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.
What does "Core" mean?
- Thread starter jwolf6589
- Start date
- Sort by reaction score
Takes longer to download the updates or to actually install them? macOS updates are often pretty large, so that could just be a limitation of your network speed rather than the speed of your processor.
If your current machine is plenty fast for everything you do except updating (a relatively rare occurrence) then you should be good to hold out a while longer on an upgrade, assuming that's why you asked the question.
Longer to install. I have a 200MP connection so downloading takes little time. But if I were to time the updating the iPhone 12, no actually the appletv would beat everything.
Compared to what? Yes, dogs are smart. Mine's a pill. I'd say she's got 1/2 core though.
FYI it is in fact actually one of the forum rules. I imagine it must've caused some problems in the past:
Forum Rules
Jump to: Instantly Bannable Offenses | Advertising/Self-promotion | Things Not to Do | Minor Problems The following rules govern use of the forums. You are also responsible for observing the Regist...
I wouldn't worry so much about the core count on your computer. All that matters is that it does what you want at a speed that's reasonable to you. Core count on its own doesn't tell you much at all.
I bet he/she's a very good boy/girl! And relative to a Pitbull, JWolf's Mac is probably fast, while your dog is likely smarter than 6502
At the core level: schedulers, dispatch, execution pipelines, cache design, core-to-core interconnect, buses...
There's a lot that goes into determining overall performance, not just core counts and frequencies. Just like life itself, a balancing act. Budgeting, time management, deciding where to focus resources to achieve desired outcomes, learning from past mistakes/failures and trying not to replicate them again, etc.
Every processor was designed to be the fastest because, generally speaking, the faster you work the more you get done. The Pentium 4 was the fastest in 2005/06. Alder Lake Core i9s are among the fastest today.
Last edited:
This is literally all that matters. When a device can no longer meet your needs, then it’s time for an upgrade.
I am just worried that in the next few years apple won’t release OS updates for intel macs.
That is a very real possibility and it’s unsure how long they will continue developing Rosetta.
They'll probably release new versions of MacOS for another two, maybe three years from when the last Intel mac was 'current'. I.e. after the whole line has been switched to Apple Silicon, we'll get two maybe three annual OSX updates. They'll provide security updates for many years beyond that though. Not having the latest OS won't be a problem though for current Intel Macs as they'll run just fine on their last supported OS. On my main Mac right now, I'm still running Mojave and it works perfectly. That's quite an old OS but it's just as capable as it was when Mojave was released.
I mean by that definition lying in bed vegetating is just about the easiest thing that you can do. If you only do what's easiest, be my guest to try. I will continue the difficult, valiant fight of gently ribbing people who don't know about search engines in the year 2022. I'm teaching this man to fish through my disapproval, you're coddling him into a lifetime of posting "what's a RAM" on a forum as if it's the first time anyone has had that thought and asked that question ?
At the end of the day, this is all that matters. The computer is an appliance; if it provides acceptable performance for you, then that is enough.
My sister for example, has a dual core i5 iMac from c. 2012 that I find intolerably slow, so much so that I started to investigate if this performance was normal, and upgraded it with an SSD. But I am used to hardware that is either 1-3 years' old or fairly high-end so my expectations of "normal" speed are higher. Other people may have different expectations for their products. I still have a 13-year-old 32" 1080 TV, and see no reason to change it because it still works and it's big enough for the room I use it in. For others, that would be a tiny low-resolution screen that would drive them nuts.
As for your original question, each core can be considered to be an independent processor that can run programs in parallel to the other cores. This allows multiple programs (both applications and the operating system) to run at the same time, thereby running faster. Many modern applications can also divide their work across multiple cores, again increasing their throughput.
However, not all applications are written to take advantage of multiple cores, so you may find no improvement from running on a 2-core or 4, 6 or 8 core machine running at the same frequency (clock-speed). You may find that a computer with fewer cores running at higher speed performs better for a particular task than a computer having a larger number of slower cores. In general the trade off is that the more cores you have, the slower they run because they generate heat proportional to their frequency, and you have to keep within the "budget" of the CPU, e.g. 30 Watts.
Prior to 2005, all consumer-grade CPUs only had a single core, and you could gauge performance by looking at the frequency they ran at. This also has limits because frequency is related to power usage & heat, so it was hard to get beyond about 5GHz. The solution is to have more cores, which is where we are today. Larger computers can have multiple separate CPUs that work in unison, each with multiple cores, and this allows scaling to very large sizes in the case of some specialised research machines. The largest supercomputer "Fugaku" has 7,630,848 cores (158,976 × 48-core Fujitsu A64FX @2.2 GHz).
There is a good beginner's introduction to this at https://www.techspot.com/article/2363-multi-core-cpu/
Usually it's best to put n threads on a n core machine. At this point the monitor would show n*100% processor load, but by no means your code runs n times faster. It's the problem of parallel scaling which is never perfect.
The idea behind Intel's hyperthreading is to stuff the processor even more by putting more threads than there are cores, usually two time more. The idea is to take advantage from those cycles which are wasted as the threads wait for each other. But let's be honest - this is not very effective and is best switched off most of the times. So in this case the processor shows something like 2*n*100% load but in practice is not faster than n times, in practice even a bit slower than that.
Well, most of Intel’s chips, even i5s and the like, support hyperthreading (which requires software to support it, or use OpenCL/GrandCentralDispatch). But even most modern single core chips without hyperthreading are capable of executing threads, thanks to the instruction pipeline. As long as a machine code instruction is in some stage of the pipeline, other instructions can be added at different stages (as long as the pipeline isn’t saturated). This is actually one of the traditional differences of CISC based processors (like the 6502, the 68k, and x86) and RISC based processors (like the PowerPC and ARM/Apple Silicon). RISC processors typically have instructions that are all the same size in processor memory, while CISC processors have instructions that may be compound instructions and do differ in size. As a result, RISC machines have historically typically done a better job of fitting more instructions into the instruction pipeline than CISC machines (Intel has tweaked its instruction language to be more RISC like over the Core and i3/5/7/9 era, even while maintaining full compatibility with x86 proper, so it’s not obvious how much of this pipeline advantage actually still exists).
As for the gains of the M1 series over Intel, that’s down to Apple having developed the best performing compute core in the business (performance per watt as well as performance per clock speed unit, an M1 core clocked at the same speeds Intel and AMD are using would likely outperform them in terms of raw benchmark numbers, it seems) and to Apple including a bundle of accelerator hardware at the processor level (GPU for graphics, audio of some sort [of course], video acceleration hardware, Neural Engine for accelerating neural network [think AI] workflows, and I’m pretty sure the Secure Enclave also accelerates crypto functions).
