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.
2014 Mac Mini CPU performance specs
- Thread starter hipnetic
- Start date
- Sort by reaction score
Even though I think it's weak, this seems to be the least-conspiratorial explanation so far:
"Apple would have to design and build two separate logic boards to accommodate both dual-core and quad-core processors. Other Macs use the same logic board across models, so I wouldn't expect Apple to make an exception for the Mac mini. Note that this wasn't an issue with the Sandy Bridge and Ivy Bridge processors, where both dual- and quad-core processors used the same socket."
I'm not disagreeing, but perhaps I'm not coming across as I'm expecting.
I've got an i3 Haswell and i7 Ivy desktop with the same SSD. I can't tell any real performance difference doing basic tasks like email/Internet/office. In sure the i7 slaughters the i3 in benchmarks, but doesn't feel all that different unless you're rendering video/gaming
If one feels the 2012 quad is a better fit, great! If you're happy with the 2014 dual, great! The 2014 isn't going to feel like going from a SSD back to a 5400RPM HDD
You're right. In all honesty my going from a dual-core i5 to quad-core i7 on the Mac Mini will not be anything like going from HDD to SSD. Since I'm doing more stuff on my Mac now in Parallels, I can use the extra cores of the i7.
The main issue for me on the new 2014 Mini is not just the supposed soldered RAM (I still won'tt believe it until I see pics), the PCI-E storage is the end of popping in a standard SATA SSD in there (in addition to the HDD). This is again not confirmed, but I think the dual SATA connectors is gone too with the move to PCI-E SSD.
I cloned my SSD over to the original 500GB HDD in the i5 Mini, in preparation of moving it over to the i7 Mini. What a torture booting off the HDD
Not sure that's entirely accurate. The i7 3615QM uses a FCBGA1224 socket, whereas the i5 3210M uses FCPGA988 socket. I didn't even know that this was the case (I too assumed it's the same motherboard across the board), until I saw your post and did some digging.
Attachments
This workload tends not to be very CPU intensive nor does it tend to scale across CPU cores. IMO the higher single thread performance would be of more benefit than higher core count. The benefit of the quad core is the OS has more cores to run each application on. But in the end you're unlikely to see any benefit of one over the other as you're not taxing either.
Given the workload you've outlined single thread performance is likely to be more important than core count. But again none of the tasks tend to be CPU intensive therefore to the end user the performance of a higher clocked, lower core count CPU is likely to be the same as a lower clocked, higher core count CPU.
The scores in red are a composite of individual scores. Given multicore processors can greatly skew the composite results you have to look at your specific needs (which you are doing) and compare them with the individual results. To do otherwise is foolish.
We can't answer this as we can't predict the future. It will all depend on what applications you'll use in the future and how well they'll scale to additional cores.
In a nutshell the applications you use can easily be met with a lower core, higher clocked CPU or a higher core, lower clocked CPU. I doubt you'll see a perceptible difference in most instances.
Also keep in mind you're comparing two different CPU architectures so clock speed comparisons are not a valid means of comparison. As you can see the 1.4GHz Haswell processor achieves a single thread performance metric of 1564. Compare this to the 1518 achieved by the 2.5GHz Ivy Bridge i5. Of course the difference could be made up with Turbo Boost (I haven't looked into the details).
Here's my summary:
The benefit of the new models is in the increased graphics performance. If graphics performance is important to you then they're worthy of consideration. The benefit of the previous models is the ability to configure a quad core option. If multicore performance is important (in your use case it is not) then the previous model is the obvious choice. If both are important than you have to decide which is more important. In this case, and I'm making a SWAG here, multicore capability is likely to benefit more than the increased graphics performance.
HTH
Well, the processor is better and the GPU is better then the 5000:
http://www.game-debate.com/gpu/inde...el-hd-graphics-5000-mobile-vs-radeon-hd-6630m
Regarding the IRIS we will see the benchmarks as soon as they are out.
This Minis update actually seems to me a "down"date.
If true, good work. I wonder why the i7-3720QM comes in both packages. It's a little confusing; maybe there's info missing or there's a typo or two. But if accurate, then we might be back to Apple conspiracy theories.
I'm sure the 1.4 would be sufficient for many people. Knowing that most of my entourage doesn't play game and do only basic things like surfing or doc editing. Having something at the 499$ price point should be a good strategy.
Only the base mini has those. The middle tier one has the 5100 which equals the ATI HD8570. It obviously beats the dedicated graphic chip from 2011, as well.
I am the first one to hate the integrated graphics solution (see my signature), but the Iris is not that bad.
Which would mean just 12 % better performance:
www.game-debate.com/gpu/index.php?gid=1495&gid2=492&compare=radeon-hd-8570m-vs-radeon-hd-6630m#
Plus a worse processor...
I can just hope that this is not the death of the Mini, but rather just a mistake from Apple while wating for a redesigned Broadwell Hardware
So based on this I guess it makes no sense to consider a new one since I own a 2011 w/AMD, upgraded to 8GB and a 256GB SSD. If the Iris dominated the AMD I would do it to also get USB3 and TB2 but......I'll save my money and start thinking seriously about a Zotac EN-760 Plus.
That helps very much, thank you! I may be cancelling my 2012 order before they ship it, but at $679 for a new one it was hard to pass up. It appears the only drawback for me on the 2014 2.6 is that they did away with the FW800 port and added another ThunderBolt 2 port. All my peripherals are FW800, but I understand I can get a TB to FW adapter so that shouldn't be an issue if the adapter works.
One thing I forgot, and I don't know if it changes anything, but I'm on 10.9.5 and running with FileVault 2 turned on. Do you know if FV2 changes anything as far as the 2012 2.3 quad core vs the 2014 2.6 dual core?
If you go by the 3DMark 11 then you get even better boost:
AMD: 887.4
Iris: 1221.5
I find the gains passable, not good, but passable.
Just a note not to forget that the extra two cores in the i7 quad core isn't it's only advantage it has over the dual core CPUs. The 2012 i7 quad core also has 6MB on board L3 CPU cache vs. 3MB (or 4MB for some of the i7 dual core) on board CPU cache of the dual core Intel CPUs. This gives the i7 quad core Intel CPUs a performance advantage at the same CPU clock speed vs. the dual core CPUs regardless of whether the extra two cores of the quad core are utilized or not.
Last edited:
I can't say with any certainty you won't have any problems. However I see no reason why FileVault would be any different on the 2012 versus the 2014. Since you've already ordered the 2012 model I'd recommend you stick with it. I don't see the 2014 benefitting you over the 2012 for what you do.
Sorry for not being clear, I didn't mean it might cause problems, but back to the original idea, if it might play a part in pushing CPU usage up to be better served by the 2012 or the 2014. In the end, it sounds like speed is going to be so close for me that it's a non-factor. The only other thing that really matters is the graphics performance, and as you said earlier, that shouldn't be an issue, given my usage. Thanks again for your reply!
So just to confirm,
is the 2012 i7 macmini and the new i5 macmini (2014), will have equall perfomance in single core speed? or the new macmini wins that?
What I observe is that, now, "some" people is saying that most daily work loads for the macmini just uses a single or dual cores.
is the 2012 i7 macmini and the new i5 macmini (2014), will have equall perfomance in single core speed? or the new macmini wins that?
What I observe is that, now, "some" people is saying that most daily work loads for the macmini just uses a single or dual cores.
Bean Counter Filth
THIS is what you get when an accountant is CEO - Apple is a monopoly of sorts and as such without Steve to keep the new Apple regime in check this is what we can expect more and more of.... Give it a few years "Pro" apps will be dumbed down or non existent, its all about $ and not much else from here on out..... PC is starting to look more open and attractive everyday [never thought I would ever think like this....but...]
THIS is what you get when an accountant is CEO - Apple is a monopoly of sorts and as such without Steve to keep the new Apple regime in check this is what we can expect more and more of.... Give it a few years "Pro" apps will be dumbed down or non existent, its all about $ and not much else from here on out..... PC is starting to look more open and attractive everyday [never thought I would ever think like this....but...]
How about GPU performance?
I sticked GTX760 & HD5770 to my MP1,1.
Guess which one now holds the water?
Am I right that AMD Radeon HD 6630M from Mini2011 is more powerful than iris5000 in Mini2014?
Both CPU & GPU performance is going down...
I was going to buy new Mini. Then soldered RAM. Then no proper 4k support. Then checked the reality behind amazing new 15W iris5000. I want a desktop mac with less than 3k$ price tag and display not glued to the computer. Seems to be that optimal would be now quad-core mini2012 with external GPUcase for proper graphics card, which could be upgraded to dp1.3 in a few months. Expensive and complicated. Also complicated, but not so expensive is finally time to take the red pill and go for the hackintosh.
I guess there won't be a new Mini in 2016. "Hardly no-one was buying them."
Last edited:
The original Mac Pro 1,1 - 3,1 are severely bottle necked on the FSB (1.33 GHz) and Memory Bus (677 & 800 MHz).
2009 Mac Pro upgraded? Those can be converted into a 2012 model Mac Pro for a reasonable cost. A hex Xeon W3680 3.33 would slaughter the 2012 Mac Mini at not much more than you will pay for a used 2012 mini now. With the 2009 Mac Pro you are still fully supported by OS X Updates.
When 4K becomes mainstream for broadcast that's when I will become concerned with it. As of now 720P is still the "standard" for HD content especially on the web.
Last edited:
How should I be able to notice these severe bottlenecks?
RAM stripe in Activity Monitor often turns to yellow, so I guess next I'm gonna upgrade RAM from 8GB to 12GB or 16GB...
Yep, will get one of those MP4,1 or MP5,1, when one of those stumbles on my way...
I've been using 2.5k displays over 7 years now. I won't wait for rest of the world for upgrade. 60Hz 4k Seikis are shipping in few months...
Where were you when Steve nixed the whole line of PPC Macs just like that? Or when he decided that 128 MB RAM is enough for the second generation iPhone? Please, get real. Tim is just continuing steves vision.