Looking for any "comparison observations" about two 1155 mobos

[BonzaiDuck]

My sig-rig runs an ASUS P8Z68-V pro 1155 board.

I have a new AsRock Z77 Extreme4 (1155) board I recently acquired to replace another system in my house.

The ASUS board is a damn good over-clocker, and I'd chosen it early after its release as a gamble which paid off. I chose it also because of its phase-power design specs.

Back in the early 1155 days, some of us were comparing our OC settings on different Z68 boards. Some folks were using the Z68 equivalent of the Extreme4. They weren't complaining, as I recollect.

How does the AsRock Z77 compare as a second-gen successor to those Z68 boards? I couldn't afford the luxury of choosing the board on "phase-power" criteria or "extra features," because there are fewer and fewer of these Z77 (1155) boards available. I couldn't even find a source for "brand-new in the box" ASUS P8Z77 boards, so I gave up on that angle.

 

[BonzaiDuck]

Wey-ullll!! 124 folks read this post, and nobody answered. S-alright. I was just hoping someone had one of these boards and they could share observations or insight. . . .

 

[Diogenes2]

I have the Z77 Extreme4 with a 2700K with a mild OC of 4g.. It been running essentially 24/7 trouble free for over a year.. It is my wife's PC - mostly surfing - but I keep Folding at home running on the GPU ..

My experience with several Asrock boards has ben pretty much trouble free.

I got turned on to Asrock when I set up a 3930K system, and they were running about $100 less than competing boards with less features....

 

[schmuckley]

With AsRock,It's OCF or nothing.
Extreme4 is a bottom-tier z77 board while
Asus PRO series is upper-mid range.
The Asus is higher quality.
(says the guy with z77 OCF)

 

[PhIlLy ChEeSe]

I had a Z77 Extreme6 with a 3570K, not sure if I'm spoiled or what but I had a lot of issue's with it.
So you know what my suggestion will be here, call me a fan boy. I'd rather spend more and have less issue's if that makes me a fan boy then so be it. I got a maximus V gene that had a few bent pins, straightened them out. Boom my 3770K under water in it, I do think that AsRock makes good board! The one i had, I guess was not made for what I wanted out of it (EXTREME)OC. I hear nothing but good stuff about there entry tier boards from people who just pop in a normal CPU and off they go........
Oh and do some research on the AsMedia controller as they seem to make or break a party.

 

[john3850]

I consider the ASRock Z77 Extreme4 a poor oc even with water it is only stable up to 4.6MHz with 1.304v with a 3770k.
The load-line calibration must be maxed out to avoid any v droop and Whea errors.
If I increase the vcore the board will run at x47.

 

[escrow4]

Asrock boards are fantastic when you keep their limitations in mind - cheaper boards are cut back (there is a chunk cut off the ATX spec say) and the VRMs may be equally cheap:

http://www.reddit.com/r/buildapc/comments/1enn0h/psa_dont_get_an_asrock_z77_extreme4/

That said, I have an Asrock X99 Extreme 4 and it works perfectly and their Z87 range is decent too. Asrock may be no frills but you are not paying for Asus's name at least.

 

[Diogenes2]

I consider the ASRock Z77 Extreme4 a poor oc even with water it is only stable up to 4.6MHz with 1.304v with a 3770k.
The load-line calibration must be maxed out to avoid any v droop and Whea errors.
If I increase the vcore the board will run at x47.

Which other board did you use to compare, to determine your CPU would clock higher on another board?

 

[john3850]

Which other board did you use to compare, to determine your CPU would clock higher on another board?

None

The only way I able to keep the board stable was to set C1E to off at x47 or higher.
For what ever reason that or my mb has trouble when jumping from 1600MHz or idle to full speed.
From what I read is that mb uses some poor PWM controllers.

 

[john3850]

http://sinhardware.com/index.php/vrm-articles/113-digital-vs-analog-pwms/74-analog-vs-digital-pwms

ASRock uses analog converters and Asus uses digital converters.

 

[BonzaiDuck]

http://sinhardware.com/index.php/vrm-articles/113-digital-vs-analog-pwms/74-analog-vs-digital-pwms

ASRock uses analog converters and Asus uses digital converters.

Well, it's good to hear all this stuff, just to avoid arousing my expectations too much. I suppose the P8Z77 boards would have been better (ASUS). I might have had a chance to pick up a P8Z77-M (mATX) or "Pro", but they may not "be there" anymore. I'll see.

All the good boards, or those I'd be more sure of -- seem to be "out of stock" and not likely to be re-stocked.

It's going to be interesting, though. I darn well know the features mentioned by respondents here were better in the ASUS implementation.

Even so! [dammit!] -- you'd think a second-gen socket-1155 would be better than a wider range of first-gen Z68's. But I suppose we "think" a lot of things, which, for various reasons -- may be unrealistic.

 

[BonzaiDuck]

I consider the ASRock Z77 Extreme4 a poor oc even with water it is only stable up to 4.6MHz with 1.304v with a 3770k.
The load-line calibration must be maxed out to avoid any v droop and Whea errors.
If I increase the vcore the board will run at x47.

And . . . since you have one, you have first-hand experience. We'd touched base before on a thread about the 2600K's back in '11. Of course, you're also rocking an IB on that sucker, so maybe I'll be lucky with the 2700K.

I don't like to eliminate all vDROOP. I've been lucky for keeping at least a 20mV droop under load-testing.

I only lack the CPU cooler -- due to get here next week. I became conflicted over lapping the processor, but I think I'll go ahead and grind it down to bare copper. Howsoever it may overclock, it's going to be "allocated and re-allocated" -- before ever trying to resell it.

If I wait too long, and something's wrong, I'll have to RMA to AsRock instead of the Egg. I'm not going to get worked up over it. Don't ever remember having problems with a new board that required return or replacement, but there's always a first, isn't there?

ADDENDUM: As I said, the Z77 offerings were sparse, and I didn't want to spend much on old tech. If I'd been more eager, a Maximus V board would have been a better choice -- even more sparse in the market-place now, and there aren't many bargains.

I may have found this HARDOCP review before, but can't remember.

http://www.hardocp.com/article/2013...e4_lga_1155_motherboard_review/1#.VGf5msJ0yUk


Like John3850, they ran their tests with an IB i7-3770K. Again -- I have some hopes with the 2700K, but we can only wait and see.

The review noted the PCB is "thinner than usual." I'm going to be putting a cooler that I recall had a weight between 500 and 600g on it, so I'm not too worried. I was a bit puzzled that the reviewer was not pleased with result from mounting a Koolance water-block on the Extreme4. But it would just seem to me that a tower-cooler secured with a backplate should not be a problem with the board, especially if I'm not even going to hang a fan on the cooler.

We can only X our fingers and see how it shakes out. Any other thoughts?

 

[john3850]

I was forced to use the lowest load-line calibration or # 1 to go above x44 with a new 750 pc power and cooling ps.
If I lower the LC to #2 I get whea errors at the same vcore.
Always need 1.304-1.312v at x46 in IBT.
I run on the PLL voltage on low side at 1.78v on 3770k.
I never tried a high 1.91v PLL voltage but I will try it out soon.
The pc will run higher but it is not worth the extra volts.
Do to the boards odd size I used wooden shims were needed like close ps connection to help reduce the flex.
With sockets and revisions changing close to every year why waste money on top of the line boards unless your after the max oc.

 

[BonzaiDuck]

I was forced to use the lowest load-line calibration or # 1 to go above x44 with a new 750 pc power and cooling ps.
If I lower the LC to #2 I get whea errors at the same vcore.
Always need 1.304-1.312v at x46 in IBT.
I run on the PLL voltage on low side at 1.78v on 3770k.
I never tried a high 1.91v PLL voltage but I will try it out soon.
The pc will run higher but it is not worth the extra volts.
Do to the boards odd size I used wooden shims were needed like close ps connection to help reduce the flex.
With sockets and revisions changing close to every year why waste money on top of the line boards unless your after the max oc.

Before you do that, you might consider this. It seemed counterintuitive to some that LOWER PLL voltage would increase stability, but in some cases it does. I'm running mine right now at 1.68V.

But we're also comparing Navel Oranges to Valencias -- between my Sandy Bridge and the Ivy Bridge. Even so, I thought I saw where the lower PLL voltage settings seemed feasible for the IB. One guy had even dropped his to 1.55. But a friend of mine who'd worked in a specialty encountering PLL voltages told me that 1.6 seemed to be a lower limit, and that something higher was a margin of safety.

What did you OBSERVE as the vDROOP with lowest LLC or none?

I can't lose a residual suspicion for benchtest reviews, since they often seem forgivingly less critical than they need to be. There was also my observation about comparison reviews at Maximum PC: they cherry-picked the contenders; the "winner" often seemed to publish a two-page color advertisement within a page or two of the comparison -- smacking of something akin to "Payola-in-kind."

I agree about "wasting money," though. I'd been coveting the 2700K chip for the good part of a year, without intending to buy another board. Like someone who'd given up smoking who visits a tobacco shop every time he goes to the grocery to "say hello," I kept tempting myself with web-searches. I was really planning to try it in my P8Z68-V Pro. Then it occurred to me I have a household of upgrades to make, wanted to test the processor without disturbing my sig-rig at all, and started looking for something like an mATX P8Z77-type board.

There ARE places out there offering what appears to be "new" Maximus V boards for approximately what they sold for in 2012 or thereabouts. They . . . are . . . scarce, though! But -- it's like you say . . .

ANOTHER AFTERTHOUGHT: You didn't say whether you de-lidded that IB, but I'm also wondering if you just got a lesser prize in the chip lottery. Only you would know the particulars about that. And the de-lidding might not have made much difference in the lottery aspect. I'll just have to try out the Extreme4 with the new chip. I know I can beat an H110 for the cooling by a couple degrees, based on neck-and-neck comparisons with my existing rig. Won't know until the new $50 cooler arrives next week.

 

[john3850]

I never de-lidded my IB but if I had a i5 IB with a hsf I would.
Both my 2500k and IB will run at x47 and both are limmited by cheap ASRock mb.
Many people here claim extreme 4 and 212 hsf are great to me there barely decent but they are a good bang for the buck.
I only saw one review with a high cpu pll on IB so I will stay low lower.

 

[BonzaiDuck]

I never de-lidded my IB but if I had a i5 IB with a hsf I would.
Both my 2500k and IB will run at x47 and both are limmited by cheap ASRock mb.
Many people here claim extreme 4 and 212 hsf are great to me there barely decent but they are a good bang for the buck.
I only saw one review with a high cpu pll on IB so I will stay low lower.

Well, the possibility of more stability at lower PLL Volts is documented. There's also the promise of a slight drop in temperatures. There was a report of a "sweet-spot" somewhere around 1.65 to 1.68. It COULD depend on the mobo and processor, but I also thought I'd found it there. You'll have to give it a go on your own.

If there are certain heatpipe coolers that can match an H110, it gives me reason to stick with air until I build an "E" system. The CM Hyper 212 models may be "good," but there are better, even if the prices are higher. For this system with the 2700K I finally just broke down and decided to pick up the EVGA "SuperClock" or now called the "ACX" cooler. I only "discovered" it in the last couple days, but it had been around since just after the D14 release. The benchies show a 6C improvement over D14 performance. What does that mean, if my "mods" and refinements match an H110? The refinements will be easier on a more compact cooler like the EVGA.

I've got rock-solid settings for 4.7 on my 2600K, but I've yet to validate the RAM XMP settings with command-rate 1. Since the XMP profiles with "Auto" timings yield that 1T/1N or CMD=1 option, it's still on a to-do list.

Compared to my P8Z68-V Pro board, I don't think I would diss your AsRock boards. Did you set Phase and Duty Control to the Extreme settings on those boards? I did, but I don't change "Current Capability" above its stock 100%. I'll have to see about the LLC settings on the AsRock board, but on the Z68 Pro I have 20mV of droop with "High" and that's below "Ultra High" and "Extreme."

I can see what the reviewer is saying about the Z77 Extreme4. But I visited this question once "back when" with a comparison of the MIVE and the P8Z68-V Pro. You MAY get better performance out of "top-end," but you're also paying for extra features on "top-end" that you might not want.

It's been good with this thread to find out what I can expect with the Extreme4 board, but the only way to know for sure is to put the damn thing together and find out myself. I just really appreciate your input, though.

 

[BonzaiDuck]

Well, the possibility of more stability at lower PLL Volts is documented. There's also the promise of a slight drop in temperatures. There was a report of a "sweet-spot" somewhere around 1.65 to 1.68. It COULD depend on the mobo and processor, but I also thought I'd found it there. You'll have to give it a go on your own.

If there are certain heatpipe coolers that can match an H110, it gives me reason to stick with air until I build an "E" system. The CM Hyper 212 models may be "good," but there are better, even if the prices are higher. For this system with the 2700K I finally just broke down and decided to pick up the EVGA "SuperClock" or now called the "ACX" cooler. I only "discovered" it in the last couple days, but it had been around since just after the D14 release. The benchies show a 6C improvement over D14 performance. What does that mean, if my "mods" and refinements match an H110? The refinements will be easier on a more compact cooler like the EVGA.

I've got rock-solid settings for 4.7 on my 2600K, but I've yet to validate the RAM XMP settings with command-rate 1. Since the XMP profiles with "Auto" timings yield that 1T/1N or CMD=1 option, it's still on a to-do list.

Compared to my P8Z68-V Pro board, I don't think I would diss your AsRock boards. Did you set Phase and Duty Control to the Extreme settings on those boards? I did, but I don't change "Current Capability" above its stock 100%. I'll have to see about the LLC settings on the AsRock board, but on the Z68 Pro I have 20mV of droop with "High" and that's below "Ultra High" and "Extreme."

I can see what the reviewer is saying about the Z77 Extreme4. But I visited this question once "back when" with a comparison of the MIVE and the P8Z68-V Pro. You MAY get better performance out of "top-end," but you're also paying for extra features on "top-end" that you might not want.

It's been good with this thread to find out what I can expect with the Extreme4 board, but the only way to know for sure is to put the damn thing together and find out myself. I just really appreciate your input, though.

WELL!! Bad news . . . and good news.

I am REALLY GLAD I chose to put together the system with this AsRock board this week. I am REALLY MORE glad that the heatsink I ordered on Saturday arrived here this morning -- spurring me to get started.

I am REALLY SAD, though, at looking at the "brand new" products I'd ordered. I took a look at the AsRock processor contact pins. ONE is missing! OR -- it was jammed into the motherboard in an odd location -- next to another contact pin which was in its normal position. Further, it's height was higher than the other pins. I suspect some sort of assembly-line error. There was a Newegg customer review with a single egg where someone had noted the same sort of problem.

I am REALLY GLAD that I've got the RMA number from the EGG and a pre-paid UPS shipping label. I am GLAD that they are giving me a refund. And glad that I have maybe 14 days.

I'm neither GLAD nor SAD about something that probably belongs over in the CPU forum. Not only did I inspect the motherboard carefully with a magnifying glass -- I took a look at the processor.

This was BRAND NEW RETAIL BOX, SEALED with the white label. On the edge of the PCB, I noticed that a little tiny bit of the green surface at the edge -- smaller than a 64th of an inch -- was chipped off. No problem! I thought.

Then I looked at the gold contact pads. Near the chip, there is one pad which seems to be missing almost half the gold plate!!

I decided to install the processor in another Z77 motherboard -- slated for replacement of my Mom's machine -- and test it.

So far, I can get into BIOS and change settings. I had put a 320GB WD Blue drive in there for my Mom's system, but never installed a new OS. Turns out, it held a clone of her existing system. The system boots to the Windows password entry. At which time, I'd run the mouse over to the red button to shut down, because her existing LGA_775 system uses an NVidia chipset and drive controller.

HOWEVER -- these machines in my office/room are all hooked up to a Belkin 4-port PS/2/VGA KVM.

I am APPREHENSIVE. When I'm in the BIOS for the Z77, the mouse works fine. When I switch over to another machine and return, the mouse is FROZEN. The rest of the systems on the KVM continue to have mouse and keyboard function. The keyboard continues to work. I find if I let the system boot to the Windows password logon, the mouse COMES BACK. WHAT . . . is GOING ON . . . with THAT?!!??

I'm wondering if this isn't a BIOS issue with the ASUS Z77-A board I'd bought for Mom.

And I'm still wondering . . . if I should get an RMA from Intel. I CAN'T IMAGINE they'd ever ship a product with a chip on the PCB or this . . . gold contact pad anomaly.

Everything in the BIOS checks out, though.

From now on, I ADVISE MYSELF -- and I ADVISE OTHERS -- DON'T BUY LAST-GEN STUFF WHEN IT HAS EVAPORATED FROM MAINSTREAM ONLINE RESELLERS. You're better buying USED. That's what I think, anyway . . .

I NEVER, NEVER had these sorts of things occur before in 20 years of building computers. Tsk, tsk, tsk . . . .