ASRock Z97 Extreme6 Review: Ultra M.2 x4 Tested With XP941
by Ian Cutress on May 24, 2014 12:00 PM ESTScientific and Synthetic Benchmarks
2D to 3D Rendering –Agisoft PhotoScan v1.0: link
Agisoft Photoscan creates 3D models from 2D images, a process which is very computationally expensive. The algorithm is split into four distinct phases, and different phases of the model reconstruction require either fast memory, fast IPC, more cores, or even OpenCL compute devices to hand. Agisoft supplied us with a special version of the software to script the process, where we take 50 images of a stately home and convert it into a medium quality model. This benchmark typically takes around 15-20 minutes on a high end PC on the CPU alone, with GPUs reducing the time.
Console Emulation –Dolphin Benchmark: link
At the start of 2014 I was emailed with a link to a new emulation benchmark based on the Dolphin Emulator. The issue with emulators tends to be two-fold: game licensing and raw CPU power required for the emulation. As a result, many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance. This benchmark runs a Wii program that raytraces a complex 3D scene inside the Dolphin Wii emulator. Performance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU. Results are given in minutes, where the Wii itself scores 17.53; meaning that anything faster than this is faster than an actual Wii for processing Wii code, albeit emulated.
Point Calculations – 3D Movement Algorithm Test: link
3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.
Encryption –TrueCrypt v0.7.1a: link
TrueCrypt is an off the shelf open source encryption tool for files and folders. For our test we run the benchmark mode using a 1GB buffer and take the mean result from AES encryption.
Synthetic – 7-Zip 9.2: link
As an open source compression tool, 7-Zip is a popular tool for making sets of files easier to handle and transfer. The software offers up its own benchmark, to which we report the result.
Facebook
Twitter
RSS
43 Comments
View All Comments
457R4LDR34DKN07 - Saturday, May 24, 2014 - link
The Gen2 M.2 x4 PCIe looks very impressive. Its too bad that Asrock did not include this in their mITX offering.457R4LDR34DKN07 - Saturday, May 24, 2014 - link
I mean Gen3smoohta - Saturday, May 24, 2014 - link
Regarding the Dolphin benchmark- you wrote:"Results are given in minutes, where the Wii itself scores 17.53; meaning that anything above this is faster than an actual Wii for processing Wii code, albeit emulated."
This should read "... meaning that anything below this is faster than..." - right?
Ian Cutress - Saturday, May 24, 2014 - link
Bad choice of words - anything 'above on the graph' would be faster based on the orientation of results. I've updated it to remove the ambiguity. Thanks for pointing it out :)Ian
r3loaded - Saturday, May 24, 2014 - link
The overarching conclusion from this review - Intel needs to put far more PCIe 3..0 lanes on their CPUs and chipsets if we want motherboards where connectors don't play musical chairs with each other for bandwidth. Also, Samsung needs to make the XP941 a retail product right now!Tunnah - Saturday, May 24, 2014 - link
I think the reasoning behind an either/or situation with M.2/SATAe is sound - the consumer most likely will not require 2 high end SSDs, and will go with one that is either M.2 or SATAe, the rest can be served via normal SATA ports.Also I think SATAe, while a great idea, will lose out to the convenience and capability of M.2; SATAe would have been great 2 years ago but now M.2 is here it is absolutely perfect for an OS drive, cutting down on not just size but cabling.
SirKnobsworth - Sunday, May 25, 2014 - link
Agreed about SATAe, especially since (a) SSDs don't really need that much circuit board area and (b) A RAID array of two SATA SSDs offers at least as much theoretical bandwidth as a SATAe while taking up the same amount of ports on your board.Babar Javied - Sunday, May 25, 2014 - link
Exactly my thought. So why have SATAe???? Its great that it uses PCI but it doesn't really offer anything that we couldn't already do.As you've said, "A RAID array of two SATA SSDs offers at least as much theoretical bandwidth as a SATAe while taking up the same amount of ports on your board".
There seems to be a lot of stupid decisions being made regarding ports. DDR4 is another example of a useless upgrade.
Death666Angel - Tuesday, May 27, 2014 - link
"A RAID array of two SATA SSDs offers at least as much theoretical bandwidth" But your are losing random performance when going RAID and increase the chance of a disk failure. Why would you even compare them?Galatian - Saturday, May 24, 2014 - link
Ian,Can you explain to me why no mainboard manufacturer is using 4 lanes from the chipsets PCIe 2.0? I mean that would be enough for the Samsung SSD and still has room left. I mean what do I gain from all those SATA and USB ports? Who is actually using all of them?
Right now I have to choose between either a slow M.2 slot because they only allocate 2 lanes or I can go with ASRock which feels like overkill and takes away CPU PCIe lanes.