ASUS X99 DELUXE (Intel LGA 2011-3)

[cadaveca]

We take our first look at ASUS's LGA 2011-3 products, starting with the ASUS X99 DELUXE. Wrapped in new clothing, the ASUS X99 DELUXE not only has DELUXE looks and DELUXE features, it also has a DELUXE socket for you to place your new Haswell-E CPU into, with an additional 48 pins.

Show full review

 

[VulkanBros]

Thanx for the review!

Holy .... - the list of "EXCLUSIVE FEATURES" is long as my bad conscience.
But a nice board - if you are into Intel stuff

 

[BorisDG]

Great review! Thank you.

 

[pokazene_maslo]

I have one question regarding CPU voltage settings. Is it possible to setup something like this: higher CPU voltage than default at overclock frequency and at the same time lower voltage than default at low frequency? Setup like this was possible on Core 2 CPUs since those CPUs supported software VID control through 3rd party software like Rightmark CPU Clock. Now i'm curious if this is possible on x99 platform.

 

[hippogriff]

It would be a nice board. If it didn't die in one week killing the CPU....

 

[Katanai]

I dunno, this seemed more like an advertisement than a review to me. I get it that you like the board but you shouldn't let your feelings dictate the pace of a review. This seems like a good board overall but it scored very badly in the SATA6 performance test. This should be listed at the end of the article as a con. 50MB/s less than the best board is not insignificant and it was in the last place as the only board that didn't go faster than 500MB/s. Seeing how many storage options this board has, that may be the main reason some buyers are interested in it and it's not something that should be skipped over so fast...

 

[LAN_deRf_HA]

Is that the best audio score to date?

 

[GhostRyder]

Great review @cadaveca, I look forward to seeing more from you on the X99 chipset!

 

[basco]

thanks and will we see a side by side overclocking comparison with oc-socket against other mobos?

 

[Hilux SSRG]

thanks and will we see a side by side overclocking comparison with oc-socket against other mobos?

This would be beneficial, especially because it's main "feature" is the oc socket.

Good review, but I think it's overpriced for what it offers. I wouldn't pay more than $300 for a new one.

 

[Dippyskoodlez]

This would be beneficial, especially because it's main "feature" is the oc socket.

Good review, but I think it's overpriced for what it offers. I wouldn't pay more than $300 for a new one.

I agree. I got the Gigabyte G1 Gaming Wifi for $300 from microcenter and it has essentially every feature here, plus a backlit I/O bracket and more useful features like being able to flash the bios without a CPU or memory (which is especially useful given DDR4s immaturity and CPU issues X99 has had.), split PCI-E slots, extra hole spacing, and M2 location.


What Wifi module is on board? What VRM is being used? Asus is really starting to cut corners on even their high end boards now it seems.

 

[cadaveca]

It would be a nice board. If it didn't die in one week killing the CPU....

Mine's still alive and kicking. Had it since before the platform launched. I've also been using mine for memory reviews, too, so it's not a matter of me just setting the board up once, testing, and then never using it again. This board sees quite a lot of usage.

I dunno, this seemed more like an advertisement than a review to me. I get it that you like the board but you shouldn't let your feelings dictate the pace of a review. This seems like a good board overall but it scored very badly in the SATA6 performance test. This should be listed at the end of the article as a con. 50MB/s less than the best board is not insignificant and it was in the last place as the only board that didn't go faster than 500MB/s. Seeing how many storage options this board has, that may be the main reason some buyers are interested in it and it's not something that should be skipped over so fast...

Very valid feedback. However....

What if I told you that using another drive of the same model, after testing, showed better numbers?

By all rights, 50 MB would be 10%, and that isn't insignificant, for sure. But I have found that my Crucial M4 drives aren't exactly consistent from install to install. I've tried various things to remove this variable from testing, and try to use the same drive for all tests, but something about going from one board to the next screws up bandwidth numbers. You will note, yeah, I'm pretty excited by this board, but it didn't get a perfect score, and only scored 9.4, and didn't get an "Editor's Choice" award. With everything else being so great, why do you think it got that score? Audio is killer, OC is killer, features are great, software is great, BIOS is fantastic... the only things that give it the award it got was due to drive performance, and ASUS is not my favorite brand. They actually have to work a bit harder than other brands to do to get good scores, since nearly everyone expects a lot from them, and not delivering reflects poorly.

thanks and will we see a side by side overclocking comparison with oc-socket against other mobos?

Yes, I will do a full OC comparison with all tested boards in a future article. It will encompass using many CPUs, and pushing to the limit, and what limits were each with each product, and not much else. Decent watercooling will be used. For those with interest in VRM designs, that will also be covered at that time. I do not remove cooling from boards any more because doing so can affect cooling performance compared to what the end user is likely to get, and can affect power consumption numbers.

 

[Katanai]

Very valid feedback. However....

What if I told you that using another drive of the same model, after testing, showed better numbers?

By all rights, 50 MB would be 10%, and that isn't insignificant, for sure. But I have found that my Crucial M4 drives aren't exactly consistent from install to install. I've tried various things to remove this variable from testing, and try to use the same drive for all tests, but something about going from one board to the next screws up bandwidth numbers. You will note, yeah, I'm pretty excited by this board, but it didn't get a perfect score, and only scored 9.4, and didn't get an "Editor's Choice" award. With everything else being so great, why do you think it got that score? Audio is killer, OC is killer, features are great, software is great, BIOS is fantastic... the only things that give it the award it got was due to drive performance, and ASUS is not my favorite brand. They actually have to work a bit harder than other brands to do to get good scores, since nearly everyone expects a lot from them, and not delivering reflects poorly.

Then you should have mentioned this fact in the article. Better yet removed the SATA results from the review altogether if they reflect the performance of the drives and not of the board itself. Anyways I'm sure this is a good board...

 

[cadaveca]

Then you should have mentioned this fact in the article. Better yet removed the SATA results from the review altogether if they reflect the performance of the drives and not of the board itself. Anyways I'm sure this is a good board...

You may very well be right about that. But like I said, it was after testing, and the review was ready to go. I have two identical M4 drives, one was used for drive testing, and holds benchmarks and such on it, while the other gets OS installed. If after more testing I can confirm this as a problem with the drive, then I'll buy another drive or a different make and brand to do testing with. It was while testing memory I decided to run the drive testing again that I noticed the oddity. I've checked things like cable used, and other stuff (this seems to have an effect as well), but I need to be able to replicate the problem on another board to be 100% sure.


I can pull the drive testing results, no problem. But in the end, it's the results I got, so it's the results I reported. And if I pull the drive testing, I'd have to give the board a higher score. Do you understand my problem?

 

[Aquinus]

I suspect that X99 doesn't have a problem with driving a single SSD. Better disk comparisons would be between different PCHs, SATA, and RAID controllers. I find little point to drive performance numbers between different boards with the same hardware driving those ports. Even more so when X79 was capable of driving two SSDs in RAID-0 at a full 1GB/s. Sometimes there are abnormalities in benchmarks, but I think those are few and far in between and comparisons against actual controllers would be a better gauge.

Nonetheless, another great review. It's my P9X79 Deluxe upgraded basically. All the modern bells and whistles as opposed to my 3 year old bells and whistles. ASUS seems to like to pack features into their DELUXE boards.

 

[springs113]

Great board review! As you already know that I couldn't wait for this one. I for one is definitely looking out for the OC sockets review, but for now I will enjoy reading it in its entirety.

 

[LiveOrDie]

It would be a nice board. If it didn't die in one week killing the CPU....

If you look back over that review you will find out it was the power supply that was the problem not the board.

Great board review! As you already know that I couldn't wait for this one. I for one is definitely looking out for the OC sockets review, but for now I will enjoy reading it in its entirety.

What type of testing did you do to see if your stable on 1.28v for 4.6Ghz?

 

[fullinfusion]

Thanks Dave for the time on reviewing this board,

The more I see X99 the more me want!

That was a slick move by Asus placing the cmos battery the way it is.

 

[Frick]

So what is this TPU thing? EPU?

 

[Aquinus]

So what is this TPU thing? EPU?

They're for overclocking and energy saving respectively. They control the power phases for the CPU, DRAM, etc.

 

[springs113]

If you look back over that review you will find out it was the power supply that was the problem not the board.



What type of testing did you do to see if your stable on 1.28v for 4.6Ghz?

I gamed, firestrike, intel extreme tuning, heaven and valley. That oc was tuned by Asus' optimisation utility on Windows, I've since updated the bios and did a manual try and hit 4.625ghz with memory@ 2750. I didn't finish stress testing it out yet but I will once I get home, vcore is set 1.29 and I didn't mess with any other voltages as of yet not to mention my chip is stable at 4.5ghz w/ a vcore of 1.25.

 

[LiveOrDie]

I gamed, firestrike, intel extreme tuning, heaven and valley. That oc was tuned by Asus' optimisation utility on Windows, I've since updated the bios and did a manual try and hit 4.625ghz with memory@ 2750. I didn't finish stress testing it out yet but I will once I get home, vcore is set 1.29 and I didn't mess with any other voltages as of yet not to mention my chip is stable at 4.5ghz w/ a vcore of 1.25.

OK please run Asus RealBench on Heavy Multitasking as i found i could run stable with the tests you listed with my chip on 4.5Ghz on as low as 1.235v, But after testing my OC with RealBench i found it wasn't stable RealBench is good to test out your OC for normal day use unlike programs like Prime which are bad for your CPU.

 

[springs113]

OK please run Asus RealBench on Heavy Multitasking as i found i could run stable with the tests you listed with my chip on 4.5Ghz on as low as 1.235v, But after testing my OC with RealBench i found it wasn't stable RealBench is good to test out your OC for normal day use unlike programs like Prime which are bad for your CPU.

Thanks but I tried downloading it yesterday but Asus'site is down. I'm new to oc'n at least on the Haswell platform, I never really messed around with things before on my z87, it was more or less try this multi using this volt and if it's stable then you're good.

 

[LiveOrDie]

Thanks but I tried downloading it yesterday but Asus'site is down. I'm new to oc'n at least on the Haswell platform, I never really messed around with things before on my z87, it was more or less try this multi using this volt and if it's stable then you're good.

Yer i am also waiting to download the new version i only have the old BETA version, Well keep us updated over on this thread i would like to see what your CPU end up being stable on .

 

[springs113]

I

Yer i am also waiting to download the new version i only have the old BETA version, Well keep us updated over on this thread i would like to see what your CPU end up being stable on .

I think I'm totally fine@4.6, I wanted to find my max and then scale back a tad. I'm curious now as to whether I should go with my manual oc or the Asus 5 way oc. The latter gives the all the power saving benefits, not having to run my chip at a constant v core when it's not needed would be great.

Is 1.3 really a safe voltage?
I know adding more volts overall decreases the lifetime of a cpu but is 1.3 bad(for a lack of a better term).