990FX Motherboard Roundup with Thuban and Bulldozer – A Second Wind for ASUS, Gigabyte, MSI and Biostar
by Ian Cutress on April 5, 2012 11:00 AM ESTUSB Speed
For this benchmark, we run CrystalDiskMark to determine the ideal sequential read and write speeds for the USB port using our 64GB Patriot SuperSpeed USB 3.0 drive. Then we transfer a set size of files from the SSD to the USB drive, and monitor the time taken to transfer. The files transferred are a 1.52 GB set of 2867 files across 320 folders – 95% of these files are small typical website files, and the rest (90% of the size) are the videos used in the Sorenson Squeeze test.
A lot of how the system deals with USB interactions is based on the hardware, including BIOS stated priorities, but also on the software and drivers supplied. This results in our spread on USB 2.0 and USB 3.0 speeds/times. On the whole the Sabertooth 990FX is a good performer, while the MSI tends to lag behind in real world testing.
SATA Testing
We also use CrystalDiskMark for SATA port testing. The operating system is installed on the SSD, and the sequential test is run at the 5 x 1000 MB level. This test probes the efficiency of the data delivery system between the chipset and the drive, or in the case of additional SATA ports provided by a third party controller, the efficiency between the controller, the chipset and the drive.
For SATA testing, again we get a processor independent spread. The interesting thing here is the SATA 6 Write results, in which the ASUS boards clearly have an advantage over the other products. This is most likely due to more agressive BIOS settings at default.
DPC Latency
Deferred Procedure Call latency is a way in which Windows handles interrupt servicing. In order to wait for a processor to acknowledge the request, the system will queue all interrupt requests by priority. Critical interrupts will be handled as soon as possible, whereas lesser priority requests, such as audio, will be further down the line. So if the audio device requires data, it will have to wait until the request is processed before the buffer is filled. If the device drivers of higher priority components in a system are poorly implemented, this can cause delays in request scheduling and process time, resulting in an empty audio buffer – this leads to characteristic audible pauses, pops and clicks. Having a bigger buffer and correctly implemented system drivers obviously helps in this regard. The DPC latency checker measures how much time is processing DPCs from driver invocation – the lower the value will result in better audio transfer at smaller buffer sizes. Results are measured in microseconds and taken as the peak latency while cycling through a series of short HD videos - less than 500 microseconds usually gets the green light, but the lower the better.
DPC Latency always shows the biggest differences between our motherboards. It is hampered by the fact that it can be disturbed by software. In particular, ASUS' AI Suite and Gigabyte's EasyTune software, when installed, initialise themselves on startup and continuously monitor the motherboard sensors (by pinging them every 3-5 seconds). This request for data has a higher priority, and thus when we do our DPC checks, those boards when the software is enabled shoot up to 3000 microseconds or more. For these tests, we disable the software (as it shows uncharacteristic spikes) to take a reading while playing HD video. In this case, we can confirm that ASUS do it best, and you get what you pay for with respect to the Biostar motherboard.
Facebook
Twitter
RSS
57 Comments
View All Comments
mmstick - Thursday, April 5, 2012 - link
The primary problem with AMD FX is that in order to use the full power of the FPU the program needs to be compiled with FMA4 support, else it is only using half of the FPUs, thus making it a quad core. Secondly, many Windows-based programs are compiled with the Intel C+ compiler, so although the FX may support AVX and many other instructions, the compiled program sees it as a non-Intel CPU so it disables those instruction sets, allowing Intel CPUs to be optimized, and AMD CPUs to remain deoptimized. This is what happens when you are up against someone with the most market share, whom has the ability to dictate what instruction sets they want programmers to use. As well, when people say they are going to buy Intel CPUs instead because they claim AMD didn't make a good processor, why do you think they can't be on top of performance? Without R&D budget there isn't much that can be done, and when you face someone who practically owns a monopoly, that makes it even moreso harder to compete.Omoronovo - Thursday, April 5, 2012 - link
The IC++ compiler has not done that since 2010 when they were forced to settle their antitrust dispute with AMD.DigitalFreak - Thursday, April 5, 2012 - link
" This is what happens when you are up against someone with the most market share, whom has the ability to dictate what instruction sets they want programmers to use. As well, when people say they are going to buy Intel CPUs instead because they claim AMD didn't make a good processor, why do you think they can't be on top of performance? Without R&D budget there isn't much that can be done, and when you face someone who practically owns a monopoly, that makes it even moreso harder to compete."Waaaaah. It's always someone else's fault.
anubis44 - Friday, November 9, 2012 - link
"Waaaaah. It's always someone else's fault."Well, sometimes it really IS someone else's fault. If the mafia had it in for you, and cut your brake cables and burnt your house down when you weren't looking, you'd say it's 'someone else's fault' too. Intel's blackmail and threats to suppliers who used AMD processors kinda screwed AMD over just a tad.
That said, I think now that Jim Keller is back at AMD and head of AMD's CPU division, it won't be too long before AMD is seriously back in the game.
Monkeysweat - Thursday, April 5, 2012 - link
I saw them on some of the benchmarks, why didn't you post them along side the AMD benchies for gaming?If we are looking at a roundup of the best of what AMD and it's partners have to offer, I'd like to see what the competing team brings to the table,, just leave em stock and even let the AMD ones get overclocked.
I wouldn't even worry about cherry picking the Intel combos,, just something random.
Beenthere - Thursday, April 5, 2012 - link
AMD has not abandoned the highend CPU market. Their focus may be broadening but that does not mean they will discontinue discrete highend desktop CPUs for at least several years. Eventually everyone except a small group will use APUs as they will deliver the best performance/value proposition. Only extremists will bother with a discrete CPU/GPU with higher power consumption, increased heat and little practical benefit for mainstream users.Articuno - Thursday, April 5, 2012 - link
It's a pretty nice chipset and the lower tier boards are quite cost-effective. Just wish Bulldozer was competitive with Intel, let alone their last gen chips.Mathieu Bourgie - Thursday, April 5, 2012 - link
Thank you for this article Ian.Are there any chance that we'll see a review of the newer FX-6200 CPU or at least have data for it in the CPU bench? Considering that it's 500MHz faster than the model that it's replacing and no major site (or any that I can see) did a review of it, it'd be interesting to see how it performs.
I'm curious to see if it's a valid alternative, in any way, for $170, vs the Intel Core i5-2300 ($180).
I don't expect any miracle for gaming performance, but for workstation workloads (Photoshop, video editing and the like), who knows?
Thanks,
Mathieu
cosminmcm - Thursday, April 5, 2012 - link
There is a review at pcper, a good one. The processor is pretty weak, nothing exciting there. Thuban walks all over it.Mathieu Bourgie - Friday, April 6, 2012 - link
Thanks, I didn't see that.Quite disappointing indeed.
Here's about that Piledriver or Trinity are more competitive.