benchmarks
Latest
HTC Jetstream hits AT&T store, gets benchmarked
Thinking about picking up an (inconceivably expensive) HTC Jetstream? After all, it just slipped into AT&T's online store this weekend. We'll just leave this link to HotHardware in the source below, in case you want to check out some comparative benchmarks before trading $700 for a two-year contract. Not that it performed poorly mind you, it raced neck-and-neck with Lenovo's IdeaPad K1 -- we just like think charts and tables are cool. Videos too -- hit the break to see the folks at HotHardware give the 10-inch slate a quick hands-on.
Macworld benchmarks: Mac mini BTO can match iMac performance
As the closest Apple comes to a "budget model" Mac, the Mac mini usually turns out to be substantially less powerful than the larger, more expensive iMac. However, Macworld discovered via recent benchmarks that a build-to-order current-generation Mac mini can meet or exceed many performance marks of a stock configuration iMac. The tale of the tape: 2011 Mac mini 2.7 GHz Core i7 256 GB SSD 4 GB RAM US$1499 versus 2011 21.5-inch iMac 2.5 GHz Core i5 Standard HDD 4 GB RAM $1199 Unsurprisingly, the Mac mini trounced the iMac at disk-intensive I/O tasks thanks to its speedy SSD. I've seen for myself the huge performance gains that can come from putting an SSD into an otherwise lackluster machine, but it comes at a significant cost; Macworld paid an additional $600 for an Apple-sourced SSD on top of the Mac mini's original $799 price. Processor-intensive tasks showed remarkable gains for the Mac mini, with several tasks completing at almost the same time as the iMac. However, the iMac's far better graphics hardware handily bested the Mac mini's performance. By now you're probably wondering why anyone would bother tricking out a Mac mini at a cost $300 higher than the iMac when the Mac mini doesn't come with a display, keyboard, and so forth. Macworld addresses this by correctly pointing out that the iMac's glossy glass display isn't everyone's cup of tea. The BTO Mac mini they tested also turns in similar performance in a far smaller form factor, which could be an important consideration if your workspace is too cramped for an iMac or Mac Pro.
WD's 9.5mm Scorpio Blue 1TB laptop hard drive gets benchmarked
These days, it's hard to shake the urge to pop a solid state drive into your next laptop, but even if you're down with dropping the requisite coin, the restrictive capacity choices may make it darn near impossible for pack rats to bite. For those fitting squarely into that category, there's Western Digital's newest laptop drive: the 9.5mm 1TB Scorpio Blue. It's one of the first in this form factor (read: the one that slips into most everything smaller than a Clevo) to hit the 1TB milestone, and at just $99, it's a veritable bargain. The benchmarking gurus over at Hot Hardware threw it through the usual gauntlet of tests, pitting it against a 500GB Scorpio Black and a 640GB Seagate Momentus. As you'd likely expect, the 1TB spinner bested the competition in SiSoftware, ATTO and CrystalDiskMark tests, though not by a tremendous margin. Still, taking performance up a notch while also boasting a full terabyte of space makes it somewhat of a no-brainer for capacity freaks, and you can hit the source for a barrage of charts proving as much.
Engadget Primed: Using benchmarks
Primed goes in-depth on the technobabble you hear on Engadget every day -- we dig deep into each topic's history and how it benefits our lives. Looking to suggest a piece of technology for us to break down? Drop us a line at primed *at* engadget *dawt* com. Staring at your smartphone, you realize that there's something missing. It does everything you want it to -- very well, we might add -- but what hole is left to fill? We'll help you out with this one: you want bragging rights. There has to be a way to face your friends with confidence, right? All you need is a little nudge in the right direction, and in this edition of Engadget Primed, we'll give you that much-needed shove by explaining benchmarks. Perhaps you've seen us talk about benchmarks in our product reviews. We'll typically use them to gauge the relative performance of various devices, but discussing a Linpack score doesn't mean much without going deeper into what it actually means. What aspects of performance do these benchmarks measure, and what techniques do they use? How much can we rely on them when making purchasing decisions? Read on after the break for the full scoop.
Qualcomm launches Vellamo browser benchmark for Android devices
Qualcomm's not exactly a novice when it comes to sizing up phones -- it's already responsible for the graphics benchmark Neocore. Now, it also wants to show you just how much your mobile browser is lagging. The company just introduced Vellamo, a suite of 11 tests designed to gauge browser performance on Android phones and tablets. In case you're curious, it takes its name from the Finnish goddess of the sea who lures away sailors web surfers (Qualcomm's joke, not ours). And yes, it'll work with any device running Android 2.0 or above, even if it doesn't pack a Qualcomm-made processor. These tests span four broad categories -- rendering, JavaScript, user experience, and networking -- with only two requiring an internet connection (even then, you can cherry pick specific tests to run). We took it for a spin on our aging, Froyo-packing, Motorola Droid, whose score of 237 landed at the very bottom of the list of results, far behind tablets and newer handsets. (As of this writing, Samsung's Galaxy Tab 10.1 took the cake.) Curious to see how your device ranks? Hit the source link to download the free APK file. %Gallery-128414%.
Bare Feats finds iMacs compare to Mac Pros running Final Cut Pro X
Final Cut Pro X users might want consider an iMac instead of a Mac Pro, according to some recent benchmarks run by Bare Feats. The graphics and speed testing site recently tested FCP X on three different Macs to see which current model was able to tame the power-hungry app the best. The contestants were a 2011 iMac 3.4 GHz Quad Core i7 with 16 GB of RAM and a Radeon HD 6970M GPU with 2 GB of VRAM, a 2011 MacBook Pro 2.3 GHz Quad Core i7 with 8 GB of RAM and a Radeon HD 6750M with 1 GB of VRAM, and a 2010 Mac Pro 3.33 GHz 6-core Westmere with 24 GB of RAM and a Radeon HD 5870 GPU with 1 GB of VRAM. The team ran four different tests using the same 32-second HQ video clip. The first test (above) was to apply the Directional Blur effect to the clip, and in this test the iMac beat both the Mac Pro and MacBook Pro by over 3 seconds. The next test applied the Sharpen Blur effect, and once again the iMac was victorious, beating the MacBook Pro by 4.3 second and thoroughly schooling the Mac Pro which came in a full 5.7 seconds behind. Two more benchmarks measured exporting and streaming speeds. Here the Mac Pro squeaked ahead of the iMac, coming in .4 second faster on a H.264 export. When the project was loaded into Compressor 4 and exported as an H.264 stream, the Mac Pro was a full 2.6 seconds ahead of the iMac, really showing off the power of the 6-core processor. The results show two things -- that the new iMacs are surprisingly capable machines for the price, and that Apple really needs to release a new Mac Pro. The latter is widely expected to happen sometime this summer. One comment about these benchmarks, though -- Final Cut Pro X has full symmetric multicore support and renders in the background, so it no longer really matters how fast rendering is done. You can continue working while your multicore Mac is crunching away on rendering. For further details on the testing, be sure to visit the Bare Feats site.
AMD Llano desktop APU gets reviewed: the best integrated graphics in town
AMD is due to release a batch of new Llano APUs next month that are specifically tailored to desktops rather than laptops. The most powerful among them will be the 2.9GHz A8-3850, which has already caused a stir on the review circuit for one simple reason: it pulls off a brutal "one shot one kill" on Intel's HD 3000 integrated graphics. AnandTech raised an impressed eyebrow at the fact that all its benchmarking games were playable on the $135 AMD chip, which roughly doubled frame rates in titles like Modern Warfare 2, Bioshock 2 and World of Warcraft compared to the more expensive Sandy Bridge i5 2500K. TechSpot declared the APU its "new budget king," with graphical performance "on another level" compared even to an i7. However, the superlatives quickly evaporated once reviewers shifted their focus to the CPU. TechReport spotted that pure CPU performance per dollar was actually lower than what you'd get from a lowly i3. Moreover, it reckoned you'd only have to spend an extra $70 to buy a much more powerful CPU and a separate graphics card -- an option that comes "awfully close to making the A8-3850 seem irrelevant." Ouch. Nevertheless, if an affordable processor with integrated graphics is what you're after, then it's fair to say this one sets the standard. Click the source links below for full reviews.
Motorola Xoom hits 1.7GHz, teeters on the edge of oblivion
Just when we thought the Motorola Xoom had hit its stride at a blazing 1.504 billion operations cycles per second, the trusty Tiamat kernel has strapped on an veritable afterburner capable of 1.7GHz. What happens when your shaking hands flip that switch and give that Tegra 2 all the jet fuel it can take? Well, anecdotal cases from the XDA-developers forums suggest it'll probably just reboot anticlimactically. If you're lucky enough to have the magic silicon, however, you'll be treated to a benchmark-blitzing rig, reportedly capable of 70 MFLOPS in Linpack, 1480ms runs in SunSpider, and Quadrant scores approaching a smooth 5,000. See just how far that rainbow benchmark bar can stretch in a screencap after the break. Update: There's a jolly discussion in comments about whether gigahertz can be directly translated to operations per second in the case of the Tegra 2 -- we'll err on the side of caution and say cycles per second instead.
NVIDIA's GTX 560 desktop GPU fills an exceedingly narrow pricing niche
With Tegra 2 hogging the spotlight, sometimes it's easy to forget that NVIDIA is still primarily in the business of making GPUs for computers. Yet, here it is with the GeForce GTX 560, another graphics chip ready to be inserted into mid-range gaming rigs. This smaller sibling of the GTX 560 Ti is designed to plug right into a small price gap in the company's lineup -- right around the $200 mark. The 336 CUDA cores inside this second-gen Fermi card, predictably, perform slightly better than the GTX 460 and fall just short of the 560 Ti, but it does eke out a victory over similarly priced competition from AMD. The only thing that kept reviewers from wholeheartedly endorsing the various (and often overclocked) flavors of the 560 was the tiny difference in price between it and its relatives -- tacking on the letters Ti and its 48 additional stream processors costs as little as $15 after a mail-in rebate. Check out the reviews below for all the benchmarks your little nerd heart can handle. And don't miss the video of a GTX 560 plowing through Duke Nukem Forever, Alice: Madness Returns, and Dungeon Siege III at the more coverage link. Read - Tech Report Read - AnandTech Read - Tom's Hardware Read - Guru 3D
Samsung Galaxy S II overclocked to 1.5GHz, used to obliterate benchmarks (video)
What do you do after you've bought the most powerful smartphone to yet grace the Android universe? Some timid folks would urge you to just enjoy it, but if you're like us, you'll probably want to know just how high that dual-core Exynos 4210 birdie could fly. The processor inside the Galaxy S II started off life at 1GHz, then got sped up by Samsung to 1.2GHz just before release, and has now been taken all the way to 1.5GHz thanks to coolbho3000 over on the xda-developers forum. He's been nice enough to provide the source code and instructions necessary to reproduce this stable overclock on your Galaxy S II, but don't rush off just yet -- there's video of the GSII romping through Quadrant and Linpack after the break. [Thanks, Mike and Sam]
Sony Ericsson Azusa outed via NenaMark, reminds us of Xperia Arc
It doesn't matter how much you warn young phones about public benchmarking -- they just don't listen. An unannounced Sony Ericsson device, dubbed Azusa, was caught peeking out from behind an anonymous submission to the NenaMark Android GPU benchmarking tool. The resulting specs outed the device as a 854 x 480 resolution handset packing the same Qualcomm Adreno 205 GPU as the Xperia Arc. Unsurprisingly, the Azusa's single-test benchmark stats are about on par with its bigger brother, leaving us wondering -- what does the newbie bring to the table?
Macworld publishes first round of benchmark results for new iMac
Macworld posted the results of its first benchmark tests for the newly updated family of iMacs this morning. The publication's tests found Apple's latest iMacs to be generally faster than previous iMacs, although certain build-to-order (BTO) models from 2010 still outperformed the new family of iMacs in some tests. The new top-of-the-line, quad-core 3.1 GHz i5 iMac bested the previous high-end model, a quad-core 2.8 GHz i5 iMac, in every test. Notably, the 3.1 GHz iMac performed 16% faster in the Speedmark 6.5 test and 22% faster in Macworld's iTunes encode test. The results were mixed when comparing the new iMac to powerfully configured BTO models from last year. When pitted against a quad-core 2.93 GHz Core i7 BTO iMac from 2010, the new 3.1 GHz iMac performed about equivalently in the Speedmark test but fell short in applications like Cinebench and MathematicaMark. The older, i7-based iMac offered Hyper Threading, a technology that virtually doubled the number of processor cores seen by the operating system; apps that took advantage of this technology worked better on the older machine. When compared to a dual-core 3.6 GHz i5 BTO iMac from 2010, the current 3.1 Ghz iMac edged out its older relative by 14% in overall performance and finished tasks in Mathematica, an app that makes use of multiple processing cores, up to 67% faster. In many cases, however, the faster clock speed of the older machine helped it complete certain tasks more quickly. Overall, the new iMacs seem to offer generally better performance in a more affordable package. But, if you frequently use software optimized for fast clock speeds or Hyper Threading technology and don't need any of the new gadgetry in the newest machines, it may be worth tracking down a 2010 BTO iMac with a 3.6 GHz i5 or 2.93 GHz i7 inside.
AMD elevates the low-end with trio of sub-$100 cards: Radeon HD 6670, 6570, and 6450
Graphics card companies don't live and die by the enthusiast market alone. That may be where the glory is, but it's the budget cards that really bring in the bacon. For the entry level, AMD just unleashed a trio of sub-$100 cards, the Radeon HD 6670, 6570, and 6450. How do they perform? Well, let's just say you get what you pay for. Reaction from reviewers has been one of mild indifference. Depending on manufacturer, fan noise does appear to be an issue, possibly precluding the cards from being a viable HTPC choice. Otherwise, even the lowly, $55 6450 is a worthy upgrade over an integrated graphics chip or a two-year-old discrete card, but it can't match the performance of NVIDIA's GT 430, which can be had for only a few dollars more. Consensus was that, with prices of the older 5000 series being slashed, purchasers can get more bang for their GPU buck by sticking with last generation cards (like the Radeon HD 5750) if they're looking for pure gaming prowess. That said, the GDDR5 flavors of the 6670 provide perfectly playable performance on most modern games (it averaged 45 FPS in Call of Duty: Black Ops) for just $99 (the 6570 runs about $79). Just beware those models shipping with GDDR3. Benchmarks galore below. Read - Hexus Read - techPowerUp 6450 Read - techPowerUp 6670 Read - Guru3D Read - Tech Report Read - Tom's Hardware 6670 and 6570 Read - Tom's Hardware 6450 Read - TweakTown Read - AnandTech Read - HotHardware
Galaxy S II benchmarked, makes other phones cry in shame
The first benchmarks of Samsung's 1.2GHz Galaxy S II -- we saw the 1GHz variant tested previously -- are rolling in, and it's fast. Almost suspiciously so. Its speedy dual-core Exynos CPU pulled off a 3,053 in Quadrant and scored double what the similarly specced HTC Sensation did in Smartbench2011. Those scores also represent a more than threefold improvement over the original Galaxy S. Of course, these are purely synthetic benchmarks and may not translate into an equally improved experience in day to day use. We'll know for sure when we get our hands on one for a proper review in the coming weeks. Full benchmarks are after the break. [Thanks to everyone who sent this in]
A5 beats Tegra 2 in benchmark tests, thanks to larger size
GLBenchmark pitted a few tablet processors against each other in a speed showdown, and Apple's custom-designed A5 chip in the iPad 2 handily beats NVIDIA's Tegra 2 processor, used in competing tablets like the Motorola Xoom. A few analysts are saying that it's the size that matters in this case -- Apple's chip is more than twice the size of the Tegra 2, and that allows Apple to pull off some better benchmarks, even though the two chips are relatively the same in terms of specifications. Usually, of course, a bigger chip would mess with the design of the overall hardware, but since Apple is doing everything itself, it can afford the extra space and the larger components. Of course, NVIDIA is set to introduce a Tegra 3 chip later on this year, and that will undoubtedly introduce a new wrinkle to the tablet lineup. Apple no doubt also has its engineers working on faster chips, which means the speed of tablet computing likely still has a long way to progress. [via 9to5Mac]
Radeon HD 6790 sneaks in at under $150, leaves reviewers wanting more for the money
As sure as snow in winter or sun in summer, AMD has yet another refresh to its graphics card portfolio this spring. The Radeon HD 6790 is only a couple of misplaced digits away from the far more illustrious HD 6970, but you should be able to tell the two apart by another, altogether more significant spec: the new mid-tier card retails at $149. Predictably, its performance offers no threat to AMD's single-GPU flagship, but the 6790's 840MHz graphics and shader clock speeds plus 1GB of GDDR5 running at an effective 4.2GHz data rate don't seem like anything to sniff at either. Reviewers agreed that it's AMD's slightly delayed answer to NVIDIA's GTX 460, and with the latter card exiting retail availability to make room for the (oddly enough) less powerful GTX 550 Ti, AMD's new solution looks set to be the better choice at the shared $149 price point. Alas, being limited to 800 Stream processors and 16 ROPs does expose the HD 6790 to being cannibalized by AMD's own Radeon HD 6850 (which can be had for sub-$150 if you're tolerant of rebates) and that turns out to be exactly what happens. A solid card, then, but one that would require an even lower price dip to make economic sense. Benchmarks await below. Read - Tech Report Read - AnandTech Read - Tom's Hardware Read - PC Perspective
Qualcomm's 1.5GHz dual-core MSM8660 destroys the competition in majestic benchmark run
Take the 1GHz Scorpion core residing in Qualcomm's current flagship Snapdragon, the MSM8x55, duplicate it, overclock the resulting pair by 50 percent, and give them improved graphics in the form of Adreno 220. What do you get? A barnburner by the unsexy name of MSM8x60. Yes, the 1.5GHz Mobile Development Platform Qualcomm loosed on the world at CES earlier this year has found itself prostrate on a test bench, where it has produced some of the finest graphical performance scores yet seen on a mobile device. The taxing OpenGL ES 2.0 test you see above shows the new Snapdragon doubling the frame rates churned out by Motorola's mighty Atrix 4G (which admittedly has to work harder thanks to its higher-res display) and completely embarrassing older generation hardware like the EVO 4G. That's a theme carried on throughout AnandTech's benchmarking, which you may explore in full at the source link. If you're wondering when this world-beating dual-core chip will be coming to market, the answer is that it's already inside HTC's imminently upcoming EVO 3D and Pyramid devices, albeit running at a tamer 1.2GHz. Exciting, eh? P.S. - Do take note that the Qualcomm dev platform was plugged into the wall during these tests and was not subject to any power management software that may have otherwise restrained its performance as on the retail chips under test. Moreover, the Egypt benchmark can only run at native resolution, which is what's causing some seemingly aberrant results such as the iPhone 4 (960 x 640) ranking below the iPhone 3GS (480 x 320).
Android vs. iPhone in 'flawed' mobile browser performance test
Post edited to clarify that the browser testing is not representative of Safari performance, and included Blaze response to CNET. –Ed. Blaze Software, a Canadian software company, today released the results of what it calls a "definitive" research effort to discover "which [mobile] browser is really faster from a user's point of view." The study concluded that Android's browser is 52% faster than the iPhone's. Before you trade in your iPhone for a device powered by Android, The Loop suggests Blaze's study is "flawed." According to its report, Blaze's testing methodology relied on "custom apps, which use the platform's embedded browser. This means WebView (based on Chrome) for Android, and UIWebView (based on Safari) for iPhone." As we've been hearing from developers of iPhone web apps over the past few weeks, Apple's improvements to the Mobile Safari JavaScript engine and other rendering speedups have not been extended to the internal browser tool used by apps, nor to standalone web apps that are pinned as icons to the home screen. It's not yet completely clear if or when the Safari performance boost will make it to the embedded browser view; John Gruber cites some security-related concerns that may be involved. The tests don't reflect performance of the official web browsers included with each platform. UIWebView did not include this performance boost; it may be "disingenuous" to conclude Android beat Safari, according to The Loop. Using an embedded browser is not the same as using the official browser where customers spend the most time interacting with websites. "Obviously someone is looking to make a mountain out of a molehill," Gartner analyst Michael Gartenberg told The Loop. "It's not an apples to apples test." Apple's Natalie Kerris was equally dismissive, speaking to CNET: "Their testing is flawed. They didn't actually test the Safari browser on the iPhone. Instead they only tested their own proprietary app, which uses an embedded Web viewer that doesn't actually take advantage of Safari's Web performance optimizations." Kerris also noted that even without the true Safari match-up, the testing only showed about a second of difference browsing pages. Blaze's CTO Guy Podjarny admitted to CNET that the testing methodology made an invalid assumption that the embedded browsers would work as fast as Safari: "This test leveraged the embedded browser which is the only available option for iPhone applications. Blaze was under the assumption that Apple would apply the same updates to their embedded browser as they would their regular browser. If this is not the case and according to Apple's response, it's certainly possible the embedded browser might produce different results. If Apple decides to apply their optimizations across their embedded browser as well, then we would be more than willing to create a new report with the new performance results." Even so, the results of Blaze's research should still disappoint Apple's fans. Apple touted significant web technology performance gains in its latest iOS release. It seems reasonable to expect these gains to appear simultaneously in both the Safari browser and the underlying UIWebView framework used in nearly any app that relies on web technologies like JavaScript. Blaze's researchers built custom apps to compare the iPhone 4 and Google Nexus S using websites from Fortune 1000 companies. Each site included in the test was loaded multiple times over several days using a Wi-Fi connection. The final results reflect a median benchmark from over 45,000 page loads. "Android 2.3 was 52% faster than iPhone 4.3, with a median load time of 2.144 seconds vs. iPhone's median load time of 3.254 seconds," Blaze reports on its blog, adding, "Android was faster than iPhone in 84% of the tested websites, and iPhone beat Android in 16% of the races. Android...provided a faster browsing experience 4 times out of 5."
iPad 2 benchmarks show impressive gains over original model
One of the first things many iPad 2 upgraders noticed was that the new device seemed very fast in comparison to the original iPad. Over at GigaOM, blogger Geoffrey Goetz ran benchmarks on an iPhone 4, an iPad and an iPad 2 (all running iOS 4.3) to show how speedy the new iPad really is. As you can see in the chart above, the iPad 2 consistently outperformed the Apple A4-powered iPad and iPhone 4 in every test. Using the Geekbench tests from Primate Labs, the iPad 2 was about 64 percent faster than its predecessor in overall results. Likewise, the Gague benchmarks showed an overall performance gain of about 56 percent, while BenchTest clocked in with a 63 percent improvement. Goetz noted that he personally believed that his iPhone 4 was faster than his original iPad, a belief that the benchmarks proved to be incorrect. He also compared iOS 4.3 benchmarks submitted by a large sampling of Geekbench users, all of which supported his assertion that the iPad 2 performance gains are "stunning." The Geekbench numbers show the iPad 2 to be roughly as powerful as a G4-based 15" PowerBook or Mac mini, but as Goetz summarizes in his post, "I really don't think that raw performance alone is where the focus needs to be anymore."
AnandTech benchmarks the iPad 2, shows huge graphics gains
AnandTech is in the midst of a lengthy analysis of the iPad 2's hardware. The results so far confirm what most of us already knew: the iPad 2's display is virtually identical to the original iPad, the cameras are nothing to write home about, and the CPU is slightly more powerful. What's most revealing in AnandTech's testing is how greatly improved the iPad 2's graphics are compared to the original iPad. AnandTech tested the iPad 2's PowerVR SGX 543MP2 GPU against the original iPad and the Motorola Xoom, and the iPad 2 outperformed both of them significantly. More than that, actually -- in terms of performance on AnandTech's benchmarks, the iPad 2 destroyed both the original iPad and the Xoom. Neither device was even close to replicating the iPad 2's graphical performance. A handful of developers have already pushed out updated apps that take advantage of the iPad 2's graphics hardware, including Firemint's Real Racing 2 HD and Chair's Infinity Blade. While the iPad 2's screen resolution is unchanged from the original iPad, its more powerful GPU has allowed developers to increase the level of detail and lighting effects in their games. AnandTech has a mouseover pic comparing Infinity Blade's appearance on the original iPad versus the iPad 2-optimized graphics, and the difference is easy to see. As more developers begin tailoring their apps' graphics to the iPad 2's impressive hardware, it's entirely possible that gaming might see a new surge on the iOS platform. Meanwhile, if you're interested in a deep dive into the iPad 2's hardware, definitely check out AnandTech's review.
