Z77 motherboards from Asus, Gigabyte, and MSI

Intel’s next-generation processor may be the worst-kept secret in the industry right now. Intel has disclosed quite a bit about Ivy Bridge, and other details have already leaked onto the web, including specifications, benchmark results, and Intel’s own marketing materials—yet I’ve been asked to refrain from mentioning her by name. Instead, she’s to be referred to only as Intel’s “3rd generation Core processor.” And she’s not ready yet.

Intel has confirmed that Ivy Bridge’s arrival has been delayed to iron out kinks in the 22-nm fabrication process used to make the chip. We may not see notebooks featuring Intel’s new hotness until June. Desktop versions of the CPU should be available before then, though, and the 7-series motherboards designed to host them are already for sale. Three of these boards have been toiling away in the Benchmarking Sweatshop for the past week. Today, we’re going to take a closer look at what they have in store.

Thanks to their LGA1155 sockets, these 7-series boards are compatible with existing Sandy Bridge CPUs. We’ve been able to run them through a full range of tests, including some new additions to our peripheral performance suite. With Ivy out of the picture, we have more time to focus her accompanying platform hub and its integrated USB 3.0 controller. We also have more time to spend on the unique attributes of the latest offerings from the big three motherboard makers. Join us as we lift the lid on the Z77 Express chipset and explore the Asus P8Z77-V, Gigabyte Z77X-UD3H, and MSI Z77A-GD65 motherboards.

Introducing the Z77 Express

Intel has a full range of 7-series platform hubs ready for its upcoming 22-nm processor. PC enthusiasts will want to pay the most attention to the Z77 Express, which effectively replaces the Z68 from the 6-series lineup. This flagship platform supports CPU overclocking, SSD caching, QuickSync transcoding, and multi-GPU-friendly PCI Express lane configurations in addition to offering a full array of display outputs for integrated graphics and Lucid’s GPU virtualization software. Unlike the Z68 Express, which didn’t hit the market until four months after the first wave of Sandy Bridge motherboards, the Z77 has arrived alongside the rest of the 7-series family.

For the most part, the Z77 is very similar to its Z68 predecessor. Its integrated PCI Express 2.0, Serial ATA, Ethernet, and USB 2.0 controllers appear to have been pulled directly from the past generation. Intel says these components will all “behave and perform” similarly to those found in 6-series platforms. There is, however, a new addition: USB 3.0 support.

In addition to a 10-port USB 2.0 controller, the Z77 has a second USB controller that ripples with four SuperSpeed ports. Finally! Intel is pretty much the last major player to adopt USB 3.0 (apart from Apple, of course). Enthusiast-oriented motherboards have sported SuperSpeed ports for years thanks to third-party controllers, and AMD’s Llano platform added native support last summer. Unlike some next-gen interfaces, there’s a real need for USB 3.0. The second-gen USB spec tops out around 37MB/s under ideal conditions, which is far too slow for modern portable storage devices.

The Z77 block diagram. Source: Intel

Intel’s implementation supports the USB Attached SCSI protocol, otherwise known as USAP. This protocol replaces the Bulk-Only Transport (BOT) approach that was designed for USB 1.1 and has persisted ever since. USAP adds command queuing to deal with multiple concurrent I/O requests, and it offers much lower overhead than the BOT protocol. Those attributes should allow storage devices to make the most of the 5Gbps of bandwidth available in the SuperSpeed USB specification.

We’ll take an in-depth look at the Z77’s USB 3.0 performance in a moment. For now, you should know that USAP support requires compatible devices, which are relatively rare. We hear that Windows 8 will offer native USAP support, so that should encourage device makers to get on board. In the meantime, the Intel controller is backward-compatible with older USB devices. It’s been blessed by the USB-IF governing body and by Microsoft’s WHQL department.

The rest of the Z77’s block diagram is mostly old news, save for next-gen CPU features we aren’t supposed to discuss (Ivy Bridge will support PCIe 3.0, for instance) and a few “responsiveness technologies” largely inherited from Intel’s mobile division. Rapid Start and Smart Connect fall under that umbrella. The former promises faster system standby and resume times for SSD users, while the latter will periodically wake a sleeping system to synchronize with email and social media services.

While it’s tempting to chide Intel for not adding more 6Gbps SATA ports to the Z77, I can understand why the company might be playing it safe with the new platform’s primary storage controller. The first batch of Sandy Bridge chipsets was famously sunk by a Serial ATA bug so serious Intel had to halt shipments and re-spin the silicon. Then, this fall, the high-end X79 platform arrived minus the expanded 6Gbps SATA and SAS connectivity many were expecting. The extra storage logic was still there, but Intel couldn’t get it working properly in time for Sandy Bridge-E’s November debut, so it was disabled in hardware.

The fact is that even most power users won’t need more than a couple of 6Gbps SATA ports. (Just because your mechanical hard drive has a 6Gbps interface doesn’t mean it needs the extra bandwidth.) Besides, the Z77 and its accompanying CPU have plenty of PCI Express connectivity for more exotic SSDs.

Pick your poison

To get a sense of how the Z77 Express has been implemented by motherboard makers, we’ve rounded up three models from Asus, Gigabyte, and MSI. Before we dive into more detail with each board, here’s a quick look at their particulars:

Asus P8Z77-V Gigabyte Z77X-UD3H MSI Z77A-GD65
DIMM slots 4 DDR3 4 DDR3 4 DDR3
Expansion slots 2 PCIe 3.0 x16

1 PCIe 2.0 x16 (x4)

2 PCIe 2.0 x1

2 PCI

2 PCIe 3.0 x16

1 PCIe 2.0 x16 (x4)

3 PCIe 2.0 x1

1 PCI

2 PCIe 3.0 x16

1 PCIe 3.0 x16 (x4)

4 PCIe 2.0 x1

Gigabit Ethernet Intel 82579V Atheros AR8151 Intel 82579V
Wireless Atheros AR9485 NA NA
Auxiliary USB 3.0 ASMedia ASM1042 VIA VL800 NA
Auxiliary SATA ASMedia ASM1061 Marvell 88SE9172 ASMedia ASM1061
Audio Realtek ALC892 VIA VT2021 Realtek ALC898
FireWire NA NA VIA VT6315N
Warranty Three years Three years Three years
Price $210 $170 $190

Each board has a slightly different slot layout. Only the MSI Z77A-GD65 supports FireWire (and then only via an internal connector), and the Asus P8Z77-V is the lone example with integrated Wi-Fi. All of the boards have at least one auxiliary storage controller, but that role is filled by a number of different chips. We also see some variety on the audio and Gigabit Ethernet fronts.

Motherboard makers have become better at differentiating their products in ways that can’t be easily distilled down to a specifications table. Let’s take a closer look at each of the boards to see what makes them unique.

Asus’ P8Z77-V motherboard

We took an early look at a bunch of Asus’ 7-series motherboards last month. Now, it’s time for a deeper analysis of one model in particular: the P8Z77-V. The lack of a suffix denotes the fact that this is Asus’ “standard” Z77 model. It doesn’t have all the extra goodies included in the pricier Pro and Deluxe models, but it’s not missing key ingredients left off the budget LE and LX members of the family. To be fair, though, the P8Z77-V is selling for $210 online, making it rather pricey for a standard offering.

The P8Z77-V looks not unlike Asus’ 6-series motherboards. The color palette hasn’t changed, but the heatsinks have a new aesthetic treatment. Asus has traded the curved lines of the old heatsinks for a jagged pattern that reminds me a little bit of Winamp’s spectrum analyzer. I like the look of the board, whose white and light-blue accents at least deviate a little from the black-and-blue motif that has blanketed the motherboard landscape.

Also popular among enthusiast-oriented motherboards is digital power delivery circuitry, which the P8Z77-V has not only for the CPU and system memory, but also for Intel’s processor-based integrated graphics. Four power phases are dedicated to the IGP, eight are reserved for the CPU, and two cover the DRAM.

While perhaps not as sexy as digital VRMs, Asus’ 7-series motherboards feature a new trace layout for their DIMM slots. This arrangement allows individual DIMMs to be accessed in parallel rather than the serial approach typically favored by motherboards. The result, Asus says, is more memory overclocking headroom with multiple modules installed. Don’t expect an increase in memory bandwidth or a reduction in latency, though.

Going to be using memory with taller heat spreaders? Be sure to check your cooler against the P8Z77-V’s socket clearances. The DIMM slots are about as close as they are on the other boards, but there might not be enough of a gap for some aftermarket coolers. Overall, Asus does a good job of keeping the socket area free of obstructions; the VRM heatsinks do crowd the socket on two sides, but they’re short enough to avoid conflicting with oversized coolers.

The socket sits just to the north of a stack of seven expansion slots. Perhaps as a concession to owners of its Xonar DG sound card, Asus provides a pair of old-school PCI slots, one of which will be obscured by double-wide graphics configurations. The top two x16 slots are linked directly to the CPU and are compliant with the PCI Express 3.0 specification (as long as a PCIe 3.0-capable CPU is installed); when both slots are occupied, each one gets eight lanes of bandwidth. The rest of the PCIe slots conform to the gen-two standard and hang off the Z77 platform hub. That chip has only eight lanes of PCIe connectivity, so the bottom x16 slot is a little starved for bandwidth. It operates in x1 mode by default and can be configured with four lanes of bandwidth if you’re willing to give up the x1 slots and the auxiliary SATA controller.

The SATA ports associated with that ASMedia controller are colored dark blue in the picture above. Even without them, the P8Z77-V has dual 6Gbps SATA ports and four 3Gbps SATA ports linked to the Z77. Asus has also cooked up a little something special just above the front-panel connector: a Thunderbolt header. This header is to be used in conjunction with a PCIe-based Thunderbolt card that Asus expects to sell for about $40. The header is necessary because current Thunderbolt chips need to see the motherboard’s display outputs, something they can’t do over PCI Express, Asus tells us.

Another surprise pops up in the P8Z77-V’s port cluster. This time, it’s an 802.11n Wi-Fi module that complements the board’s Intel Gigabit Ethernet controller. Asus has cooked up some nifty software to go with the wireless controller, but the universal Bluetooth support from its 6-series motherboards has been dropped. I wouldn’t be surprised if users preferred Wi-Fi over Bluetooth. Still, it would have been nice for Asus to include both.

Half of the USB 3.0 ports at the rear (the ones under the Gigabit Ethernet jack) are connected to the Z77 Express, while the others stem from a two-port ASMedia chip. The Z77’s remaining USB 3.0 ports are routed to an internal header meant for front-panel connectors. Asus has special software designed to accelerate the performance of both USB controllers, and we’ll take a closer look at how well it works in a moment.

Before then, I have to give Asus a shout-out for equipping the P8Z77-V with robust integrated audio. Although the Realtek codec is standard fare, it’s paired with DTS Connect and UltraPC II software. UltraPC provides virtualization mojo, while DTS connect allows folks to encode multi-channel digital audio on the fly. If you have a compatible receiver, you can get pristine surround sound in both games and movies without resorting to a sound card or analog outputs.

Asus’ excellent motherboard firmware needs no introduction. The interface remains largely unchanged, and it’s still as responsive and as slick as ever.

Unfortunately, the firmware will still take liberties with your CPU. The “Asus MultiCore Enhancement” mode kicks in automatically if you change something like the memory multiplier. This feature tweaks the CPU’s Turbo behavior, applying the single-core multiplier to all load levels, including when all four cores are occupied. That’s overclocking, according to Intel, and it shouldn’t be done without the user’s knowledge and consent. Asus doesn’t ask permission to change this behavior, although it does make a token effort to give the user a heads up by changing the all-core Turbo speed displayed in the firmware interface. Proper Turbo behavior can be restored by disabling the MultiCore Enhancement, which is better than the old workaround, which required setting the correct per-core Turbo multipliers manually.

Changing the system’s memory speed shouldn’t overclock the CPU. Indeed, altering one system setting should never affect another, especially one that’s not in any way related. We’ve expressed our reservations about this behavior to Asus countless times since noticing it with 6-series motherboards, and the company continues to defend the practice stubbornly. It’s becoming increasingly difficult to view this practice as anything other than a thinly veiled attempt to inflate benchmark scores artificially.

With that off my chest, we can get back to the firmware, with is otherwise excellent, right down to how it can be updated. The P8V77-V includes a BIOS Flashback feature that allows updates to be applied to the board with only a USB stick and a power supply—no processor or DRAM required.

Like the other boards, the P8V77-V’s firmware is loaded with overclocking and tweaking options. It also features an excellent collection of temperature-based fan speed controls for the CPU and all four system fans. Users can change the minimum and maximum temperatures and speeds associated with each fan header. The CPU header’s speed control only works with four-pin PWM fans, but that’s really the only limitation.

Fire up Windows, and there’s even more fan control goodness. Asus’ Fan Xpert 2 application allows the user to define multiple points along the speed profile associated with each fan. Fans can be given distinct names and classified based on their position in the system. There’s also an auto-tuning mechanism that ramps fans from zero to full blast to determine their exact rotational speed at each step along the way. No other motherboard we’ve seen comes close to the level of fan control offered by Asus’ firmware and software tag team.

The P8V77-V comes with Windows-based overclocking software, too, and that’s just the tip of the iceberg. Asus has bolstered its AI Suite software substantially over the past couple years, adding extensive power regulation controls and, most recently, a networking component with bandwidth prioritization and other goodies. I tend to avoid the software that comes bundled with most motherboards, but AI Suite is worth installing.

Gigabyte’s Z77X-UD3H motherboard

Like Asus, Gigabyte has a dizzying selection of Z77 boards. And, like the P8Z77-V on the prior page, the Z77X-UD3H is sort of the standard model. Gigabyte’s approach is quite different, a fact that’s illustrated in not only the hardware, but also the firmware and software that comes with the board. Alas, it seems Gigabyte couldn’t resist sticking to the same tired palette being overused by every other motherboard maker for their enthusiast-oriented products.

To Gigabyte’s credit, the UD3H is a black heatsink swap away from the murdered-out look that seems to be popular with kids these days. If blue is a must, it should at least match the turquoisey shade that defined the company’s entire motherboard lineup for so many years. Perhaps that would help the UD3H from looking so generic. The chunky, bright red power button is all kinds of awesome, though.

The UD3H gets by with a single heatsink for its voltage regulation circuitry, which is all-digital, of course. Digital VRMs are the new solid-state capacitors. Nine power phases feed the CPU, while two keep the memory juiced.

With such a barren socket area, there’s loads of room for aftermarket coolers that bulge out like finned-aluminum muffin tops. The lone heatsink is only about an inch tall, and everything else in the region stays low to the board.

The socket area isn’t as empty as it looks. Over to the right in the picture above sits an mSATA slot designed for mini SSDs like the kind one might find in an ultrabook. mSATA drives have a low enough profile to avoid creating clearance problems for the PCIe x1 slot. The appeal for ATX systems with plenty of room for standard SSDs seems a little dubious, though. mSATA drives tend to be more expensive than their 2.5″ counterparts, and they typically have lower performance ratings.

The mSATA slot shares a link to the Z77 platform hub with one of the 3Gbps SATA ports. That’s a good thing, because it allows mini SATA drives to serve as Smart Response caches or boot drives for Intel’s Rapid Start tech. Those luxuries wouldn’t be available if the mSATA slot were tied to an auxiliary SATA controller. The Z77X-UD3H has one of those, too, but it supplies the eSATA ports in the rear cluster.

Gigabyte follows a similar expansion slot layout to Asus with one exception: there are three PCIe x1 slots but only a single PCI slot. The x1 slots share bandwidth with the lowest (left-most in the picture above) PCIe x16 slot, which gets either one or four lanes of connectivity, depending on whether the x1 slots are needed. Those PCIe lanes come from the chipset, while the other two x16 slots are fed directly by the CPU. If you’re going to be running a pair of graphics cards, Gigabyte recommends hooking a SATA power connector to the auxiliary plug located next to the SATA ports. That’ll purportedly improve system stability for CrossFire and SLI setups by providing more power through the PCIe slots.

Blue USB 3.0 ports pepper the rear cluster, but it’s hard to tell which ones are connected to the Z77 chipset and which branch off the board’s auxiliary VIA controller. The manual isn’t much help, either. We had to check the Device Manager to sort out which ports were routed where, details that somehow didn’t make it into my notes. Sorry. Four of the ports are definitely connected to the VIA controller, because half of the Z77’s USB 3.0 ports are reserved for the front-panel headers.

The Z77X-UD3H’s use of a VIA USB controller isn’t uncommon. However, it is unusual to see a VIA audio codec on a modern enthusiast board. The implementation isn’t that exciting, to be honest. There’s no support for surround-sound virtualization or on-the-fly multichannel encoding.

Gigabyte scores more diversity points by using a Gigabit Ethernet chip from Atheros. Most mobo makers are migrating towards Intel networking chips, and we’ll see in a moment how the Atheros one compares.

When the first 6-series motherboards hit the market, Gigabyte’s offerings were conspicuous for their lack of next-generation firmware. The old-school BIOS remained, and it was rather slow. Thankfully, Gigabyte’s 7-series motherboards feature the 3D BIOS firmware interface that debuted with the company’s X79 boards. It makes all the difference in the world.

Were it not for the slightly laggy mouse input, Gigabyte’s firmware would feel more advanced than the Asus EFI that has been the benchmark for more than a year. One may click on areas of the motherboard to bring up related variables, which are tuned with sliders rather than punching in numbers. The interface feels more like a Windows app than a firmware interface. There are still some rough edges, like settings windows that block the user’s view of the monitoring panel, but Gigabyte is on the right track.

The 3D interface is easy to navigate with the mouse, but it’s faster to flip through the firmware’s secondary interface, which puts a coat of fresh paint on a more familiar arrangement of tweaking options. There are more options here than in the 3D interface. The feature parity between them is impressive, though, especially considering the limited scope of Asus’ EZ interface.

To be fair, the Gigabyte EFI does have some limitations. The fan controls confine users to setting the slope of the fan speed curve. It’s possible to set a different slope for the CPU and three system fans (which are governed by a single profile), but that’s about it. At least the temperature-based speed control works with both three- and four-pin CPU fans.

Gigabyte’s EasyTune6 software offers somewhat more fan control functionality in Windows. Still, there are only two points to adjust on each fan control curve. If only motherboard makers put as much effort into their fan controls as they do into overclocking or, most recently, power tweaking options.

So important is power tuning that Gigabyte came up with a whole new 3D Power application just to handle it. The app is admittedly much improved over the early version we used when testing one of Gigabyte’s X79 boards, but the interface still feels clunky, and the sluggish transition animations don’t do the app any favors. You’re better off tweaking the power settings in the firmware.

At the Consumer Electronics Show earlier this year, Gigabyte told us it plans to offer a more cohesive collection of tweaking utilities with its motherboards. There’s certainly plenty of room for improvement, and Gigabyte would do well to draw inspiration from its latest EFI interface rather than expanding on 3D Power.

MSI’s Z77A-GD65 motherboard

Motherboards makers often slap a couple of stickers onto their boards to highlight specific features or attributes. MSI takes that to the extreme with the Z77A-GD65, which is covered by three massive stickers and a couple of smaller ones. There’s a board under there somewhere:

Features like third-gen PCI Express lanes, Military Class III components, and OC Genie II overclocking are already pimped on the box, so it’s unclear why users need to see the same logos again. Presumably, a sale has already been made. The stickers wouldn’t be so annoying if they were easy to remove. However, a number of them left behind sticky residue that needed to be scraped off with a fingernail. If you’re going to blanket a board with silly stickers, at least ensure they won’t leave behind a mess.

Remove the stickers, and the Z77A-GD65 looks like a typical enthusiast board, right down to the black-and-blue color scheme. Ugh. Am I the only one who misses the days when MSI’s motherboards came in a shade of red that wouldn’t look out of place on a Ferrari?

Perhaps digital camo would be more fitting today. The GD65 does, after all, feature components that meet the Department of Defense’s MIL-STD-810G standard for humidity, vibration, temperature, shock, and low pressure tolerance. The board has digital power circuitry, too. MSI could even go with an arctic-camo theme to match the growing number of white enclosures trickling onto the market.

The VRMs are covered by a pair of stylish heatsinks linked by a single pipe. Amazingly, this is the only board of the three to feature a heatpipe. MSI used to run networks of the things across its motherboards, complete with rollercoaster-style loops. People used to wear acid-wash jeans, too.

With heatsinks flanking it on two sides, the GD65’s socket looks a little crowded. Only the heatsink to the left really comes close, and it’s still short enough to steer clear of most aftermarket coolers. Clearance for fancy memory modules will probably be a more pressing issue; our Corsair Vengeance DIMMs are much taller than the motherboard’s heatsinks.

The Z77A-GD65 is entirely devoid of PCI slots, a move I can fully endorse now that there are a number of decent PCIe sound cards on the market. Unlike the other boards, all three PCI Express x16 slots are linked to the CPU. With a Sandy Bridge processor, your options are limited to 16 lanes to the first slot or 8 lanes each to the first two. The third slot is essentially useless unless you’re running Intel’s next-gen CPU. With it on board, there are three configuration options: x16/x0/x0, x8/x8/x0, and x8/x4/x4.

If the Z77 Express chipset lacks enough 6Gbps SATA ports to meet your needs, good news. The GD65 has two more 6Gbps ports courtesy of the same ASMedia chip found on the Asus board. Peripheral controllers get around.

MSI hasn’t seen fit to add an auxiliary USB 3.0 controller, though. The Z77A-GD65 gets by with the four ports built into the Z77 Express; two are routed to the rear cluster, while the others are accessible via front-panel headers.

The GD65 is the only board of the three without a DisplayPort video output. It’s also the only one with both coaxial and S/PDIF audio jacks. MSI combines the board’s Realtek audio codec with THX TrueStudio Pro software, which offers speaker virtualization but doesn’t support on-the-fly multichannel encoding for those digital outputs.

Kudos to MSI for putting a CMOS reset button right in the port cluster. It’s nice to be able to clear the CMOS without having to crack open the case.

MSI’s first attempt at next-generation firmware was a flop, but the company went back to the drawing board and has produced a much better second effort that bears little resemblance to what shipped on its first 6-series motherboards. ClickBIOS II skips the dual interfaces of the Asus and Gigabyte firmware for a single GUI. The interface is devoid of sliders and flashy effects, and it and operates more like the advanced modes available on the other two boards. I quite like the layout, and the mouse tracking feels much smoother than on the Gigabyte board. It does take a little time to get used to having to double-click on everything, though. I also wish it were possible to key in values directly rather than scrolling through lists or hammering the +/- keys. At least the mouse wheel works for scrolling.

The overclocking options provided in the firmware are numerous. All three boards are on pretty even footing on that front, provided you’re not one of those extreme overclockers who totes around a canister of liquid nitrogen.

Unfortunately, MSI’s fan speed controls are underwhelming at best. Intelligent speed control is limited to the CPU fan, for which the user can set only a target temperature. Manual speed control is provided for the system fans, but they won’t change speed based on system temperatures. It’s becoming increasingly obvious MSI hasn’t improved its fan speed controls for years. In fact, it feels like they’ve regressed.

The Control Center software that ships with the Z77A-GD65 fails to expand on the fan controls, although it does offer a host of overclocking settings. There’s also a version of the ClickBIOS interface that runs in Windows, providing a familiar atmosphere for system tuning. Unfortunately, it takes forever to load and apply changes, which can require a reboot. You’re better off modifying system settings directly in the firmware.

Digging into the details

Congratulations for making it this far already. As a reward for not skipping ahead to the conclusion after the first page, I’m going to let you in on a little secret: this page isn’t all that interesting. Unless you like seeing firmware options, board specifications, and system configurations laid out in painstaking detail, you might want to skip ahead to the beginning of our performance results.

Curious how the three boards stack up in the most important firmware categories? Here’s a quick rundown of the key overclocking and fan control settings:

Asus P8Z77-V Gigabyte Z77X-UD3H MSI Z77A-GD65
Clock speeds Base: 80-300MHz

DRAM: 800-2400MHz

IGP: 1350-3000MHz

Base: 80-133.33MHz

DRAM: 800-2133MHz

IGP: 400-3200MHz

Base: 0-655.35MHz

DRAM: 800-2133MHz

IGP: 1350-3000MHz

Multipliers CPU: 35-63X CPU: 16-59X CPU: 16-63X
Voltages CPU: 0.8-1.92V

IGP: 0.8-1.92V

DRAM: 1.2-1.92

VCCSA: 0.8-1.7V

CPU PLL: 1.2-2.2V

PCH: 0.8-1.7V

CTRL ref A/B: 0.395-0.63X

CPU: 0.8-1.9V

CPU VTT: 0.8-1.7V

CPU PLL: 1.2-2.2V

IMC: 0.715-1.4V

IGP: + 0-0.35V

DRAM: 1.1-2.1V

Data TX: 0.695-0.81V

Data term.: 0.595-1.939V

Data A/B ref: 0.595-1.939V

Address A/B: 0.595-1.939V

CPU: 0.8-2.155V

CPU I/O: 0.95-1.55V

DRAM: 1.08-2.464V

IGP: 1-1.52V

System Agent: 0.925-1.585V

CPU PLL: 1.4-2.53V

PCH: 0.775-1.724V

VREF CA A/B: 0.435-1.125V

VREF DA A/B: 0.435-1.125V

Fan control CPU: Min, max temp

Min, max duty cycle

System 1,2,3: Min, max temp

Min, max duty cycle

CPU: slope

System 1-3: slope

CPU: Target temp

System 1,2: Manual speed

There’s more difference in the fan controls than anywhere else. For most folks, all the boards have sufficient clock, multiplier, and voltage headroom for overclocking.

Next, we’ll take a quick tally of expansion slots and ports.

Asus P8Z77-V Gigabyte Z77X-UD3H MSI Z77A-GD65
DIMM slots 4 DDR3 4 DDR3 4 DDR3
Expansion slots 2 PCIe 3.0 x16

1 PCIe 2.0 x16 (x4)

2 PCIe 2.0 x1

2 PCI

2 PCIe 3.0 x16

1 PCIe 2.0 x16 (x4)

3 PCIe 2.0 x1

1 PCI

2 PCIe 3.0 x16

1 PCIe 3.0 x16 (x4)

4 PCIe 2.0 x1

Storage I/O 2 SATA RAID 6Gbps

4 SATA RAID 3Gbps

2 SATA 6Gbps

2 SATA RAID 6Gbps

4 SATA RAID 3Gbps

1 mSATA RAID 3Gbps

2 SATA RAID 6Gbps

4 SATA RAID 3Gbps

2 SATA 6Gbps

Audio 8-channel HD 8-channel HD 8-channel HD
Wireless 802.11n Wi-Fi NA NA
Ports 1 PS/2 keyboard/mouse

1 DisplayPort

1 HDMI

1 DVI-D

1 VGA

4 USB 3.0 w/ 2 headers

2 USB 2.0 w/ 8 headers

1 Gigabit Ethernet

1 analog front out

1 analog bass/center out

1 analog rear out

1 analog surround out

1 analog line in

1 analog mic in

1 optical S/PDIF out

1 PS/2 keyboard/mouse

1 DisplayPort

1 HDMI

1 DVI-D

1 VGA

6 USB 3.0 w/ 2 headers

6 USB 2.0 headers

2 eSATA 6Gbps

1 Gigabit Ethernet

1 analog front out

1 analog bass/center out

1 analog rear out

1 analog surround out

1 analog line in

1 analog mic in

1 optical S/PDIF out

1 PS/2 keyboard/mouse

1 HDMI

1 DVI-D

1 VGA

2 USB 3.0 w/ 2 headers

4 USB 2.0 w/ 6 headers

1 FireWire header

1 Gigabit Ethernet

1 analog front out

1 analog bass/center out

1 analog rear out

1 analog surround out

1 analog line in

1 analog mic in

1 optical S/PDIF out

1 coaxial S/PDIF out

If you’ve been paying careful attention over the preceding pages, you’ve already picked up all the details in the table above.

Our testing methods

Since we can’t test with her little sister just yet, we’ve run all the boards through the gauntlet with a Sandy Bridge CPU. A Z68 motherboard has been included for comparison, but time constraints prevented us from testing additional platforms.

We used the following system setups for testing. With few exceptions, all tests were run at least three times, and we reported the median of the scores produced.

Processor Intel Core i7 2600K 3.4GHz
Motherboard Asus P8Z77-V Gigabyte Z77X-UD3H MSI Z77A-GD65 MSI Z68A-GD80
Bios revision 0906 F7 V10.3B6 V17.5
Platform hub Intel Z77 Express Intel Z77 Express Intel Z77 Express Intel Z68 Express
Chipset drivers Chipset: 9.3.0.1020

RST: 11.1.0.1006

Chipset: 9.3.0.1020

RST: 11.1.0.1006

Chipset: 9.3.0.1020

RST: 11.1.0.1006

Chipset: 9.3.0.1019

RST: 10.8.0.1003

Memory size 8GB (2 DIMMs) 8GB (2 DIMMs) 8GB (2 DIMMs) 8GB (2 DIMMs)
Memory type Corsair Vengeance DDR3 SDRAM at 1333MHz Corsair Vengeance DDR3 SDRAM at 1333MHz Corsair Vengeance DDR3 SDRAM at 1333MHz Corsair Vengeance DDR3 SDRAM at 1333MHz
Memory timings 9-9-9-24-1T 9-9-9-24-1T 9-9-9-24-1T 9-9-9-24-1T
Audio Realtek ALC892 VIA VT2021 Realtek ALC898 Realtek ALC892
Graphics Asus EAH5870 1GB with Catalyst 12.2 drivers
Hard drive Western Digital Caviar Black 1TB

Samsung 830 Series 256GB

OCZ RevoDrive 3 X2 240GB

Power Supply Corsair AX850 850W
OS Microsoft Windows 7 Ultimate x64

We’d like to thank Asus, Corsair, PC Power & Cooling, and Western Digital for helping to outfit our test rigs with some of the finest hardware available. Thanks to each of the motherboard makers for supplying their boards, too, and to Intel for providing the CPUs.

We used the following versions of our test applications:

The test systems’ Windows desktop was set at 1920×1200 in 32-bit color at a 60Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

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.

USB 3.0 performance

We’re going to start with USB 3.0 performance, since that’s the most compelling addition to the 7-series platform. The latest version of Thermaltake’s BlacX docking station, combined with a Samsung 830 Series 256GB SSD, will serve as the USB device for the following tests. The solid-state drive is probably faster than what most folks will be plugging into their USB ports, and the docking station supports USAP.

The two actually look quite good together. The brushed finish on the top of the dock is almost a perfect match for the SSD’s sexy skin.

Since the Z68 doesn’t have built-in USB 3.0 support, we’ve pitted the Z77’s USB controller against the third-party chips found on the various boards. We’ve also added some results from the integrated USB controller in AMD’s A75 chipset using an Asus F1A75-V PRO motherboard.

CrystalDiskMark

First up is CrystalDiskMark, which will give us a sense of the how the USB controllers perform in a handful of key metrics. The Samsung SSD was secure erased before each batch of tests on each board to ensure consistent results. We used the test’s 1GB transfer size with random data.

Intel’s USB 3.0 controller appears to have been worth the wait. It pushes higher sequential read and write speeds than the competition and has a substantial lead over its next-closest rival in the read test.

The controller’s random read performance closely matches that of the others. However, its 4KB random write speed is a good 10-16MB/s higher.

Hammering our USB-attached SSD with random I/O requests at a queue depth of 32 puts more strain on the USB controllers. The Z77’s position doesn’t change much, though. It’s competitive in the random read test and faster than anything else in the write test.

TR FileBench

Concocted by resident developer Bruno “morphine” Ferreira, FileBench runs through a series of file copy operations using Windows 7’s xcopy command. Using xcopy produces nearly identical copy speeds to dragging and dropping files using the Windows GUI, so our results should be representative of typical real-world performance. We’ve combined our USB dock with OCZ’s RevoDrive 3 X2 PCIe SSD, which will serve as the target for our read tests and the source for our write tests. We’ve tested with two file sets. The movie set includes 4GB of the sort of videos one might download off BitTorrent, while the mix set contains 5GB worth of movies, MP3s, JPEGs, RAW images, spreadsheets, and HTML files.

Score another one for the Z77’s USB 3.0 controller—make that two, actually. The Intel controller reads the files from both sets much faster than the others. Something is up with the VIA VL800, though. It produced inconsistent results when copying the mixed file set from our USB drive. That behavior persisted through multiple tests, with the VL800 hitting upwards of 160MB/s in some runs but less than 40MB/s in others.

The VL800’s issues don’t appear to affect write speeds, which were very consistent. In these tests, the Z77 again looks very fast. It essentially ties the ASMedia controller for the lead in the movie test and enjoys a 7MB/s lead when writing the files from our mixed set.

Serial ATA performance

Although the Z77’s Serial ATA controller is similar to the one in the Z68 Express, we thought we’d double-check its performance to be sure. We are using slightly different driver revisions for the two platforms, so keep that in mind as we move through our test results. The MSI Z77A-GD65 was used to represent the Z77 in our SATA tests; it’s the best match for our MSI Z68 board.

These tests were conducted with the same Samsung 830 Series SSD as our USB tests, just with the drive plugged into a 6Gbps SATA port instead of the Thermaltake docking station. We went through the same secure-erase routines and used the same benchmark settings.

CrystalDiskMark

Looks like the same controller logic to me. The Z77 Express scores slightly higher in CrystalDiskMark’s sequential read speed test, but the results are virtually identical otherwise.

TR FileBench

Our FileBench results offer a little more variety, although the two platforms remain closely matched. The Z77 is a little bit faster when reading our movie files and writing our mixed set. However, its predecessor pulls ahead when reading the mixed set and writing the movies.

TR DriveBench

DriveBench simulates disk-intensive multitasking scenarios by playing back traces of real-world disk activity. You can read more about DriveBench here. We’re using our file copy and virus scan workloads because they’re the most demanding ones we have.

The two platforms achieve performance parity here. Let’s move on.

IOMeter

IOMeter subjects storage devices to an increasing number of concurrent I/O requests. We’ve limited the load to 32 simultaneous requests, which matches the depth of the Serial ATA spec’s command queue.

The Z77 Express largely shadows the Z68 in IOMeter. I’d be more worried about the lower transaction rate with a single I/O in the Database test if the Z77 hadn’t already proven itself capable elsewhere.

PCI Express performance

Remember the RevoDrive PCIe SSD from our USB tests? We thought it might be useful for teasing out any differences in the Z77’s PCI Express implementation, so we ran a few additional tests.

The RevoDrive is built on a PCIe 2.0 x4 card, so it’s perfect fit for a motherboard’s third PCIe x16 slot, which typically offers up to four lanes of bandwidth to the platform hub. We’ve also run the card in the secondary x16 slot usually reserved for graphics cards. The PCIe lanes for that slot come directly from the CPU, so they don’t have to negotiate the chipset or hop over its DMI link to the processor.

We used HD Tune to probe the RevoDrive’s read speed and accompanying CPU utilization on each platform. Since the MSI Z77 board doesn’t offer a four-lane PCIe slot connected to the chipset, the Gigabyte Z77X-UD3H will fill in for that test.

The Z77 and Z68 offer comparable PCIe throughput, but the RevoDrive is slightly faster when hanging directly off the CPU on both platforms. In that configuration, the Z77’s CPU utilization is a little bit higher than its predecessor.

Memory performance

Now that we’ve explored the Z77 platform in depth, let’s take a closer look at motherboard performance. We’ll start with memory bandwidth, which should be pretty consistent from one board to the next. We are, after all, using the same Sandy Bridge memory controller and Corsair Vengeance DIMMs on all the boards.

That’s a wash, folks.

Application performance

Can any of the boards differentiate themselves in our handful of application tests?

Not really. All the boards are running the same CPU and graphics card, and those components are more important limiting factors for application performance.

Power consumption

We measured system power consumption, sans monitor and speakers, at the wall outlet using a Watts Up Pro power meter. Readings were taken at idle and under a load consisting of a Cinebench 11.5 render alongside the rthdribl HDR lighting demo. We tested with Windows 7’s Balanced power plan, which is how most folks run their systems. Win7’s High Performance mode didn’t yield better performance or a meaningful change in power consumption.

Most of the motherboards have additional power-saving features that can be controlled via their firmware. The names of these power-saving schemes are all different, of course, but they attempt to achieve the same thing: lowering power consumption without restricting performance. Each board was run with its power-saving features disabled and enabled, with the latter config detailed in parenthesis.

The MSI Z77A-GD65 has lower idle power consumption than its peers, which are pretty closely matched. The field spreads out a bit under load, but the MSI remains one of the most power-efficient options. Only the Asus board has lower power consumption under load, and then only when its EPU mojo is enabled. Enabling the EPU setting cuts the board’s power draw by nearly 10W under load, although it has little impact at idle. Gigabyte’s Lite power-saving scheme does virtually nothing to our system’s power consumption at the wall socket.

Overclocking

Using a Sandy Bridge CPU hasn’t hampered our testing thus far. Overclocking is a different story, however. Motherboard makers spend countless hours tuning their products for specific CPUs, and Z77 boards have been optimized for Intel’s next-gen CPU rather than last year’s model—rightfully so.

Overclocking is a big part of the motherboard picture, so we’ve pushed our Core i7-2600K to its limits on each board to gauge the overall experience. We started with each board’s auto-overclocking scheme to establish a quick baseline before moving on to manual tuning. At each step along the way, we tested stability with AIDA64’s torture test running alongside the rthdribl HDR lighting demo.

Asus’ auto-overclocking feature is activated via the firmware, and it took our CPU to 4.4GHz using a 43X multiplier and a 103MHz base clock. CPU-Z reported a 1.312V CPU voltage, and the system was perfectly stable. The auto-clocking mechanism worked its magic quickly, giving us plenty of time for manual tuning.

Next, we reset the base clock to 100MHz and confined our manual tweaking the CPU multiplier, leaving the rest of the settings at “auto.” The board dutifully increased the CPU voltage as the multiplier ramped up, and by the time it hit 50X, we were up to 1.528V. That config proved stable, but BSODs started appearing at 5.1GHz, so we called it a day.

Gigabyte’s hands-off overclocking is invoked via the EasyTune software, and there’s a new wrinkle. Instead of relying solely on pre-baked profiles that automatically pick a speed based on your CPU’s model number, there’s a new auto-tuning mechanism that steps up the clock speed and tests for stability along the way, much like how one might overclock a system manually. This auto-tuner was quite aggressive, pushing our CPU to 4.9GHz using a 47X multiplier and a 104.6MHz base clock. EasyTune increased the CPU voltage to a whopping 1.56V to hit that speed, but it bluescreened just a couple of minutes into our stability test.

The culprit turned out to be the base clock. When we dropped it to 100MHz for manual tweaking, the board made it to 5GHz without crashing. BSOD errors returned at 5.1GHz, though. Gigabyte would do well to beef up EasyTune’s stability test and have the auto-tuner focus on multiplier rather than base clock enhancements. It would be nice to be able to define a maximum voltage that can be applied to the CPU, as well. 1.56V is a little aggressive for an auto-tuning scheme.

MSI’s Z77A-GD65 can be overclocked with the touch of a button—the OC Genie button, which is located right on the circuit board. OC Genie now includes the ability to define a preset overclocking profile, but we decided to let the board overclock itself, instead. Turns out the default profile is rather conservative; OC Genie settled on a 42X multiplier and didn’t touch the base clock, resulting in a CPU speed of 4.2GHz on 1.34V. That config was perfectly stable, so we kept turning the screws on the CPU multiplier.

The system proved stable up to a 47X multiplier, but 48X required more juice, and the board didn’t increase the voltage automatically. Out of curiosity, we cranked the CPU voltage to 1.5V manually and managed to get the system stable at 4.9GHz. I suspect 5GHz would have been attainable with a smidgen of extra voltage. Since we didn’t resort to manual voltage tuning on the other boards, we moved on.

Motherboard peripheral performance

Congratulations on making it this far. You’re almost to the conclusion, but before we finish up, there are a few more motherboard-specific attributes we need to test. We’ve distilled most of the results into a series of condensed tables, but you’ll have to endure one more page of graphs first.

USB 3.0 performance

The P8Z77-V ships with Asus’ USB Boost software, which enables USAP for the board’s ASMedia controller and offers a “turbo” mode for controllers that lack support for the USB Attached SCSI standard. USB Boost only presented us with a turbo option for the Intel USB 3.0 controller, though. Intel’s drivers may not yet expose the controller’s USAP compatibility, which won’t get native OS support until Windows 8, Asus says.

Turbo-charging the Asus board’s Intel USB controller makes it even faster in the sequential tests. However, it can’t catch the P8Z77-V’s ASMedia controller running in USAP mode, which delivers by far the best random I/O performance at high queue depths. If no additional software is involved, the Intel controller offers about the same performance on all three Z77 boards.

Asus USB Boost software pays dividends in our real-world file transfer tests, as well. Thanks to USAP, the ASMedia controller scores higher than anything else in our mixed read test. Enabling USAP does drop the speed at which the ASMedia controller writes those mixed files, though. The controller’s performance is improved enough in the other tests that I’d enable USAP mode for the ASMedia controller and turbo for the Intel one.

Next to those results, the VIA USB controller on the Z77X-UD3H looks a little slow. The Intel implementation is faster, and the VIA chip stumbles dramatically when reading our mixed file set. Even without its Boost magic, the ASMedia controller appears to be more robust than the VIA solution.

Serial ATA performance

CrystalDiskMark Serial ATA performance (MB/s)
Sequential Random 4KB Random 4KB QD32
Read Write Read Write Read Write
MSI Z68A-GD80 505 416 25 88 306 151
Asus P8Z77-V 508 415 24 86 305 152
Asus P8Z77-V (ASMedia) 366 367 20 48 208 145
Gigabyte Z77X-UD3H 513 416 25
78 306
151
MSI Z77A-GD65 518 415 25 90 305 151
MSI Z77A-GD65 (ASMedia) 364 364 21 46 211 147
FileBench Serial ATA performance (MB/s)
Movie Mix
Read Write Read Write
MSI Z68A-GD80 340 433 213 238
Asus P8Z77-V 333 407 205 241
Asus P8Z77-V (ASMedia) 305 387 183 241
Gigabyte Z77X-UD3H 336 417 211 248
MSI Z77A-GD65 350 414 209 244
MSI Z77A-GD65 (ASMedia) 309 393 190 240

There are two lessons to be learned from our Serial ATA results. The Z77’s SATA performance is pretty consistent from one board to the next. It’s also notably faster than the auxiliary ASMedia solution found on the Asus and MSI mobos. Only use those auxiliary SATA ports as a last resort.

Ethernet performance

NTttcp Ethernet performance
Throughput (Mbps) CPU utilization (%)
MSI Z68A-GD80 942 3.2
Asus P8Z77-V 935 2.1
Gigabyte Z77X-UD3H 942 4.5
MSI Z77A-GD65 942 2.0
FileBench Ethernet performance (MB/s)
Movie Mix
Read Write Read Write
MSI Z68A-GD80 114 114 69 39
Asus P8Z77-V 115 113 71 39
Gigabyte Z77X-UD3H 115 112 77 44
MSI Z77A-GD65 114 113 71 39

FileBench works for network transfers, too. There isn’t much difference in movie transfer speeds. However, the Atheros Gigabit Ethernet controller on the Gigabyte board offers better performance in the mixed tests. That’s particularly notable because the Asus and MSI boards both use Intel GigE chips. The system at the other end of the CAT6 cable in these tests is also based on an Intel network controller.

There is a catch, of course. The Z77X-UD3H’s CPU utilization is a little higher than the others in the NTttcp throughput test. I’d happily trade 2.5% of the Core i7-2600K’s CPU resources for faster network transfer speeds.

PCI Express performance

HD Tune PCIe performance
CPU Chipset
Read (MB/s) CPU utilization (%) Read (MB/s) CPU utilization (%)
MSI Z68A-GD80 1303 7.2 1258 7.6
Asus P8Z77-V 1275 8.4 1276 7.3
Gigabyte Z77X-UD3H 1333 7.3 1262 7.9
MSI Z77A-GD65 1300 10.2 NA NA

Our RevoDrive PCIe SSD is insanely fast on all the boards. The MSI is the only one to push CPU utilization into double-digit territory, but we’re only talking about a difference of a few percentage points versus the rest of the field.

Analog audio signal quality

RightMark Audio Analyzer audio quality: 24-bit/192kHz
Frequency response Noise level Dynamic range THD THD + Noise IMD + Noise Stereo Crosstalk IMD at 10kHz Overall score
MSI Z68A-GD80 4 4 4 5 3 4 5 5 4
Asus P8Z77-V 5 4 4 5 3 5 5 5 4
Gigabyte Z77X-UD3H 5 4 4 5 3 5 5 5 4
MSI Z77A-GD65 5 4 5 5 3 4 5 5 5

According to RMAA’s loopback test, the MSI Z77A-GD65 has better overall analog signal quality than the others. The individual test results are pretty close, though. If you’re really worried about analog signal quality, you’re better off with a discrete sound card. If you’re not, features like speaker virtualization and real-time encoding support may be more important to you than minute differences in frequency response or dynamic range.

Conclusions

Without Intel’s next-generation processor, the arrival of 7-series motherboards is a little anticlimactic. Perhaps that’s why Easter Sunday was chosen as the official launch day. Everyone will be too busy hunting for eggs to notice that Z77 motherboards are here and Ivy Bridge isn’t. Well, everyone but me. I’m stuck trying to come up with a placeholder conclusion until we can see how Z77 boards fare with the CPU most folks will be running in them.

Until then, it’s difficult to pass judgment. We can, however, draw some conclusions about the Z77 Express platform hub and how each motherboard maker has approached it. We’re pleased that this chipset finally brings USB 3.0 support to the table. Intel’s SuperSpeed controller is exceptionally fast in both standard benchmarks and real-world file transfers. It offers better all-around performance than auxiliary controllers, and we prefer it as a result.

The rest of the platform hub resembles the Z68 Express, which isn’t necessarily a bad thing. We’re just happy that Ivy Bridge buyers should be able to grab boards based on Intel’s high-end desktop platform starting on day one. Whether you want to overclock the CPU, run a Smart Response SSD cache, accelerate video Transcoding with QuickSync, or tap into Lucid’s Virtu software, the Z77 has you covered. The same can’t be said for other members of the 7-series family.

So what about the motherboards? Let’s start with Asus, whose 6-series boards set the bar. The P8Z77-V raises it, adding Wi-Fi and a host of new software features, including even better fan speed controls and USB Boost functionality that can vastly improve transfer rates. Asus’ firmware remains excellent, and while it doesn’t have any new bells and whistles, we’re told additional features are on the way now that the base code has been finalized for Intel’s next-gen CPU. While Asus is adding those extras, it really needs to fix the firmware’s Turbo behavior, which overclocks the CPU without the user’s consent. That’s the sort of shenanigan we’d expect from a bush-league motherboard maker, not the biggest name in the business.

The P8Z77-V’s $210 asking price is also rather high, at least compared to the other boards. You get a lot more functionality from the Asus board, but it doesn’t come cheap.

We’ll move through the field alphabetically, which means Gigabyte is up next. The Z77X-UD3H surprised me the most over the past few days. Gigabyte’s new firmware is just a few tweaks—and more extensive fan speed controls—away from being my favorite of the three. The accompanying Windows software needs a complete overhaul, though.

The UD3H has an interesting mix of features, including a nifty mSATA slot and a different mix of peripheral chips than we usually see. While the performance of the Atheros Gigabit Ethernet controller is impressive, the VIA USB 3.0 controller appears to have a few quirks. At least there’s an Intel one to back it up. We can forgive a few transgressions given the price tag attached to the Z77X-UD3H. It’s currently selling for just $170 online, 12 bucks less than the suggested retail price we were quoted a couple of days ago.

MSI’s Z77A-GD65 costs $20 more at $190, and it’s hard to see why. Apart from its FireWire support, which is missing from the other boards, the GD65 doesn’t really offer anything above and beyond. It’s the only board of the three without a secondary USB 3.0 controller, and it easily has the worst fan speed controls. That’s a shame, because MSI has really cleaned up its firmware interface. The GD65 has impressively low power consumption, too, and I like the fact that the third PCIe x16 slot is linked directly to the CPU rather than the platform hub. This configuration should provide considerably more bandwidth when Intel’s next-gen CPU arrives.

That day is surely coming soon, but until then, it’s hard to pick a favorite from the three boards we’ve looked at today. Stay tuned for a second look in due time.

Comments closed
    • abloc
    • 8 years ago

    Was difference in minimum fps a fluke, or these mobos behave differently under stress?

    • rjseo1
    • 8 years ago
    • rjseo
    • 8 years ago
    • merryjohn
    • 8 years ago
    • d0g_p00p
    • 8 years ago

    Page 6 states “the docking station supports USAP.” That should be UASP right?

      • merryjohn1
      • 8 years ago
    • Bensam123
    • 8 years ago

    Curious, why didn’t you guys bother manually tweaking voltage in the manual portion of the OC tests? Isn’t one of the defining factors between boards (or purportedly) their voltage management circuitry?

    It’s nice to see you guys doing PCI-E bandwidth comparisons now. It sort of looks like your PCI-E SSD was the limiting factor though… I wonder if a IB enabling PCI-E 3 will reduce the CPU overhead in the tests.

    • dpaus
    • 8 years ago

    I saw a Gigabyte GA-Z77-D3H (i.e., no ‘U’) for just $130 Cdn, so I snapped it up for my new HTPC. I hope to have it all put together later this week or early next week.

    • LoneWolf15
    • 8 years ago

    Tell me why I’d spend money on a Z77 mainboard when I have a Z68 mainboard with USB 3.0 that I can put an Ivy Bridge processor in.

      • derFunkenstein
      • 8 years ago

      Native USB 3.0 is faster than the crap external controllers that companies have been shoving on our boards, but otherwise I get where you’re coming from. My Z68 board already supports PCI-e 3.0 in Ivy Bridge, supports Lucid’s Virtu software, and it already supports RapidStorage if I wanted to go that route with my SSD (which I don’t, but I could).

        • LoneWolf15
        • 8 years ago

        I have an NEC/Renessas controller, should be fine. I do agree that not all of them are great though.

        I think PCIe 3.0 is unnecessary for at least one more generation.

      • Airmantharp
      • 8 years ago

      Even without PCIe 3.0, I see no reason to do so- and that would be my only reason to actually upgrade. Looking over at Anand’s benchmarks on the various USB 3.0 controllers, it seems that even the add-in chips do quite well overall, though not quite as fast as Intel’s. But given that I don’t have any external USB 3.0 SSD drives to take advantage of that speed, I don’t see the reason for it now.

      As I said above, PCIe 3.0 might be worth it, but that would only be after going with a pair of the newly released higher-end cards, which I have a hard time justifying. My HD6950 2GB pair seems to be doing just fine, even if I can’t turn BF3 all the way up in 64-man maps at 2560×1600.

        • derFunkenstein
        • 8 years ago

        Even then, the 6000 cards are not PCIe 3.0. You’d have to upgrade to a 7000 or 680 to get anything extra out of those lanes.

      • JustAnEngineer
      • 8 years ago

      Because your Z68 motherboard is already populated with a Sandy Bridge processor and you’re building a new system.

      • Bensam123
      • 8 years ago

      For the same reasons someone who originally bought a AM2 motherboard would upgrade to a AM3+ after all the years.

      I really do envy the ultra long upgrade cycle of the AM2 socket. I have a 1156 with a i7 in it and I have to replace it with a 1155 if I plan on upgrading to a IB. There isn’t really any difference between the two sockets or their coinciding chipsets I would want to spend a extra $175 on at this point in time. The memory hasn’t even changed. I don’t currently need PCI-E 3 and probably wont even if I get a graphics card two generations newer then the current ones on the market. USB3 is nice, but I don’t use external drives and all my peripherals are still USB2 (including my gaming ones).

      The only real kicker is Lucid’s Virtu software, which looks pretty cool, but other then that…

        • derFunkenstein
        • 8 years ago

        There’s not “really any difference” between the sockets on a superficial level, but go back and read TR’s Sandy Bridge review – there’s plenty of difference in the way the CPU interacts with pretty much everything else.

          • Bensam123
          • 8 years ago

          I’ve read TRs review… I understand architecture changes… but AMD acomplished that without ditching compatability. That’s what I was marveling at.

            • Airmantharp
            • 8 years ago

            AMD did the HyperTransport thing- but they’re also considerably behind on CPU tech. Compatibility is cool no doubt. I just wonder that if they’d been able to stay close to lock-step with Intel on CPU performance they’d also be able to keep up the compatibility.

            • Bensam123
            • 8 years ago

            I don’t think they’re considerably behind… Bulldozer is the ugly duckling right now, but people are using a hyperbole every time they discuss it. They have their own versions of all the major tech in Intels CPUs except for Virtu, but that’s made by Lucid.

            Looking back a generation to Phenom 2 vs the original i7, they were around roughly the same area and the Phenom 2 was backwards compatible with AM2 as long as motherboard makers chose to support it. The Phenom 2 can still slip into a AM3+ socket too. Things not only go backwards, but forwards too… I think that’s pretty amazing.

            I personally think Intel is choosing not to keep up socket compatibility solely to push product segmentation. Heck that’s what the whole socket 1156 vs 1366 was, now 1155 vs 2011. You really think 1155 style processors needed one less pin?

            • UberGerbil
            • 8 years ago

            Well, there’s obviously segmentation going on wrt 1366/2011 vs 1155/1156. But the shift within each of those lines isn’t about segmentation, it’s about revenue maximization in the chipset business — and the tension between Intel and its mobo partners on that point. If it was up to Intel, they’d change sockets with every generation, and there’d be no chance you’d be able to swap an Ivy Bridge into a Sandy Bridge motherboard. In fact, Intel planned to take things even further in the Arrandale era by having separate sockets for the IGP-enabled and sans-IGP CPUs of the same generation. But the mobo makers balked at that, and also pushed back on churning sockets with every generation. So it seems like every-other-gen is the compromise for the time being.

    • deathBOB
    • 8 years ago

    Mmmm… Controllers [i<]rippling[/i<] with ports...

    • chuckula
    • 8 years ago

    Interesting motherboards but Intel is being a little bit silly by finally “officially” releasing these things several weeks before IB is ready… They have some nice features but nobody is going to run out to buy one just to slap a sandy bridge chip into it. I eagerly await the conclusion of this article where you finally get to slap the Ivy Bridge chips in 🙂

      • NeelyCam
      • 8 years ago

      [quote<]"Buy a motherboard now, and get ready to be awesome with a brand-new super-CPU!"[/quote<] Yeah, they took a page off AMD's evil marketing playbook.

        • BobbinThreadbare
        • 8 years ago

        Someone could buy this just for future proofing. They buy a Sandy now, and in a year or something get an Ivy.

      • Peldor
      • 8 years ago

      Eh, I’d much rather have a few weeks for the channel to fill up with motherboards than have to scramble for both when the CPUs come out.

    • Moha
    • 8 years ago

    Ugly back I/O ports for Asus P8Z77-V comparing with P8Z68-V, no eSATA, has less USB 3.0/2 ports and PS/2 port in 2012, you kidding me?

      • derFunkenstein
      • 8 years ago

      The VGA port could easily be replaced by 4 USB 2.0 ports and make the board more useful at least.

        • Lazier_Said
        • 8 years ago

        6 inconvenient USB 2 ports, to go along with the 4 inconvenient USB 3 ports, useful for what exactly?

        The only thing back USB ports are good for is attaching a hub that’s somewhere you can reach it.

          • derFunkenstein
          • 8 years ago

          I have lots of stuff that’s always plugged in to USB ports. 2 USB hard drives, keyboard, mouse (look there’s 4 already), a USB cable for my phone, an Apple iDevice sync/charge cable, iLok, a MIDI keyboard, and an audio interface. Up front I’m using a total of 4 USB ports – 2 for the case and 2 for a media reader.

          Way more utility than a VGA port that I don’t need since I could put a DVI-> VGA adapter on a DVI-I port (rather than the included DVI-D).

            • axeman
            • 8 years ago

            Perhaps OP’s keyboard and mouse connect with magic.

            • derFunkenstein
            • 8 years ago

            I can’t see how you’d be against more connectivity as long as that connectivity isn’t artificially slowed down. These would be native ports that otherwise you have to spend more money to use (in the form of front-panel drive bay modules or slot brackets) vs. having 4 extra ports for “free”.

            This is different than my crusade against subpar USB 3.0 controllers; why you’d bother with them when Intel’s USB 3.0 ports are so fast is mind-boggling. It’s not like you need your keyboard, mouse, MIDI controller, phone, iDevice, etc. plugged into them – just USB 3.0 peripherals like HDD enclosures and the occasional very fast flash drive.

            • axeman
            • 8 years ago

            I was just saying back USB ports are far from useless 🙂

    • ronch
    • 8 years ago

    This reminds me of when mobo makers released their 9-series boards for Bulldozer and no Bulldozer to plug into them. You could’ve bought a 9-series board with a K10-based chip and simply be prepared for your 8-core future.

      • chuckula
      • 8 years ago

      It is similar. One difference is that due to the delays, the AM3+ boards were out for a LONG time before Bulldozer showed up. The delay was so long that a bunch of the AM3+ boards required firmware updates just to support Bulldozer, and some of those boards can’t simultaneously support Phenom II and Bulldozer.. you have to choose a firmware for one or the other.

      Hopefully the fact that Ivy Bridge is so plentiful that it’s already openly on sale in the Asian gray markets means that these same motherboards will have no issues supporting IB when it comes out in 3 weeks.

    • phez
    • 8 years ago

    Regarding PCIe 3.0,

    I just thought it was nice we now have the [i<]equivalent[/i<] of 16x/16x for sli/cfx instead of 8x/8x from the previous spec 🙂

      • Nevarre
      • 8 years ago

      Or even PCIe 3.0 @ x8/x4/x4 (equivalent to PCIe 2.0 x16/x8/x8) giving you the flexibility to run SLI/Crossfire *and* a RAID card at the same time, *or* Triple SLI/Crossfire.

      Sandy Bridge is a lot less flexible if you stress out the PCIe bus.

    • Visigoth
    • 8 years ago

    I miss the Intel motherboard in this review. I would’ve liked to see how competitive they are with the established competition.

    • pedro
    • 8 years ago

    It kind of makes me sad to see VGA ports on motherboard port clusters these days.

      • ronch
      • 8 years ago

      I know, but it could be handy for when you accidentally smash your LCD monitor and you have a spare CRT up in the attic.

        • Duck
        • 8 years ago

        No because you can use a passive DVI to VGA converter…

          • Bensam123
          • 8 years ago

          You CAN’T use a passive DVI to VGA adapter on a DVI-D port, which these boards have. The four pins necessary for DVI to VGA are missing. You would need an active converter that would cost $30-40.

          Displayport and HDMI also cannot use an adapter to go to VGA. You need a converter for those as well.

        • MadManOriginal
        • 8 years ago

        Or you could use a DVI-VGA adaptor, assuming the connector is DVI-I on the motherboard.

          • Bensam123
          • 8 years ago

          All the boards in this review are DVI-D.

            • derFunkenstein
            • 8 years ago

            True, but that can be changed – just re-purpose the appropriate traces from the 15-pin D-sub connector.

            • Bensam123
            • 8 years ago

            You’d lose the ability to power two displays then… That aside I’m not entirely sure how DVI>VGA mode works, I’m pretty sure it involves more then just resoldering wires as there are only four extra pins on a DVI-I connector.

            • derFunkenstein
            • 8 years ago

            I don’t see why you’d lose the 2nd display, at least theoretically. I could see you’d lose it if you had 1 DVI and 1 VGA monitor and no other alternatives. But that’s true to an extent now – you can’t have more than one HDMI or more than one DisplayPort without adapters.

            • Bensam123
            • 8 years ago

            If the DVI port is occupied by a DVI to VGA adapter, you can’t use the DVI port and vice versa.

    • MadManOriginal
    • 8 years ago

    Pg 13 –

    1) Any chance you could make some purty graphis out of this data going forward? (Aside from maybe the audio quality tests.) A chart of numbers doesn’t lend itself to comparison as easily as a nice graph.

    2) Serial ATA Performance as the heading for the Ethernet performance section.

    • codedivine
    • 8 years ago

    How does the USB 3.0 performance compare to AMD’s A75? I am suspecting the Intel solution is faster, given that in the A75 review the performance of A75 seemed only slightly better than ASMedia, but here a different test is being used so would have liked to see the numbers.

    • Duck
    • 8 years ago

    All of these waste too much money. I Want to see boards based on Z75 with no 3rd party controllers at the very least.

      • Nevarre
      • 8 years ago

      Then look for the MSI Z77A-GD55. It’s the cheaper little brother, but still has the core features of Z77.

      Honestly that was something missing from the review. If you’re price-sensitive, that’s the board to put up against the Gigabyte for the $170 price point. The lower-end MSI gives up the 3rd party SATA 6gbps controller, the on-board diagnostics and a few other minor features, but should otherwise be quite comparable to both its big brother and the Gigabyte.

      I haven’t personally seen any Z75 boards, but Z77 and H77 are both readily available. Z75 adoption is lagging, with newegg having one lone Biostar board right now. I’m not convinced that Z75’s going to be cheap enough to offset the few feature differences with Z77.

    • derFunkenstein
    • 8 years ago

    These things honestly all make me a little apprehensive. Why are they bothering with extra USB 3.0 controllers? As expected, Intel’s solution is more than a tiny bit faster, and some peripherals are finicky with certain USB 3.0 controllers (particularly stuff that relies on relatively low latencies), as well. Quite literally, a waste of money.

    MSI gets bonus points for not throwing in this needless extra. I’m sure ASUS’ and Gigabyte’s lower-end offerings will exclude them as well.

      • spuppy
      • 8 years ago

      Complaining about TOO MANY USB 3.0 ports?? That makes zero sense, sorry

        • Visigoth
        • 8 years ago

        Well f*cking said.

        • Duck
        • 8 years ago

        Waste of bloody money. All those 3rd party controllers cost money you know.

          • derFunkenstein
          • 8 years ago

          That, and the fact that they’re actually slower than the Intel ones, is my point.

        • derFunkenstein
        • 8 years ago

        It’s not the ports, it’s the extra controller. Totally nonsensical. They’re slower than the native ports and they cost extra.

    • thill9
    • 8 years ago

    So very anticlimactic. Guess that makes for one less read on the real launch day.

Pin It on Pinterest

Share This