ASRock X99 WS LGA 2011-v3 Motherboard Review @ [H]

ASRock X99 WS LGA 2011-v3 Motherboard Review - ASRock comes to us with its "Work Station" version Haswell-E motherboard. This time our out-of-box experience with its X99 WS was as rock solid as it could be and did leave us with feelings of getting to work with a quality component. As you all know, we are much more interested in how it performs at high clocks while under stress.

Dunno man. A board that won't work out of the box with supported devices is a big fat fail in my book, "tuner" board or not.

Strange. I am guessing they should have spent a bit more time on the BIOS before releasing it.

I've been running ASRock boards for a while (1366 and 2011) and they always worked great out of the box.

The 1366 board was a somewhat poor overclocker, but it is still working fine after a few years.

The 2011 board I have is a much better board overall, although I am not sure that the RAM sub-timings are quite correct.

Overclocking is generally really good, but it does lack some of the extended settings other mfgs have.

Sad to see ASRock potentially releasing a board that needed some BIOS bug fixes right off the bat.

I'm all for ASROCK, but this is an EPIC FAIL in my book!

Seems to be the norm lately. Corsair is releasing unfinished products, everyone is releasing half borked games. They just want to make a quick buck nowadays it seems. I love my ASRock HTPC board though, rock solid and plenty of features, hopefully this board gets its required updates fast as well.

Looks like it need a few BIOS updates to get up to its potential.

Hardware wise, it looks like a pretty good system that might be worth a second look. Interestingly, Asrock went with Fairchild Mosfets this round. MSI did too, although they use 55A Mosfets.

There is one huge issue though I see, even with a BIOS update. If you are already spending so much for a CPU, and a premium for DDR4 on top of that, what is another $75 or so USD? You might as well get one of the following:

- MSI X99S XPower
- Gigabyte X99 SOC Force (only $25 more)
- Asus X99 Deluxe

Seeing that these boards are already well regarded and stable. Build quality wise, there might also be an argument again to go with say, MSI. I mean X99 here is supposed to be the best literally money gets you.

What I'm interested to see is how the Asrock X99 Extreme 11 turns out. They are armed with 2x PLX 8747 chips (so 72 lanes here) and have an LSI storage on top of the Intel X99 storage options, Probably will be very pricey though.

HeavensCloud said:

Seems to be the norm lately. Corsair is releasing unfinished products, everyone is releasing half borked games. They just want to make a quick buck nowadays it seems. I love my ASRock HTPC board though, rock solid and plenty of features, hopefully this board gets its required updates fast as well.

Click to expand...

Yeah I'm waiting until at least next year spring or perhaps late summer before considering an X99 build. I want to make sure that mos problems are ironed out, although some of the motherboards like the Rampage V Extreme, X99 XPower, and a few others are looking stable as is, for the most part.

cyclone3d said:

Strange. I am guessing they should have spent a bit more time on the BIOS before releasing it.

I've been running ASRock boards for a while (1366 and 2011) and they always worked great out of the box.

The 1366 board was a somewhat poor overclocker, but it is still working fine after a few years.

The 2011 board I have is a much better board overall, although I am not sure that the RAM sub-timings are quite correct.

Overclocking is generally really good, but it does lack some of the extended settings other mfgs have.

Sad to see ASRock potentially releasing a board that needed some BIOS bug fixes right off the bat.

Click to expand...

And hardware wise I really like this motherboard and the feature set it offers. Unfortunately the BIOS requires allot of time to sort out if your going to use one today. At least that is what my experience was like. It might behave better with other memory modules and a retail CPU, but I can't be certain of that.

Was there any update on the BIOS from Asrock? I think it happens from time to time that these board makers have a quick BIOS update after release right?

ohenryy said:

Was there any update on the BIOS from Asrock? I think it happens from time to time that these board makers have a quick BIOS update after release right?

Click to expand...

I don't know. I'd assume so, but I have no idea what it addresses. I rarely revisit a motherboard after I've moved onto the next ones. In some cases vendors update me on BIOS updates right after a review is finished but ASRock isn't one who does that. In fact I've never had a conversation with anyone from ASRock ever.

All the mobo manufacturers have their issues, AR is no different... There will be always a black sheep in the family, where ever you go

L1amrob said:

All the mobo manufacturers have their issues, AR is no different... There will be always a black sheep in the family, where ever you go

Click to expand...

I never said that wasn't the case.

Hi,

I'm trying to figure out what is the next step in finding the cause of my system freezing at random times, irrespective of workload.

System configuration:

ASRock X99 WS
i7 5820K
Team Elite DDR4 PC4-19200C16
Zotac Titan X
Samsung 850 EVO
EVGA SuperNOVA 1000 G

I've run a number of tests, for several days over a two weeks period.

MemTest86+ (latest version) does not bring up any errors. The system passed all the tests, several times, with success. I tested the system with one DIMM at a time as well as with all DIMMs present. No error whatsoever.

The CPU was tested using Intel's Extreme Tuning Utility, for 8+ hours per session, several times. Memory Stress tests and CPU Stress tests passed, no errors.

The GPU was stress tested for several hours using MSI Afterburner's stress test utility. No errors or freezes either.

At some point, the system would not POST, so I decided to have the motherboard replaced. The new motherboard is more stable, but the system still freezes every few hours.

This could be a power supply issue, so I replaced the PSU with a Coolermaster Silent Pro M 700 Watt. Now the system is more stable, it runs for 48 hours before freezing.

I have also disabled C6/C7, just in case, no improvement.

I reinstalled Windows 10 Pro, just in case it's an old broken driver (e.g. a leftover). No improvement.

I used Samsung's utility to check the condition of the SSD. It doesn't report any issue.

I've been left clueless, as I've never had to deal with such a nasty, well-hidden issue before.

I'm sharing my experience here, in case the forum's collective knowledge and experience could help me debug the system and figure out how to fix this.

Thanks and regards.

Have you manually set the RAM timings and voltages? If you haven't, I'd do that. The next step in my opinion is to try different RAM all together. You might want to replace your SATA cable while you are at it. Believe it or not I had stability issues once and that's what it turned out to be.

Dan_D said:

Have you manually set the RAM timings and voltages? If you haven't, I'd do that. The next step in my opinion is to try different RAM all together. You might want to replace your SATA cable while you are at it. Believe it or not I had stability issues once and that's what it turned out to be.

Click to expand...

Where would I find the correct settings for RAM timings and voltages? My understanding is that's now using the XMP profile.

Also, I'm wondering, is it to be expected nowadays that you have to manually set the timings and voltages of RAM for a standard configuration (i.e. without overclocking) ?

Also, forgot to mention, latest Dr Debug error message was 68. That's: "Chipset initialization error. Please press reset or clearCMOS".

I'll try the SATA cable, just in case.

Thanks.

ntsarb said:

Where would I find the correct settings for RAM timings and voltages? My understanding is that's now using the XMP profile.

Also, I'm wondering, is it to be expected nowadays that you have to manually set the timings and voltages of RAM for a standard configuration (i.e. without overclocking) ?

Also, forgot to mention, latest Dr Debug error message was 68. That's: "Chipset initialization error. Please press reset or clearCMOS".

I'll try the SATA cable, just in case.

Thanks.

Click to expand...

Modern memory modules usually contain two profiles which determine their basic speeds, timings and voltages. The two profiles are: Serial Presence Detect (SPD) and Extreme Memory Performance. (XMP) SPD values conform to JEDEC specifications. JEDEC specifications for DDR4 RAM can vary based on the type, but the most common JEDEC specification is DDR4 2133MHz @ 1.2v with CL-tRCD-tRP timings of 14,14,14. XMP goes beyond JEDEC specs and is always a form of overclocking. XMP RAM profiles often exchange latency for raw clock speed. The Corsair Dominator Platinum DDR4 3600MHz RAM I have on my test bench for example runs at DDR4 3600MHz @ 1.35v with CL-tRCD-tRP timings of 18,19,19.

The memory controller on Haswell-E (Core i7 5820K, 5930K, 5960X) officially only supports DDR4 2133MHz RAM at 1.20v. This conforms to JEDEC specifications. Motherboard manufacturers on the other hand can make use of XMP profiles for configuration. ASUS motherboards and a few others have an EZ-XMP type feature which is controlled via an external switch on the motherboard, or in the UEFI BIOS. The latter method being the most common. When XMP is set, sometimes the values aren't read and configured properly. This results in instability and other problems. This is very common. I've had to update the UEFI BIOS on virtually every motherboard I've reviewed in 2015 to resolve this problem. XMP modules are usually fine with default non-XMP configurations, whatever the system sets them at. The timings will often be a lot tighter, but the clocks are greatly reduced compared to their XMP values.

Just because your RAM may be rated for a given set of XMP timings and clocks, doesn't mean it can do it on your CPU. Each CPU's memory controller is a little different. Each motherboard is a little different with regard to X99 as that platform still has memory dividers. (Base clock times a fixed multiplier = clock speed) As you ramp up CPU clock speed, the memory timings may have to be further relaxed beyond the XMP values or the modules may need to have their clocks reduced. For example, my test bench Core i7 6700K can do 4.7GHz at speeds over DDR4 3000MHz on some motherboards, but at 4.8GHz, the most I can do is DDR4 2666MHz. Haswell-E is much the same way as Skylake is in this regard. Although Skylake can more reliably achieve greater than DDR4 3000MHz speeds.

The RAM you mentioned in your post is DDR4 2400MHz PC19200 @ 1.2v with timings of 16-16-16. This is actually another JEDEC spec, although it's beyond the standard 2133MHz spec the CPU supports officially. If you have XMP timings on that RAM I'd be surprised. In either case, your problems might be over if you simply configure the RAM manually and disable XMP. (Assuming you can enable it in the first place.)

Hope that helps.

thinking of getting this board from a fellow [H] member, anyone have it perhaps with an updated bios? did that fix things from the review?

Hi,

I'm quite confused with regards to whether this motherboard is any good for overclocking, since the article suggests that its components are getting too hot.

Has anyone achieved a good, stable overclock on the ASRock X99 WS using the i7 5820K CPU? If so, can you please share your hardware configuration and the BIOS settings that helped you achieve that?

It does get a bit hot but the thing got the job done. It was always stable while overclocked. We don't have too much in the way of fans blowing on the board during testing so your mileage may vary. With a good case and decent fans it should be fine for most people.

Dan_D said:

It does get a bit hot but the thing got the job done. It was always stable while overclocked. We don't have too much in the way of fans blowing on the board during testing so your mileage may vary. With a good case and decent fans it should be fine for most people.

Click to expand...

Thanks for your response.

I already have the motherboard for a year and a half now and I'm happy with it. It's installed in a Zalman Z12 with 4 x 120mm Noctua fans ( 1 in front as intake, 1 at the back and 2 more at the top as extractors), the PSU drawing cool air from the bottom. The PSU doens't leave enough space for a fan at the bottom, but there's still a filtered openning there.

I am considering overcloking my i7 5820K to improve the single threading performance of the CPU (currently capped at 3.6GHz using default settings), but I'm worried if the motherboard's coolers aren't up to the task...

I'm not asking you to guarranty anything, just asking if you found it worryingly hot, to affect the longevity of the motherboard.

Thanks in advance.

An X99 motherboard doesn't really get hotter when overclocking unless you have to mess with the power phase controls. I rarely have to.