Mac mini (Late 2014) 2.6 GHz review

Apple refreshed the Mac mini range in October 2014, causing a stir in the process by releasing models that were slower than the previous 2012 generation that they replaced. That’s what we found when comparing the best of 2012 against the best of today, with the shortfall from the latest Macs squarely pointing to the retraction of any quad-core processor models. Check out the Mac mini on Apple’s web store: here.

Read our other Mac mini reviews:  Apple Mac mini (Late 2014) 2.8GHz review |  Mac mini (Late 2014) 1.4 GHz review

And if you’d like to know about the next Mac mini update and when it’s likely to be released, turn to our article  New Mac mini 2016 update rumours

As much ire was raised by members of the Mac mini appreciation society from Apple’s decision to solder the memory to the logic board, removing the popular option of buying basic models then maxing out the RAM with upgrade kits from the likes of Crucial Memory. Or simply acquiring the Mac with the memory you required today, safe in the knowledge it could be expanded in the future when and if software memory requirements demanded.

We did some sums in our first 2014 Mac mini review, and calculated that Apple’s high tax on memory upgrades is not as steep as it once was. In the case of the move from 8 GB to 16 GB, as you may be tempted to do with this middle-order Mac, going the Crucial route currently costs £98, while Apple charges £160.

So while that’s £60 more than some people feel that should have to pay for the same end result, it’s not the several hundred pound difference it used to be. And the fixed-memory policy should mean that RAM of guaranteed quality is being securely installed. It thereby reduces the time-consuming problem of tracking down bad RAM as a cause of system instability – even the big-name memory vendors have been known to suffer from occasional quality issues.

Read: 2014 Mac mini v 2012 Mac mini comparison review

Mac mini (Late 2014) 2.6 GHz: Positioning

The middle Mac mini costs £569 in standard trim, and sits neatly between entry model at £399 and the top option at £799.

(When we say ‘top’, that’s just the best off-the-shelf version. If you buy from the online Apple Store, as with any other Mac, you can select your own configuration with limited tweaks to the processor, storage and memory. In fact the top Mac mini today would be one with a 3.0 GHz Intel Core i7 dual-core processor, 16 GB memory and 1 TB PCIe-attached flash drive. It costs an eye-watering £1759.)

There’s actually a smaller gap in price between the most basic and middle, than the middle and top. You’ll see this more easily if you round up the distracting 99s from the swing tags: £400, £580 and £800.

So what is the difference between entry and middle Mac mini models? Looking at the specs alone, quite a lot. The entry model runs the same processor and memory as the cheapest MacBook Air, but with a slow 500 GB hard disk instead of the fastest storage in consumer computers that is the PCIe-attached flash drive. Meanwhile the new entry-level Mac mini also runs the same processor and 500 GB storage as the cheapest iMac, although this budget desktop PC is at least fitted with 8 GB of memory.

Step forward the middle Mac mini. This sees a healthy kickstart in processor clock speed, from the 1.4 GHz Intel Core i5-4260U dual-core processor to a 2.6 GHz Intel Core i5-4278U. Do note that most of Apple’s consumer PC range is using U-suffix Intel chips today, denoting ultra-low power consumption (by Intel standards), and designed with power economy in mind.

The middle Mac mini sees a doubling in storage capacity, from 500 GB to 1 TB, but keeping the same old hard-disk technology. This remains the weakest part of the system, holding back much potential, but unlike with the soldered memory issue, this one is at least fixable – provided you’re technically adept at stripping down intricate PC hardware. We won’t deny, though, that since Apple barricaded the way with security screws, it’s not as easy as it used to be.

Memory has also doubled, from 4 GB to 8 GB. Compared to the last generation of Mac mini, it has evolved subtly from standard 1600 MHz DDR3 SO-DIMM modules, to a low-power variant (LPDDR3) that is famously and quite permanently soldered to the circuit board.

The third and final difference between cheapest and middling Mac mini models is in the graphics processor specification. All Mac minis since the previous Late 2012 generation have relied upon the built-in graphics capability of the Intel Core series mobile chips, and here we see a change from an Intel HD Graphics 5000 to Intel Iris Graphics 5100.

Given the impressive performance leap we found with the Intel Iris Pro graphics in the last 15-inch MacBook Pro with Retina display, you may be forgiven for expecting a substantial upgrade in graphics performance when selecting the middle Mac mini. Sadly while they both have the word ‘Iris’ in common, the graphics processor in the middle Mac is little better than that found in the entry-level. Let’s take a closer look.

Read next: Magic Trackpad 2 review | Mac accessory reviews

Mac mini (Late 2014) 2.6 GHz: Graphics performance

When we tested the 1.4 GHz model, we found it just about capable of playing some Windows games ported to OS X – provided you selected a low screen resolution and reduced quality settings.

For example, Batman: Arkham City could play with an average framerate of 31 fps, when set to 1280 x 720 pixels and Medium render quality. The middle 2.6 GHz model with its Iris Graphics was found to be able to lift this to just 33 fps. Those are average framerates; but on both Macs, the game slowed to minima of around 15 fps, which will be noticeable as occasional graphical stuttering. In other words, we wouldn’t recommend either Mac mini for this game at least.

A similar story was told when we ran Tomb Raider 2013 on both Mac minis, but with a slightly happier ending. The basic Mac mini could average 35 fps when set to 1280 x 720 and Low quality (Legacy OpenGL mode), but usefully it never dipped below 23 fps, which is closer to the 25 fps threshold usually considered a workable minimum.

The middle Mac mini in the same game and settings recorded 36 fps average but a minimum framerate now at 27 fps. While this Mac is still borderline it may tip the balance toward some occasional gameplay.

Read: Mac mini or MacBook Air: low cost Macs compared

Mac mini (Late 2014) 2.6 GHz: Processor and memory speed

If not the graphics, then perhaps the processor upgrade may hold some useful extra power for those that need it. But despite the near doubling in processor clock frequency, you shouldn’t expect twice the performance.

In Geekbench 3, the processor and memory test app rated the cheapest 1.4 GHz model with an average of 2803 points in single-core mode, and 5401 points in multi-core.

This 2.6 GHz Mac mini here scored 3184 points single- and 6793 points in multi-core mode. That’s a useful boost in raw speed, and not an insignificant one: we’re looking at almost 14 percent faster for one core but closer to 26 percent with four virtually hyperthreaded cores working away.

As a digression, it’s worth remembering that the Primate Labs Geekbench 3 benchmark test was actually calibrated with an Apple Mac mini back in 2011. The model used was also the middle of three (‘better’ in the marketing hierarchy of good/better/best), with 2.5 GHz Intel Core i5-2520M processor and 4 GB 1333 MHz DDR3 memory. This sets a useful comparison, given that that model and the middle mini we’re testing here have almost the same clock frequency (2.5 vs 2.6 GHz) and same configuration of two physical cores with Hyper Threading Technology to approximate four cores.

That bench-reference Mac mini of 2011 was normalised to score 2500 points in single-core mode, and it then scored 5740 points running in multi-core mode. So comparing middle models of 2011 and 2014, we can calculate a percentage increase in raw processing power of 27 percent and 18 percent for each mode of operation, when assessing core processor and memory performance at least.

Given the advances in performance-per-clock, with the seemingly backward but surprisingly capable 1.4 GHz Intel Core i5-4260U as exemplar, we were actually surprised that there wasn’t a clearer gap between 2011 and 2014 mid-range Mac minis, especially given the minor clock advantage of the later 2.6 GHz Mac.

Turning to the Cinebench tests of processor performance, we saw similar gains between the entry-level 1.4 GHz and middle 2.6 GHz Mac mini models of 2014.

In Cinebench 11.5, scores moved from 1.1 points to 1.3 points, an 18 percent improvement. Running multi-core mode, the scores increased from 2.49 to 3.13 points, or a 26 percent increase.

In Cinebench 15, the scores went from 97 to 112 points with one core, and from 236 to 274 points with both cores. This indicates a less convincing advantage of 15 and 16 percent faster performance for the middle Mac mini.

Read: iMac versus Mac mini

Mac mini (Late 2014) 2.6 GHz: Power consumption

The peak power consumption of the 2.6 GHz Mac mini (Late 2014) was around 30 percent higher than that of the 1.4 GHz model (52 versus 40 W). But to be clear, this is running at top speed with both CPU and GPU pushed to maximum.

When sat idling at the desktop, both Mac mini models were seen to draw just 5 W of mains power, making it by a considerable margin the most economical x86-based PC we have ever tested.

Mac mini (Late 2014) 2.6 GHz: Against the best…

We didn’t test the £1759 configured Mac mini listed earlier, but how does the 2.6 GHz Mac mini in the middle stand against the best Mac mini of 2014 with its 2.8 GHz processor?

Differences between these models are numbered as two. There is a slightly faster main processor, by 200 MHz. This also incorporates an increased Turbo Boost Technology headroom by the same amount, moving from 3.1 GHz to 3.3 GHz. And there is a hybrid storage system on the best Mac mini, supplementing the 1 TB notebook hard disk with a 128 GB PCIe-attached flash drive. This composite device is the Fusion Drive which offers much of the speed advantage of solid-state storage, with the mass storage capacity of cheap hard disks.

Looking at the CPU differences, the increase in benchmark scores were small but quite commensurate with the delta change of the processor clock frequencies. While the CPU is running around 6 to 8 percent faster, the scores similarly rose by around 6 to 7 percent.

So Cinebench single-core ratings went from 1.30 points to 1.38, and 112 points to 120; or 6.2 and 7.1 percent increases respectively. Multi-core ratings changed from 3.13 points to 3.36, and 274 points to 292; giving a 7.4 percent and 6.6 percent increase.

Geekbench 3 moved from 3184 to 3363 points (5.6 percent), and for multi-core, from 6793 points to 7208 points (6.1 percent).

Graphics performance between better and best Mac minis was almost negligible. They both feature the same Intel Iris Graphics 5100 engine, and in our tests the difference was no more than 1 frame per second advantage to the faster Mac, and that only in around half the tests.

Mac mini (Late 2014) 2.6 GHz: Fusing the storage

For the user of the Mac mini, a far greater impression of speed on the best model will predominantly be due to the use of fast flash storage. Performance of the PCIe-attached Fusion Drive is tricky to benchmark due to the complex way it is knitted to the larger hard disk by OS X’s Core Storage technology. But we can give a flavour of the speed you can expect, from a benchmark test of a solitary 256 GB PCIe-attached flash drive found in the MacBook Air (Early 2014).

The top sequential read and write speeds of this SSD were around 730 MB/s read, and 620 MB/s write. Compare those top speeds with the 104 and 103 MB/s results of the Mac mini’s hard disk.

(This is not a directly comparable test we realise, not least because a 128 GB SSD is likely to have reduced peak write-speed performance than a 256 GB sample due to reduced internal parallelism.)

More important to the overall user experience is the small-file transfer performance. The hard disk averaged 16 and 18 MB/s for random reads and writes respectively, using data sized from 4 kB up to 1024 kB. Our SSD returned results in the same test of 172 MB/s and 273 MB/s – that’s greater than an order of magnitude of difference.

Not measured here but a crucial difference between storage technologies is the ease with which a flash drive can swiftly juggle stacked threads of data at the same time, while a hard disk is still caching and churning when presented with the same multiple streams of data simultaneously.

In practice the use of a flash drive simply means that apps launch much faster, files open and can be saved in next to no time; and when you’re multi-tasking yourself on various projects on your Mac, it’s unlikely to randomly slow down and present a colour wheel for a cursor.

Mac mini (Late 2014) 2.6 GHz: Final comparison, honest

How does today’s middle-rank Mac mini compare to the last generation middle mini with 2.3 GHz quad-core Intel Core i7 processor? On the whole, not well. It’s slower, by as much as 50 percent.

In the Geekbench 3 test, the single-core result was 7 percent higher – but after that the numbers start to tumble. In the multi-core test the 2014 model was 42 percent slower (6792 points versus 11,752 points).

With Cinebench 11.5, the new Mac mini  was 1.5 and 50 percent slower in single- and multi-core modes. Using Cinebench 15, it was 2.6 and again 50 percent slower in the same tests.

We don’t have direct comparison figures for gaming tests, but we can compare the 2.6 GHz quad-core Mac mini of 2012 with the 2.6 GHz dual-core of 2014. Batman: Arkham City averaged 42 fps (1280 x 720, Medium) against 31 fps for the new model; that’s 26 percent slower than once was. And in Tomb Raider 2013, the older Mac played at 35 fps against 36 fps for the 2014 middle Mac mini. Hence despite the mooted improvements to Intel’s integrated graphics between the HD Graphics 4000 of 2012 and the Iris Graphics 5100 of today, there’s little in it based on this simple comparison. And indeed the older graphics trumped the new in one game test.