benchmarks
Latest
Three WD VelociRaptors get setup in RAID 5 array, testing ensues
Western Digital's hasty VelociRaptor already got reviewed by its lonesome, but for those thinking of getting a RAID system into their rig, HotHardware has taken a trio of 'em, setup a RAID 5 array and put the drives through their collaborative paces. The configuration was made possible thanks to an Areca PCIe X8 hardware RAID card, and the results were rather impressive -- to no one's surprise, might we add. Across the entire volume, performance was generally linear save for a few small valleys along the way, burst speed was 598MB/sec and average read speed was 209.4MB/sec, which pretty much blew the doors off of everything that came before it. Number lovers, there's more where this came from in the read link below.
NVIDIA's 512MB GeForce 9800 GTX+ hits the bench
Yeah, that bench. NVIDIA's just-announced-yesterday GeForce 9800 GTX+ has already been used and abused for your satisfaction, and the folks over at PC Perspective have the benchmarks to prove it. The 55nm-based card was pit against an 8800 GT and AMD's extraordinarily fresh Radeon HD 4850, but we're not going to insert any spoilers in this space (okay, so it fared well... really well). All the graphs and screen captures you crave are waiting just down there.[Thanks, Ryan]
MSI's GX600X TurboBook gets inspected, reviewed
Surely you remember that G600 that MSI showcased at CeBIT, right? We know you're sitting in stunned silence just thinking back, so we'll move forward anyway. The crew over at TweakTown managed to get ahold of the GX600X TurboBook, and rather than just frolicking in fields of flowers with it, they put this puppy to the test. Unlike the vast majority of MSI lappies, this particular one manages to stand out somewhat, with a relatively loud design and a fairly potent set of components to boot. Reviewers found it to be a "peak performer" with "excellent overclocking capabilities," and while the logos may catch a few eyes, these critics felt that those very decals may deter folks from buying an otherwise solid machine. Nevertheless, head on down to the full review -- you may find yourself surprisingly attracted to an MSI. Imagine that.[Via I4U News]
ASUS M50S multimedia laptop seen in the wild
Ooh, shiny! ASUS' CES-announced M50S has finally found its way out of whatever lair it has been tucked away in, and thankfully, the shutterbugs over at PC Perspective snapped a few shots for the world to see. Initial impressions are that it should handle most gaming needs with ease, yet won't destroy your back when toting it from LAN party to LAN party. The built-in Blu-ray burner and multi-function trackpad are quite nice, too. Hit up the read link for the full preview -- a bona fide review should be following shortly.[Thanks, Ryan]
MBP vs. MBA benchmarking showdown
Bare Feats continues to pit Macs against each other in a no-holds-barred, up-against-the-wall benchmarking breakdown, and this time around it's the MacBook Pro and the MacBook Air in the octagon together. And things go about as planned -- while the solid state drive in the MBA has its advantages, the processor and GPU (what little there is of it) in the MBA don't even really compare to the MBP. Sure, it's a small, super thin, super portable computer, but not only are you paying in cash for the portability, you're paying in performance, too.This doesn't mean the MBA is a bad computer at all -- for most tasks, it'll work just fine. But complicated 3D graphics (both Halo and Unreal Tournament 2004 were tested) will be almost unplayable in games, and there will be a substantial wait, sometimes up to minutes more, for certain processor intensive tasks. If speed is a high priority (at least higher than portability), the MBA isn't for you.I would like to see how the MacBook compares, though. It's not surprising that the MBP is a fast machine, but where does the MBA line up compared to the cheaper model?[Via IMG]
Intel's Core 2 Duo E8500 Wolfdale gets tested
If you'll recall, we already witnessed a pre-production Wolfdale CPU outpace its competition back in August of last year, but the real deal has finally arrived. Over at Hot Hardware, the folks were able to crack open a retail Core 2 Duo E8500 processor, which comes in at 3.16GHz and features 6MB of L2 cache, a 1,333MHz FSB speed and is built on 45-nanometer technology. When put to the test, reviewers found this luscious piece of silicon to be quite the performer, noting that it was probably a better choice for gamers than a low-end quad-core chip. Still, with an expected price of just under three bills, it wasn't exactly seen as the greatest of values, but if you've got the coin and happen to fit the market, the "Recommended" badge tells you everything you need to know.
Apple MacBook Pro Penryn tests: a little more speed, a lot less heat
Sure, more speed is great, but on OS X most bumps don't have an appreciable effect on everyday work (unless all you're doing is video encoding). We were more interested in what Intel's Penryn processors can do for the MBP's heat envelope, so we maxed out both CPU cores in this brand new machine at let 'em crank for a half an hour. Our test machine may or may not compare to a later 3rd-gen MBP in terms of heat efficiency, but here's the quick rundown:Core Duo (2.16GHz) Peak: 195° F, fans over 5000 RPM Average: 185° F, fans about 4900 RPM Enclosure (bottom): very hot to the touch Enclosure (top): fairly hot to the touch Penryn Core 2 Duo (2.5GHz) Peak: 176° F, fans about 3400 RPM Average: 173° F, fans about 2500 RPM (much quieter) Enclosure (bottom): warm to the touch Enclosure (top): warm to fairly hot to the touch The usual Xbench tests are after the break.
Intel's Skulltrail QX9775 x 2 never met a benchmark it didn't like
Forget that shrinking nonsense Intel keeps talking about -- squeezing battery-sipping processors inside minuscule form factors -- it's totally played. We got here some numbers on Intel's new Skulltrail gaming platform, featuring dual quad-core QX9775 processors and other ridiculous specifications designed for besting the likes of AMD and that pesky Crysis frame rate. If you want to get into all the tech nitty gritty, the read links have all the bullet points to satisfy, but here's the long and the short of it: those eight Xeon cores and various other server-inspired innards blow away every sort of competition in multi-threaded applications, and other single-threaded CPU-heavy tests put the QX9775 near or at the front. Unfortunately, the board is held back by its use of DDR2 800 FB-DIMMs, which landed it a bit behind the QX9770 and QX9650 Core 2 Extreme in Crysis benchmarks -- one of the few benchmarks that Skulltrail even felt any competition from the rest of the pack. It seems like the board mainly shines when it has multiple graphics cards to back it up, and it's also clear that the Crysis Everest won't be bested by CPU juice alone. There aren't any specifics on price or release date yet, but expect to pay $600 or more per CPU.Read - Hot HardwareRead - PC PerspectiveRead - TechgageRead - Custom PC
First benchmarks: MacBook Air is the slowest Apple machine on the block
No one should be surprised, but if you were looking for a speed freak of a laptop, you're going to want to glaze those lustful eyes over the MacBook Air. While the new hotness has a speedy memory bus, overall its performance is dramatically smushed by its integrated graphics, anemic 1.6GHz Core 2 Duo processor, and 1.8-inch hard drive. Yes, you can officially call the MacBook Air the slowest machine Apple makes -- but you already figured as much, right? No one buys an ultraportable for its real ultimate power.
CNET Labs benchmarks Parallels and Fusion virtualization products
Whenever two options are presented, it is often a smart move to weight the pros and cons of each and make a decision based on your needs. If you need to push Windows in a virtualization environment, Daniel A. Begun at CNET Labs has benchmarked Windows Vista running in both Parallels Desktop and VMware Fusion. The short of the long? If you really need to push Windows, do it in Boot Camp if at all possible. If Boot Camp is out of the question, VMware Fusion was the winner in heavy duty, multimedia tasks by a landslide, due largely in part to its support of multiple core processors (Parallels, as of this writing, only supports one core per CPU). On the flip side, however, Parallels Desktop is the only virtualization product that supports 3D gaming right now, but as a Parallels owner, I'll still vouch for running your games in Windows via Boot Camp.At the end of the CNET Labs day, both apps still have their pros and cons. Begun hails Parallels Desktop for being the more usable and intuitive of the two (and during my initial tests of both before deciding to buy, I agreed), though I had a much easier time getting non-Windows OSes installed in VMware Fusion (to this day, Parallels Desktop still won't install Ubuntu, and I have received no response as to why). Still, the moral of this particular story is that if performance is at the very top of your virtualization shopping list, VMware Fusion is, at least for now, the reigning champion. Check Begun's full article for more details on the tests and how they were performed.
iPhone news roundup: benchmarks, Facebook, and (obviously) rumors
There was a lot of iPhone chatter this week -- although Apple might have done all it can to lock the little bugger down, it seems people are still finding ways to extend and explore its capabilities, while AT&T might have finally gotten the hint about those ridiculous paper bills. Logic3 unveiled the i-Station Traveler (pictured), the first speaker dock we've seen specifically for the iPhone. While most iPod docks tend to work fine with the iPhone, the $60 Traveller is designed specifically around the horizontal orientation, allowing you to watch movies while the iPhone is docked. Meebo and Facebook both launched iPhone-specific versions of their sites, allowing you to IM your friends and stalk your exes with all the swoopy-slidey flair you'd expect. Orange continued to act all coy about potentially being Apple's partner in France, saying only that it had "no comment" on the iPhone, even as rumors heat up. The iPhone got straight-up benchmarked for the first time: Craig Hockenberry whipped out his stopwatch and discovered that Javascript in MobileSafari runs right around eighty times slower than on a 1.83GHz Core 2 Duo Mac. He also whipped up a little app using that pirate toolchain we love so much and discovered that native ARM code runs right around 200 times faster than Javascript in the iPhone. Looks like that Safari sandbox might not be so "sweet" after all. AT&T seems to have decided that its vendetta against the trees of the world might be a little misplaced, and is in the process of moving to "summary billing," according to a call center employee. Either that, or they're trying to guilt people into switching to e-billing by sending out ridiculous bills. Really, that's what the email says. All in all, a pretty busy week for the iPhone -- kinda makes you wonder how much action there'll be when Apple finally releases that official SDK, eh? Read - i-Station Traveler Read - Meebo Read - Facebook Read - Orange declines to comment on the iPhone Read - iPhone benchmarks Read - AT&T reducing paper bills[Thanks, risingsonn and The Boy Genius]
TabletKiosk's Sahara i440D Slate PC reviewed, benchmarked
Just over a fortnight after the TabletKiosk's Sahara i440D Slate PC began shipping out to interested buyers, the niche machine has now been reviewed and benchmarked for those still mulling a purchase. The folks over at TabletPCReview noted that this machine was a breeze to tote around, had a solid feel, sported a satisfactory (though not awe-inspiring) display, and performed sufficiently enough for those not looking to actually replace their (likely more robust) laptop. Additionally, the pen functionality was highly praised as was the respectable array of ports, but the amount of noise and heat that this bad boy emitted was a slight turn-off. Still, the Sahara i440D benchmarked well against similar alternatives, and while we wouldn't recommend 3D gaming or motion picture rendering on it, it seems like a solid choice for those in the market. Check out a videoed extension of the review after the jump.
Meet the New Macs, A Little Faster Than the Old Macs
Personally, I crave speed. No, not the kind that can often be purchased on some local street corner in that part of town on the wrong side of the tracks. No, I'm referring to processors, front side buses, RAM and other factors that go into calculating the raw, unbridled, number-crunching power of the latest and greatest Macs.No doubt you are aware that Apple recently released an updated version of its flagship prosumer desktop iMac as well as an updated Mac Mini too. The new iMacs feature changes both external and internal with increases to the clock speed of its Intel Core Duo 2 processors as well as a switch to the faster Santa Rosa chipset. Yes, that all sounds great but what does it mean in terms of the new iMac's ability to crunch those numbers?Well, according to Primate Labs' tests on the new 24" iMac, as featured at Apple Insider, it doesn't help a great deal unless you're using one of those math-intensive apps like Photoshop or Aperture. In that case, according to the article: "If you're running memory-intensive applications you'll certainly notice an increase in performance with the new 24-inch iMac. Heck, the fact that the new 24-inch iMac supports 4GB of RAM while the old 24-inch iMac supports 3GB of RAM might be enough to convince you to get one." Good point. More RAM is a good thing.
Apple's 2.4GHz 24-inch aluminum iMac gets benchmarked
Although we certainly noticed that Apple's latest iMac felt snappy enough, the folks over at Primate Labs have cranked out a few numbers for the data freaks in the crowd to chew on. Granted, these benchmarks do not include the Core 2 Extreme iMac nor are they the most complete set of tests we've ever seen, but they do seem to give those on the fence a decent look at what level of performance increases they'll be dealing with. Put simply, the new 2.4GHz 24-inch iMac posted "modest gains" across the board compared to the previous iterations that clocked in at 2.33GHz / 2.16GHz, and while the results don't seem earth-shattering, those who rely on "memory-intensive applications (like Aperture or Photoshop)" would likely benefit most from the improved "integer, floating point. and memory / stream performance." As you'd expect, the full skinny on the test results await you in the read link.[Via AppleInsider]
SSD shootout: pricey HDD replacements do battle
Considering that solid state discs -- although still noticeably more expensive that traditional HDDs -- are quickly plummeting in price, it's about time a brief roundup was executed in order to pinpoint which SSD was best for you. Granted, CustomPC's evaluation only included a handful of options, but the in-depth testing process found a few glaring losers and even fewer runaway winners. As expected, reviewers used a number of read / write testing applications to judge the speed differences and real-world performance increases on seven SSDs ranging from 8GB to 32GB in size. Coming out on top was Samsung's 32GB drive, which proved both "quick and silent" in their testing; however, the 18GB STEC Zeus-IOPS proved the hands-down winner in terms of sheer speed. 'Course, we highly doubt many of you will be snagging the latter after witnessing its £7,050 ($14,461) pricetag, but a number of slightly slower performers including PQI's Turbo Plus 2.5 could very well provide the boost you desire without (totally) breaking the bank.
IBM kicks out energy-efficient 4.7GHz POWER6 processor
Nah, it's no BlueGene L supercomputer, but IBM's latest dual-core microprocessor runs at a cool 4.7GHz while sporting 8MB of total cache per chip. The device reportedly runs "twice as fast" and packs four times the cache as the POWER5, and boasts a processor bandwidth of 300Gbps. Interestingly, the massive power increase doesn't seem to come with a boost in energy requirements, as IBM claims that the 65-nanometer POWER6 somehow ups its game while "using nearly the same amount of electricity" as its predecessor. The company plans on shoving the new darling into the System p570 server, and preliminary testing showed that all four of the "most widely used performance benchmarks for Unix servers" were shattered by its CPU. Unfortunately, there's no word on pricing nor availability just yet, but we're anticipating a bit of sticker shock when it does finally land.[Via LinuxDevices]
Apple's $1099 2GHz MacBook gets benchmarked
If you'd been pinching your pennies waiting for Apple to unleash an update to the MacBook side of things, your wait was ended last week, but if you've been wondering just how much difference the increase from 2MB to 4MB of L2 cache (and the 0.17GHz bump, too) made, here's your sign. MacLife was able to get its hands around a $1,099 base unit, and rather than carefully coaxing into the world, it broke out the whipping stick and put this bad boy to work. The new machine was pit against the old base MacBook (at 1.83GHz) and the previous 2GHz BlackBook, and while increases in efficiency weren't exactly considerable, reviewers found them at least noticeable. The latest machine shaved seconds off here and there from iMovie exporting and Photoshopping, and while most tests showed just single digit improvements, the iPhoto test did display a 22-percent gain compared to the previous 1.83GHz MacBook. Of course, these preliminary tests aren't anything out of the ordinary, but keep an eye on the read link for more fleshed out testing (including the refreshed BlackBook) in the coming days.[Via MacSlash]
Penryn CPUs benchmarked, don't slouch
So, 45nm sounds hip-cool and all, but what does that mean for performance? Intel's coming forward with some Penryn numbers -- with the help of Reg Hardware looking over its shoulder to make sure the tests are legit -- and it seems we're in for a decent speed bump, but not necessarily a revolution. Dual core and quad core 3.33GHz Penryns, each with a 1333MHz frontside bus, were pitted against the current desktop-leading 2.93GHz Core 2 Extreme QX6800 on a 1066MHz bus. Tests were run on top of a well decked test system, running Vista Ultimate. Naturally, the quad core Penryn took the top spot in all the tests, with varying margins depending upon which app was used -- no surprises there. The dual-core Penryn took third behind the spendy QX6800 for the most part, other than the single-threaded Half-Life 2 and the SSE 4-capable DivX test, which capitalized upon the media-friendly instructions Intel is adding to its new processors. Looks like we'll be seeing much more from these processors as more SSE 4 apps hit the market, but for now you shouldn't be feeling too bad about buying a QX6800 -- it's not obsolete by a long shot.
Apple TV benchmarked, not a supercomputer in disguise
So it's true after all, Apple has no magical wand for squeezing unrealistic performance out of off-the-shelf components, but they have managed to put together a halfway decent "Mac" for the price. Tipster Matt wiped his Apple TV's drive and ran a clean install of 10.4.9, just to throw it to the Xbench dogs. It came out surprisingly well, with a 55.75 score, about half that of a medium-range Intel Mac mini. In most operations it actually managed to score fairly close to its taller cousin, just falling behind significantly in certain operations, specifically 3D graphics. Of course, Xbench scores are user submitted and not all that scientific, but it should at least give you a general idea of how this $299 Mac in disguise stacks up.[Thanks, Matt]
Gaming mice get objective rankings
Evaluating high-end gaming mice has always been a somewhat subjective process, relying more on a reviewer's opinion of the mouse's "feel" than any quantifiable measures of performance. Now, a writer at ESReality seems to have solved this problem with an inventive setup that can compare mice objectively on their own merits.By clamping a mouse over a controlled-speed turntable, the author has created a system to accurately gauge the exact speed at which a mouse goes from "perfect control" to malfunctioning randomness. How much sensitivity is necessary depends largely on your play style, but it goes without saying that a higher top-speed is generally preferable to a lower one.When the final results were tabulated and weighted the overclocked Logitech MX500 reigned supreme, giving perfect control even when zipped along the surface at a blazing meter per second. If you're moving your mouse any faster than that, may we suggest looking into some sort of program for caffeine addiction.[Via Digg]
