• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FEATURED *OFFICIAL* Kaby Lake Overclocking results thread!!!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Did a bit more playing last night with the new APEX board and managed to shave a bit of V_Core off. Ran Aida64 @ 5.0 with 1.184v XMP enabled for 20 minutes before I stopped the test. Reallly nice board for playing with memory too. My previous Z170 board couldn't make these sticks dance like this one does. Set XMP for 3600 with 1.45v managed to get it to 4133 almost stable the test crashed about 5 minutes in and I was out of time.

Johan45 / 7700k / 5.0 @ 1.184v / 3600 CL17 / Max IX APEX ver. 401 / Chilled H2O / Yes

5.0 1.184 20 minutes.JPG

4133  17.JPG
 
Thank you for the update!

It does look fantastic. Apex may have advanced power management for the MOSFET or better power regulation in general? I'm sure Woomack will let us know.

I wonder if Earthdog is getting a review board?

Thanks Again!:)
 
I just said I have an Apex. I plan on going cold with it for the review so it will be a week or two before you see it posted. But its a beast, as you are seeing already.
 
I just said I have an Apex. I plan on going cold with it for the review so it will be a week or two before you see it posted. But its a beast, as you are seeing already.

Yes understood, I did NOT know you were planning on going cold with it. The should quite a review!

Thanks in advance.
 
Looks great Woomack, what voltage were you using?
 
you have that somewhere on hangouts :p ... don't remember exactly 1.6V or something
 
I'm testing lower voltages now and I think that my other Trident Z kit is a bit better. If I find time then will make a thread about Trident Z 3200 c14 memory this week.

at 1.35V it looks like this:
3200 14-14-14 ( didn't try cl13 )
3600 16-16-16 ( cl15 is crashing at the end of hyperpi )
3866 17-17-17
4000 19-19-19
4133 20-20-20
 
Just thinking out loud. It seems that the Kaby Lake chips can run higher speeds and or can OC memory higher then the Sky Lake chips. Just wondering if it's just due to the maturity of the chip or the boards.
 
In theory yes. In real I see no difference but ASUS really improved their memory support comparing to Z170. Also other brand motherboards have at least DDR4-3600 support when on Z170 max guaranteed was 3200. I mean boards like lower gaming series for MSI have now 3600-3866 when it was 3200-3600, for ASRock it's about 1-2 memory ratios higher in dual channel and probably all popular series have 3733-4000 support, for Gigabyte I'm not sure as they had issues with some boards and I wasn't really following but Z270 has for sure higher guaranteed clock.
If you take a look at the motherboards then one of the "special features" which were not so popular in Z170 is now better DIMM signal ( DDR4 boost or something similar ).

ITX boards were always better overclocking because of shorter traces and only 2 slots. Apex is like ITX board on large PCB with better power section and additional PCIE slots. ASRock made the same but on micro ATX and it was OCF-M. Apex is EATX. Afaik there won't be Z270 OCF ( info from ASRock rep ) so Apex seems like the best option for overclocking now. ASRock OCF/OCFM is still great and has KL support.
 
Last edited:
All the DDR4 memory was not out when the motherboard manufactures tested memory configurations on the Z170s before they were released. The Z170 motherboard manufactures did not have a great range of DDR4 memory to test with back in 2015.

I say the distance of the tracers are about the same on all motherboards because of Intel's specification on the cooler to memory layout distance. From what I see it is the sub-timings that the different board partners use to run unspecified by Intel OC memory.
 
All the DDR4 memory was not out when the motherboard manufactures tested memory configurations on the Z170s before they were released. The Z170 motherboard manufactures did not have a great range of DDR4 memory to test with back in 2015.

I say the distance of the tracers are about the same on all motherboards because of Intel's specification on the cooler to memory layout distance. From what I see it is the sub-timings that the different board partners use to run unspecified by Intel OC memory.
The traces definitely are shorter on the APEX, how many motherboard run dual channel in slots 1 and 2?
Woomack was right about the ASUS mem support even when I sold the Hero last month there wasn't a BIOS that would let me run this memory over 4000 consistently. Besides Haswell "E" (DDR4) debuted in summer 2014 so board makers have had plenty of time to tune their BIOS. Now the Samsung "B" die was available in NOV-DEC 2015 the biggest problem was Manufacturer support for BIOS updates. Some did like Asrock and some didn't like ASUS
 
Samsung B was available when Z170 boards were available. Since then we haven't seen any new Samsung IC. New were only some batches of Micron/Hynix which were not even overclocking better. Samsung D 4133 kits were available when Z170 boards appeared on the market and the same IC was in earlier kits for X99 boards. Max frequency DDR4 kits designed for Z170 which were not released in mass market were ~DDR4-4500. We could see them in press releases but never in stores.

Intel specs has nothing to do with improved traces layout, thicker PCB and other things. Manufacturers only stick to space required for stock Intel cooler and that's all. Some motherboards have slots a bit closer to the CPU. Also 2 slot motherboards have closer traces of starting DIMM as it's actually the 2nd DIMM counting from CPU side. You can look at that as 2 memory slot boards have only DIMM 2 and 4 so it's shorter by the length of 1 memory slot.
 
Back