POST A COMMENT

58 Comments

Back to Article

  • ShieTar - Thursday, June 27, 2013 - link

    Is there a special reason not to test the POST times and DPC latency of the Gigabyte Board? Its power consumption is quiet impressive, and whatever design measures have been used to achieve it do not seem to negatively affect the overall performance. So it would be interesting to complete the picture with the two measurements which are missing. Reply
  • IanCutress - Thursday, June 27, 2013 - link

    DPC Latency on the Gigabyte during testing was jumping around a fair bit, hitting 800+, though that is more likely due to the early BIOS revision. I need to run the POST test (as the results are strangely missing from my database) as well as the DPC test on a newer BIOS. Since I started testing almost every manufacturer has released newer BIOSes (as is always the way coming up to a launch) and I really have to lay the hammer down as testing a whole new BIOS takes a good 30 hours or so start to finish, so when I'm locked in that's it. That in a way does give an unfair advantage to the board I test last, but there's not a lot else I can do. I am still getting emails of BIOS updates for these boards as of yesterday.

    Ian
    Reply
  • tribbles - Thursday, June 27, 2013 - link

    Am I wrong in thinking that Gigabyte hasn't been doing well in the DPC Latency Test since Z77? If so, that's kind of surprising, since Gigabyte seems to be a "go-to" brand for digital audio workstation builders. Reply
  • IanCutress - Wednesday, July 10, 2013 - link

    I retested the UD3H on the F5 (public) BIOS, and it scored 164. The two next boards I have in for review got 160 and 157, which points fingers to the DPC on Haswell being 150+ regardless of motherboard. This might be a fundamental issue. Reply
  • Timur Born - Saturday, July 27, 2013 - link

    Run Prime95 (or turn off CPU power features) while measuring DPC latencies to see how much CPU power saving features affect DPCs. Reply
  • jhonabundance - Thursday, August 28, 2014 - link

    great share http://asus.com Reply
  • jhonabundance - Thursday, August 28, 2014 - link

    this is the best share http://asus.com Reply
  • Rick83 - Thursday, June 27, 2013 - link

    Maybe a bad choice to use two different ways of graphing the Rightmark results.
    Being consistent with regard to cutting off the irrelevant bit of the graph makes it a much easier read.
    Now it appears at first glance as though the Gigabyte board is much better in THD+N, simply because the differences were so minuscule in the dynamic range bit.

    On another note: Shouldn't it be more interesting to use a standardized input instead of the input of the board? In the current protocol a good output could be handicapped by a bad input, and conversely. For most users the output is much more important than the input, so it might be better to test it independently? I would recommend using a USB soundcard as an easy means of doing this test on the same machine, without changing the setup protocol too much.

    And finally - I seem to remember Rightmark results for earlier reviews - it would be interesting to have those (or maybe a reference soundcard?) as comparison in the same graph. After all, for DPC you maintain a large cross-platform table as well.

    Nice thorough initial review, those nitpicks withstanding.
    Reply
  • IanCutress - Thursday, June 27, 2013 - link

    Unfortunately I can't adjust the engine to represent from 0 for negative values, I don't have access to the low level options. I forwarded it on as an issue.

    I'd love to use a standardized input with RMAA. I guess it would be good to get a sound card with an input that supercedes the output of the motherboard and put it through that way, and hopefully there won't be a driver conflict along the line. I'll see what I can do to get in the hardware for that, although many soundcards are designed more for output and the input dynamic range/distortion might be the limiting factor as is the case on motherboards. Something like the Xonar Essence STX has a 118 dBA input with -113 dBA THD+N which might be a good starting point.

    Our RMAA testing for Z87 has changed a little from Z77 to make it more of an efficiency test rather than an out-the-box test as audio is such a varied playing field. RMAA is very sensitive to certain windows settings and volumes for example such that with the right combination it was very easy to show A>B or B>A depending on how the OS felt it should be set up. The new testing regimen for RMAA should iron out those issues but the results are not exactly comparable to Z77 for that reason. There are so many wrong ways to set up RMAA it can be difficult (and a learning experience) to get it right.

    Ian
    Reply
  • popej - Thursday, June 27, 2013 - link

    Nice to see you are planning steps in right direction. Using reference card for measurements is a proper solution. Be aware, that separate card add complications to the test, for example you will have to take care about ground loops and signal level matching. Professional card with balanced input could help a lot. Reply
  • Timur Born - Saturday, July 27, 2013 - link

    Balanced inputs would only help when the outputs to be tested offered balanced outputs, but you won't find these on mainboard solutions. Still a professional solution might be preferable, because don't just concentrate on listing theoretical spec numbers of the converter chips, but also make sure to get the most out of it in practical implementation.

    The English ASUS site lists little (and partly wrong) information about the Xonar Essence STX for example, there is much better information on the German site, though. Still they list frequency response at -3 (three!) dB points, while usually you would choose -0.5 dB or -1 dB points to give a real picture.
    Reply
  • repoman27 - Thursday, June 27, 2013 - link

    While the Intel block diagram for the DZ87KLT-75K does appear to show 20 lanes of PCIe 3.0 coming from the CPU, the 8 PCIe 2.0 lanes from the PCH are used for the GbE controllers (2 x1), Thunderbolt controller (1 x4), the PCIe mini card / mSATA slot (1 x1), and a "PCIe Hub" (1 x1). We can presume this is really a conventional PCIe switch, and the Marvell 88SE9172, the 3 x1 slots, and the PCIe to PCI bridge are all connected to that.

    What's the deal with FDI now that there are display connections coming directly from the CPU? The block diagrams still show FDI and the Thunderbolt controller being fed DisplayPort from the PCH. Are the CPU display outputs DP 1.2 and the PCH connections still DP 1.1a?
    Reply
  • repoman27 - Thursday, June 27, 2013 - link

    Sorry, I should have kept reading before commenting about the display interfaces. Reply
  • repoman27 - Thursday, June 27, 2013 - link

    The article does however say, "leaving the VGA behind", which should read, "leaving LVDS and SDVO (like anybody cares) behind." The PCH does still support VGA via the FDI x2 link as long as those lanes aren't being co-opted by Port D for eDP.

    Also, I meant to point out in my original comment that the additional USB 3.0 ports on the Intel board appear to be provided by a pair of USB 3.0 hubs, not a discrete controller.
    Reply
  • Jaaap - Thursday, June 27, 2013 - link

    Great Review Ian.
    I'd also be very interested in the minimal power consumption of Z87 motherboards without videocards and a PSU efficient at low powers (a PicoPSU or a light Seasonic).
    Reply
  • IanCutress - Thursday, June 27, 2013 - link

    I think Anand hit 34W idle on his 4770K with IGP?
    http://www.anandtech.com/show/7003/the-haswell-rev...

    I have 500W Platinums for mini-ITX reviews, but I that might be too much for IGP idle at sub-10%.
    Reply
  • DanNeely - Thursday, June 27, 2013 - link

    500W probably is too much for anything other than relative comparison scores. IIRC most of the 80+ standards only require specific efficiency levels between 20 and 80% loads.

    ex The 400W Seasonic X-400 2 is 89.5% efficient at an 86W load; but only 82.6% at 37W.

    http://www.jonnyguru.com/modules.php?name=NDReview...
    Reply
  • igxqrrl - Thursday, June 27, 2013 - link

    Am I the only one confused by the video out options?

    I'm looking for integrated graphics that can drive 2x30" (2560x1600) or 2x27" (2560x1440) displays. Can any of these motherboards do that?
    Reply
  • repoman27 - Thursday, June 27, 2013 - link

    The Gigabyte, Asrock and Asus boards theoretically can, however in each case you'll need to drive one of the displays via the HDMI output. AFAIK there aren't a ton of displays that support 2560x1440 or 2560x1600 over HDMI. Reply
  • Earballs - Thursday, June 27, 2013 - link

    The max resolutions are listed with typical refresh rates, but not max refresh rates. Just thought I'd throw that out there while talking about confusing video out options. Reply
  • Rob94hawk - Thursday, June 27, 2013 - link

    The MSI Gaming board would not do 2560x1440 off the DVI-D port. Reply
  • Aikouka - Thursday, June 27, 2013 - link

    One thing to note is that ASRock does have a version of the Extreme6 that does NOT have 802.11ac built in. I think it's about $20 cheaper.

    I thought about going with the Extreme6 when I built a Haswell machine, but I figured that some of those features would go unused. Unused chips that aren't disabled still need to be initialized during POST, so I get an increased POST time and spend more money. :P I went with the Extreme4, and honestly... I wish that I hadn't. The UEFI BIOS on my system is so flaky that it randomly hard locks a few seconds after the GUI appears. I don't know if it's a bad board, a bad batch, or an issue with that model, because I'm not the only one with that issue. I've seen other reports on the Anandtech forums and a quick Google search turns up results on other forums and in the Newegg reviews. I'm assuming that the Extreme6 did not have this problem?
    Reply
  • DanNeely - Thursday, June 27, 2013 - link

    Check your BIOS; you should be able to disable unused controllers to boost post time. This shortened my i7-920/930 post times from ~30s to ~20s. Reply
  • Aikouka - Thursday, June 27, 2013 - link

    Oh, I know. I just figured that it wasn't worth spending more on a board just to disable the unused controllers. Although, then I ran into the Extreme4's BIOS problem, and I wish I would have spent a little more! =P Reply
  • James5mith - Thursday, June 27, 2013 - link

    Since when is $200 a "reasonable price bracket"? In my mind $50-$100 = budget. $100-$160 = mainstream. $160+ is enthusiast pricing. It is not reasonable at all. Reply
  • WeaselITB - Thursday, June 27, 2013 - link

    $200 is a reasonable price bracket for the enthusiast chipset family that they're reviewing. If you're looking for lower price, check out the lower families, such as H87 for the mainstream segment. Reply
  • MrSpadge - Friday, June 28, 2013 - link

    A solid Zx7 board without any BS used to be around 100€ (-> 100$ without VAT for you). Now that parts of the voltage supply have transitioned into the CPU the boards should not become any more expensive.

    I may be enthusiastic about tweaking my PC for efficiency and value, but I'm certainly not enthusiastic about paying big bucks for my toys.
    Reply
  • crimson117 - Thursday, June 27, 2013 - link

    Ian, this line/paragraph sounds a bit too much like it was lifted from marketing materials: "Whenever a user considers purchasing an ASUS motherboard they know that it is built to perform and has a great feature set" Reply
  • IanCutress - Thursday, June 27, 2013 - link

    Haha I can assure you those are my words, although it does sound generic given that I was meant to be referring to the board in the review. Perhaps a bit subjective for a review, but I had a good experience with the board and that filtered through. Changed it a little...

    -Ian
    Reply
  • snakyjake - Thursday, June 27, 2013 - link

    I've had an ASUS mother board for years, and the EZ-Update never downloaded updates from the Internet. It never was able to connect with any of the update servers. I always had to go to the Asus website and download the updates, and then update the BIOS from disk.

    I do wish the ASUS software was all integrated, and not separate programs on my system tray.
    Reply
  • clyman - Tuesday, November 26, 2013 - link

    In my personal experience and from ASUS tech support, the safest way to update bios is by first downloading it. This mobo has an excellent update program in bios that only looks at local drives, not online. I found it quite simple at each bios update. Reply
  • silenceisgolden - Thursday, June 27, 2013 - link

    Great job Ian, really looking forward to the super high end motherboard review as well! I'm curious though, did you use the VGA port at all in any of these reviews, and also when was the last time you used a VGA port? Reply
  • IanCutress - Thursday, June 27, 2013 - link

    My Korean 1440p panels are all via DVI-D. But the VGA comes in use when you have to use a DVI-I to VGA converter for DVI-I cables. Otherwise you need a DVI-D cable. Reply
  • JeBarr - Thursday, June 27, 2013 - link

    Great review Ian. I very much appreciate your taking the time to explain the PCIe slot assignment and repeating the less-than-x8-no-good-for-SLI truth that needs to spread far and wide.

    What I took away from this review when analysing my own usage patterns and component choices is that each of the boards in this review would be better suited to the mATX form factor. There is no point in purchasing these mid-high end boards for multi-GPU, multi-display configs. There are only a handful of full size z87 boards that in my opinion earn their full-size status.
    Reply
  • Shadowmaster625 - Thursday, June 27, 2013 - link

    If it has fully integrated voltage regulators then why in blazes does it still need $20 worth of components buried beneath $3 heatsinks surrounding the cpu socket? Yeah, that's what I thought.... Reply
  • DanNeely - Thursday, June 27, 2013 - link

    What FIVR means to Intel isn't quite what you think it means. What's on package is the collection of VRs needed to provide power to all the separate sections of the CPU and to vary them as the chip clocks itself up and down. The motherboard itself still needs to do the heavy lifting to convert the 12V from the PSU to the voltage used by the DRAM and to a single input voltage the that CPU converts to the other levels it uses internally (IIRC this is the full load core voltage).

    I believe the reason why those regulators can be squeezed into the package while the ones still on the mobo are much larger is that the uncore/cache have relatively low power levels and the lower core voltages are only used at low CPU loads and thus don't need to push nearly as high of a peak current level.
    Reply
  • WeaselITB - Thursday, June 27, 2013 - link

    Wonderful review, Ian. Any chance of a Z87 vs Z77 vs ... comparison chart? It seems like functionalities of the processor are highly publicized, and the individual motherboard reviews chart the differences between chipset models within that family, but I don't recall seeing a comparison between chipset families. I know the chipset seems to be taking a smaller and smaller role these days, but it would still be helpful to exactly see the differences between generations.

    Thanks!
    Reply
  • Kougar - Thursday, June 27, 2013 - link

    The only downside for me is that it unfortunately does not work with Korean 1440p monitors (!), but Gigabyte is looking into this.


    It doesn't work with my 30" U3011 monitor either, yet another forum user with the same monitor didn't have any problem. Only difference was he used a Radeon and I use a GTX 480.
    Reply
  • Creig - Thursday, June 27, 2013 - link

    I only had time so far to skim this article, so I apologize if this question was already answered in the body. Is there any way to find out when each board starts shipping with the C2 stepping of the Z87 chipset? As some people are already aware, the C1 stepping has been shown to have issues with certain USB 3.0 controllers disconnecting when the computer awakes from sleep mode. The C2 stepping is apparently already shipping to manufacturers, but it would be handy to know a way to ensure that a person who orders a board in the near future receives the updated chipset. Reply
  • blackie333 - Friday, June 28, 2013 - link

    I'm not sure that C2 stepping shipping already started, according to public available plan from May only samples have been sent to manufacturers. Mass production of C2 should start on 1st of July and manufacturers should start receiving them from the 30th of July. Boards based on C2 stepping should be available for end users from middle August.
    But you maybe have some more actual/insider information.
    Reply
  • Mr Perfect - Friday, June 28, 2013 - link

    I was wondering the same thing. If all shipping Haswell boards have the faulty USB3, then this is a non-starter. Reply
  • Avalon - Thursday, June 27, 2013 - link

    Is it just me, or are these boards too expensive?

    The Asrock Z77 Extreme6 is $155 on Newegg, $169 w/Thunderbolt. Asrock Z87 Extreme6 is $220-$20 MIR.

    The Gigabyte Z77 UD3H is $140 on Newegg. Gigabyte Z87 UD3H is $180.

    You get a couple extra USB 3.0 and SATA 6Gb slots and Haswell support, but I don't understand how that makes mid range boards at best command low high-end prices.
    Reply
  • Mr Perfect - Friday, June 28, 2013 - link

    My guess is it's just new vs old products at this point. The Z77 boards are old news and have had a year to fall in price. Meanwhile, the Z87s are shiny new toys that some people will pay a premium for. Reply
  • Rob94hawk - Thursday, June 27, 2013 - link

    "As it stands the MSI BIOS looks like a higgledy-piggledy jumble to a new overclocker."

    Going from X38/775 to this I still haven't figured out what everything does.
    Reply
  • nsiboro - Friday, June 28, 2013 - link

    Ian, kindly provide info/links to ASUS Z87-Pro 3xxx series BIOS.
    The website product page (download) is only showing 1xxx series BIOS.

    Thanks.
    Reply
  • blackie333 - Friday, June 28, 2013 - link

    Could someone please check/confirm whether USB 3.0 S3 wakeup bug also affects devices connected via additional(ASMedia 1074) onboard USB 3.0 hub ports available on Asus Z87-PRO stepping C1 board?
    Some people are suggesting that only USB 3.0 ports directly connected to Z87 chipset are affected by the bug.
    Reply
  • chizow - Friday, June 28, 2013 - link

    The PCIe lane config was the biggest deciding factor for me. I will only ever run 2-way SLI, so I wanted to maintain x8/x8 config for my 2x primary GPUs but wanted the flexibility of that 3rd slot for a PCIe SSD or PCIe PhysX card.

    Only the Asus and Gigabyte options offered that lane config, from what I saw both the MSI and Asrock designs go with x8/x4/x4 3.0 rather than x8/x8 3.0 + x4 2.0

    The Gigabyte UD range was pretty vanilla, but I was OK with that, the Asus boards, although solid, offered a lot of features I would never need or use, like Wi-Fi.

    I ended up with the Gigabyte Z87X-UD4 as it was cheaper than the comparable Asus offering Z87-Pro by quite a bit.
    Reply
  • pandemonium - Saturday, June 29, 2013 - link

    I love the thoroughness of these articles. Excellent job, guys!

    "ASUS’ reasoning is such that some of the Haswell i7-4770K CPUs, the ones that only just get into this category, will throttle the CPU speed when using the default Intel CPU cooler when MCT is enabled."

    Who - buying a 4770K - will be using a stock cooler? What kind of rationality is that garbage? >.>
    Reply
  • blackie333 - Saturday, June 29, 2013 - link

    There can be some, maybe those waiting for a better cooler. But the question is why Intel is including cooler which isn't capable to cool the CPU and we still have to pay for it? It should be able to do it's job at least on default frequency.

    Anyway this problem is IMHO more an effect of Haswell heat transfer issue than poor quality of the stock cooler. If Intel could fix the CPU overheating issue the cooler should be good enough.
    Reply
  • ven - Sunday, June 30, 2013 - link

    why is that PCIe hub is present, many will prefer having a single device that will utilize all the bandwidth than having multiple devices choking with shared bandwidth, six SATA 6Gpbs is enough for most, with flex i/o and that hub removed gives x7 lanes and given this a Desktop board,msata will not be missed that much, so we can get tri-way SLI, i am little surprised that no manufactures choose this configuration. Reply
  • HexiumVII - Sunday, June 30, 2013 - link

    Hey Ian! Maybe run the Asrock Z87 in water and let us know how waterproof it is! Reply
  • Hrel - Monday, July 08, 2013 - link

    You guys DO talk about how motherboard manufacturers have less and less to do now; with Intel moving more and more things onto the CPU die. Yet for some reason you still fail to ask WHY motherboards are getting MORE expensive. Not less. I got an AMAZING motherboard with SLI and massive overclocking ability that's rock solid back in 2007 for 125 dollars. Yet now that the northbridge isn't even ON the motherboard anymore I have to pay 175 for the SAME level of performance... WTF is that! Reply
  • adridu59 - Monday, July 08, 2013 - link

    It looks like you are playing the marketing game, because ALC1150 (nominally ALC900) is just a tweaked ALC898 (same as ALC889 and ALC892).

    More info: http://www.overclock.net/t/1398739/is-realteks-alc...
    Reply
  • Rafalus - Sunday, July 28, 2013 - link

    Hi, Is it possible to stop TPU tuning eg when it reach 4.5GHz as I did not want to raise it higher? Reply
  • SilentRyder - Tuesday, October 22, 2013 - link

    Honestly i believe this integrated Voltage Regulator will cause many problems. If we check the previous Ivy Bridge main boards, They do have a huge power regulator designed on the top side of processor socket. Now the whole idea of power supplied from the mainboard is gone off. May be this technology will benefit laptops or any other mobile devices. but this is certainly a bad idea for desktop users.

    We would certainly have HEAT problems, which i am truly afraid of. I do not think haswell will support enough for overclocking. i never over clock my self but i am sure INTEL will change this concept on its new generations.

    Even laptops running in high temperature can be a issue.

    How many of us are satisfied with the intel built in Graphics processor?
    Reply
  • clyman - Tuesday, November 26, 2013 - link

    So far, it meets my needs just fine. I am sure it would meet the needs of all my customers. I did put a fluid filled cooler on the processor, but that was only needed while running OCCT. I will add a video card should i ever need one. Reply
  • clyman - Tuesday, November 26, 2013 - link

    I have the ASUS Z87-PRO mobo and was hoping someone had a few answers here for me, haven't seen anything related though. My problem is that no matter what I do, the multiplier will not go above 39 and I cannot find out why. Is that due to having 1600 MHZ memory?

    Another issue is that when I update AI Suite 3 from the original on the supplied CD, it will not recognize my WIFI Engine adapter, however it finds it with the original. ASUS techs have been useless on both of these points as they say have no information regarding overclocking and no one has a clue about the WIFI Engine. I have been given all kinds of advice, all fruitless. I can't get them to understand it simply will not recognize the device on the updated version. I think it is a bad update, they want to RMA it.

    Any help on these issues would be appreciated.
    Reply
  • LoCk3d - Friday, December 20, 2013 - link

    A problem encountered someone USB Charger + function? I do not detect the device, I enabled ERP in bios but still does not work. Help me please ! Reply

Log in

Don't have an account? Sign up now