News - ASUS teases upcoming Z87 Haswell motherboards

[HEXUS]

Does the word “Gryphon” mean anything to you?

Read more.

[krazy_olie]

If my motherboard tries to fly off I'll be asking for a refund

[crossy]

The ASUS Gryphon Z87 Rev 1.01 motherboard is pictured above. This motherboard is the successor to the Z77 Sabretooth. Following this teaser image several ASUS fans voiced their disappointment on Facebook that they have just bought Z77 Sabretooth boards. ASUS replied, that is why it has published these early teaser images; to let fans see what’s on the horizon.

Can't see why the "fans" are disappointed - after all, it's not as if Haswell's availability this year was a big secret. In which case surely it's obvious that there'd be a successor to Sabretooth?

Hmm, Z87 Deluxe could worth saving for - my PhenomII 1090T is beginning to look a tiny bit long in the tooth. Intel cpu or AMD replacement is up for grabs, but I know that the board hosting that processor will say "ASUS" on it.

[Willzzz]

Exactly, any fan that is disappointed is quite frankly an idiot. A Z87 board is of no use until Haswell is released.

[harryj1995]

Sweet, Can't wait until Haswell comes out

[zoomee]

So I'll assume these will be released with the dodgy usb3 chipset problem? - Can't see intel getting a new spin and these boards being released in time. Sounds like another early sandybridge debacle waiting to happen.

[Willzzz]

I thought that was going to be fixed before release?
Not that I'm personally very bothered by it.

[Agent]

The bug affects the USB 3.0 ports on a Haswell machine; when your machine wakes up from an S3 sleep state the bug means that you may have to disconnect and reconnect a USB 3.0 device to access it once more. It’s not a major thing but could be quite an annoyance.

Are you kidding? Given that the default for almost every machine sold today uses S3 in some way or another, this is a *major* issue.

It's the entire temporary failure of a component until you intervene. How is this not major in any way?

[Andi-C]

"with the head, talons, and wings of an eagle and the body of a lion" This going to be one ugly Mobo...

[crossy]

Are you kidding? Given that the default for almost every machine sold today uses S3 in some way or another, this is a *major* issue. It's the entire temporary failure of a component until you intervene. How is this not major in any way?

I think you're panicking unduly. From the article:

In related new Fudzilla has reported that Intel will be making a new chipset revision to address a hardware problem on Haswell motherboards

and

The bug affects the USB 3.0 ports on a Haswell machine; when your machine wakes up from an S3 sleep state the bug means that you may have to disconnect and reconnect a USB 3.0 device to access it once more.

(my underline) Note that you "may" have to disconnect and reconnect - not necessarily that you will. And it's a problem that Intel is already in the process of addressing.

Maybe I'm being naive but I can't really see the major mobo manufacturers being happy to accept known-faulty parts just so they can get out on launch day, (because they'd then have to process a good few RMA's and there's damage to reputation also). We're in March and I though Haswell wasn't supposed to launch until June - which I'd guess/hope is enough time to get a new revision done and out to the OEMs?

From my own limited experience - the devices that need USB3.0 speed tend to be external disk crates, in which case I'd politely suggest that if your machine goes into S3 sleep then it's probably not doing anything, so having to unplug/replug your external HDD is merely going to be an annoyance rather than an "unfit for purpose".

The Fudzilla article linked to also categorises this issue as "nothing major".

[Agent]

I think you're panicking unduly. From the article:

I'm not panicking, It's just annoying to see issues like this being sugar coated given the costs of motherboards these days. There is a fault at the hardware level - plain and simple.

(my underline) Note that you "may" have to disconnect and reconnect - not necessarily that you will. And it's a problem that Intel is already in the process of addressing.

The language really means very little. Take Intels B2 chipset recall: "Intel Chipset With Potential Problem" from their own site, even though they knew exactly what the issue was and that degradation would happen as a result. You can label anything as "Potential" or "may" if you find that one extreme situation. "You may not get this bug if you don't use the USB3 ports" for example. It's used all the time by the multi-million/billion dollar companies.

We're in March and I though Haswell wasn't supposed to launch until June - which I'd guess/hope is enough time to get a new revision done and out to the OEMs?

Some places are saying they won't get new revisions in time. Others are saying OEMs will only get new silicon. I guess we'll find out on launch

From my own limited experience - the devices that need USB3.0 speed tend to be external disk crates, in which case I'd politely suggest that if your machine goes into S3 sleep then it's probably not doing anything, so having to unplug/replug your external HDD is merely going to be an annoyance rather than an "unfit for purpose".

Well S3+other management is often the default on systems, depending on the OEM. You hit standby and the PC 'powers off' so no fans are spinning? That's S3. It's not so much a case of consciously choosing to use standby over shutdown now. The vast majority of people I know who have a laptop just close the lid to 'turn it off'.
That to one side - what if that HDD is your backup drive? or a NAS with external storage? or you do realtime recording / streaming and you need USB3? The people who are using USB3 are not solely HDD users, there are quite a few more USB3 products out there now and emerging. I'm currently using 3: External HDD * 2 and a Bluray RW. Unplugging and re-plugging all 3 devices after each standby seems a bit barmy, given that Intel are supposed to have a reputation for stability. My machine is under the desk, probably like a lot of people, so there is no chance I'm clambering around on the floor each time.

It's also worth noting that this hits the ports, not the protocol you negotiate over that port. It appears that USB2 devices on USB3 ports will also be hit, unless this is very bad wording. Keep in mind that internally, a lot of laptops have devices mounted on a USB bridge chip. Things like Bluetooth and TV cards often use bridge chips.

The Fudzilla article linked to also categorises this issue as "nothing major".

Fudzilla
You can't honestly quote Fudzilla with any seriousness? They are the hardware guys equivalent of the Onion

Either way, there is no clear indication on if this will hit end users, that's all that really matters. I wonder if HEXUS could find out?

[crossy]

It's just annoying to see issues like this being sugar coated given the costs of motherboards these days. There is a fault at the hardware level - plain and simple.

Utterly agree, in this case there's no such thing as "slightly wrong", it's either working or it's faulty.

Some places are saying they won't get new revisions in time. Others are saying OEMs will only get new silicon. I guess we'll find out on launch

Maybe I'm still being naive but I've got to wonder if - as I said in my original posting - that motherboard manufacturers will chose to ship with "known bad" chipsets merely to get product out there on launch day. Personally speaking I'd see that as a pretty cynical practice, and I'd certainly avoid any product that doesn't claim to have the updated/fixed chipset.

Then again, maybe we're all worrying unduly - if the Intel-supplied fix is already to hit production then maybe this'll be a non issue. If I was Asus, Gigabyte, etc I'd think it worth losing that "Launch Day" cachet just to ensure that I got a fully stable end product. In which case there's maybe even PR mileage in pointing this out, e.g. "Get your Asus Z87 Deluxe on pre-order - the board with the most up to date Haswell chipset"

Either way, there is no clear indication on if this will hit end users, that's all that really matters. I wonder if HEXUS could find out?

+1 on this. At least you're more likely to get this on Hexus than on The Register.

[Willzzz]

Well everyone can make their own decision, if you want to wait that's fine.

Even if the fix isn't in time for launch would you really expect them to throw away millions worth of stock for a bug that 99% of users won't even notice?

[kompukare]

Though I heard that Intel is demanding that it's 'partners' have to accept this fault if they want parts allocation. Which all sounds like an abuse of their dominance - but 'parts allocation' is the big weapon which Intel uses to bully any 'partners' who do anything Intel don't like (like pushing good laptop design with AMD APUs etc.).

March>June is still a long time but I suspect that if Intel cannot guarantee a new revision in time, motherboards manufacturers will have no choice but to not use Intel's USB3 and will have to include a third-party USB3 chip in any USB3 capable boards.

Still it does demonstrate something: even if you're Intel and able to outspend AMD 10:1 on R&D, this doesn't meant no mistakes. In fact, that's three major bugs in three years. In order of importance/cost SB SATA-III was the worst, this looks like being #2 and #3 was the mysterious disappearance of VT-d in the first stepping of SB-E. All were handled differently too. SB chipset was very major and handled very well. SB-E was handled by almost hiding the problem (although people who asked nicely did get stepping 2). See how this is handled. At the minimum you'd think Intel would reduce the price of their chipsets to their 'partners' to cover the cost of having to buy a third-party USB3 chip. Guess this is one reason why AMD hasn't released any new chipset for AM3+ for ages: no changes = no problems.