Asus’ Ivy Bridge-powered N56VM notebook

At last, Ivy Bridge is out. Surely that means a tidal wave of next-generation ultrabooks is about to hit. Surely we’re about to slice our fingers on Ivy Bridge-powered systems so thin they can be classified as bladed weapons.

Well, maybe we are. Just not quite yet.

In a disappointing turn of events, last month’s big Ivy Bridge launch didn’t bring us dual-core, 17W chips ready for ultrabooks. Those are coming, but later this year—perhaps early June, if we’re to believe the rumor mill. For now, the best we’ve got are mobile Ivy Bridge CPU models based on the same silicon as their existing desktop brethren.

In short, these mobile Ivy Bridges are fabbed on a 22-nm process with four cores and up to eight threads. They have as much as 8MB of L3 cache, DirectX 11-capable integrated graphics, PCI Express 3.0, and of course, the same architectural improvements that make Ivy Bridge faster, clock for clock, than Sandy Bridge. To top it off, Ivy Bridge has higher performance per watt than its predecessor.

We explored all of these improvements in detail in our Ivy Bridge desktop review last month. That’s where you should head if you’re dying to know about the nitty-gritty side of things.

What’s on our agenda today is a look at a real, honest-to-goodness notebook PC based on one of those new quad-core Ivy Bridge chips. The machine has a 15.6″ display and weighs in at just over six pounds, so in today’s ultrabook-crazy world, it might seem like the notebook equivalent of the fat girl at the prom. Well, okay; maybe not. This Asus N56VM is quite elegant by full-sized laptop standards, and it has some neat perks, like a 1080p display with a matte finish. It’s fast, too. The Core i7-3720QM processor ticking under the hood is the third-fastest mobile Ivy Bridge variant available today. Here’s how it compares to the rest of Intel’s mobile Ivy Bridge lineup:

Processor Core i7-3612QM Core i7-3610QM Core i7-3615QM Core i7-3720QM Core i7-3820QM Core i7-3920XM
TDP 35W 45W 45W 45W 45W 55W
Cores/threads 4/8 4/8 4/8 4/8 4/8 4/8
Base speed (GHz) 2.1 2.3 2.3 2.6 2.7 2.9
Peak Turbo speed (GHz) SC 3.1 3.3 3.3 3.6 3.7 3.8
DC 3.0 3.2 3.2 3.5 3.6 3.7
QC 2.8 3.1 3.1 3.4 3.5 3.6
L3 cache 6MB 6MB 6MB 6MB 8MB 8MB
Intel HD Graphics 4000 Yes Yes Yes Yes Yes Yes
Base IGP speed (MHz) 650 650 650 650 650 650
Peak IGP speed (MHz) 1100 1200 1200 1250 1250 1300
Package rPGA, BGA-1224 rPGA BGA-1224 rPGA, BGA-1224 rPGA, BGA-1224 rPGA
Price N/A N/A N/A $378 $568 $1096

With a 45W power envelope, there’s little chance of the i7-3720QM showing up in anything remotely thin or light. Still, if you don’t mind a little bit of bulk in your backpack (or you’re buying a laptop to serve as a glorified desktop machine), then the chip has undeniable appeal. Turbo Boost can take one of its cores up to a blistering 3.6GHz to speed up single-threaded tasks, and even when all four cores are busy, it’ll run as fast as 3.4GHz if thermal constraints allow. The chip comes with the full-featured version of Intel’s brand-spanking-new HD Graphics 4000 IGP, and like its mobile partners in crime, it supports DDR3 memory speeds up to 1600MHz.

The i7-3720QM even has the full array of enterprise-friendly features, like AES, TXT, and vPro support. The Core i7-3820QM and Core i7-3920XM do, as well, but slower members of the mobile Ivy family aren’t so lucky.

 

Our host for this beast of a processor is, as we’ve said, the Asus N56VM. As I’m sure you can tell from the pictures above, we weren’t lying about its aesthetic appeal. I guess if we’re drawing comparisons to full-figured ladies, then this is more Christina Hendricks than anything. Rrrrrrawr.

This laptop came to us directly from Intel, but it’s considerably closer to a real, honest-to-goodness retail product than the Sandy Bridge test vehicle we looked at last year. Asus actually plans to sell the N56VM eventually, although it’s going to release a “very similar configuration,” the N56VZ-DS71, beforehand. We’re told that config will pack a slower Core i7-3610QM processor and a quicker GeForce GT 650M GPU. I don’t know if pricing is final yet, but just to give you an idea, Asus mentioned figures in the $1149-1249 range.

 

Processor Intel Core i7-3720QM 2.6GHz
Memory 8GB DDR3-1600 (2 DIMMs)
Chipset Intel HM76 Express
Graphics Intel HD Graphics 4000

Nvidia GeForce GT 630M (2GB DDR3)

Display 15.6″ TFT with 1920×1080 resolution and LED backlight
Storage 750GB Seagate Momentus 7,200-RPM hard drive

Matshita UJ141AF Blu-ray combo drive

Audio HD audio via Realtek codec
Ports 4 USB 3.0

1 VGA

1 HDMI

1 Ethernet via Atheros AR816x controller

1 analog headphone/digital SPDIF port

1 analog microphone port 1 analog speaker port

Expansion slots 1 SD card reader
Communications 802.11b/g/n Wi-Fi via Intel Centrino 2230

Bluetooth 4.0

Input devices Chiclet keyboard

Elan touchpad

Internal microphone

Camera 2.0-megapixel webcam
Dimensions 15.0″ x 10.1″ x 0.5-1.4″ (381 x 257 x 12-35 mm)
Weight 6.02 lbs (2.73 kg)
Battery 56 Wh battery

Retail system or not, the N56VM has its bases covered on the hardware front. It delivers everything from Bluetooth to Blu-ray, with fancy Bang & Olufsen built-in speakers for good measure. All the USB ports are SuperSpeed-enabled, and in addition to a generous 8GB of system memory, it features GeForce GT 630M graphics with 2GB of DDR3 RAM. Nvidia’s Optimus real-time switching functionality lets the GeForce kick in when it’s needed and kick back the rest of the time, limiting power drain and maximizing battery life.

So, yes, not bad. Not bad at all.

Ready to see how well the N56VM fares under closer scrutiny? Read on.

The display, speakers, and controls

Displays with matte coatings are a rare sight in modern notebooks. Displays with matte coatings that eschew the infernal 1366×768 resolution in favor of something a little nicer—say, 1920×1080—are even harder to come by. The N56VM’s panel is a rare gem indeed, because it ticks both boxes.

This isn’t one of those extremely rare displays that features VA or IPS panel technology, though. As nice as it may be, the N56VM’s display is of the TN variety, and it has the (limited) viewing angles to match:

 
 

To be fair, color shift isn’t particularly bad when you look from the side or down from above.

Since we wanted to check color accuracy, we ran X-Rite’s Eye-One Match v3.6.2 software and used it to calibrate the display. We don’t expect average joes to go around calibrating notebook displays, of course, but the tool spits out some very useful information after the process. In the screenshot below, the graph on the left shows the correction curves required to achieve “correct” colors per the specified gamma and color temperature settings (2.2 and 6500K, respectively). The diagram on the right shows the panel’s color gamut. We specified a luminosity target of 120 cd/m² and attempted to match it as closely as possible using the laptop’s brightness controls.

Bear in mind, by the way, that our N56VM didn’t come pre-loaded with any Asus software. If the company applies custom color profiles to its retail systems via software, then our test machine may not have been using such a profile.

Those are some very decent results. The default calibration is a tad too blue, but not by a whole lot. Compare the curves on the left to those we measured on Asus’ Zenbook UX31—the N56VM requires much less correction, and its curves follow a smoother progression, suggesting more uniformity in how colors are rendered out of the box.

Next, we cranked up the display’s backlight to its maximum setting and measured luminance at nine points along the panel’s surface. Luminance readings are presented both as cd/m² figures, which were produced by the calibration software, and as percentages of the most luminous point we measured. Note that luminance and perceived brightness follow different scales, so the display appears more uniform than the chart below might suggest.

311 cd/m²

(80%)

376 cd/m²

(97%)

384 cd/m²

(99%)

327 cd/m²

(84%)

389 cd/m²

(100%)

371 cd/m²

(95%)

296 cd/m²

(76%)

344 cd/m²

(88%)

318 cd/m²

(82%)

I guess backlight uniformity isn’t so great. The left side of the display is measurably darker than the right side, and the bottom is darker than the top, too. The differences aren’t quite as stark to the naked eye as the diagram above might suggest, though.

To recap: here we have a nice 1080p display with a matte coating and decent color reproduction, which unfortunately suffers from poor backlight uniformity and the kind of color shifting problems typical of TN panels. You might want to shop elsewhere if you’re planning to do serious image editing work, but I think most users should be pleased. This is such a nice step up from the glossy, 1366×768 status quo.

Above the keyboard are the N56VM’s Bang & Olufsen ICEpower speakers. They dwell under sets of little holes drilled through the aluminum, which are (quite elegantly) arranged in fading concentric circles. The speakers sound surprisingly decent by notebooks standards, with passable separation between highs and mids, and plenty of volume headroom. As you’d expect, though, they sound a tad tinny. Also, they make the system’s casing vibrate even at low volumes, which has the potential to get uncomfortable. You’ll probably be best off plugging in a nice pair of headphones.

Asus has taken cues from Apple in making the N56VM’s keyboard: keys poke straight out of the aluminum surface covering the top half of the notebook’s body. The metal seems to give this keyboard a little bit of extra resilience, because I noticed very little flex when typing, and keys had nice, crisp tactile feedback.

  Total keyboard area Alpha keys
  Width Height Area Width Height Rough area
Size 276 mm 103 mm 28,428 mm² 167 mm 52 mm 8,8684 mm²
Versus full size 96% 94% 90% 97% 91% 89%

In terms of size, the keyboard is a wee bit smaller overall than our non-chiclet reference. In fact, the keys feel a tad small even by chiclet standards. Perhaps that’s because Asus needed to leave room for the numeric keypad on the side. As someone who does a lot of data entry in Excel, I can appreciate a good numpad—but I wouldn’t mind some slightly fatter keys, either.

Asus tells us retail N56VM systems will boast keyboard backlighting. That feature wasn’t enabled on our test system, however.

Let’s now talk about the N56VM’s multi-touch touchpad, an Elantech model that looks like something right off a MacBook Pro. Apple does make some truly excellent touchpads—I’d go so far as to call them the best in the industry—so this would be a good thing… if the imitation weren’t only skin-deep. Unfortunately, this Elantech contraption doesn’t have the right friction coefficient; it feels too tacky, and it rattles a little when you tap to click. The software suffers from the same shortcomings we noticed on our pre-release Zenbook UX31 last year. Try to click-and-drag with your thumb positioned too far forward, and the cursor will freeze, thinking you’re trying to perform a two-finger resize. It’s a frustrating design flaw that I haven’t encountered on any other multi-touch touchpad designs, save for the UX31’s.

I’ll give Asus a tentative pass here, since this isn’t a retail sample, and the company ended up fixing the problem on retail Zenbooks. Besides, everything else, from resizing and rotating to scrolling, works just fine. I can’t say I wasn’t disappointed, though.

Connectivity and expansion

Much of the N56VM’s flanks are occupied by a large exhaust vent and a Blu-ray drive, so there aren’t as many ports as one might hope. Asus still took care of the essentials, though:

You’ve got VGA, Ethernet, and HDMI on the left side, a couple of audio jacks (analog headphone/SPDIF and microphone) on the right side, and two USB 3.0 ports on either side. There’s also a 2.5 mm connector on the left that’s apparently designed to accommodate an external subwoofer. Our notebook didn’t ship with such a bass box, but certain other Asus systems do—the N55SF, namely. Perhaps retail N56VM systems will, as well.

Despite the shiny aluminum around the keyboard, the bottom of the notebook is all plastic. Asus offers relatively easy access to the notebook’s internals. One screw opens the compartment where the RAM, Wi-Fi adapter, and hard drive are located. The battery is removable with a single latch.

We complete our tour of the N56VM with a look at its jumbo-sized AC adapter. Yes, it’s big; and yes, it’s heavy. With cables and all, it weighs in at just over 1.5 lbs—heavier than an iPad. Since the laptop tips the scales at about 6 lbs by itself, you’ll be lugging 7.5 lbs of equipment if you plan to take the thing on the road.

Our testing methods

We’ve run a lot of laptops through our test suite, so for the sake of informativeness (and entertainment), we’ve included all the results in the graphs on the following pages. To make things readable, we’ve greyed out the results for everything but the N56VM and its Sandy Bridge-based counterpart, the N53S.

The N53S was kindly provided by Asus for this review. As you’ll in the chart below, it resembles the N56VM in a lot of ways—it, too, has a quad-core, eight-thread CPU with a 45W power envelope. It features the same hard drive model, the same amount of memory, and the same GPU. It even has the same battery specs: 56Wh, 5200 mAh, lithium-ion. The few internal differences (besides the CPU and chipset, of course) are relatively minor. The N56VM’s GeForce GT 630M is clocked at 475MHz, while the N53S’s runs at a speedier 675MHz. Also, the N56VM runs its RAM at 1600MHz instead of 1333MHz, but it has more relaxed memory timings.

Before we go forward, we should talk about the machines we tested in more than one state. The A53T, N82Jv, U33Jc, UX31, Eee PC 1015PN, and T235D were all tested using special battery-saving profiles, and the A53T, N82Jv, U33Jc, UX31, and 1015PN were run in high-performance mode, too.

With the N82Jv, we recorded our Battery Saving results with Asus’ Super Hybrid Engine on, which dropped the CPU clock speed from 2.4GHz to 0.9-1GHz depending on the load. The U33Jc also has a Super Hybrid Engine mode, but we didn’t enable it for testing. The UX31’s battery-saving profile, meanwhile, limits the CPU speed to 60% of its maximum.

On the U33Jc, the high-performance profile included by Asus raises the maximum CPU clock speed from 2.4 to 2.57GHz. On the N82Jv, the same profile leaves the CPU running at default speeds, i.e. up to 2.66GHz when Turbo Boost kicks in. Finally, with the Eee PC, the low-power profile limits the CPU to about 1GHz and disables the Nvidia GPU, while the high-performance profile raises the CPU speed by a whole 25MHz.

The Samsung Series 9 was tested both plugged in and unplugged, since the default battery profile limits the CPU clock speed to 50% of its maximum when the machine is running off the battery.

With the exception of battery life, all tests were run at least three times, and we reported the median of those runs.

System AMD A8-3500M test system Acer Aspire 1810TZ Acer Aspire 1830TZ Acer Aspire One 522 Asus A53T Asus K53E Asus Eee PC 1015PN Asus N56VM Asus N53S Asus N82Jv Asus U33Jc Asus UX31 HP Pavilion dm1z HP ProBook 6460b Intel Core i7-2820QM 17″ review notebook Samsung Series 9 (900X3A) Toshiba Satellite T235D-S1435
Processor AMD A8-3500M APU 1.5GHz Intel Pentium SU4100 1.3GHz Intel Pentium U5400 1.2GHz AMD C-50 1.0GHz AMD A6-3400M 1.4GHz Intel Core i5-2520M 2.5GHz Intel Atom N550 1.5GHz Intel Core i7-3720QM 2.6GHz Intel Core i7-2670QM 2.2GHz Intel Core i5-450M 2.4GHz Intel Core i3-370M 2.4GHz Intel Core i5-2557M 1.7GHz AMD E-350 1.6GHz Intel Core i5-2410M 2.3GHz Intel Core i7-2820QM 2.3GHz Intel Core i5-2537M 1.4GHz AMD Turion II Neo K625 1.5GHz
North bridge AMD A70M FCH Intel GS45 Express Intel HM55 Express AMD Hudson FCH AMD A70M FCH Intel HM67 Express Intel NM10 Intel HM76 Express Intel HM65 Express Intel HM55 Express Intel HM55 Express Intel QS67 AMD Hudson FCH Intel HM65 Intel HM67 Express Intel HM65 Express AMD M880G
South bridge Intel ICH9 AMD SB820
Memory size 4GB 3GB (2 DIMMs) 3GB (2 DIMMs) 1GB (1 DIMM) 6GB (2 DIMMs) 6GB (2 DIMMs) 1GB (1 DIMM) 8GB (2 DIMMs) 8GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 3GB (2 DIMMs) 4GB 4GB (2 DIMMs) 8GB (2 DIMMs) 4GB (2 DIMMs)
Memory type DDR3 SDRAM DDR2 SDRAM at 667MHz DDR3 SDRAM at 800MHz DDR3 SDRAM at 1066MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 667MHz DDR3 SDRAM at 1600MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 1066MHz DDR3 SDRAM at 1066MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 667MHz DDR3 SDRAM at 1600MHz DDR3 SDRAM at 1333MHz DDR3 SDRAM at 800MHz
Memory timings N/A 5-5-5-15 6-6-6-15 N/A 9-9-9-24 9-9-9-24 6-5-5-12 11-11-11-28 9-9-9-24 7-7-7-20 7-7-7-20 9-9-9-24 9-9-9-25 9-9-9-24 11-11-11-30 9-9-9-24 6-6-6-15
Audio IDT codec Realtek codec with 6.0.1.869 drivers Realtek codec with 6.0.1.6043 drivers Conexant codec with 8.41.0.0 drivers Realtek codec with 6.0.1.6373 drivers Realtek codec with 6.0.1.6273 drivers Realtek codec with 6.0.1.6186 drivers Realtek codec with 6.0.1.6537 drivers Realtek codec with 6.0.1.6463 drivers Realtek codec with 6.0.1.6024 drivers Realtek codec with 6.0.1.6029 drivers Realtek codec with 6.0.1.6446 drivers IDT codec with 6.10.6302.0 drivers IDT codec with 6.10.6328.0 drivers Conexant codec with 4.127.0.60 drivers Realtek codec with 6.0.1.6271 drivers Realtek codec with 6.0.1.6072 drivers
Graphics AMD Radeon HD 6620G + AMD Radeon HD 6630M

with Catalyst 8.862 RC1 drivers

Intel GMA 4500MHD with 15.17.11.2202 drivers Intel HD Graphics with 8.15.10.2057 drivers AMD Radeon HD 6250 AMD Radeon HD 6520G + AMD Radeon HD 6650M 1GB

with Catalyst 8.861.0.0 drivers

Intel HD Graphics 3000 with 8.15.10.2321 drivers Intel GMA 3150 with 8.14.10.2117 drivers

Nvidia Ion with 8.17.12.5912 drivers

Intel HD Graphics 4000 with 8.15.10.2696 drivers

GeForce GT 630M with 296.54 drivers

Intel HD Graphics 3000 with 8.15.10.2462 drivers

GeForce GT 630M with 296.54 drivers

Intel HD Graphics with 8.15.10.2189 drivers

Nvidia GeForce 335M with 8.17.12.5896 drivers

Intel HD Graphics with 8.15.10.2119 drivers

Nvidia GeForce 310M with 8.17.12.5721 drivers

Intel HD Graphics 3000 with 8.15.10.2476 drivers UX33 AMD Radeon HD 6310 with 8.821.0.0 drivers Intel HD Graphics 3000 with 8.15.10.2361 drivers Intel HD Graphics 3000 with 8.15.10.2246 drivers Intel HD Graphics 3000 with 8.15.10.2266 drivers AMD Mobility Radeon HD 4225 with 8.723.2.1000 drivers
Hard drive Hitachi Travelstar 7K500 250GB 7,200 RPM Western Digital Scorpio Blue 500GB 5,400-RPM Toshiba MK3265GSX 320GB 5,400 RPM 500GB Seagate Momentus 7,200-RPM Seagate Momentus 640GB 5,400-RPM Toshiba MK2565GSX 250GB 5,400 RPM Western Digital Scorpio Blue 500GB 5,400-RPM Seagate Momentus 750GB 7,200-RPM Seagate Momentus 750GB 7,200-RPM Seagate Momentus 7200.4 500GB 7,200-RPM Seagate Momentus 5400.6 500GB 5,400-RPM Adata XM11 128GB solid-state drive Hitachi Travelstar 7K500 320GB 7,200-RPM hard drive Hitachi Travelstar 7K500 320GB 7,200 RPM Intel X25-M G2 160GB solid-state drive 256GB Samsung MZ8PA256HMDR solid-state drive Toshiba MK3265GSX 320GB 5,400 RPM
Operating system Windows 7 Home Premium x64 Windows 7 Home Premium x64 Windows 7 Home Premium x64 Windows 7 Starter x86 Windows 7 Home Premium x64 Windows 7 Ultimate x64 Windows 7 Starter x86 Windows 7 Home Premium x64 Windows 7 Professional x64 Windows 7 Home Premium x64 Windows 7 Home Premium x64 Windows 7 Home Premium x64 Windows 7 Home Premium x64 SP1 Windows 7 Professional x64 Windows 7 Ultimate x64 Windows 7 Professional x64 Windows 7 Home Premium x64

We used the following versions of our test applications:

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

Application performance
SunSpider JavaScript benchmark

The SunSpider benchmark is often used to test the JavaScript performance of different web browsers. Here, we’ve pressed it into service to see how different notebooks perform with the same version of Firefox (3.6.9).

We realize this is an old version of Firefox. However, the point of this benchmark is to compare web browsing performance across multiple systems, and we can do a good job of that now that we’ve accumulated a reasonable data set.

Ivy’s off to a decent start in SunSpider, trouncing the Sandy Bridge-based Asus N53S and most other configs. The only exceptions are three Asus systems at the top; they all have slower CPUs, but they were all run with Asus’ “High Performance” battery profiles enabled. We suspect those battery profiles game the results somehow—at least on the UX31, the High Performance setting prevents the processor from lowering its clock speed to save power. SunSpider’s workload is fairly staccato, so clock speeds may be varying some on the other configs.

7-Zip

7-Zip has a handy built-in benchmark that lets us test both compression and decompression performance.

x264 video encoding

The x264 video encoding benchmark doesn’t call on the GPU to accelerate the encoding process, leaving us with a good look at how the various mobile CPUs stack up.

Our data compression and video encoding results are less surprising. Ivy is hands-down faster than the competition, with the older Core i7-2820QM and Core i7-2670QM—both Sandy Bridge variants—in close pursuit.

TrueCrypt

This latest version of TrueCrypt makes use of the AES-NI instructions built into Intel’s Westmere and Sandy Bridge CPUs.

The Asus N53S’s Core i7-2670QM chip lacks the hardware AES acceleration present in the i7-2820QM and our Ivy Bridge CPU, which makes a big difference in TrueCrypt. As expected, Ivy leads the pack again by a comfortable margin.

Gaming
Call of Duty 4: Modern Warfare

We tested the original Modern Warfare by running a custom timedemo at 1366×768 with everything cranked up except for vsync, antialiasing, and anisotropic filtering, which were all left disabled. With the Eee PC and Aspire One 522, we opted for respective native resolutions of 1024×600 and 1280×720 instead of 1366×768.

Both of the graphics processors in the A53T’s Dual Graphics team, the Llano IGP and the discrete Radeon GPU, were enabled throughout our gaming tests. However, since Dual Graphics only supports games that use DirectX 10 and 11, some of the games we tested couldn’t make use of both GPUs.

Far Cry 2

In Far Cry 2, we selected the “Action” scene from the game’s built-in benchmark and ran it in two configurations: first at 1366×768 in DirectX 10 mode with detail cranked up, and then at that same resolution in DX9 mode with the lowest detail preset. Vsync and antialiasing were left disabled in both cases. Again, the Eee PC and Aspire One 522 were run at 1024×600 and 1280×720, respectively.

We tested integrated and discrete graphics performance, and in both cases, our Ivy system performed better than its Sandy counterpart—even though, as we noted earlier, the Ivy machine has a slower GeForce GPU. I suppose the faster CPU makes up for the slightly slower GPU.

The differences between the two systems’ IGPs are particularly stark in Far Cry 2 with the detail turned up, where Ivy’s HD Graphics 4000 offers nearly double the performance of Sandy’s HD Graphics 3000. Neither solution yields frame rates that could be considered playable, though, and they’re much closer overall in our other tests.

Off the beaten path

What happens when we take the N56VM for a spin in some newer games? To find out, we installed Battlefield 3, The Elder Scrolls V: Skyrim, and Bulletstorm, and we tried to find playable settings for both the IGP and the discete GeForce GPU. Then, we played each game while keeping an eye on frame rates using Fraps. In the interest of maximizing in-game eye candy and performance, we lowered the resolution from the default 1920×1080 to 1366×768. It should come as no surprise that 1080p gaming isn’t a sensible proposition for IGPs in modern, graphically intensive titles.

In Battlefield 3’s Kaffarov mission, the N56VM’s discrete GPU handled the game smoothly at 1366×768 using the “low” detail preset. Frame rates were in the 40-60 FPS range, with occasional dips below 40 during heavy action. We tried the “medium,” preset, but that took frame rates down to the mid-30s, which felt a little sluggish considering the fast pace of the action.

Intel’s HD Graphics 4000 also made the game playable at 1366×768 using the “low” preset, but it didn’t manage frame rates above 25-35 FPS—again, too sluggish for this type of game. Getting a really smooth experience would probably involve lowering the resolution further, which wouldn’t look too great. (1366×768 is already a little fuzzy on the N56VM’s 1080p display.)

Our walk through Whiterun in Skyrim was much smoother with both graphics solutions. At 1366×768 using the “medium” preset, the GeForce managed around 50 FPS on average, with occasional drops to 30 FPS. Getting the same level of smoothness out of the Intel IGP required a trip back to the graphics settings and a drop to the “low” preset. Once we did that, frame rates were back in the 30-50 FPS range. The game was smooth and definitely enjoyable, even if it didn’t look particularly pretty.

Our experience with Bulletstorm‘s “Hideout” echo was pretty close to what we saw in Skryim. At 1366×768 with the “medium” preset and no antialiasing, the GeForce seemed to average around 50 FPS with occasional dips to 40 or so during heavy action. We had to step down to the “low” preset to get similar performance out of the Intel IGP. Dips during firefights were closer to 30 FPS there, though.

All in all, kudos to Intel. Battlefield 3 may be somewhat off-limits to the IGP, but still, the N56VM would be a passable gaming machine even without its discrete GPU. Not only was performance solid, but we also didn’t encounter any visual artifacts. Serious gamers will still want some sort of GeForce or Radeon graphics, though, for obvious reasons.

Video playback

Video decoding performance was tested using the Iron Man 2 trailer in multiple formats. Windows Media Player was used for the H.264 QuickTime clips, while Firefox hosted the windowed YouTube test. Windows 7’s Performance Monitor was used to log CPU utilization for the duration of the trailer, but we played each video three times and grabbed the lowest numbers for each. This method should provide representative numbers largely untarnished by CPU utilization from background processes.

  CPU utilization Result
Iron Man 2 H.264 480p 0-2.7% Perfect
Iron Man 2 H.264 720p 0-2.9% Perfect
Iron Man 2 H.264 1080p 0-3.5% Perfect
Iron Man 2 YouTube 720p windowed

(Flash 11.2)

4.1-9.7% Choppy

Ivy performs pretty much as you’d expect when confronted with high-def H.264 videos—flawlessly. Windowed Flash playback is another story. We tested Flash 11.2 in Chrome and were dismayed to find that playback wasn’t particularly smooth despite the low CPU utilization. Switching to full-screen mode seemed to resolve the problem entirely, though. I expect we can chalk up that unexpected kink either to Flash itself or to Intel’s drivers.

Battery life

To gauge run times, we conditioned our systems’ batteries by cycling them two times. For the web browsing test, we used TR Browserbench 1.0, which consists of a static version of the TR home page that cycles through different text content, Flash ads, and images, all the while refreshing every 45 seconds. Then, we tested video playback in Windows Media Player by looping an episode of CSI: New York encoded with H.264 at 480p resolution (straight from an HTPC). Wi-Fi and Bluetooth were enabled for the web browsing test and disabled for movie playback.

We attempted to keep the display brightness consistent across all systems, choosing levels corresponding to a readable brightness in indoor lighting. A 40% brightness setting was used on the Acer 1810TZ, Asus A53T, Asus K53E, Asus N82Jv, Asus N53S, Eee PC 1015PN (in its “Super Performance” mode), HP Pavilion dm1z, Toshiba Satellite T235D. We used a 50% setting on the Aspire One 522, Eee PC 1015PN in “Battery Saving” mode (since disabling the Nvidia GPU seemed to reduce brightness), as well as on the U33Jc. Because of their dim, matte displays, the HP ProBook 6460b and AMD A8-3500M systems were tested at 70% brightness settings. Conversely, because of their high display luminosities, the Zenbook UX31 was tested at a 25% brightness level, and both the N56VM and the Series 9 were tested at 30%. The Series 9’s adaptive brightness setting was disabled, as well.

Well, that’s a little disappointing. With a similar supporting cast of hardware, Ivy and Sandy offer similar battery run times. The former does slightly better in video playback, but not by a whole lot—and in any case, the results aren’t anything to write home about. Folks seeking all-day battery life are going to want to wait for Ivy Bridge’s dual-core incarnation.

Surface temperatures

We measured temperatures using an infrared thermometer at a distance of 1″ from the system after it had been running TR Browserbench 1.0 for about an hour.

27°C

80°F

 
25°C

78°F

 
28°C

82°F

 
26°C

79°F

 
27°C

80°F

26°C

78°F

 
26°C

79°F

 
29°C

84°F

 
27°C

80°F

 
26°C

80°F

The NV53M’s Ivy Bridge CPU doesn’t dissipate too much heat, and Asus’s chassis appears to handle that heat dissipation well. The N56VM felt barely lukewarm even after running our browsing test for just over an hour. When we left Skyrim running for an hour, temperatures around the keyboard and palm rest climbed only to 30-35°C—still barely warm to the touch.

Conclusions

Let’s talk about Ivy Bridge first and the N56VM second.

Ivy Bridge delivers greater performance per watt than the previous generation, and it shows in our performance results. On top of that, the new HD Graphics 4000 IGP is a substantial improvement over Intel’s prior attempts at game-worthy integrated graphics. It’s fast enough to handle modern games like Skyrim and Bulletstorm at lower detail settings, which is excellent. An entry-level discrete GPU is always going to be faster, but at least gaming without one is a very reasonable option.

Our battery life results were a little more surprising. The new silicon may pack a stronger punch than Sandy Bridge within those thermal constraints, but it doesn’t appear to offer tangible benefits for battery run times within the same power envelope. Then again, we were looking at a laptop with a 45W Ivy Bridge variant. If it’s longer battery life you seek, you’ll want to look for a processor with a 35W or lower TDP rating.

As before, I’d say we’re more excited about upcoming ultrabooks with 17W, dual-core Ivy Bridge chips than anything. Big honkin’ notebooks have some appeal if you absolutely need a portable system as your primary computer. In that case, it’s hard to beat getting a full-featured Ivy Bridge CPU, 8GB of RAM, a discrete GeForce, a 1080p display, Blu-ray, and USB 3.0 in something that fits in your backpack. For the average laptop user, though, a more slender, lightweight, and less feature-packed system might be more enticing.

Speaking of which, what do we make of the N56VM? Well, considering the system’s size and weight, Asus has done a good job of achieving some level of elegance and design flair. The matte 1080p display is a very nice touch, as is the snappy keyboard and the liberal application of aluminum panels. Our only issue is with the touchpad, which doesn’t feel ready for prime time. Again, though, we weren’t looking at an actual retail system, so the issues we encountered with clicking and dragging may be fixed when the N56VM hits stores. We’ll have to withhold our final verdict until then.

Comments closed
    • steaphaniez339
    • 7 years ago

    I can’t believe this laptop is outperforming the one reviewed here..

    goo.gl/dDekA

    How does that even make sense?

    • Palek
    • 7 years ago

    Cyril,

    There are a couple spots in the article where you wrote N53VM instead of N56VM.

    Page 4
    [quote<]The [u<]N53VM[/u<]'s GeForce GT 630M is clocked at 475MHz, while the N53S's runs at a speedier 675MHz. Also, the [u<]N53VM[/u<] runs its RAM at 1600MHz instead of 1333MHz, but it has more relaxed memory timings.[/quote<] Page 7 [quote<]The [u<]N53VM[/u<] felt barely lukewarm even after running our browsing test for just over an hour.[/quote<]

      • Cyril
      • 7 years ago

      Fixed. Thanks for the heads up.

    • ish718
    • 7 years ago

    For teh folks who want to Decode/Encode videos, compress and decompress files. Not to mention, run CPU hungry Flash apps…

    • WhatMeWorry
    • 7 years ago

    Tech Report has just lost their fat prom girl readership.

    • yogibbear
    • 7 years ago

    ANAND has UX21A preview up…. I want that one!

    • ShadowTiger
    • 7 years ago

    My guess is that they shrank the battery to cut cost (and weight) instead of giving you a better battery life number than other laptops (despite a gain in performance/watt from using Ivy Bridge).

    Overall I think laptops are improving which is nice to see. Thanks for the write up!

    • link626
    • 7 years ago

    ivy bridge igp is fcking nuts !

    • NeelyCam
    • 7 years ago

    [quote<]"We tested Flash 11.2 in Chrome and were dismayed to find that playback wasn't particularly smooth despite the low CPU utilization. Switching to full-screen mode seemed to resolve the problem entirely, though. I expect we can chalk up that unexpected kink either to Flash itself or to Intel's drivers."[/quote<] I've seen the same thing with "GPU-easy" games like Audiosurf and Civ4. Fullscreen the games look fine, windowed they get choppy (this with Clarkdale and Sandy Bridge) Maybe Intel drivers are cr*p after all...

      • Goty
      • 7 years ago

      I don’t have any problem running windowed flash at 720p on my SB integrated graphics on my laptop, though that is after being passed through AMD’s automatic GPU switching layer.

    • donkeycrock
    • 7 years ago

    Can we start adding Diablo 3 to the bench mark suite for laptops.

    As that is the game mostly likely played on a laptop.

      • grantmeaname
      • 7 years ago

      They’d have to throw out all the test results they have already. When the review is written, the review laptop goes away. So to test Diablo III they’d have to do it only on laptops coming in from here forward, plus one or two that the staff members individually own. Seems like a bum deal for one game.

      Is Diablo III even that graphically intensive? Couldn’t you estimate it from the Skyrim results or something?

        • UberGerbil
        • 7 years ago

        They do add new tests from time to time, and they don’t necessarily go back and retest older hardware when that happens (in fact, they rarely do). The test suites are locked for all time.

        That said, it’s certainly valid to question / investigate Diablo’s applicability (though I’m not sure about “that is the game most likely played on a laptop”). Does it push any part of the hardware in new and interesting ways, or are its demands approximated by one of the other tests already in the suite? More to the point, is it possible to use it as a benchmark at all (ie with reproducible results)?

    • flip-mode
    • 7 years ago

    Matte 1920×1080. Yes, Asus, you did this right.

    • pogsnet
    • 7 years ago
      • Goty
      • 7 years ago

      It probably has something to do with not having any real Trinity products to test just yet, as well as this review’s proximity to the Trinity review.

        • dpaus
        • 7 years ago

        Hunh?? Cyril did this review, and Cyril co-wrote the Trinity review, which included ‘real Trinity’ hardware. Why not compare them, or, at least, why not include the Trinity results in this article??

    • derFunkenstein
    • 7 years ago

    So between the two tables on page one, I’m confused:

    CPU table says the 3720QM runs at 2.6GHz with a peak 3.6GHz turbo.
    Notebook specs table says 3720QM at 2.3GHz.

    Did ASUS underclock the chip?

      • Cyril
      • 7 years ago

      Ah, no, that was a typo. The 3720QM does run at 2.6GHz.

    • Chrispy_
    • 7 years ago

    Yay \o/
    A reasonably priced mobile workstation without a turd of a screen.

    It’d even look nice and minimalist from the front too, if it didn’t have that pesky Windows sticker like a cold-sore on it’s wrist rest.

      • superjawes
      • 7 years ago

      I’ve been looking for something to replace my old college laptop…and yeah, this one looks like it’d be a pretty good choice.

      As for sticker, I think those are there to subsidize some of the cost (Microsoft cuts the price on a Windows license for system builders). But that’s not to say that you couldn’t remove that once you get it home…

        • UberGerbil
        • 7 years ago

        Yeah, I’m not sure why people complain about something that both lowers their costs and is so easy to remove.

          • dpaus
          • 7 years ago

          [quote<]and is so easy to remove[/quote<] ?? Most of the ones I've tried to take off computer products required a hair dryer at full blast combined with lots of patience and lots of acetone. But then, I'm one of those who's still complaining that all my fresh fruits and vegetables are covered in little stickers. I hate stickers. I especially hate the 'security' stickers that are virtually impossible to remove from something I just bought as a gift for someone. I usually ask the store staff to remove those stickers. [/OldManRant]

            • cygnus1
            • 7 years ago

            [quote<] Most of the ones I've tried to take off computer products required a hair dryer at full blast combined with lots of patience and lots of acetone.[/quote<] wtf?? i've never had them be that tough. hell, half of them fall of with a month if i don't try to take them off first. and when i do take them off something new, i yank the sticker off and just rubbing the spot with a paper towel to get any residue. this had included dell, hp, acer and asus... and the stickers look the same on all the other brands i've seen

            • dpaus
            • 7 years ago

            Hmm, just tried a not-statistically-valid test on a 2-year-old Dell laptop. The ‘Windows 7’ sticker came off easily and left no residue. The ‘Intel Inside’ sticker and the ‘Energy Star’ sticker both required serious fingernail action, and both left a nasty residue behind that’s going to require acetone/varsol/solvent of some sort – certainly not paper towel. But no hairdryer was harmed in this experiment.

            • sweatshopking
            • 7 years ago

            the windows stickers are easy, it’s the other ones that aren’t.

            • grantmeaname
            • 7 years ago

            Goo gone does a great job of getting it off. I imagine lighter fluid would too.

            • Chrispy_
            • 7 years ago

            My biggest problem with Goo Gone and solvents is that on plastic wrist rests as found on most laptops, they remove the glue AND some paint/plastic/texture.

            • OneArmedScissor
            • 7 years ago

            It has “melted” plastic I’ve put it on before. That’s not something I’d use for anything but removing labels on cheap file organizers.

            • indeego
            • 7 years ago

            Just use the sticker itself to remove the residue. I do this for every new machine I place where I work/friends/family. Works 100%.

            • SPOOFE
            • 7 years ago

            Hooray for narcissistic adhesives!

            • Washer
            • 7 years ago

            How about just rubbing it a little with your finger? I just removed the stickers from a new T420. Energy Saver, Intel and Windows 7. It took all of a minute.

    • Arag0n
    • 7 years ago

    Am I the only one that sees the A6-3400M beating or equaling the new Ivy Bridge 4000HD GPU?

      • cosminmcm
      • 7 years ago

      Yes, the others are probably seeing the 300-400% better processor performance of Ivy.

        • Alchemist07
        • 7 years ago

        the performance of the A6-3400M is good enough for every day tasks

          • cosminmcm
          • 7 years ago

          Of course, the “good enough” argument. So is a ULV i5/i7. This thing is just much more powerful. Is it bad?

            • willyolio
            • 7 years ago

            no, but depending on the usage, the slower GPU [i<]is[/i<] a bottleneck, whereas the slower CPU [i<]isn't[/i<] limiting the possible uses for this machine nearly as much.

          • UberGerbil
          • 7 years ago

          For “every day tasks” you wouldn’t be buying something with a discrete GPU anyway, so the comparison is a bit flawed.

        • Arag0n
        • 7 years ago

        I think you miss the point, the A6-3400M has a much weaker GPU than the A8 series (320 vs 400, and 400Mhz vs 444Mhz). My point is that if the 4000HD was almost 20% slower than Trinity, and Trinity is 50% faster than Llano…. how can it be possible? Seems that Intel definitely doesn’t use the same setup for the 4000HD despite using the same name. So people claiming that ULV Ivy Bridge will have the same GPU performance than the desktop version seems to be wrong.

          • cosminmcm
          • 7 years ago

          You didn’t say any of these in your first post. Either way, there is no point in comparing the integrated graphics on this laptop in games because it has a discrete card that will be used in that case (I think it is thrown in there just for information).
          The ULV won’t have the same performance, because with its limited TDP it will be harder to turbo to the maximum video frequency, and either way the huge difference in processing power will be felt at 1366×768.
          Trinity’s situation at 17W will be much worse though compared to the desktop variant, and even to Ivy.
          But all that I wrote here has nothing to do with the article, and you can’t compare an A6 to an i7 just in graphics performance, as much as AMD would love that.

            • Arag0n
            • 7 years ago

            Seems so, but I remember many people during trinity review claiming that performance of ULV will be lower cuz lower core number and frequency so intel would win in the ultrabook segment in graphics. This seems to prove they will be wrong.

            • Alchemist07
            • 7 years ago

            think you can compare an A6 to an i7, since there will be LOTS of laptops selling an i7 without discrete graphics

            • cosminmcm
            • 7 years ago

            The only thing where the A6 is close is gaming, but who will buy an i7 to play games at 1366×768?
            And because A6 is so close, an A8 is even better than an i7, right?
            Oh, in gaming. What about rendering? Photo editing? Archiving? Video encoding? But I guess you don’t do these things on a laptop, right? The quad i7 makes them all possible at very good speed (many times faster than the competition), but any processor is good enough now, because Rory Read said so…

            • Alchemist07
            • 7 years ago

            im sure not very many people render, archive or video encode, and very few people photo edit. There are far more people who game.

          • maroon1
          • 7 years ago

          “Trinity is 50% faster than Llano”

          What ?

          Trinity GPU is only 20% faster on average than Llano
          [url<]http://www.anandtech.com/show/5831/amd-trinity-review-a10-4600m-a-new-hope/6[/url<]

        • Lans
        • 7 years ago

        Well, to complicate things more:

        [url<]http://www.anandtech.com/show/5831/amd-trinity-review-a10-4600m-a-new-hope/7[/url<] While Anandtech "only" shows N56V (same i7-3720QM) being about 190% faster than A8-3500M (instead of out performing a A8-3500M by 360% more MIPS in 7-Zip compression), and while A10-4600M (Trinity) doesn't fare that much better (N56V still ~150% faster), things look much better with heterogeneous compute (unfortunately there is no 7-Zip heterogeneous results). Anyhow, it seems like going forward, the choice will be even less clear cut.

    • MadManOriginal
    • 7 years ago

    duh…didn’t read down all the way on the display page. Soo…glad to see TR is finally doing objective display measurements! Do you think you could include contrast ratio for easy reference so readers don’t have to do the math?

    p.s. First post, oooh yeah

Pin It on Pinterest

Share This