Sitting here with a brand new purple iMac. I do mainly office style work. Apps open quicker, but at the end of the day, this M3 iMac at 16 GB RAM is very much the same experience wise as my M1 iMac at 16GB RAM. I am happy I made the switch as I now have more storage, and I am guessing this new computer will be supported for longer (I can also add AppleCare to it, which I could not do to my old iMac). But at the end of the day, M1 iMac, when used for their intended purpose and especially if spec'd do 16 GB of RAM, will last for several more years, and those on the fence don't need to upgrade. The just illustrates how strong the M1 + 16GB RAM combo is for general work.
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.
FYI - M3 iMac is a slight improvement, at best, from M1
- Thread starter Adelphos33
- Start date
- Sort by reaction score
interesting to me because my 27" 2014 is at a point where no further OS upgrades are available over Big Sur and it is getting a little sluggish as I surf and do a few Excel spreadsheets and emails. it just wants to crawl now and then. Was waiting for the M3 to have more positive reviews released as I didn't want to be too soon to pull the trigger and regret it later. Maybe a M1 would be sufficient for home use for emails and light office stuff. I have not really seen any M1's on a super sale yet. I also am confused on 8Gb or 16GB Ram. Storage is not such a big issue with all the cloud storage available. But then again local backup's make a larger HD seem more necessary. You put it all together and a M3 16GB with a larger 1TB HD and it gets to $2K quickly. And the M1 is not much less expensive.
I've got an M1 MBP of the 16G / 512G variety and an M2 Mac mini of the 8G / 256G. I can't tell the difference day to day if I'm honest.
I think the problem is they did such a good job of it initially and had a fairly tight software stack, that all gains since M1 are incremental. I mean high load stuff the M2 is 20% quicker but that isn't noticeable for the average desktop machine.
I think the problem is they did such a good job of it initially and had a fairly tight software stack, that all gains since M1 are incremental. I mean high load stuff the M2 is 20% quicker but that isn't noticeable for the average desktop machine.
I have a M1 Mini 16GB as well as a M2 Pro Mini base (16 GB) I use these in my office using Word, Numbers or Excel, Mail, iMessage, Safari, iTunes and some 2D CAD stuff. The M2 Pro "may" be a little faster????? I really can't tell. I suspect unless you're doing photo or video editing you would not notice a difference either.
Not sure of what is confusing you, but the amount of RAM needed is dictated by the job required. If all you're going to do is email, websurf, watch YouTube, or do a Zoom call, the 8GB is quite fine. If you intend to do serious photography or video stuff, then 16GB will help (in some cases quite a bit.)
The Refurb Store has many M1 iMacs in stock. Price starts at $1,389.00 for the 16GB model, though that only has 256GB storage. If you want 512GB storage then the price is $1,559.00. That's $220 less than an M3 version with 16/512.
Looking at Apple web site a M3 is $1699 not $1599 for 16/512. Looking just now a refirbished M1. 16/512 is $1599 so the M3 is looking like a better deal. A M1 8/512 for $1049 offers more savings if it performs well enough. With cloud storage and/or a cheap USB HD for storage could make a nice bundle.
Be sure to compare similarly spec'd iMac.
The base models do not have ethernet ports, and do not have Touch ID. So there is more than just RAM and SSD to check when comparing prices.
Plus the discount prices (Veterans, Educational, some corporate) are different than the list price.
I’m sorry, but we’ve passed the need for faster computers to do standard office stuff a long time ago. You can’t blame Apple (or other PC manufacturers) for having crossed that threshold a decade ago. Going forward it will be all about power efficiency and silent computing for the most part. Generic computing is ubiquitous, most phones these days can handle that stuff easily. There is no need to upgrade from M1 to M3, it’s just that when you buy a new iMac it has the M3 in it.
People that really need computing power still have an unquenchable thirst for more, be they gamers, AI developers, AR/VR developers, programmers, scientists, animators, graphics artists… That thirst will remain for a long time, but the group of thirsty people will shrink slowly but surely.
Oh I agree ( and thanks for your wisdom) ....I can do everything with my 2014 27" that I need to do except it is stuck in Big Sur and no way to keep current with OS upgrades or software that upgrades to keep up with the OS improvements. I will most likely grab a deal on a M1 but I'm sure I will have a little buyers remorse stuck in my head for a long time.
Well for me and that's just me, if I had an M1 with 24GB of RAM, 2TB HD and 4 Ports, I would not buy the new iMac M3 to replace my iMac M1. I'm would be fine with it for at least 10 years.
I only waited for the next version because of the possible 27" screen. It didn't happen so I bought the M3, also the 24" is gorgeous to me. Not having a 27" isn't a deal breaker. Even if Apple releases a 27", 32" or whatever this year, I'm happy with my 24" M3 for another 10 years. Then again, that's just me.
I only waited for the next version because of the possible 27" screen. It didn't happen so I bought the M3, also the 24" is gorgeous to me. Not having a 27" isn't a deal breaker. Even if Apple releases a 27", 32" or whatever this year, I'm happy with my 24" M3 for another 10 years. Then again, that's just me.
FYI, as someone who "do mainly office style work," you are not in a position to advise others that
"FYI - M3 iMac is a slight improvement, at best, from M1"
Your usage does not challenge the M1 or the M3 or the Intel Macs that preceded M1. So your comment headline is not just fully meaningless, it is misleading; because in many ways M3 is more than a slight improvement. But of course your "mainly office style work" will not identify those improvements.Note that we fully agree that users with low end usages can often suffice with older generation hardware. However that fact does not logically extend to your headline conclusion that M3 is unimproved.
While you are correct, technically, no one can seriously argue that the improvement from Intel to M1 was not FAR FAR greater than the improvement from M1 to M2 or M3. Creators all over YouTube attest to the fact that even for demanding tasks like video editing the main improvement for the M3 iMac over the M1 is the availability of 24GB RAM. (Even MKBHD, who does demanding 8K video work on his M1 Max MacBook Pro, said he saw no reason to upgrade, despite the better benchmarks.) So, no, I would not consider the OP's original post misleading. The purpose of this thread is not to give advice to people designing the next Space Shuttle -- they know exactly what they need.
Let me be really clear. Claiming that "M3 iMac is a slight improvement, at best, from M1" is wrong.
Low end users and arbitrary low end usages do not change the truth. And the big improvements M1 made over Intel are also beside the point.
You're painting with too broad a brush. One needs to distinguish between light and heavy office work here.
I have a 2019 i9 iMac (Apple's fastest consumer desktop from 4 years ago--i.e., much less than a decade). And I frequently get spinning beachballs in Word, Excel, and Adobe Acrobat Pro. The Office issues are mostly because Office isn't well-optimized for Mac. Thus, unfortunately, when doing heavy office work, one still needs a very fast processor to overcome this.
I haven't tested my 2021 M1 MBP for regular office work yet. But I've done couple of quantitative tests with office operations. Note: Before you dismiss these as unimportant, let me acknowledge that it's the routine stuff (the spinning beachballs) that are most annoying; but I think it's good to put some numbers to this, and these are operations that could be easily timed:
1) I have an ≈300-page Word document that takes an annoyingly long time to fully open on both my i9 iMac and my 2021 MBP. It's 34 s on the iMac, and > 20 s on the M1 Pro (I don't have the exact number in front of me). This is not a disk access issue (it's 220 MB, which takes only a few seconds to transfer to RAM)—it's the time it takes Word to fully process the document and thus finish opening it. When you want to quickly open the doc and take a look at something, this is a longer wait than one would like.
2) Converting a multi-page PDF to readable form (using Adobe's Optical Character Reader), to make it searchable. I haven't tested this with my MBP, but here are results from members on this forum and others on the same 56-page patent application. Now imagine your office work requires you to do this with several documents...
M3 Max: 28 s
M1 Ultra 36 s
M1 Max: 45 s
2019 i9 iMac: 74 s
Again, it's not about whether these specific operations are imporant. Rather it illustrates that, even with AS, there can still be significant wait times for heavy office work. Thus heavy office work could likewise benefit from significantly faster speeds than even the M3 Max can provide.
I strongly disagree with the part I bolded. Those groups have been consistently pushing the bounds of what's possible with computing, and thus will happily continue to gobble up any additional computer power you give them.
That's certainly been the case with scientists, going back to the dawn of computing. Indeed, computing in the sciences is becoming increasingly ubiquitous, and thus the size of that "thirsty" group is increasing, not decreasing.
Last edited:
And RAM often is also hugely important. Unfortunately many folks today (perhaps because the Mac OS does such a good job of coping with lesser RAM, albeit sub-optimally) seem to think of RAM as an afterthought, instead of as the essential component that it is.
Were people really expecting huge differences with each new generations?
Past experience tells me that wouldn't happen, no matter who is making the CPU/GPU.
When we moved from 68040 to PowerPC 601, there was a big difference. When we moved to Intel, there was a big difference. When we moved to Apple Silicon, there was a big difference. In the middle of each, there wasn't much difference.
Past experience tells me that wouldn't happen, no matter who is making the CPU/GPU.
When we moved from 68040 to PowerPC 601, there was a big difference. When we moved to Intel, there was a big difference. When we moved to Apple Silicon, there was a big difference. In the middle of each, there wasn't much difference.
Yeah, I've got 128 GB for scientific work, but had to temporarily downsize to 32 GB while I replaced some bad RAM sticks. That's still considered a generous amount of RAM, yet even with that, and even when doing only office work, I was noticing some swap.
At the same time, Apple's entry level configs for their most popular machines, and high RAM upgrade pricing, ensures that 8 GB will continue to be the most common RAM size on Macs—at least if you consider Steam's hardware survey reasonably representative. If you look at the results for Dec. 2023 (https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam?platform=mac), which shows Steam users are 70% on AS Macs and 30% on Intel Macs, you can see the most common RAM size on OSX is 8 GB, followed by 16 and 32. For Windows and Linux, it's 16, followed by 32 and 8.
What's striking is that 44% of the Macs have 8 GB RAM, while only 14% and 8% of the Windows and Linux boxes do. On the latter two, it seems 8 GB is mostly a relic of the past. Using the Wayback Machine, I found the last survey that had 8 GB as the most common RAM size was four years ago (Oct 2019).
Last edited:
What would you gain from having gone with AMD or Intel instead? Speed gains are incremental, short of adding more cores, and with most additional things like video decoding or AI, Apple seems to be keeping up. The real arguments are probably just memory and price at this point, and the former will be addressed in time (At least for the common people).
Heavy office work is not what I was referring to at all. The type of things you describe are cpu hungry for various reasons (don’t get me started on the gimped MS Office software) and do benefit from faster computing. As to the shrinking pool of heavy users: I stand by my opinion that this group is shrinking. The market for ‘super’ computers is rapidly shrinking, whereas mid tier computing power is now within reach (m3 max pro/latest intel Nvidia PCs) of a lot more people.
New developments such as AR, quantum computing and AI will require another leap in computing power (the next 10 years or so), but in that same decade your 300 page PDF will be dealt with in no time on your compute device of choice. I’m sure of it. 😅
I doubt this is a concern for many iMac purchasers. I can't think of a Mac-using segment less likely to update annually.
It also may be an "issue" for the Macbook Air, Mac Mini, etc lines. The point being Apple may need to rely on non-chip upgrades to generate additioonal interest and sales in these computers
Do you have data to support this? The projections I've seen indicate the opposite--that the supercomputer market continues to grow:
Supercomputer Market is Likely to Rise at 11.06% CAGR By 2030
The global supercomputer market size was evaluated at USD 9.44 billion in 2022 and it is expected to touch over USD 21.9 billion by 2030 with a CAGR of 11.06%.
www.precedenceresearch.com
Hopefully! But it's actually a 300-page Word doc, and I expect that Word's overhead will continue to increase, using up some of the increase in computing power. I recall that Word c. 2011 on my 2011 MBP was actually slower to open docs than Word c. 2008 on my 2008 MBP, because the increase in the program's overhead was larger than the increase in computer speed!
I believe you correct but I wish there was more data on capacity and # units. This type of data is available but seems to be sparse with the bulk of the data in term of monetary value. I am not disagreeing just frustrated with emphasis of the data