What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

ASUS Sabertooth X79 Motherboard Review

Status
Not open for further replies.

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
BIOS Rundown p.2

BIOS Rundown


Moving over one more tab to the Advanced tab are a number of configuration subscreens for CPU, System Agent, PCH, SATA, USB, Onboard Devices, and APM. The CPU Configuration page contains CPU information as well as settings for hyper-threading and active cores.

Sbios11.png
Sbios12.png

The main subpage you will need under Advanced is Onboard Devices as this is where you can disable a number of devices to save CPU cycles when benching and speed up boot time. Also worth mentioning is the CPU Power Management Configuration page which has power management options as well as CPU Ratio and Turbo Mode.

Sbios13.png
Sbios14.png

The Monitor section contains the system temperature/power status, and adjustable fan settings. Since we are running a watercooling system without any fans attached to fan headers on the board, fan controls were all set to ignore or disable. Fan speeds and customizable based on temperature, RPM, or there are profiles included in the BIOS.

Sbios15.png
Sbios16.png

The Boot tab contains boot priority options and POST options. Here you can also change the default BIOS mode from EZ to Advanced. Moving over to the Tools tab there are several useful tools available including DRAM SPD Information, EZ Flash 2, and OC Profiles. DRAM SPD Information shows all installed memory with subsections showing SPD.

Sbios17.png
Sbios18.png

ASUS OC Profiles allow for saving customized OC Profiles suited for different situations. This is a very handy tool for any overclocker but keep in mind that if you flash your BIOS your profiles are gone. Last but not least we have EZ Flash 2. This utility is used for flashing the board’s BIOS from a in a safe and easy manner.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
Test Setup and Testing Methodology

Test Setup and Testing Methodology

testsetup.png

Our test setup consists of an Intel Sandy Bridge-E 3960X, ASUS Sabertooth X79 motherboard, two GTX 470s, 16GB of G.Skill RipjawsX DDR3-2133 9-11-9, and a Crucial M4 128GB SSD. All this is powered by a Corsair AX1200w PSU. Here are a few shots of the setup and hardware:


First we have the whole test bench up and running. As you can see we used watercooling (Apogee XT) for the CPU along with two heatercores. A second GTX 470 was also added for SLI testing.


Overclocking Methodology

At Hardware Canucks, we understand we have a diverse reader base and to represent a variety of user types, so we put the Sabertooth X79 through two types of tests.
  • Beginner Overclocker - To represent a beginner overclocker or a mainstream user that wants to have a quick and easy way to get some extra performance we used the OC Tuner setting found in the UEFI BIOS.
  • Experienced Overclocker - To represent an experienced overclocker that is looking for the optimal 24/7 overclock to maximize system performance while keeping voltage and temperatures in check we overclocked the Sabertooth X79 manually and stopped when we started to get concerned with voltage levels and temperatures.

We did stability testing a little differently for the Sandy Bridge-E platform than usual. The main stability test used was Linpack (LinX version 0.6.4) with memory usage set to 2,048MB and 20 loops run. In the enthusiast world, Linpack is a benchmark designed to measure performance on Intel CPUs in GFlops. However, it's also a very useful tool for checking the stability of a CPU and memory. LinX picks up very quickly and if you are able to complete a 20 loop test with the specifications above your system is likely stable or very close to it. For testing memory stability we used HyperPI as this bench picks up memory errors quickly.

After LinX / HyperPI stability was achieved, 2 runs of 3DMark 11 were run to test 3D stability. Once an overclock passed these tests, this is the point deemed as “stable” for the purposes of this review.

Windows 7 Service Pack 1 was installed to take advantage of the Advanced Vector Extensions (AVX) with Sandy Bridge processors. Intel AVX is a 256-bit SIMD floating point vector extension of Intel architecture. The BIOS used for overclocking and benchmarking was version 0906.

ocpic.jpg


Benchmarking Methodology

Benchmarks in the System Benchmarks section will be a comparison of the i7 3960X and Sabertooth X79 at stock speed, at auto overclock speed as set by OC Tuner, and at maximum 24/7 overclock to give an idea of how much performance a user can gain when overclocking the Sabertooth X79.

For stock testing, optimized defaults were loaded putting the CPU at 3,602Mhz (36 x 100.1) and memory at 1,600Mhz and 11-11-11-28 1T timings. Optimized defaults do not enable Turbo fully by default hence the 3,600Mhz instead of 3,900Mhz like we saw with the Rampage IV Extreme. The Sabertooth does however clock to 3,600Mhz from 3,300Mhz which is the stock speed of the processor. The auto overclocked speed on the 3960X (OC Tuner) was 4,297Mhz (34 x 126.4) with memory at 1011.1Mhz and 9-11-10-28 2T timings. The overclocked speed on the 3960X for 24/7 stability was 5,000Mhz (50 x 100.00) with memory at 2,133Mhz and 9-11-9-28 1T timings. Windows 7 Ultimate 64 bit was used with SP1 installed.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
Overclocking Results

Overclocking Results

As we typically do for reviews, we put the Sabertooth X79 through countless hours of overclocking and testing; from auto overclocking through OC Tuner to manual overclocking where tweaking is king. Since the Sabertooth is more of a mainstream board than the likes of the Rampage IV Extreme, we did not include sub-zero testing but instead added sections on voltage regulation and ASUS SSD Caching.


Auto Overclocking

ASUS has included a handy utility with the Sabertooth to automatically overclock the system: OC Tuner. OC Tuner allows for quick and easy CPU overclocking simply by enabling the feature in the BIOS. The board resets once you select OK and both a CPU and memory overclock are applied. The result is a boost from stock speed (optimized defaults) of 3,600Mhz to an instant 4,297Mhz on an i7 3960X with no effort whatsoever! Not only do we have a very nice CPU overclock with OC Tuner, but the Sabertooth also overclocked the RipjawsX to 2,022Mhz (9-11-10) as well. This is the best auto overclock we have seen to date of all the boards reviewed thus far.

Soc1.png
Click for full screenshot

4.3Ghz and the system was rock stable while using the OC Tuner feature and we were able to run all stress tests and benchmarks without any issue. For reference load Vcore was at 1.288v for the auto OC of 4.3Ghz which isn't all that bad either.


Manual Overclocking

octitle.jpg

As we mentioned in the Rampage IV Extreme review, manual overclocking was a bit of an adventure on Sandy Bridge E compared to the previous Sandy Bridge generation. The basics were the same but add quad channel memory, straps, overclockable BCLK, and a ton more factors that can affect stability and you have different beast altogether.

Starting with the CPU overclock and leaving the RAM at stock (or below) we went directly for the overclocker’s holy grail of 5Ghz. After a fair amount of work, (a lot more time than SB) we were able to achieve the 5Ghz we were after and realized that overall the Sabertooth was very easy to work with. One of the biggest advantages of the ASUS BIOS (especially with X79) is that their boards can easily recover from an instable overclock. After weeks of testing we did not have to clear CMOS once and if the OC wasn't stable the Sabertooth simply tells you so in the POST screen so adjustments can be made accordingly.

Achieving 5Ghz was possible with a number of BCLK and multi combinations but in the end we ended up going with 50 x 100.00 so we could run our RipjawsX kit at 2133Mhz. Vcore required for 5Ghz at 50 x 100 was 1.445v with and LLC at Ultra High however using different BCLK and multi combinations requires different levels of Vcore so keep that in mind when overclocking.

Soc2.png
Click for full screenshot

After finding that point of CPU stability, the next step is always to maximize the memory overclock. Since the RipjawsX kit is rated for 2133Mhz at 9-11-10 and 1.6v that was a natural first step. We achieved this speed without issue with only 1.0v VCCSA and 1.15v VTT.

Soc3.png
Click for full screenshot

Since the spec speed is never enough, we decided to drop the CPU clock and move up on the RipjawsX overclock to see how well it could hit the top end. The 16GB kit continued to clock up to 2400Mhz at 9-11-10-32 1T and scaled nicely up to 1.825v while still keeping VCCSA and VTT below 1.2v. This is much more voltage than we recommend using for 24/7 operation but it gives you an idea what the system is capable of if you want to use it for benching.

Soc4.png
Click for full screenshot

Loosening timings to 10-12-10 2T we were able to push to 2484Mhz but the looser timings yielded less performance so we have to conclude that 2400Mhz at 9-11-10 1T is about the best we will see with this particular memory and CPU.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
System Benchmarks

System.jpg

In the System Benchmarks section we will show a number benchmark comparisons of the 3960X and Sabertooth X79 using the stock speed, applying the OC Tuner overclock, and our manual overclock. This will illustrate how much performance can be gained by overclocking the 3960X using the Sabertooth. For full comparisons of 3960X vs. a number of different CPUs have a look at the Intel Sandy Bridge-E Core i7-3960X CPU Review.


SuperPI Benchmark

SuperPi calculates the number of digits of PI in a pure 2D benchmark. For the purposes of this review, calculation to 32 million places will be used. RAM speed, RAM timings, CPU speed, L2 cache, and Operating System tweaks all effect the speed of the calculation, and this has been one of the most popular benchmarks among enthusiasts for several years.

SuperPi was originally written by Yasumasa Kanada in 1995 and was updated later by snq to support millisecond timing, cheat protection and checksum. The version used in these benchmarks, 1.5 is the official version supported by hwbot.


Pi.png

Results: A 17% increase in SuperPI 32M is noted going from stock speed of 3600Mhz (optimized defaults) to the OC Tuner speed of 4300Mhz on the i7 3960X and Sabertooth X79. Jumping up to the manual overclock speed of 5000Mhz a 27% gain is noted.


CINEBENCH R11.5

CINEBENCH is a real-world cross platform test suite that evaluates your computer's performance capabilities. CINEBENCH is based on MAXON's award-winning animation software CINEMA 4D, which is used extensively by studios and production houses worldwide for 3D content creation.

In this system benchmark section we will use the x64 Main Processor Performance (CPU) test scenario. The Main Processor Performance (CPU) test scenario uses all of the system's processing power to render a photorealistic 3D scene (from the viral "No Keyframes" animation by AixSponza). This scene makes use of various algorithms to stress all available processor cores. The test scene contains approximately 2,000 objects which in turn contain more than 300,000 polygons in total, and uses sharp and blurred reflections, area lights, shadows, procedural shaders, antialiasing, and much more. The result is displayed in points (pts). The higher the number, the faster your processor.


Cb.png

Results: The CINEBENCH R11.5 results show an impressive increase in performance in rendering moving from a stock system to two levels of an overclocked system. For CPU rendering, an 19% and 39% improvement (in points) is noted when moving to auto OC and manual OC speeds.


Sandra Processor Arithmetic and Processor Multi-Media Benchmarks

SiSoftware Sandra (the System ANalyser, Diagnostic and Reporting Assistant) is an information & diagnostic utility. The software suite provides most of the information (including undocumented) users like to know about hardware, software, and other devices whether hardware or software. The name “Sandra” is a (girl) name of Greek origin that means "defender", "helper of mankind".

The software version used for these tests is SiSoftware Sandra 2012. In the 2012 version of Sandra, SiSoft has updated the .Net benchmarks and the GPGPU benchmarks have been upgraded to General Processing (GP) benchmarks, able to fully test the new APU (CPU+GPU) processors. The two benchmarks used are the Processor Multi-Media and Processor Arithmetic benchmarks. These three benchmarks were chosen as they provide a good indication of three varying types of system performance. The multi-media test shows how the processor handles multi-media instructions and data and the arithmetic test shows how the processor handles arithmetic and floating point instructions. These two tests illustrate two important areas of a computer’s speed and provide a wide scope of results.


Sandraarith.png


Sandramm.png

Results: Sandra processor arithmetic and multi-media benchmarks also show impressive improvements on an overclocked system, with 19% to 39% gains in performance across the board for the OC Tuner and manual overclock.


MaxxMem Benchmark

Created by MaxxPI², the MaxxMem benchmark tests your computer’s raw memory performance, combining copy, read, write and latency tests into one global score. This memory benchmark is a classic way to measure bandwidth of a memory subsystem.

MaxxMem uses continuous memoryblocks, sized in power of 2 from 16MB up to 512MB, starting either writing to or reading from it. To enable high-precision memory performance measurement, they both internally work with multiple passes and averages calculations per run.

Further, the main goal was to minimize (CPU) cache pollution on memory reads and to eliminate it (almost completely) on memory writes. Additionally, MaxxMem operates with an aggressive data prefetching algorithm. This all will deliver an excellent judge of bandwidth while reading and writing.


Maxxmembw.png
Maxxmemlat.png

Results: Moving from stock speed to the OC Tuner overclock on the Sabertooth, we see nice gains of 20-28% in memory read, write, and copy. When we up to the manual overclock, bandwidth goes up even more with improvements of 38-48%!

The results are similar when looking at latency; 19% is gained by OC Tuner and 33% with the manual overclock. OC Tuner overclocks the memory in addition to the CPU so even with little effort nice gains are possible on the Sabertooth.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
ASUS SSD Caching

ASUS SSD Caching

Many users have found caching to be a key feature for the Z68 chipset as it allows for installing a smaller (and much cheaper) SSD which allows for faster boots and application load times when paired up with a mechanical drive or RAID array. Since Intel did not provide this feature with the X79 chipset, ASUS went ahead and developed their own version of SSD caching using a Marvell controller and a utility built into AI Suite II. The question is how much additional performance does the ASUS version provide and is it worth using?

First of all, ASUS claims up to three times faster performance with ASUS SSD Caching but we weren’t really sure what that number was even referring to so it may be difficult to test this claim. We can however compare ASUS SSD Caching performance to mechanical hard drive performance as well as standard SSD only performance.

Before jumping into test results first let’s have a look at what goes into setting up ASUS SSD Caching.

cache1.png

Step one of ASUS SSD Caching setup is to connect a mechanical drive and an SSD to the white Marvell ports on the Sabertooth. Next, boot into Windows, open AI Suite and go to ASUS SSD Caching under the Tools menu. You should see a screen similar to the one above. We should mention here that there are no capacity limitations for the SSD as there are with Z68 (64GB) so our 128GB M4 worked without a partition. The benefit of this is a higher limit on the amount of programs that can be cached on the SSD.

cache2.png

In the bottom right corner of the caching screen there is a “Caching Now!” button which can be simply click on and the system will automatically set up ASUS SSD Caching. It will format your SSD so keep that in mind before proceeding.

cache3.png

Once initialization is complete you are done! There are no drivers to deal with, no registry hacks, no installing Windows in RAID mode, nothing. You get a popup on the taskbar indicating initialization is complete and caching is ready to go. There is a typo in the software as shown above but in spite of that big points to ASUS for ease of use - SSD Caching is foolproof.


ASUS SSD Caching Performance:

In order to test ASUS SSD Caching, we paired up a 128GB Crucial M4 SSD with a 1TB WD Black 6Gb/s drive and ran tests with no caching, ASUS SSD Caching, and then with the SSD only. All tests were repeated 3 times and the best result was graphed. For ASUS SSD Caching 5 runs were made before the 3 that were recorded in order to allow time for the cache to be populated.

pcm11.png

The first test we ran was PCMark 7. PCMark7 combines more than 25 individual workloads covering storage, computation, image and video manipulation, web browsing and gaming so is very reflective of day-to-day PC usage. As you can see above ASUS SSD Caching performs notably better than a mechanical drive alone however not quite as good as an SSD only solution.

W7.png

Regarding Windows boot time, we found the Sabertooth to be similar to the Rampage IV Extreme which means it was very slow booting even with an SSD installed. Granted we have a ton of devices connected but the board sits for ages in POST mode before starting to load Windows so boot times are slower than what we are used to from Z68. Looking at the ASUS SSD Caching results however we see that it performs on par with a SSD in booting and slightly slower in shutdown.

Scrystal.png
Click to enlarge

CrystalDiskMark performance for ASUS SSD Caching is still a massive improvement over a mechanical drive. However, it just can't touch the SSD performance in reads and writes.

All in all we feel ASUS did an excellent job in implementation but looking at our testing with a 128GB SSD, we feel like users in this category would be much better off buying an SSD up front rather than combining a mechanical drive with a large capacity SSD. However, if money is tight, this feature combined with a lower capacity SSD could provide an excellent starting point for a faster system.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
Load-Line Calibration Testing

Load-Line Calibration Testing

Load-line Calibration is defined by Intel’s VRM spec and affects CPU voltage. In short, the CPU’s working voltage will decrease proportionally to load so higher load-line calibration nets higher voltage, increased stability and better overclocking performance. However, in some cases LLC has been known to apply too much voltage in order to compensate for these decreases while other boards have been known to apply too little voltage and thus, stability remains a bridge too far.

The Sabertooth X79 has 5 levels load-line calibration, less that Gigabyte’s 10, but more than most other manufacturers provide. In this section we will test each level of LLC at three different voltages (1.3v, 1.4v, 1.5v) so we can give users an idea of what to expect from each level. Both actual readings (measured via a multimeter) and BIOS set voltages are displayed below to give you an idea of how much voltage you get vs. how much you set.

vmeasure.jpg

Unfortunately the Sabertooth X79 does not include voltage measure points so the best bet was to solder some wires underneath the board on the capacitor thru legs. With both the positive and the ground on the same cap, Vcore measurement is quite accurate.

llc.jpg

For Vcore measurement we used our digital multimeter connected to the wires we soldered to the cap through legs. To get a good idea of how the Sabertooth will react to different levels of voltage we tested all 5 levels of LLC at 1.3v, 1.4v, and 1.5v and measured idle and load voltages with a digital multimeter accurate to three decimal places. Each load voltage was recorded at the 30 second mark using p95 and all tests were run with the CPU speed at 4Ghz and memory at 2133 9-11-10.

llc1.png

Starting at 1.3v, the idle voltage starts as low as 1.289v and moves up to 1.331v with the Extreme LLC setting. Load voltage starts at the same point but climbs more rapidly up to 1.400v with the 1.300v BIOS setting. The closest level of LLC to the BIOS set voltage at 1.3v is LLC Regular which gives 1.289v under load.

llc2.png

At 1.4v results are similar to 1.3 but with the higher LLC settings voltages increase by even more, as high as 1.53v with the Extreme setting. The closest level of LLC to the BIOS set voltage at 1.4v is once again LLC Regular which gives 1.389v under load.

llc3.png

Moving to 1.5v, LLC results become even more extreme with load voltages going to dangerous levels. We see as high as 1.671v, a point that we didn’t like to test for more than our 30 second measure interval.

llctable.png

Above is a table summarizing all LLC testing results and as you can see ASUS essentially eliminated all vDroop on the Sabertooth even at the lowest setting for LLC. At any setting above Regular the board overvolts by a fair margin and this effect intensifies as more voltage is used. Looking at these results we have no choice to recommend only the Regular setting be used so you have a better idea of the actual voltage that your chip is getting. Even for a user that stabilizes an OC at 1.4v, up to 1.53v could be pumped through the CPU depending on the level of LLC used. We always recommend caution with overclocking and these results emphasize how important that recommendation actually is.
 
Last edited by a moderator:

Eldonko

Well-known member
Joined
Apr 5, 2007
Messages
5,416
Location
Calgary, AB
Conclusion

Conclusion

The release of the Sandy Bridge E processors and motherboards a few months ago marked a new generation of hardware out there to replace the former enthusiast chipset X58. With the Sabertooth line of boards, ASUS has targeted mainstream enthusiasts; those who are looking for a durable product that provides features they need without charging a premium for features they may never use a la RoG line. Benchers will likely skip the Sabertooth and jump directly to the Rampage IV Extreme for the additional benching-related features but Sabertooth will be on the shortlist for everyone else.

For a gamer, the Sabertooth X79 will certainly be appealing for its ability to overclock the system to 4.3Ghz with next to no effort or overclocking experience. When you include the ability for automatic memory overclocking through ASUS' OC Tuner, we really can't think of a better motherboard for anyone new to the SB-E platform. Add SLI capabilities, PCI-E Gen 3 as well as eight DIMM slots and we’re talking about a match made in heaven for anyone worried about future proofing. The Sabertooth will also appeal to general users that may not upgrade too often since it offers peace of mind by including one of the best warranties around.

Ever since the UEFI standard was implemented, ASUS has pulled into the forefront of BIOS development as our 5Ghz overclock will attest to. The BIOS on the Sabertooth X79 is not only a joy to work with and highly user friendly but during our two weeks of manual overclocking we never had to clear CMOS once. That's really saying something considering some of the disasters that we've seen from ASUS' competitors.

Without a doubt the 40 PCI-E lane X79 chipset lives up to the hype as being the platform to have for multi-GPU configurations. Running SLI on the Sabertooth uses 16 lanes for each GPU and there are still another 8 remaining for other devices. Users will also not have to worry about upgrading their board as higher end PCI-E 3.0 video cards drop. Multi-threaded tasks such as gaming (where applicable), 3D rendering, and audio / video transcoding are where LGA2011 really shines and in these tasks even LGA1155 looks dated.

cons.jpg

Similar to the Rampage IV Extreme, we found it difficult to find much to complain about with the Sabertooth X79. ASUS has done an excellent job with their X79 boards and competitors will have to up their game to even stay close. The only points of contention for us was the degree to which LLC overvolts and relatively long POST times. At 1.5v and the Extreme setting we saw nearly 1.7v, a death sentence for a CPU with prolonged use. At around $350, this board is certainly not cheap but with the extra warranty included, the ASUS BIOS and a great feature set we would find it difficult to say the Sabertooth is not worth the money. Taking all this into consideration we are happy to award the Sabertooth X79 a well-deserved DAM Good award.


Pros

- Excellent mainstream alternative to ROG series
- Industry leading UEFI BIOS
- 4.3Ghz + a memory overclock in 3 seconds
- PCI-E 3.0 ready
- x16 SLI with 8 lanes to spare for other devices.
- TUF Thermal Armor and Thermal Radar for those that love monitoring
- TUF build quality (Military standard chokes, caps, MOSFETs)
- 5 year warranty
- ASUS SSD Caching is foolproof and will be useful for those with smaller SSDs
- USB BIOS Flashback allows for BIOS flash without CPU or memory
- 8 DIMM slots allow for 32GB of memory on a budget


Cons

- LLC overvolts more than many will expect
- $350 price range could be a deterrent for some

 
Last edited by a moderator:
Status
Not open for further replies.
Top