Mac Performance Guide

I completely agree with the article an will recommend reading Apple Core Rot: http://macperformanceguide.com/AppleCoreRot-intro.html

> The guy says he's going to list bugs, then he really doesn't, he just bitches in bulletpoint form. Did you miss the pages following the first? He provides numerous specific examples. He even says in the original article (though it's easy to miss, which makes me think you missed it): "The pages that follow are those personally encountered— not 3rd party reports." For example, the next two pages: Impossible to Complete a File Copy on Any Drives http://macperformanceguide.com/AppleCoreRot-Finder-copy-error-DS_Store… Time Machine Silently Excludes Critical Data http://macperformanceguide.com/AppleCoreRot-TimeMachine-excludes-criti… etc. Here is a list of all the pages in that specific series: http://macperformanceguide.com/index_topics.html#AppleCoreRot There is also a list for "Lion Hairballs" http://macperformanceguide.com/index_topics.html#LionHairballs

Apple used to put effort into the details, that was the main reason their products were great. That attention to detail has been lost. >When macOS updates delete the user's documents directory we can start comparing it to Windows. https://tidbits.com/2009/10/13/apple-acknowledges-guest-account-data-l… >When logging into the guest account, if the computer hangs, it is possible that, upon returning to your primary account, you’ll find that all of the files and folders in your user account have been erased and that your account has been reset to default settings. Your account’s path still exists on the hard drive, but everything has been erased from within it. https://www.macrumors.com/2018/02/19/apfs-bug-macos-data-loss-disk-ima… >Two related problems are identified by Bombich, above. The first is that the free space on the APFS-formatted sparse disk image doesn't update as it should when the free space on the underlying physical host disk is reduced. The second problem is the lack of error reports when write requests fail to dynamically grow the disk image, resulting in data being "written" into a void. https://macperformanceguide.com/blog/2019/20190305_1246-macOS-AppleCor… >Copy a folder with dot files to another destination (I used ".icj"). Change one of the dot files. Repeat the copy. When the Finder offers Cancel or Replace, choose Replace. >PROBLEM: the copied folder contains old data in the invisible ".icj" file.

>This is a case of “works on my machine.” Nope, I even have had a couple of issues (wifi related) which were solved in .2 update. But I also know that I had this or that issue everytime since 10.2 when I started using the OS. I also know I had tons of issues in Linux (been using it since 1997) and Windows (since 3.1), and I'm also a programmer, so I don't have BS expectations from something as huge as a complete OS and apps distro. So, it's mostly a case of "have been following this space since 2002, including reading most forums, trade mags and blogs, so have a feel for what people compain about." >There are serious[1] documented[2] issues[3] with Yosemite[4] that were not there in the previous OS X release, and, after multiple point updates still haven not been fixed. Nothing special about those issues you posted. I post below a list of issues for Mavericks (tried to find similar or even the same). I can find for you similar items even down to 10.2, including lamenting articles about "buggiest release ever", etc. As for the "and after multiple point updates still haven not been fixed", first we're at 10.9.3. So multiple is just 3 releases. Usually we get up to 7 or 8. And some even carry over or get fixed in the next version, especially if they don't affect many people. And before dismissing "works for me" casually because of some issues online, note that with a user base in the 100s of millions, even a bug affecting 0.1% (with some weirdo network setup, bad HD…

According to this paper[0] ~1600GB/s raw memory bandwidth, ~590GB/s in practice. I haven't seen any actual benchmarks of M2 Ultra CPU bandwidth, but the M1 Ultra has been benchmarked at ~240GB/s[1]. [0] https://www.ixpug.org/images/docs/ISC23/McCalpin_SPR_BW_limits_2023-05… [1] https://macperformanceguide.com/blog/2022/20220618_1934-AppleMacStudio…

> Anyone that has hooked up a external drive to an M1 or M2 based notebook knows why: the performance of external devices dropped off a cliff with Apple’s own silicon. This is false in my experience. I bought a Thunderbolt 3 to Dual NVMe M.2 SSD enclosure (Sabrent EC-T3DN), set up the SSDs in a RAID 0 array, and Blackmagic Disk Speed Test benchmarked that configuration at 2,500 MB/sec read and write. That's very close to the maximum real-world usable bandwidth for TB3¹. ¹ https://macperformanceguide.com/blog/2019/20190128_1352-understanding-…

There is apparently a bug in the buffer cache, that can be alleviated by regular purges: https://macperformanceguide.com/blog/2021/20211029_0832-diglloydTool-I…

> On these new Macs, Target Disk Mode will be retired in favor of Mac Sharing Mode. Rather than turning your Mac into a disk, the new Mac Sharing Mode will turn your Mac into an SMB file server. This is disappointing, especially given all of the trouble with migrating data via WiFi or even Ethernet[1,2] (as opposed to a direct Thunderbolt/FireWire/USB connection). Would much prefer to have direct access to the disk, which opens up many more data recovery options than an SMB server (assuming it can even start). [1] https://macperformanceguide.com/blog/2017/20170707_2222-howto-Migratio… [2] https://www.podfeet.com/blog/2016/11/how-not-to-do-migration-assistant…

Answering this question means you have to understand a bit about how computers work. More cores doesn't automatically translate to faster performance. Photoshop is lightly threaded for most operations. A cursory search brought up this link, but you should read up on threads and the difference between CPUs and GPUs: https://macperformanceguide.com/blog/2017/20171223_2311-why-more-cores…

So, I wonder about this: https://macperformanceguide.com/topics/topic-AppleCoreRot.html I also think OSX have been less stable last versions. I need to kick Spotlight indexing almost each 1-2 weeks. I only note it when install a new app and after a while it fail to find it. Also small things here and there...

APFS, although not buggy per se, also has its disadvantages: 'macOS High Sierra APFS Performance is Inferior to HFS on Apple’s Fastest SSD: all cost, no benefit, at least not yet.' https://macperformanceguide.com/blog/2017/20171124_2015-macOS-HighSier…

Apple could do a lot worse than looking at Lloyd Chamber's 'Apple Core Rot' series of posts and start addressing everything on that list... http://macperformanceguide.com/topics/topic-AppleCoreRot.html

It looks like the real-world read speed of 1362MB/s http://www.thinkscopes.com/blog/2015/01/06/lenovo-thinkpad-x1-carbon-g… is only around 10% more than the 1184MB/s speed of the 2013 Mac Pro: http://macperformanceguide.com/MacPro2013-performance-SSD.html

I don't think he's talking about removable drives - he claims internal drives are getting added to the exclusion list. Details here: http://macperformanceguide.com/Mac-TimeMachine-exclude-bug.html

Someone pointed me to the OWC Mercury Extreme Pros, which are extremely fast and maintain performance without TRIM: http://macperformanceguide.com/SSD-RealWorld-Seasoning.html http://www.barefeats.com/mbpp19.html I took the plunge and am quite happy I did.

I meant the process described here: http://macperformanceguide.com/Storage-SSD-Reconditioning.html And then the "restoring performance" part on the bottom.

> moving the Mac to a new processor architecture (for the second of three times) Four times kinda — maybe five if you want to count PPC32 and PPC64 separately but I usually don't since the Intel transition happened so soon afterward that there is really no PPC64 lineage to speak of. I definitely count 32-bit and 64-bit Intel separately though due to the number of years taken to transition, all of the annoying early-Intel-Mac 32-bit EFI issues, and the need to manually opt in to the 64-bit kernel on many machines. In fact Snow Leopard was the first OS to let you do so! The “no new features” tagline was snappy but it's really not true at all :p https://apple.stackexchange.com/questions/261749/in-which-versions-did… sez — “Mac OS X Snow Leopard and above could only be installed on Macs with Intel processors, and introduced a number of fully 64-bit Cocoa applications (e.g. QuickTime X), and most applications were recreated to use the 64-bit x86-64 architecture (although iTunes was a notable exception to this!) This meant these applications could run in 32-bit mode on machines with 32-bit processors, and in 64-bit mode on machines with 64-bit processors. And yes, the kernel was updated so that it could run in 64-bit mode on some limited hardware, and only by default on Mac Pros, while other newer Macs were capable of running the kernel bit did not do so by default.” Relevant articles: - “Mac OS X v10.6: Macs that use the 64-bit kernel”…

>say a ubuntu desktop, a macbook air and a windows [...] -- but they all can read exFAT at decent performance. Fyi, it's somewhat common knowledge in tech circles that exFAT is an unreliable way of permanently sharing data between different operating systems: https://www.google.com/search?q=exfat+corruption+macos+windows exFAT is ok to use as a secondary file system for transferring/copying data between systems (aka "sneakernet") but not as a critical primary disk drive. EDIT reply to : >the more reliable file system to use if different operating systems need to use it is UDF UDF like exFAT does not have journaling feature in HFS+ and NTFS. Without journaling as a self-check feature, it makes the filesystem very fragile and prone to corruption. Similar to this explanation: https://apple.stackexchange.com/questions/7609/what-are-the-difference… For permanent data sharing between macOS/Windows/Linux, the more reliable setup than exFAT is network shares such as SMB/NFS mounts on a NAS with ZFS/btrfs filesystems. Unfortunately, that costs more money because of the extra hardware. EDIT reply to: >If you just want to share files between 2-3 computers, by far the fastest and cheapest will be to wire the computers together over Thunderbolt. When I researched the idea of daisy-chaining Thunderbolt with mac + Linux + Windows, I ran into stories saying it was not reliable. …

Yes. I saw quality slipping last year around hardware and software when I dealt with a lot of different items from Apple. The perspective also depends on how you look at your devices and the timeframe of expected use. Apple equipment used to be the kinds of devices you could have in production for many years. But, they are really now much more a consumer device company and part of the device churn. If this is acceptable to you (as a user) then you probably don't care because you will just skip the next hardware revision and get the one after (timed with some sort of Apple software obsolesce incentive) and be fine. Last year around this time, I wrote this: http://mergy.org/2012/12/irecognition-i-am-no-longer-apples-target-mar… But then Lloyd Chambers did a detailed piece on the decline here: http://macperformanceguide.com/AppleCoreRot-intro.html But there are many more examples before me and others. Before the holy war kicks-in, I think it really is an expectation thing.