Best of the best: 10 top high-performance computing upgrades on test
18th Apr 2012 | 12:30
High-end PC processors and motherboards rated
The high performance PC is dead. If you're a gamer, that certainly seemed like one of the take home facts from last month's astonishing performance comparison exposé.
An exotic PC with £3,000 worth of top-end components took on a gaming-focused £300 system and a quartet of keen gamers couldn't tell the difference. Sobering stuff, but actually very good news for those of us living in the real world and subject to financial realities.
Building a PC on a limited budget is the reality for most. The thing is, our gaming PC comparison was all about subjective performance. That means broader system performance was only tangentially part of the mix. Even more importantly, it means outright performance wasn't a factor.
Find out more about PC Format magazine
That fact matters because while the two rigs often served up a similar experience, the benchmark results showed that the high-end system pumped more than twice the number of frames per second. Objectively, by the benchmark numbers, it was miles faster.
That's precisely the sentiment we're taking into this much broader look at high performance computing. Without question, outright performance matters when it comes to the sort of heavy duty number crunching involved in video encoding, image rendering and full on multi-tasking. With gaming, there's arguably a cut-off point at around 60 frames per second on average. Anything above that is probably performance wasted - you won't be able to see or feel a difference.
Not so for something like video encoding. The faster an encode job completes, the sooner you can watch the video or start the next encode project. Similarly, if you've a limited amount of time available, the more performance your PC has to offer, the better the quality of encode it can execute.
Theoretically, there's no limit to the additional benefit brought by more performance. However, that doesn't mean you should simply go out and spend the absolute maximum.
The question of bang for buck remains: which platforms deliver the most for the least? Likewise, what should you make of the promise of hardware acceleration offered by graphics cores? If the graphics chip inside a cheap Intel CPU is faster at the jobs that matter most to you, what's the point in forking out for a seriously expensive multi-core beast?
Then there's the minor matter of offloading compute-intensive tasks to the cloud. There's plenty to consider.
Spend and save
So, you're in the market for a performance PC. We've established that high performance computing is a very different box of SATA cables from the pure gaming performance, which only needs to be good enough for smooth frame rates. Question is, how and where do you spend you money?
15 best graphics cards in the world
Best PC upgrades for gamers
Best CPU: the 8 top processors today
Best gaming laptop: 8 reviewed
The first conundrum to conquer is the CPU vs GPU question. Plenty of new technologies for the PC have experienced troubled births.
The first solid state drives, for instance, were absolutely awful, but it's getting on for a decade since GPU makers began bigging up the idea of using graphics chips for general purpose processing, and we're still waiting for the number of so-called GP-GPU applications to hit critical mass. That's not just disappointing, it's also a little peculiar.
On paper, general-purpose processing on the GPU makes an awful lot of sense. For starters, CPU themselves are increasingly turning to parallel processing to improve performance. Which just so happens to be the thing GPUs are really good at. Similarly, many of the most CPU-intensive applications around today are heavy in floating point maths, which is essentially the core task at which modern GPUs are designed to excel.
Now, that's not to say there's no GP-GPU-compatible software out there. GPUs are doing some pretty stunning work currently when it comes to industrial and scientific computation. To take just one example, the world's largest human genome sequencing centre (BGI in Shenzhen, China) slashed the time taken to analyse a genome from four days to just six hours thanks to a switch from CPUs to GPUs.
Likewise, if we were to print a list of every PC application with at least some GP-GPU support, it would look pretty darn impressive. Highlights include a healthy list of video encode, transcode and enhancer applications, along with professional rendering, image editing and password cracking tools, plus a few specialised favourites like Folding@home.
However, courtesy of a combination of mixed performance and a less than crystal clear situation on the support side, both in terms of hardware and software support - questions over whether AMD or Nvidia GPUs work with a given application, for instance - GP-GPU still hasn't taken off.
Unfortunately, that's not something that's changed thanks to Intel's QuickSync technology, which arrived with the Sandy Bridge generation of Core i3, i5 and i7 CPUs and offers specialised circuitry within the integrated graphics core to accelerate video encoding. It's a promising development but currently boasts a very modest list of supporting applications. The day when you can rely on a fast GPU for all your high performance computing needs seems as far away as ever.
With that in mind, it becomes a question of platforms and, in turn, the key components you slot inside them. Currently, it's a choice of three platforms: two from good ol' Intel and one from AMD.
We weren't happy when Intel originally switched to two platforms on the desktop with the LGA 1156 and LGA 1366 sockets back in 2009, and our attitude hasn't changed a great deal with the more recent introduction of LGA 1155 and LGA 2011. It makes life a lot more complicated for serious PC users.
Pick a platform
But what are the differences, and how do they affect overall performance? In simple terms, LGA 2011 is really a thinly disguised server socket and is only available with a single chipset, known as X79. It therefore majors in superfast interconnects and memory bandwidth.
First up, you get a quad-channel memory controller. That's great news in terms of bandwidth, and with memory prices as they are, it's not even that expensive to ensure there's a DIMM in each channel. On the other hand, it's doubtful whether desktop applications really need so much raw memory throughput. Even the old triple-channel arrangement on the LGA 1366 socket was overkill. While four channels is a boon for multi-socket servers, it's borderline silly on the desktop and only serves to increase cost and complexity.
Much more relevant are the 40 PCI Express 3.0 channels served up by X79. That's enough to guarantee optimal operation of just about any combination of add-in cards you can imagine, including multi-GPU setups, super-fast PCI Express storage solutions and more.
Next up is full support for the latest 6Gbps version of the SATA interconnect, which is critical for getting the most out of the latest and fastest solid state drives. Unfortunately, you only get six ports as standard. That may sound reasonable. But in a scenario where you're mixing SSDs and magnetic drives and perhaps chucking an optical drive into the mix, six ports get saturated pretty quickly.
Still, while Intel's own X79 motherboard is limited to six ports, several third party boards bung in an additional SATA controller and entirely sidestep the issue.
Another obvious X79 shortcoming is the lack of native USB 3.0 support. Again, pretty much all retail X79 boards add USB 3.0 courtesy of a daughter chip, but it's a feature that really ought to have been part of the native feature set on what is, after all, Intel's premium, highest performing platform.
The last big piece of the puzzle is, of course, the CPU. And this is where things get really intriguing, both for the better and for the worse. First, it's absolutely undeniable that the fastest currently available PC processors can only be had in LGA 2011 trim. Those will be the two new six-core Core i7 3900 series based on the latest Sandy Bridge E processor die: the 3930K and the 3960X.
15 best graphics cards in the world
As you'll know from our coverage, while the 3960X is the undisputed heavy champ of the PC processing world, it still managed to disappoint us. That's because it's really an eight-core chip with two cores disabled. At launch time, Intel told us that's because the balancing act between operating frequency and core count shook out in favour of six cores. Put simply, the more cores you enable, the greater the power consumption and heat dissipation. That in turn puts the kybosh on clock speeds. Sounds reasonable, doesn't it?
The problem is, Intel has just released a series of Xeon-branded chips based on the very same 32nm Sandy Bridge E die with all eight cores enabled. The fastest is the Xeon 2687W. And it clocks in at 3.1GHz. So that's 33 per cent more cores running six per cent slower.
Intel will no doubt argue that the mean old eight-core Xeon is rated 20W higher than the Core i7 3960X at 150W. But again, that's an increase in power consumption of just 15 per cent in return for that extra pair of cores.
If all that makes Intel's argument seem seriously specious, here's the good news. In theory, you can drop a 2687W straight into any X79 motherboard and let rip. The only snag is the price, which we expect to be in the region of £1,500. Yes, £1,500 for a processor.
Intel's Xeon pricing structure is scary territory for a desktop user. The Xeon 2687W isn't just an eight-core beast with 20MB of cache, it's also compatible with Intel's dual socket server platform and that commands an even greater price premium.
There's one more CPU-related issue to consider when it comes to the LGA 2011 socket and the X79 chipset: you don't get Intel's QuickSync hardware encode acceleration. This is only available on chips for the LGA 1155 socket.
As we said earlier, QuickSync hasn't yet developed into a killer technology, but it's nevertheless useful for some video encode tasks and it would certainly be galling to pay £750 or more for an LGA 2011 chip, only to find that it lacks a feature that comes with measly £100 Core i3 processors.
Next up is Intel's LGA 1155 socket. In our view it's Intel's bona fide desktop platform, and it differs from LGA 2011 in a number of crucial ways beyond the minor matter of pin incompatibility.
First, instead of just one chipset, there are several. In the high performance computing context, the best choice is clearly the Z68. It does everything the P67 and H67 chipsets can manage and throws in a few extras to boot.
One of the most important features is full access to overclocking settings. Overclocking might sound like a dubious practice in the context of serious computing, but trust us on this, Intel's CPUs are very, very conservatively clocked.
Assuming CPU support - which we'll come to in just a moment - you can almost always add an extra 500MHz and maintain complete stability. It's often possible to add a full 1GHz and remain stable.
The Z68 also offers Smart Response, Intel's SSD caching technology. The idea is to combine a small SSD with a larger magnetic hard drive and enjoy most of the performance of the former and all of the capacity of the latter while avoiding the punitive expense of a really large solid state drive. In practice, it delivers decent results even if we'd prefer a full-on SSD wherever possible.
Still, it's worth considering if you're building on a tight budget and want to absolutely maximise your bang for buck. Critically, the Z68 does all of that at the same time as allowing full access to the integrated HD Graphics core and QuickSync video acceleration engine found in all Core i3, i5 and i7 chips for the LGA 1155.
You can't say the same for the P67 and H67 chipsets. The former offers overclocking support but not HD Graphics and QuickSync, while the latter does the graphics part but not the overclocking.
Common to all three chipsets is the choice of CPUs. In a high performance computing context, we think maximising the number of threads is all-important here.
For that reason, your choice here is a simple one. You want one of the quad-core, eight-thread Core i7 chips: either the 2600, the 2600K or the 2700K. Unfortunately, Intel doesn't offer anything more than four cores for the LGA 1155 socket, despite the fact that the dual-channel memory configuration could easily handle more.
Pricing is similar on all three - in and around the £250 mark - so we think it makes sense to go with the 2700K with its 3.5GHz basic clockspeed and overclocking-friendly unlocked multiplier.
That's right, we're not recommending that champion of gaming chips, the Core i5 2500K. It's actually based on the same silicon as the 2700K, but HyperThreading isn't enabled so it only serves up four threads.
Beginner's guide to overclocking
Complete guide to underclocking
Overclocking guide: overclock your CPU, graphics card and RAM
Going any further down the Intel price list doesn't make much sense either. Instead, it's better to shift your attention to AMD and its very competitively priced quad and six-core chips. Thanks to the recent introduction of AMD's FX-branded Bulldozer processors, the number of old school Phenom II processors has dwindled significantly, but you can still pick up the Phenom II X4 960T quad core chip for an attractive £95.
The six-core Phenom II X6 is still available in a few flavours too. As with the Core i7 chips, pricing is similar across the range, so you may as pick up the quickest of the remaining X6 chips, the 3GHz 1075T.
But what about Bulldozer, you ask? Surely with its eight-threaded architecture it's a no brainer for high performance computing? Certainly, the weak per-thread performance of the Bulldozer architecture and the FX chips it sired matters less here than it does in games. The top FX 8150 chip makes a good case for itself against the more expensive Intel LGA 1155 chips. Much depends on the particular mix of applications you plan on playing around with.
For motherboards and sockets, AMD's strategy of broad compatibility means there's only one socket you need to worry about: AM3+. Chipset wise, things have also simplified with Nvidia pretty much pulling out of the desktop market. You're therefore left with a choice between AMD's own 7, 8 and 9 series chipsets.
In terms if the underlying silicon, there's little or no difference between the three ranges. Given the choice, we'd go for the latest 9 Series boards, which come in three flavours: 970, 990X and 990FX. There's not a huge amount between these three, the only major differentiator being ever more PCI Express lanes as you scale up through the range.
One thing you don't get with any of them is native USB 3.0 support, so we recommend you keep your scanners peeled for boards with a USB 3.0 upgrade chip. If that's all the different platforms covered, it's time for a tour of our favourite components for each.
Components on test
1. Asus P9X79 Pro
For many Intel's top-rung X79 platform will be right at the ragged edge of what your wallet will tolerate, so every little thing will help to make the numbers add up. That's where the Asus P9X79 Pro comes in.
It's not the cheapest choice of motherboard you can buy based on the Intel X79 chipset, but the extra £30 over the likes of the Gigabyte X79-UD3 buys you a number of attractive features.
The most conspicuous feature is the support for solid state drive caching. In an ideal world, you'd simply plug in the biggest SSD you could find and enjoy some serious solid state speed. The problem is, large SSDs come with equally hefty price tags. The ability to pair a smaller SSD with a large conventional hard disk, therefore, makes for a very sensible compromise between storage capacity and cost aiming for maximum bang-per-buck. We're still years away from solid state drives that are both large and affordable.
The P9X79 Pro also benefits from Asus's graphical UEFI BIOS, which might just be the best in the business. Apart from the snazzy looking and responsive interface, you get screenshot capability, easy updates via USB and an auto-overclocking option that ramped our Intel Core i7 3960X processor up to 4.3GHz. The latter is a function also available via the physical TPU switch, so you don't even need to jump into the BIOS.
2. Gigabyte X79-UD3
Plotting a performance PC? Then snag a high-end motherboard, right? That's the conventional wisdom directly challenged by the new Gigabyte X79-UD3.
Of course, any board based on Intel's X79 chipset hardly rates as a budget item. At £175, the Gigabyte X79-UD3 ain't exactly cheap, but it is within a fiver of the cheapest X79 motherboards on the market. Everything, therefore, is relative.
Consequently, the Gigabyte X79-UD3 is flagrantly frills-free, but with so many features now finding their way onto the CPU die itself - including the memory controller and PCI Express bus - you could argue that motherboards in general are less critical.
Gigabyte's task is to deliver quality and performance where it matters, without going overboard on the corner-cutting compared with more expensive X79 models such as the Asus P9X79 Pro and MSI X79A-GD65 8D.
Ultimately, the verdict on the Gigabyte X79-UD3 will come down to what it does and doesn't do. What it undoubtedly does is deliver performance pretty much indistinguishable from any other X79 board in the known universe. These days, we're accustomed to pretty consistent performance across motherboards sharing the same chipset. The X79-UD3 is no exception to this.
3. MSI X79A-GD65 8D
Can there be such a thing as too much system memory? In the context of the new MSI X79A-GD65 8D motherboard that's the first question that leaps to mind.
As an X79 board compatible with the latest Intel Core i7 processors for the LGA2011 socket, it forms part of the highest performing PC platform on Earth, but you still have to wonder whether support for 128GB of DDR3 memory split over eight DIMM slots is really rational. Sure, for server PCs running multiple virtualised operating systems and a whole hill of applications, that much memory is a boon. But for desktop PCs, even those running heavy duty content creations apps, 8GB or 16GB is usually plenty.
In its fisticuffs with the Gigabyte X79-UD3 and Asus's P9X79 Pro, the MSI X79A-GD65 8D will need a few more tricks up its sleeve. Moreover, the X79A-GD65 8D costs that little bit more than the P9X79 Pro, which in turn is priced at a marginal premium to the X79-UD3.
Is it a case of incremental upgrades all along the line, or has MSI done enough to put this X79-based board in another class altogether? Maybe that old MSI favourite, the OC Genie button, can make the difference?
4. Intel Xeon 2687W
Socket: LGA 2,011
Say hello to the real Sandy Bridge E, for it is this Xeon processor that truly reveals what Intel's ultimate 32nm technology is capable of. Up to a point, anyway.
Yes, the Core i7 3960X is an impostor - a cut-down shyster of a chip. Back when Intel launched the six-core 3960X, we were told that the decision to switch off two cores in the shiny new eight-core Sandy Bridge E die was all part of a balancing act. Intel had weighed up the conflict between clockspeed and cores, and decided that the best overall compromise was six cores at 3.3GHz with a little Turbo action on top.
Running eight cores would have meant a significant drop in clock speed and therefore compromising per-core performance. At the time, frankly, we weren't buying it. Even if opening out all eight cores would mean a big drop in the clocks with all cores heavily loaded, surely the whole point of Intel's Turbo technology is that the chip could still clock up when only a handful of cores were doing the heavy lifting?
Well, now the fastest eight-core Xeon iteration of precisely the same Sandy Bridge E processor die has arrived and the truth is out. The Xeon 2687W is rated at 3.1GHz, just 200MHz slower than the six-core Core i7 3960X. Thus, the 3960X runs just six per cent faster while the Xeon 2687W has 33 per cent more cores.
5. Asus P8Z68-V LX
A budget board with a premium chipset - is that the most effective combination for achieving maximum bang-for-buck? If that is so, then Asus P8Z68-V LX is positioned perfectly.
It sells for as little as £75 but it packs Intel's Z68 chipset. Okay, that means at best you're stuck with mainstream LGA 1155 processors and a quad-core cap, rather than the six and eight-core (the latter in the form of Xeon CPUs) beasts available for the monstrous LGA 2011 bucket of pins.
But as LGA 1155 chipsets go, the Z68 is easily the pick of the bunch. You get full access to overclocking features, the ability to run a discrete graphics card and still use Intel's QuickSync video transcode engine and some nice little extras, including Intel's SmartResponse SSD caching technology.
The prospect of a Z68 board for just £75, then, is a question begging for an answer. Asus may have managed to to squeeze in a few of our own particular treats. But what has been chopped from the P8Z68-V LX?
6. Gigabyte GA-Z68XP-UD3-ISSD
Double your pleasure. Double your fun. With Gigabyte's double-priced Z68 board. Deftly dropping the DoubleMint gum ditty into a motherboard review may be beyond our wits, but the real problem is whether Gigabyte has a shot at justifying the fact that the tediously-monikered GA-Z68XPUD3- iSSD costs twice as much as Asus's P8Z68-V LX.
One thing's for sure, Gigabyte isn't going to get the job done based on raw performance. There's virtually nothing in it, and even when a small differential is detectable, it isn't always in Gigabyte's favour. We couldn't discern a significant difference during overclocking either. Both boards extracted the same 4.5GHz from our Core i7 2700K test chip.
Gigabyte is, therefore, left with one remaining hope: ye olde feature set. Out of the box, things don't look great. We're pretty partial to hardware power, reset and Clear-CMOS switches - it's the least you'd expect from a £150-plus board. We can't help noticing the lack of DVI or eSATA ports on the back panel, too.
Dig a little deeper though, and you begin to identify where the extra cash is going. For starters, Gigabyte gives you proper multi-GPU support with one 16-lane socket and one eight-lane. Both Nvidia's SLI and AMD's CrossFireX are supported.
7. Intel Core i7 2700K
Socket: LGA 1155
This is awfully familiar - Intel's Core i7 2700K is saddled with Intel's mainstream LGA 1155 socket. In theory, the exciting stuff all happens on LGA 2011.
What's more, in a gaming context, we've never jived with the 2700K and its dual threads per core. That's because games don't scale enormously well across multiple threads, and the Core i5 2500K has all the per-core, single-threaded oomph of its much more expensive Core i7 cousins, be they in LGA 1155 or LGA 2011 format.
It's all the gaming chip you'll ever need. But is the 2700K back in the hunt if you shift the context to high performance computing? It's certainly a much more cost effective option than the LGA 2011 alternatives.
You can pick up a very respectable LGA 1155 motherboard for just £75, whereas a basic LGA 2011 item is more like £175. Then there's the chip cost itself. Even at £255, you're looking at half the cost of the cheapest six-core i7 model, much less the preposterously pricey Xeon eight-core.
8. Sapphire Pure Black 990FX
Your eyes do not deceive you. Do not adjust your set. The Sapphire PURE Black 990FX motherboard really does have six PCI Express x16 slots.
That doesn't mean, of course, that you can run six graphics cards in parallel. Not in the conventional multi-GPU sense, since the maximum number of cards supported by AMD's CrossFireX tech is four, and SLI isn't on the menu at all.
Still, if running a veritable army of displays is your bag, the sextet of slots offers plenty of potential. More to the point, it means you have endless options in terms of how you arrange your add-in boards, though the fact that the final two are only x4 electric does limit your options somewhat.
The lack of PCI connectivity of any kind also rules out any dusty old legacy cards you might be thinking about bunging in.
Expansion aside, the PURE Black 990FX is arguably up against it from the off. That's because it's an AMD AM3+ board, and neither the new AMD FX nor ye olde AMD Phenom II chips really threaten the top of the performance table. If we're honest, they're not exactly terrorising the mid-range either.
9. AMD Phenom II X6 1090T Black Edition
AMD's shiny new Bulldozer FX processors have been released into the wild for some time now, so you might be wondering what this crusty old Phenom II processor is doing sullying this showcase of white-hot, high performance computing platforms.
Well, it all depends on pricing and product availability. We're not sure if AMD is still cranking out the Thuban processor dies that form the basis of the Phenom II X6 1090T Black Edition, though we believe that to be the case, if only for manufacturing server processors.
10. AMD FX 8150
Sadly, in what appears to be an increasing, and worrying, tradition for the 'other' CPU manufacturer, the launch of the AMD FX processor was something of a debacle. AMD's much feted new eight-threaded modular architecture delivered in neither in real-world performance nor efficiency.
In fact, it was so underwhelming, it left us with the nagging suspicion that an eight-core Phenom II would actually be a better bet. That's quite a disturbing notion, given the general consensus that the Phenom II's micro-architecture was no longer competitive.
Things went from bad to bizarre when it emerged that the FX's transistor count was not actually two billion, as originally claimed, but closer to 1.2 billion.
Bench analysis: Making sense of an awful lot of numbers
This is all rather different to our traditional, gaming-centric set of benchmarks, and shows the productivity prowess of the real top-end platforms on the market. The eight-core Intel Xeon dropped into a desktop X79 board, like the excellent Gigabyte X79-UD3, makes for the most powerful desktop solution around.
For any video encoding, database-crunching or image manipulation tasks you can't get quicker. That said the cheaper 3960X does a good job of getting close and the AMD FX 8150 actually takes the lead in our Photoshop tests. You can put together a decently priced AMD workstation, but the top-end Xeon-powered beast will cost you.
And the winner is... Gigabyte X79-UD3
Last month our gaming performance exposé blew apart expectations, slaughtered sacred cows and generally caused a kerfuffle. We proved there was very little benefit to be gained from unloading a whole hill of cash on the finest PC components on the market.
When it comes to high performance computing and particularly digital content creation, however, things are very different. The argument goes something like this: If you're playing a game, the performance metrics are framed in terms of fast enough. When you're cranking out 60 frames per second, an extra 30 gets you absolutely nowhere.
Not so with, say, video encoding where every frame per second gets the job done quicker. That's useful if you're honing your own video, iteration after iteration, and find yourself twiddling thumbs during the encode process.
The same goes for tasks, such as professional rendering or heavy duty office apps and big database processing in Excel. The faster your PC, the more work you can get done.
In theory, the money spent on a more powerful machine is an investment in long-term productivity. Put it this way, when we told one of our back-office number-crunching brainiacs just how fast the Xeon eight-core processor tore through the huge Excel spreadsheet he gave us for testing, he nearly cried. The Xeon was 11 times faster than his Core 2 Duo box. Remarkable.
The right combo
That said, not all applications scale so nicely. Photoshop is a perfect example of a mixed workload application; not all of its filters and transformations are well threaded.
Still, the first conclusion we can draw is that the high end clobber does have something to offer. Okay, some aspects of the Intel X79 platform make no sense on the desktop. We compared, for instance, a Core i7 3960X running in dual-channel and quad-channel memory modes. The latter offered no real-world benefit - it only showed a difference in memory bandwidth tests.
But overall, there's no doubt that the X79 combined with an Intel six-core i7 or eight-core Xeon processor is by far the fastest system you can buy. Whether the Xeon is worth double the i7, however, is another matter.
If you're willing to overclock the Core i7, you can close most of the performance gap on the locked-down Xeon. If you're unwilling to compromise stability and reliability, the Xeon processor is unbeatable.
Further down the scale, the Intel Core i7 2700K combined with the aggressively priced Asus P8Z68-V LX makes an impressive mid-range solution at around £325 combined. Throw in a little overclocking and you have a very quick content creation system.
As for the AMD alternative, the FX is very handy with efficiently threaded software. If, for instance, you want to build an affordable video encoding box, you could do a lot worse. However, if you are after a multi-purpose PC, we'd argue you either spend a little less and go with the cheapest AMD Phenom II X6 or step up to the Core i7 2700K.