An 8GB M1 system running the same version of macOS and applications using the same data set showing the same memory pressure as a 16GB x64 system would be a good start (assuming the memory requirements really are 16GB). This is something that doesn't require specialized equipment or testing. Just two equal systems (save for processor architecture and memory), applications, and data. Do such tests exist?
I think it’s not too much of a stretch given the amount of swapping that seems to occur on M-series Macs. maybe I shouldn’t make that assumption tho
JFC, so you want it spoonfed to you? Go down in the thread, and it applies in Rosetta as well. The M1 Macs can release unused memory more quickly than Intel ones - even under Rosetta. M1 uses a unified memory architecture, so no need to copy data between main RAM to VRAM, etc. All of the memory is available to the GPU, GPU, and Neural Engine. Less duplication of data, and no wasted time copying between pools of memory. Even with an iGPU sharing memory, the pools are completely separate and data must be copied back and forth between the two.
No, this does not add up to 8GB == 16GB, but M1 is more efficient with memory overall. This isn't necessarily restricted to arm64 vs x64, as the iPhone will still outperform other devices with more memory.
Anyway, thanks to those who have responded and tried to address this. Some of us really appreciate your contributions.
Yes. That said all the follow on doesn't support 8GB M1 = 16GB x64. If you don't know then admit as much.
There really is not much point , judging by your aggressiveness in this thread you will only be combative to anything one posts.
the argument 8gb = 16gb, that's obviously not true , mathematically. but efficiency plays a very large role ,but memory management plays the largest role.
you can easily run activity monitor and see that ARM does use more memory than intel apps. comparing like apps. but that's not the whole story. its much more complicated.
I certainly cannot summarize how its more efficient without giving a massive lesson on the architecture of ARM itself. but you can certainly look it up and read the pages and pages of reasoning.
Well, you also have to understand that there are a lot of things mixed together. First, nobody really supports the sentiment “8GB on M1=16GB on Intel”. That is just silly talk and people who claim it quickly back down when it’s made clear how ridiculous this sounds. And the more reasonable - yet much more fussy - sentiment of ”M1 appears to need less RAM because it seems to have more agile memory management” is much more difficult to quantify. You say yourself that you don’t accept comparisons between M1 Mac and Intel Windows here (and it’s reasonable), but that’s pretty much the only publicly available test out there, since you also disregard the user testimonies. Science is not just about delivering proof, it’s about looking at the evidence, as circumstantial as it might be.
Really? That's not the way it looks from my position. Perhaps I overlooked something so I'll say this: Assuming true the issue is solved. 8GB M1 does not equal 16GB x64. PERIOD. Why you wrote anything more is puzzling given your highlighted statement.
And your point is....what exactly? The bottom line is the slowest M1 with the smallest amount of RAM is faster than most other machines out there. This isn't just about what I found personally. Read the articles and watch the video reviews from LTT and others. It is simply fast--even with 8GB.
And yes, of course it isn't JUST the RAM in the M1. It is the unified architecture, the 8 cores, the shared memory, and other factors. That is the point.
You got ARM smartphones with 16gb of RAM, so there is nothing special about RAM on ARM.
If you load 16 GB of data both on an ARM machine or x86 machine, it will both be 16 GB of RAM loaded in memory.
That is unless Macrumors is implying ARM uses an on the fly compression algorithm (which would actually slow down the machine).
If you load 16 GB of data both on an ARM machine or x86 machine, it will both be 16 GB of RAM loaded in memory.
That is unless Macrumors is implying ARM uses an on the fly compression algorithm (which would actually slow down the machine).
Last edited:
The only thing puzzling is you demanding proof that “8GB ≠ 16GB” in a thread where virtually everyone argues that “8GB ≠ 16GB”. I don’t understand your obsession with this silly myth.
M1 memory management is a different topic and one that merits discussion, but you seem to be set on on ignoring it.
There is a memory paging benchmark tool: https://link.springer.com/chapter/10.1007/978-3-319-54978-1_79 but it appears to be linux-only. Maybe someone would be interested into porting it to Mac 
here you are, right from the other thread :
Pipeline complexity,
x86 - 5+9×N
ARM - 4×N
Above assuming zero cache misses,
ARM can execute instructions without waiting for condition checks
ARM Requires alot less registers to move memory around
x86 everything has to be stored in memory , most of the code we had was moving around data, arm not so much.
an example :
PCAP traffic required us to write it all in memory as it was coming off the line , than flush to nvme array. ARM we can push data off the line right into the disk array by using significantly less memory , i think we are around 12GB total vs 200ish gb before.
inspecting that traffic can be done right from the array , vs loading large chunks into memory to inspect.
i would never argue more memory is bad , my argument is how arm handles memory in the first place , how the code is designed. how it works in a technical aspect. any developer in the apple ecosystem does not get a complete picture of how it works since apple is doing alot of this on their own. if you go down the road of building arm applications and understanding how it works. it becomes a much clearer picture about how it is more efficient.
all can be easily proven right in the white papers for AARCH.
I have been developing on Graviton for years now.
macOS does compress inactive RAM since Mac OS X 10.9 Mavericks. Related technical article, in case you're interested. But this happens on Intel Macs too.
Wait, I don't follow. ARMv8 uses condition flags jut like x86. What exactly do you mean here? Also, how does ARM use less registers to move memory? Did you mean to say that it does not suffer as much from register pressure since it has more registers?
Again, I have difficulty following this. Where do you store things if not in memory?
What about x86 prevented you from doing the same? NVME is NVME no matter whether its x86 or ARM.
In short, your post left me very confused...
Different architectures or implementations may have different branch prediction units or strategies, but AFAIK all modern CPUs absolutely do execute instructions from one of the possible branches while waiting for the result of the condition, so I'm also confused about what he meant here.