ASUS Maximus IX Formula LGA 1151 Motherboard Review @ [H]

[FrgMstr]

ASUS Maximus IX Formula LGA 1151 Motherboard Review - ASUS is nothing like Hollywood. ASUS can actually turn out sequels which not only match the originals, but surpass them. ASUS Republic of Gamers Maximus IX Formula is another sequel in the long line of Maximus motherboards. Can ASUS continue its long history of awesome sequels? One things for certain, it’s no Robocop 3

 

[UnrealCpu]

You should mention in the reviews which SATA ports are disabled when using m.2 or u.2 etc. It is different on all motherboards and sometimes even the same manufacture which can completely change the characteristics of a board when upgrading storage devices.


From newegg.

1 x M.2 Socket 3 with M Key, type 2242/2260/2280/22110(PCIe 3.0 x4 and SATA modes_M.2_1)*
1 x M.2 Socket 3 with M Key, type 2242/2260/2280(supports PCIe 3.0 x4 mode_M.2_2)**

* When the M.2_1 Socket 3 is operating in SATA mode, SATA port 1 will be disabled.
**When the M.2_2 Socket 3 is operatin in PCIe 3.0 x4 mode, SATA ports 5 and 6 will be disabled.


Personally i feel Asus was a big let down on Sata upgradability with only 6 sata ports when you can jump into a Z270 taichi with 10 sata ports for 184.00 on newegg right now. Has 12 VRM power solution and a free High bandwidth bridge. It has been proven with the 1080 that this bridge does help with a few FPS in games , nothing dramatic but it does help . Also they retail for 39 bucks

 

[macksomerville]

At this point who honestly uses more than 3-4 sata ports on a desktop board? with all the extra features and additional crap ASUS tacks on their boards 6 sata ports and 2 pcie m.2's is more than adequate. Even in raid Im not sure how much I would trust an intel 5 drive array or something along those lines.

 

[lollerwaffle]

At this point who honestly uses more than 3-4 sata ports on a desktop board? with all the extra features and additional crap ASUS tacks on their boards 6 sata ports and 2 pcie m.2's is more than adequate. Even in raid Im not sure how much I would trust an intel 5 drive array or something along those lines.

I find extra ports useful because I'm kinda a dinosaur. I have an optical drive (BluRay writer) for some backups, I have a few old HDDs that I never really fully migrated or backed up perfectly, and a number of SSDs starting from when Intel 320 128GB drives hit $200.

So having 8 drives hooked up at a time isn't a strange scenario to me. Is 6 ports really limiting? No. But 8 is nice at times.

 

[UnrealCpu]

At this point who honestly uses more than 3-4 sata ports on a desktop board? with all the extra features and additional crap ASUS tacks on their boards 6 sata ports and 2 pcie m.2's is more than adequate. Even in raid Im not sure how much I would trust an intel 5 drive array or something along those lines.

At this point who honestly uses more than 3-4 sata ports on a desktop board? with all the extra features and additional crap ASUS tacks on their boards 6 sata ports and 2 pcie m.2's is more than adequate. Even in raid Im not sure how much I would trust an intel 5 drive array or something along those lines.

at this point who needs to spend 389.00 + tax for 6 sata ports ,especially when they get disabled;
when i can have 10 Sata ports and 3 m.2 for 184.00 and the motherboard will overclock just as good

 

[FrgMstr]

at this point who needs to spend 389.00 + tax for 6 sata ports ,especially when they get disabled;
when i can have 10 Sata ports and 3 m.2 for 184.00 and the motherboard will overclock just as good

Yeah, as I specifically pointed out TWICE in the review, if the feature set was not good for you, it was not the board for you. So please tell me again about Sata ports.

 

[STrAYeR]

Asus keeps jumping the price on the Hero boards. It was $209 then $219 for the next generation and now this generation is $229. These boards are suppose to be middle tier with great performance and they keep upping the price tag. I have a feeling the Hero X will be $239. Now you have this board that is $389(NE) excluding the tax. Asus is getting really expensive.

 

[SixFootDuo]

I personally do not use and haven't used mechanical hard drives or a DVD burner in my personal PC for 3+ years. I moved over to M.2 ssd's a year a go.

My box is so clean inside of cables, it's a thing of beauty. I have a HD Dock I plug into and a usb dvd burner anytime I need go that route.

 

[FrgMstr]

I use all 8 ports on mine. The OS is on a pair of SSDs in RAID0, two more ports have a pair of 4TB WD enterprise drives in RAID1, there is a TB caching drive, a blu-ray burner, and the two add-in chip controlled SATA3 ports are each connected to an older 256GB SSD for steam games that benefit from being on an SSD, but which I don't want to place on my primary drive.



How about the feature set of the waterblock? Any expectation that it cools as well as advertised?

My two cents on this.....I personally do not care. That cooling is not on my want/need list personally, but I know there are a bunch of folks around that do very much appreciate that feature. If you are going to build a custom loop, and go gangbusters on it, then go for it. So while it is not a feature that appeals to me, I sure it will to others.

I do not believe that it is going to increase the boards abilities to overclock better.

 

[Xero717]

I have the code version of this motherboard, which is basically the formula without the water block, and can confirm the issues Dan ran into with memory. Maybe I do have a bad chip, but any RAM OC beyond default gave me issues, not crashes, but issues, even with the i7 7700k at stock speeds. Turning off HT helped, but like you said, what's the point in having an i7 without HT. After ditching XMP I could run my cpu at 5.0 ghz with HT enabled at 1.36v without issues.

The main sticks I tested with were Trident Z 3200 Mhz with Samsung B die ICs. I also had similar issues with with some Corsair lpx a friend lent me, but unlike the Trident Z, I was able to run them above default, but not at XMP like my friend could in his skylake system(this was at stock and OC'd).

Going to give the Apex a try, it may not have the fancy armor, but from all the OC records and one of the reviews I read, something special is going on with it. Not to mention it's cheaper too.

 

[Dan_D]

You should mention in the reviews which SATA ports are disabled when using m.2 or u.2 etc. It is different on all motherboards and sometimes even the same manufacture which can completely change the characteristics of a board when upgrading storage devices.


From newegg.

1 x M.2 Socket 3 with M Key, type 2242/2260/2280/22110(PCIe 3.0 x4 and SATA modes_M.2_1)*
1 x M.2 Socket 3 with M Key, type 2242/2260/2280(supports PCIe 3.0 x4 mode_M.2_2)**

* When the M.2_1 Socket 3 is operating in SATA mode, SATA port 1 will be disabled.
**When the M.2_2 Socket 3 is operatin in PCIe 3.0 x4 mode, SATA ports 5 and 6 will be disabled.


Personally i feel Asus was a big let down on Sata upgradability with only 6 sata ports when you can jump into a Z270 taichi with 10 sata ports for 184.00 on newegg right now. Has 12 VRM power solution and a free High bandwidth bridge. It has been proven with the 1080 that this bridge does help with a few FPS in games , nothing dramatic but it does help . Also they retail for 39 bucks

Noted concerning mentioning this in the reviews. It hasn't come up as a significant topic in the reviews because every manufacturer does the same thing but I'll try to be clearer for people that may not be aware of this.

As to the subject of SATA ports getting disabled by M.2 devices. First off, this isn't an "ASUS" issue and the ASRock Tachi does the same thing ASUS Maximus IX Formula or any other motherboard I've seen since SATA Express became part of the chipset specification for the mainstream platform. In the old days, the SATA ports went through a SATA controller which by itself only took one or two lanes. Now, that's not done because these are all SATA Express ports with dedicated lanes. Some manufacturers, like ASUS don't always include the physical SATA Express connectors anymore as they don't get used but they are still essentially SATA express on the back end. So instead of sharing all your drives across a couple of lanes, the ports have their own lanes. The problem is that the Z270 Express platform doesn't have enough physical PCIe lanes to cover the volume of integrated devices motherboard manufacturers include in their designs.

So let's examine the ASRock Tachi in detail. I went ahead and looked up the manual for it on the manufacturer's website and found this:

- 6 x SATA3 6.0 Gb/s Connectors, support RAID (RAID 0, RAID 1, RAID 5, RAID 10, Intel® Rapid Storage Technology 15 and Intel® Smart Response Technology), NCQ, AHCI and Hot Plug*
- 4 x SATA3 6.0 Gb/s Connectors by ASMedia ASM1061, support NCQ, AHCI and Hot Plug
- 2 x SATA Express 10 Gb/s Connectors**
- 3 x Ultra M.2 Sockets, support type 2230/2242/2260/2280 M.2 SATA3 6.0 Gb/s module and M.2 PCI Express module up to Gen3 x4 (32 Gb/s)***

*M2_1, SATA3_0 and SATA3_1 share lanes. If either one of them is in use, the others will be disabled.
M2_2, SATA3_4 and SATA3_5 share lanes. If either one of them is in use, the others will be disabled.
If M2_3 is occupied by a SATA-type M.2 device, SATA3_3 will be disabled.

I highlighted the important stuff. Essentially, the Tachi is the same as any other motherboard I've run across in the last few years. You can lose up to 5 ports out of the 10 if you load up on M.2 devices. Think about it this way: Z270 supports 6 drives at a minimum. If you use M.2 drives which are faster, you will lose SATA ports. It's that simple. Everyone does it because they have to. I wouldn't prop up one manufacturer as the standard for how to do things properly when all they've done is made you pay for an additional controller that you can only use under very specific circumstances. MSI, ASUS and even GIGABYTE don't generally add third party SATA controllers for two reasons: A.) They don't perform worth a shit unless they use expensive controllers. B.) Those controllers take up valuable PCIe lanes and even worse, require additional PCIe switches to function further increasing costs and complexity with no benefit.

Intel's mainstream platform is built the way it is for a reason. It's called "market segmentation." If you have needs beyond what the mainstream platform can provide then it's time to step up to the HEDT platform. It's not as bad as it might sound. Many people think that you have to spend a fortune to go with an HEDT system but it isn't so. There are some less expensive X99 motherboards out there and processors like the 6820K which offer more cores and decent overclocking potential. We've shown that the IPC difference isn't that big of a deal. Even the clock speed you will lose won't matter much if at all in games. You don't necessarily have to increase your memory costs either as you can go with a dual channel setup on X99 if you just refuse to spend more than you would on Z270. If your only goal is to get more PCIe lanes and 10x SATA ports, you can do so without spending a whole lot more than you would with a mainstream LGA 1151 setup. The slight overlap in processor pricing and motherboard pricing is there for reasons like this.

Getting back to the Z270 Tachi and your comment about power phases, I haven't reviewed one or even seen one in person, but 12 power phases doesn't mean shit. I've seen 8-phase solutions that were far more capable than some solutions that had double or even quadruple the number of power phases. With power phases its about quality, not quantity. How they are cooled, how they are engineered and controlled are what makes the difference.

 

[Nicepants42]

Just in case anyone wants more detail on the water block composition, EK's block looks pretty good. It also looks like EK worked with Gigabyte as well. Having asked about motherboard water block composition several times over the years, it's nice to be able to find some clear info about it with a simple Google search.

Dan, you note that the board has 1 pump header, 1 AIO pump header, and 1 high amp fan header. Do you know if these are just naming differences, or are there real differences in the capabilities of these headers? I want to assume that one 2A header is as good as another (and thus, I could use any of these headers to control multiple D5 varios, for instance) but would appreciate any addition detail you might have.

 

[Dan_D]

This is the first time I've seen the distinction. The water pump headers are different in at least the management side of things compared to the 2A fan headers. I couldn't tell you what the difference is between the AIO and the standard water pump header. I'll reach out to ASUS and see if I can get some details on that.

 

[robble]

My Rog Maximus VIII has the water chipset cooler too. I do use it. Does it help? Beats me - I never didn't use it. Looks cool though and I'm sure the temps of everything it cools are lower than they would be on air.

 

[thesmokingman]

Just in case anyone wants more detail on the water block composition, EK's block looks pretty good. It also looks like EK worked with Gigabyte as well. Having asked about motherboard water block composition several times over the years, it's nice to be able to find some clear info about it with a simple Google search.

Dan, you note that the board has 1 pump header, 1 AIO pump header, and 1 high amp fan header. Do you know if these are just naming differences, or are there real differences in the capabilities of these headers? I want to assume that one 2A header is as good as another (and thus, I could use any of these headers to control multiple D5 varios, for instance) but would appreciate any addition detail you might have.

While the IX block is looking like an improved version thermally, I'm not keen on the folded fin design with respect to long term wear.

As for the pump header afaik its just another header. I don't use mine, just the pwm/rpm signal off the cpu header for my pwm pump. That said, I'm not powering my pump/pumps off the mb, no thanks.

 

[UnrealCpu]

Noted concerning mentioning this in the reviews. It hasn't come up as a significant topic in the reviews because every manufacturer does the same thing but I'll try to be clearer for people that may not be aware of this.

As to the subject of SATA ports getting disabled by M.2 devices. First off, this isn't an "ASUS" issue and the ASRock Tachi does the same thing ASUS Maximus IX Formula or any other motherboard I've seen since SATA Express became part of the chipset specification for the mainstream platform. In the old days, the SATA ports went through a SATA controller which by itself only took one or two lanes. Now, that's not done because these are all SATA Express ports with dedicated lanes. Some manufacturers, like ASUS don't always include the physical SATA Express connectors anymore as they don't get used but they are still essentially SATA express on the back end. So instead of sharing all your drives across a couple of lanes, the ports have their own lanes. The problem is that the Z270 Express platform doesn't have enough physical PCIe lanes to cover the volume of integrated devices motherboard manufacturers include in their designs.

So let's examine the ASRock Tachi in detail. I went ahead and looked up the manual for it on the manufacturer's website and found this:

- 6 x SATA3 6.0 Gb/s Connectors, support RAID (RAID 0, RAID 1, RAID 5, RAID 10, Intel® Rapid Storage Technology 15 and Intel® Smart Response Technology), NCQ, AHCI and Hot Plug*
- 4 x SATA3 6.0 Gb/s Connectors by ASMedia ASM1061, support NCQ, AHCI and Hot Plug
- 2 x SATA Express 10 Gb/s Connectors**
- 3 x Ultra M.2 Sockets, support type 2230/2242/2260/2280 M.2 SATA3 6.0 Gb/s module and M.2 PCI Express module up to Gen3 x4 (32 Gb/s)***

*M2_1, SATA3_0 and SATA3_1 share lanes. If either one of them is in use, the others will be disabled.
M2_2, SATA3_4 and SATA3_5 share lanes. If either one of them is in use, the others will be disabled.
If M2_3 is occupied by a SATA-type M.2 device, SATA3_3 will be disabled.

I highlighted the important stuff. Essentially, the Tachi is the same as any other motherboard I've run across in the last few years. You can lose up to 5 ports out of the 10 if you load up on M.2 devices. Think about it this way: Z270 supports 6 drives at a minimum. If you use M.2 drives which are faster, you will lose SATA ports. It's that simple. Everyone does it because they have to. I wouldn't prop up one manufacturer as the standard for how to do things properly when all they've done is made you pay for an additional controller that you can only use under very specific circumstances. MSI, ASUS and even GIGABYTE don't generally add third party SATA controllers for two reasons: A.) They don't perform worth a shit unless they use expensive controllers. B.) Those controllers take up valuable PCIe lanes and even worse, require additional PCIe switches to function further increasing costs and complexity with no benefit.

Intel's mainstream platform is built the way it is for a reason. It's called "market segmentation." If you have needs beyond what the mainstream platform can provide then it's time to step up to the HEDT platform. It's not as bad as it might sound. Many people think that you have to spend a fortune to go with an HEDT system but it isn't so. There are some less expensive X99 motherboards out there and processors like the 6820K which offer more cores and decent overclocking potential. We've shown that the IPC difference isn't that big of a deal. Even the clock speed you will lose won't matter much if at all in games. You don't necessarily have to increase your memory costs either as you can go with a dual channel setup on X99 if you just refuse to spend more than you would on Z270. If your only goal is to get more PCIe lanes and 10x SATA ports, you can do so without spending a whole lot more than you would with a mainstream LGA 1151 setup. The slight overlap in processor pricing and motherboard pricing is there for reasons like this.

Getting back to the Z270 Tachi and your comment about power phases, I haven't reviewed one or even seen one in person, but 12 power phases doesn't mean shit. I've seen 8-phase solutions that were far more capable than some solutions that had double or even quadruple the number of power phases. With power phases its about quality, not quantity. How they are cooled, how they are engineered and controlled are what makes the difference.

I understand what you are saying,
The point i was trying to make is that every manufacture is different when it comes to m.2. /u.2. and which ports are disabled. There are certain boards that will disable one sata port or even two sata ports when in use and even an entire PCIE slot. As an example The new EVGA Z270 that was just released a few days ago has 8 sata ports and two m.2. slots. Using m.2. 80mm will disable two sata ports , using m.2. 110mm will disable the u.2 slot. Using u.2. will disable m.2 . Believe it or not IMO this is one of the best storage configuration i have seen on a Z270 since you can get away using m.2 110mm along with 8 sata 6G ports.



All the Asus Z270s ROG have researched have two m.2 , 110mm disables 1 sata port , leaving you with 5 sata ports and when m.2. 80mm is used disables sata ports 5 & 6 , leaving you with an outstanding 3 Sata ports and two m.2.



Personally i think this type of information needs to be shared in detail in review articles so the reader can get a better understanding of that characteristics of a board when upgrading storage devices. Asus really dropped the ball on adding additional sata ports IMO and for future upgrade options when using SSD's , M.2. etc.

At least with the Taichi Z270 for 184.00 after rebate gets you 9 sata ports and 1 m.2. port after two m.2. is being used. Personally for what Asus offers is im sure great engineering and software but like others have mentioned is it really going to improve overclocks or justify a 200.00 price premium? I dont think so unless you like fancy RGB lighting or a crazy looking waterblock loop or using LN2


I also think the engineers have come to the conclusion more people are using m.2. now since the release of the Z170 which was poorly optimized for storage configurations when using m.2. and also its quite obvious you want a Z270 for a 7700k to hit 5ghz+ Z170

 

[FrgMstr]

Personally i think this type of information needs to be shared in detail in review articles so the reader can get a better understanding of that characteristics of a board when upgrading storage devices.

Thanks for your suggestion.

 

[Dan_D]

Asus really dropped the ball on adding additional sata ports IMO and for future upgrade options when using SSD's , M.2. etc.

Naturally, you are entitled to your opinion. Everyone's feature priorities are different and I get that. I still do not agree with you. I don't think ASUS dropped the ball here at all. As I said the extra controllers add cost, complexity and don't benefit you al that much for reasons I'll dive into shortly. Very few people need all six SATA ports and more than one M.2 port on an LGA 1151 motherboard.

At least with the Taichi Z270 for 184.00 after rebate gets you 9 sata ports and 1 m.2. port after one m.2. is being used.

O'rly? 9 or 8 you say? What you mean to say is that you get to disable five good SATA ports and keep three or four shitty ones.

From the ASRock manual, use of the M.2 slots disables SATA ports the same as it does on the ASUS motherboard. Sure ASRock gives you one more M.2 slot and a few more SATA ports but leveraging them has some big caveats. As a result, I think you are overestimating their value. Using the M.2 slots disables the integrated Intel chipset based SATA ports leaving you with the inferior ASM1061 controller. Not only is the ASM1061 slower in most test, it is also much less flexible than it's Intel counterparts. The ASM1061 has no RAID support, no support for ATAPI / Optical devices and has no support for Optane. I wouldn't ordinarily bring up Optane but you mentioned future upgrade options. The ASM1061 is an ancient controller with a very limited feature set. You are assigning value to a controller that most people wouldn't use and would probably find lacking if they did.

ASUS does a lot of market research before making these decisions. They correctly assume that few people would want to load up on that many SATA devices. There is so much market shift towards M.2 devices that SATA ports are becoming less and less relevant. With large mechanical disk drive prices being in the toilet, buying a couple of large drives to handle your storage needs in an effort to replace smaller drives isn't that costly. Again, it's exceedingly rare to require that many SATA ports. If you do, the Maximus IX Formula simply isn't for you. There are plenty of other choices that would serve you better if that's the case. Another option would be to add a PCIe SAS / SATA RAID controller of your own. I think most people who know what they are talking about would agree that this is a better solution than using those AHCI only ASM1061 ports.

Those SATA ports are largely there because ASRock wanted to integrate support for 3x M.2 slots. Those slots come at a cost because they require so many PCIe lanes. They tacked on some shitty controller so you'd be able to throw on some SATA drives instead of losing all but one SATA port. Having three M.2 slots is of limited value as well because they are stuck behind the DMI 3.0 bus. 2x drives in RAID 0 damn near saturate this bus by themselves. A third M.2 drive won't do shit for you. It has some arguable benefits in writes, but the benefits of having that much extra write performance in a desktop usage scenario over two drives is debatable at best.

Personally for what Asus offers is im sure great engineering and software but like others have mentioned is it really going to improve overclocks or justify a 200.00 price premium? I don't think so unless you like fancy RGB lighting or a crazy looking waterblock loop or using LN2

You are evidently not the primary target of this motherboard offering. The people who build custom loops or use liquid nitrogen cooling are a hardcore group that know you have to pay to play. That said, those aren't the only reasons why ASUS' premium offerings are so expensive. To put it bluntly ASRock motherboards aren't usually as nice as their ASUS counterparts. ASRock uses fewer PCB layers than ASUS usually does. The Maximus IX Formula has a far more complex audio implementation than the Tachi. ASRock doesn't have the same MOSFET cooling or the fancier chipset cooling. There isn't the same amount of RGB lighting between them or all the plastic cladding, thermal armor etc. I'm not saying these features change how the motherboard overclocks on a simple air or AIO based cooling solution. These features do cost more money. I've handled a lot of motherboards from a lot of brands. They differ in many subtle ways with regard to quality. ASUS doesn't even necessarily build the highest quality motherboard out there. However, their attention to detail is something many enthusiasts appreciate. Details like the added IC's that allow you to flash the BIOS without a CPU and RAM, the brushed aluminum I/O panel, socketed BIOS ROM, IC's to control the fans and pump headers, front panel controllers, or the additional thermal probes along the PCB all add up to higher costs.

Other features in the ROG line have a value that's much harder to quantify because we don't really know how much they cost to implement. ROG motherboards have several memory presets and overclocking presets that are the result of many hours of QVL testing. This is the type of feature that adds value to a very specific crowd which are harder to see because its in the firmware.

Motherboards really have little effect on performance if any outside of extreme cooling. What separates these motherboards are integrated features, fit and finish. For whatever reason, you are absolutely hung up on SATA as seemingly the only feature of importance. That may very well be your thing but it doesn't paint a complete picture. Essentially, the $200 ASRock motherboard is a good value on paper, but it isn't in the same league as the Maximus IX Formula. That doesn't make it a bad motherboard or anything but they aren't the apples to apples comparison you seem to think they are. Its somewhat like comparing a Shelby GT 350R to a Ferrari. The latter is certainly more expensive and has a nicer fit and finish. Yet, if you stick close to the speed limit the only real difference in day to day usage is how cool you look in one vs. the other. That's to say that the extra performance or quality isn't always important depending on how you use it. The Mustang in this example is still a fine car and will get the job done but it isn't quite the same thing when you get into specifics.

At the end of the day, those extra SATA ports and the extra M.2 slot are check box features few people ever use. Their purpose is more to make people take notice of a particular motherboard model than to be of actual use. Where the rubber meets the road they are a bit of a red herring.

 

[Nicepants42]

O'rly? 9 or 8 you say? What you mean to say is that you get to disable five good SATA ports and keep three or four shitty ones.

From the ASRock manual, use of the M.2 slots disables SATA ports the same as it does on the ASUS motherboard. Sure ASRock gives you one more M.2 slot and a few more SATA ports but leveraging them has some big caveats. As a result, I think you are overestimating their value. Using the M.2 slots disables the integrated Intel chipset based SATA ports leaving you with the inferior ASM1061 controller. Not only is the ASM1061 slower in most test, it is also much less flexible than it's Intel counterparts. The ASM1061 has no RAID support, no support for ATAPI / Optical devices and has no support for Optane. I wouldn't ordinarily bring up Optane but you mentioned future upgrade options. The ASM1061 is an ancient controller with a very limited feature set. You are assigning value to a controller that most people wouldn't use and would probably find lacking if they did.

Those SATA ports are largely there because ASRock wanted to integrate support for 3x M.2 slots. Those slots come at a cost because they require so many PCIe lanes. They tacked on some shitty controller so you'd be able to throw on some SATA drives instead of losing all but one SATA port. Having three M.2 slots is of limited value as well because they are stuck behind the DMI 3.0 bus. 2x drives in RAID 0 damn near saturate this bus by themselves. A third M.2 drive won't do shit for you. It has some arguable benefits in writes, but the benefits of having that much extra write performance in a desktop usage scenario over two drives is debatable at best.

I agree with the sentiment of your post, and you just did a fantastic job reinforcing this:

Personally i think this type of information needs to be shared in detail in review articles so the reader can get a better understanding of that characteristics of a board when upgrading storage devices.

Given how (in)frequently I upgrade, M.2 RAID is a real possibility on this platform at some point. I guess I should start reading user manuals prior to purchase.

 

[thesmokingman]

Given how (in)frequently I upgrade, M.2 RAID is a real possibility on this platform at some point. I guess I should start reading user manuals prior to purchase.

M2 RAID is generally not worth the hassle and downsides. I normally suggest getting a 950/960 Pro and calling it a day.

 

[Nicepants42]

M2 RAID is generally not worth the hassle and downsides. I normally suggest getting a 950/960 Pro and calling it a day.

Yeah, but I'm not going to call it 5 years. There's 'general' PC users and there's 'general' [H].

Regardless, the shit's convoluted AF. I was just trying to (probably unnecessarily) reinforce UnrealCpu's point.

 

[thesmokingman]

Yeah, but I'm not going to call it 5 years. There's 'general' PC users and there's 'general' [H].

Regardless, the shit's convoluted AF. I was just trying to (probably unnecessarily) reinforce UnrealCpu's point.

Nods. I would read this thread before taking up the torch for him.

https://hardforum.com/threads/z270-why-no-10-sata-ports.1923751/

 

[MV75]

Hey just wondering on this board, a mate got one and I set it up for him. He also got some corsair ddr4 3200 memory, but how to get it running at that? I found the voltage and timing settings on the corsair site, but the board still just locked and auto recovered itself and just won't run memory above 2100mhz. Now I don't mind too much as it's still fast enough, but he did blow a ton of money on that corsair, err, stuff. Was there something I may have missed? I set the voltage and the timing. Now I can navigate a bios blind folded, but having no reason to overclock for the last few years, I've totally not paid much attention to uefi. To me it's bloated, slow and in the way, also non uniform. Total mess really.

 

[Dan_D]

Hey just wondering on this board, a mate got one and I set it up for him. He also got some corsair ddr4 3200 memory, but how to get it running at that? I found the voltage and timing settings on the corsair site, but the board still just locked and auto recovered itself and just won't run memory above 2100mhz. Now I don't mind too much as it's still fast enough, but he did blow a ton of money on that corsair, err, stuff. Was there something I may have missed? I set the voltage and the timing. Now I can navigate a bios blind folded, but having no reason to overclock for the last few years, I've totally not paid much attention to uefi. To me it's bloated, slow and in the way, also non uniform. Total mess really.

Seth the RAM voltage to 1.38 - 1.41v. Make sure latency values match the XMP or SPD timings at those speeds. You can check these values in the tools menu. Often times a motherboard will use tighter timings than it should.

 

[MV75]

Seth the RAM voltage to 1.38 - 1.41v. Make sure latency values match the XMP or SPD timings at those speeds. You can check these values in the tools menu. Often times a motherboard will use tighter timings than it should.

Ah, it was probably because I'd set the voltage to 1.35, so just a bit higher then. Also I did set the timings to the corsair website had them listed at for that speed on those modules, but yea, I'll look around for a tools menu to see what the actual reported non jedec rom settings on the modules are actually at too. Thank you for the tip, I'll do that when I next see him.

 

[SadTelevision8558]

Hello,

This part really caught my eye regarding Vcore settings:

In any case, a manual voltage setting still has the same effect in which the actual voltage depends on the CPU workload with the manually set voltage seemingly acting as a voltage cap.

If that's the case, then it sounds like the CPU is allowed to ramp down vcore and clock speeds assuming you have EIST/C-states activated? Then how is that any different than an adaptive voltage or offset mode? I was always under the impression manual vcore = constant vcore and if you want to use any of the VID curves then offset/adaptive are the needed settings.

 

[Dan_D]

Hello,

This part really caught my eye regarding Vcore settings:



If that's the case, then it sounds like the CPU is allowed to ramp down vcore and clock speeds assuming you have EIST/C-states activated? Then how is that any different than an adaptive voltage or offset mode? I was always under the impression manual vcore = constant vcore and if you want to use any of the VID curves then offset/adaptive are the needed settings.

You are exactly right. This is the reason why there isn't a manual mode on most motherboards. The voltages are basically adaptive whether you like it or not. I'm not certain but I think the manual mode / offset / adaptive modes are really there for Skylake, not Kaby Lake. The Z270 motherboards support both CPUs. I'd have to pop a Skylake CPU into competing Z270 motherboards and Kaby Lake to see but I think other manufacturers just hide those values with Kaby Lake CPUs. Its similar to X99 motherboards hiding the AVX offset mode on Haswell-E CPUs and displaying it for Broadwell-E. I'll try and remember to get some clarification from ASUS on this front. I haven't thought about it much as I've been looking at MSI and GIGABYTE boards lately and all the Ryzen stuff.

 

[SadTelevision8558]

You are exactly right. This is the reason why there isn't a manual mode on most motherboards. The voltages are basically adaptive whether you like it or not. I'm not certain but I think the manual mode / offset / adaptive modes are really there for Skylake, not Kaby Lake. The Z270 motherboards support both CPUs. I'd have to pop a Skylake CPU into competing Z270 motherboards and Kaby Lake to see but I think other manufacturers just hide those values with Kaby Lake CPUs. Its similar to X99 motherboards hiding the AVX offset mode on Haswell-E CPUs and displaying it for Broadwell-E. I'll try and remember to get some clarification from ASUS on this front. I haven't thought about it much as I've been looking at MSI and GIGABYTE boards lately and all the Ryzen stuff.

That's really odd because my Gigabyte Z270X-Gaming K7 definitely has a manual mode and setting that results in fixed voltages even if EIST/C-States are turned on. I lack adaptive mode though on my board.

Anyhow, I wonder if the BIOS teams are just swamped trying to cope with Z270 and Ryzen platform boards simultaneously that resulted in the behavior on the ASUS board you noted--after all Skylake and Kaby Lake aren't all that different and nor are Z170/Z270. Does manual voltage have strange behavior on older Asus boards also?

Edit: Reading ASUS' Kaby Lake overclocking guide says that Manual/Offset/Adaptive modes should still be working the way they are supposed to.

 

[Dan_D]

That's really odd because my Gigabyte Z270X-Gaming K7 definitely has a manual mode and setting that results in fixed voltages even if EIST/C-States are turned on. I lack adaptive mode though on my board.

Anyhow, I wonder if the BIOS teams are just swamped trying to cope with Z270 and Ryzen platform boards simultaneously that resulted in the behavior on the ASUS board you noted--after all Skylake and Kaby Lake aren't all that different and nor are Z170/Z270. Does manual voltage have strange behavior on older Asus boards also?

Edit: Reading ASUS' Kaby Lake overclocking guide says that Manual/Offset/Adaptive modes should still be working the way they are supposed to.

Under load the voltage is still not fixed. It's variable. This is true even when you place things in manual mode. On older systems, setting the variable to "manual / fixed mode" would result in the voltage remaining constant. That's not what happens with Kaby Lake on any motherboard I've tested it with so far.

 

[SadTelevision8558]

I understand what you're saying but that's not my experience with the Gigabyte Z270 Gaming K7 (F4 BIOS). My Vcore is set at 1.16 manually.

I only notice Vcore drop if I use the Normal+DVID mode or Auto modes. Anyhow, thank you for clarifying what you observed on the ASUS board. I am certainly curious about switching to ASUS and just wanted to make sure I understand all the overclocking options.