I'm not sure if this is posted already, but it seems like the i7-6700K is beating the i7-5820K and 5960X in games.
https://www.youtube.com/watch?v=ocwwaVGUFtk
https://www.youtube.com/watch?v=ocwwaVGUFtk
New congatec COM Express module with Intel® Xeon® processor and Intel® Iris Pro graphics
Deggendorf, Germany, 23 February 2016 * * * congatec, a leading technology company for embedded computer modules, single board computers (SBCs) and embedded design & manufacturing (EDM) services, has added a particularly powerful graphics version to its COM Express Basic Server-on-Module portfolio. The new module features the Intel® Xeon® Processor E3-1515M v5, fast DDR4 memory and Intel® Iris Pro graphics. The GPU of the new SoC module provides 128 MB eDRAM and with 72 execution units it has three times more parallel execution power than the Skylake architecture without Iris graphics. Developers of tightly packed COM Express based system designs now have access to a new performance class that would have previously required a dedicated graphics unit.
I'm not sure if this is posted already, but it seems like the i7-6700K is beating the i7-5820K and 5960X in games.
https://www.youtube.com/watch?v=ocwwaVGUFtk
There are some genuine surprises and upsets in the results here - not least that very, very few games actually seem to utilise anything like the full power of the eight-core i7 5960X. By and large, it hands in showings commensurate with clock-speed as opposed to the number of cores, meaning that the performance profile of the $1000 CPU is almost entirely like-for-like with the far cheaper 5820K. But perhaps more surprising still is the showing of the 6700K. At stock speeds it's extremely competitive with the more expensive processors. Overclocked, it is the fastest chip in the line-up in all but one title - Crysis 3, where 5960X is faster.
The biggest shock comes from Far Cry 4, where Skylake positively annihilates both Haswell-E chips. We know that it uses one or two cores to power all of the others, and for this one title we see a 20fps advantage over the 5820K. Interestingly, despite the 5820K's overclock only being 200MHz higher, it achieves a full 10fps advantage over the eight-core 5960X. There are some other interesting results - Assassin's Creed Unity really looks GPU-limited here as none of the overclocks look particularly effective, and yet Skylake still manages to pump out a 5fps advantage. It seems that faster architecture can push the GPU limit a little further in a way that clock-speeds can't.
...The takeaways here seem pretty self-evident though - virtually all modern games utilise at least eight threads, and this translates into higher and often smoother performance when a Core i7 processor is compared to its i5 counterpart - in Skylake's case, the Core i5 6600K. However, we went into these tests thinking that the eight-thread support might result in even higher performance when run across more physical full cores. Crysis 3 and Grand Theft Auto 5 do show that there are some gains here, but they're quite rare - and they seem to be eclipsed by the pure single-core brute-force offered by a more modern architecture. In effect, the advantage that kept Intel ahead of AMD in the CPU gaming performance battle also serves to limit the effectiveness of the firm's own many-core designs.
Xeon E3 1515M v5 on SFF board.
http://www.congatec.com/congatec-ag/press-releases/article/congatec-triples-graphics-performance-of-its-server-on-modules.html
Good to see that more of GT4e parts find their way to OEMS.
Skylake with GT4 is just a bit over 210mm^2, I just scaled the die below with the 77mm^2 eDRAM:
Pretty long chip, btw over 65% of the die is IGP
If integrated graphics ever had a chance of beating low end discrete GPUs this is it; that or AMD going nuts and using twice the shaders + HBM on Zen apus...
Now I think I can see why Intel could not make more than 72 EUs iGP, the die would simply too long to fit on that package LOL. Only if Intel would be a bit smarter and re-arrage their die so that they could fit more EUs in KBL, or wait for a 10mm die shrink. *sigh* This does not bode well for KBL GT4e against nVidia.
It's 24EUs in Skylake vs. 20EUs in Haswell. Then there's also the small issue of pitting an i3 vs an i7, afaik the bigger L3 cache helps the iGPU. We should also check for any memory bandwidth differences between the Skylake and the Haswell system.Well rather than fit even more EUs I'd work on the architecture: just look at the previous page tests were 6700k iGPU is well above the i3 4160 one and both have 24EUs.
If generation 9 is such a big jump over Haswell graphics I'd expect no less from generation 10 (?) Kabylake.
Now I think I can see why Intel could not make more than 72 EUs iGP, the die would simply too long to fit on that package LOL.
On the Gen 9 generation though, Intel at least have caught up to Maxwell chips in performance when you roughly normalize to equal specs(Flops/GTexels/Bandwidth). Broadwell was a bit behind. I think the issue is they ultimately treat their graphics as they did with "HD Audio", with bland GUI and no outstanding features.Then there's still the little issue about video drivers... let's hope DX12 helps here too
My calculation based on the whole package (42x28) is 219,4 mm².
Better than Geforce 960 performance in an iGPU.
ASRock is implementing a novel workaround to Intel's recent base-clock overclocking restrictions for its 6th generation Core "Skylake" processors. The company announced the Fatal1ty H170 Gaming K4 HYPER and Fatal1ty B150 Gaming K4 HYPER motherboards, which are based on a common PCB, but differ with chipset - H170 Express and B150 Express. The two use an external clock-generator chip to give cadence to the CPU, circumventing the on-die clock-generator Intel has been using for the past several generations of Core processor chipsets.
With this contraption, you can increase base clock frequency on non-K Core processors in fine increments of 0.0625 MHz, as well as tweak memory clock. Its star attraction aside, the two "HYPER" series motherboards offer mid-range feature-sets, that are nearly identical to those of the Fatal1ty H170 Gaming K4 and the Fatal1ty B150 Gaming K4 boards.
Imagine how everyone who bought into the last great news must feel: now all they have to do is buy a new motherboard every few months :\Big news, non-K BCLK overclocking is back for ASRock motherboards!
And still no native HDMI2.0?
That's pretty nice. I've calculated Broadwell-C to be roughly 200mm2. So for a 10% die increase they managed to take the GPU from Gen 8 to Gen 9, and increase the slice count by 50% at the same process.
I think there's a possibility they can make a 300mm2 Kabylake with 144EUs if they can make it far better in perf/watt.
Just for fun:
-144EUs: 2.3TFlops @ 1GHz
-72GTexels/s @ 1GHz
-200GB/s eDRAM with 2-Hi stacks 256MB capacity
Better than Geforce 960 performance in an iGPU.
Actually it's better than before, not restricted to Z170 motherboards anymore. And hopefully the previous limitations (iGPU, AVX, C-states) aren't there. Just a matter of time till others (MSI, ASUS, Gigabyte, etc.) come up with similar motherboards.
Very impressive, and certainly opens the door for >72 EUs in Kabylake-H GT4e. Adding a fourth slice for 96 EUs (+200GB/s 2x 128MB eDRAM) makes sense IMHO. Would be a solid bump and still leave a lot of room for the 10nm generations.
Would be fun to see a monstruous iGPU with lower clocks, probably more efficient than the current ~1GHz Iris Pro models.
And just a matter of time until intel finds a way to nerf it again. Honestly, I dont see why anyone would mess around with this until it is officially sanctioned by intel.
What do you guys think will be faster? intel hd530 or nvidia gt730 gddr5?
I know this is the skylake thread but the build I'm in the process of uses a haswell pentium g3258 and some people say I can't overclock with windows 10 is this true?
Ok thanks my board is not a z it's a B85-m-g r2.0. But it being asus it should still work right?