Coffee Lake i7-8700K modded to work on Z170

SPARTAN VI

SPARTAN VI

[H]F Junkie
Joined
Jun 12, 2004
Messages
8,760
Some folks at HWbot have got their i7 8700K working on Asrock Z170 boards with a pencil mod and custom UEFI. Instructions are in the HWbot OP linked. This isn't the same mod from December with an i3 8350K on a Z170 board, there were some issues/sacrifices in regards to PCIe functionality.

Mod point is below, connect both pads with tin/silver ink etc.

Flash bios http://picx.xfastest.com/nickshih/asrock/Z17MOCF751A.rar with kabylake CPU, shut down insert coffee and go.

XP working fine
Click to expand...

Would be tempted to try this if it can ever work with my Asrock Z170 ITX board.. it's like the 771 to 775 mod, which I couldn't resist either. :D

z170m_8700k.jpg z170m_UEFI.jpg

Yes, there's an old (i.e. stale) discussion on this in the motherboard sub-forum, but I feel we may get more takers from folks willing to try this mod in the Intel sub-forum.
 
Last edited:
Work with Z97 yet? lol

I just hope that their next generations do make you update the MB every single time.. annoying
 
I wouldn't get too excited though.... I bet you anything Intel will have Microsoft patch it out in a soft-microcode update.

They did the same thing with the Pentium G3258 when people were overclocking them on H97 boards. One fateful Windows update comes along and suddenly your Asrock H97M Anniversary board wouldn't boot if it was overclocked even the slightest.
 
NoxTek said:
I wouldn't get too excited though.... I bet you anything Intel will have Microsoft patch it out in a soft-microcode update.

They did the same thing with the Pentium G3258 when people were overclocking them on H97 boards. One fateful Windows update comes along and suddenly your Asrock H97M Anniversary board wouldn't boot if it was overclocked even the slightest.
Click to expand...

Yeah that was some dirty shit, not surprised that FTC or whoever never went after them though because they are all already worthless at controlling intel and microsoft illegal activity.
It was really telling that identical systems could still boot linux and not give a fuck but some mysterious magical windows update makes you think the computer is bricked.
 
Fuck Intel...we all knew this was the case. I've been saying so since July '17 when I got a couple 3 day bans for calling out the IDF on this very issue.
 
There is a danger the mobos VRMs could overheat when overclocking, depending on the class of motherboard.
50% ish higher current is not to be sniffed at.
I wouldnt be surprised if the better boards are fine.
 
Nenu said:
There is a danger the mobos VRMs could overheat when overclocking, depending on the class of motherboard.
50% ish higher current is not to be sniffed at.
I wouldnt be surprised if the better boards are fine.
Click to expand...

I was mulling over the idea of trying an 8700K on my current board. One can still dream right?
 
Nebulous said:
I was mulling over the idea of trying an 8700K on my current board. One can still dream right?
Click to expand...
I cant see it being a problem if you keep the voltage down and dont push clock speeds much.
Mobos have a fair headroom for overclockers anyway so you can push quite a lot higher.
Just remember at stock speeds its at +50% ish power use already.

Power use increases % wise with clock speed and to the square of voltage used.
Its when you start pushing voltage that power use increases the most.
If you can run it at lower voltage you have more headroom for clockspeed, assuming voltage is high enough to support it.

This assumes there is a problem with the VRMs getting too hot.
Unless there is a temp sensor its hard to know.
If there is, you are golden.

You may be able to use extra cooling on the VRMs.
Having not researched this at all (and having my thick sunday head on) I cant provide any info beyond that, yet.
 
Nenu said:
I cant see it being a problem if you keep the voltage down and dont push clock speeds much.
Mobos have a fair headroom for overclockers anyway so you can push quite a lot higher.
Just remember at stock speeds its at +50% ish power use already.

Power use increases % wise with clock speed and to the square of voltage used.
Its when you start pushing voltage that power use increases the most.
If you can run it at lower voltage you have more headroom for clockspeed, assuming voltage is high enough to support it.

This assumes there is a problem with the VRMs getting too hot.
Unless there is a temp sensor its hard to know.
If there is, you are golden.

You may be able to use extra cooling on the VRMs.
Having not researched this at all (and having my thick sunday head on) I cant provide any info beyond that, yet.
Click to expand...



antok86 said:
hmm....interesting
Click to expand...

Hells yeah. I'm sure my board has a robust VRM section if my memory serves me right. From what I remember it's 12Phase Taken from here: https://www.tweaktown.com/reviews/8009/asrock-z270-extreme4-motherboard-review/index3.html:


"The motherboard's VRM for the CPU is a 6+4+1+1 phase VRM."

The issue here is if i have the balls to do it for [H] :D
 
So does the same go for Z270 chip set motherboards? If so then it will save me the money of needing to buy a Z370 Motherboard.
 
If you care about things actually working - I would not rely upon any of these mods. It's an interesting experiment, but I would not think of it as anything other than that.

If you want a CFL - get something designed to work with it.
If this scenario pisses you off - don't buy it, and either skip Intel until they do what you want, or buy AMD. This is a real "vote with your wallet" scenario. If you feel is is BS - don't kinda work around the BS, don't give them any money.

Most of the impetus for a chipset roll and the claimed incompatibility is because there is a lot of crap stuff in the channel which barely met the original specs and absolutely cannot be modded to work. I suspect most of these are OEM customs, where every corner possible is cut, and some which should not be cut still are.
Good boards vastly exceeded these needs and clearly can be modded to work - but man, you're throwing some dice if you want to make a working system.
 
PhaseNoise said:
If you care about things actually working - I would not rely upon any of these mods. It's an interesting experiment, but I would not think of it as anything other than that.

If you want a CFL - get something designed to work with it.
If this scenario pisses you off - don't buy it, and either skip Intel until they do what you want, or buy AMD. This is a real "vote with your wallet" scenario. If you feel is is BS - don't kinda work around the BS, don't give them any money.

Most of the impetus for a chipset roll and the claimed incompatibility is because there is a lot of crap stuff in the channel which barely met the original specs and absolutely cannot be modded to work. I suspect most of these are OEM customs, where every corner possible is cut, and some which should not be cut still are.
Good boards vastly exceeded these needs and clearly can be modded to work - but man, you're throwing some dice if you want to make a working system.
Click to expand...

Yup. I'm sure my Z170 board would have been fine if not pushing too hard, but given the uncertainty I understand why Intel pushed out a new revision. The other choice was to pull an AMD and let their customers sort it out...
 
Nenu said:
There is a danger the mobos VRMs could overheat when overclocking, depending on the class of motherboard.
50% ish higher current is not to be sniffed at.
I wouldnt be surprised if the better boards are fine.
Click to expand...

Exactly. The problem wasn't that some boards could support 50% loads above the spec, but some boards couldn't. So a new spec had to be defined.
 
PhaseNoise said:
If you care about things actually working - I would not rely upon any of these mods. It's an interesting experiment, but I would not think of it as anything other than that.

If you want a CFL - get something designed to work with it.
If this scenario pisses you off - don't buy it, and either skip Intel until they do what you want, or buy AMD. This is a real "vote with your wallet" scenario. If you feel is is BS - don't kinda work around the BS, don't give them any money.

Most of the impetus for a chipset roll and the claimed incompatibility is because there is a lot of crap stuff in the channel which barely met the original specs and absolutely cannot be modded to work. I suspect most of these are OEM customs, where every corner possible is cut, and some which should not be cut still are.
Good boards vastly exceeded these needs and clearly can be modded to work - but man, you're throwing some dice if you want to make a working system.
Click to expand...

You mean to tell me an OEM z170 board couldn't have worked with an OEM chip like a i3-8100 or an i5-8400? I just don't buy it. The differences aren't that big either on the 14nm process or the power requirements. This was a money grab by Intel.

And I did vote with my wallet after Intel made new Z270 chipset boards obsolete after 7 months. Fuck that.

Edit...I just saw the date of that posting. Sorry for the necro.
 
kirbyrj said:
You mean to tell me an OEM z170 board couldn't have worked with an OEM chip like a i3-8100 or an i5-8400? I just don't buy it. The differences aren't that big either on the 14nm process or the power requirements. This was a money grab by Intel.

And I did vote with my wallet after Intel made new Z270 chipset boards obsolete after 7 months. Fuck that.

Edit...I just saw the date of that posting. Sorry for the necro.
Click to expand...

They specifically do not want to get into the "this chip works on this board, not this one, etc." That's a support nightmare for them, with basically no return. Again - many big PC OEM boards are absolutely utterly awful. They will, have, and do cripple boards to save a quarter. I have fought with some vendors at length about this very issue in the past. The corners which get cut are often unbelievable. I can't name names here, but i don't think you'd be surprised by the list of biggest offenders.

A reasonable aftermarket retail board would almost certainly be fine. But again, Intel doesn't want to get into the business of having to determine one by one which works and which doesn't.

I'm not saying it is okay, they could have done a better job requiring more headroom from the start, but caved to OEMs who wanted to save that quarter. And then this is the fallout of that decision.
 
  • Like
Reactions: Nenu
like this
PhaseNoise said:
They specifically do not want to get into the "this chip works on this board, not this one, etc." That's a support nightmare for them, with basically no return. Again - many big PC OEM boards are absolutely utterly awful. They will, have, and do cripple boards to save a quarter. I have fought with some vendors at length about this very issue in the past. The corners which get cut are often unbelievable. I can't name names here, but i don't think you'd be surprised by the list of biggest offenders.

A reasonable aftermarket retail board would almost certainly be fine. But again, Intel doesn't want to get into the business of having to determine one by one which works and which doesn't.

I'm not saying it is okay, they could have done a better job requiring more headroom from the start, but caved to OEMs who wanted to save that quarter. And then this is the fallout of that decision.
Click to expand...

This really only affects OCing and MCE, etc. If it is running within the 65W or 95W specs like a Dell/Lenovo/Acer, etc. prebuilt, I don't see any difference between KBL and CFL in terms of board requirements. A H110 board could run a 9900k when running at 95W. Would you? No, but it's as simple as making the bios and ME compatible with the newer chip and it would work assuming it would run a 95W CPU to begin with. You just wouldn't have as much turbo flexibility on lower end boards (theoretically).
 
https://www.pcbuildersclub.com/en/2...ntels-lga-1151v2-z370-and-z390-are-pointless/

Tape up all the extra power pins and it still works fine. A complete money grab by Intel.

Link to Youtube vid:


Roman Hartung has proven very clearly that the LGA-1151v2 socket is basically completely unnecessary. The pins withstand even a very limited power supply without any problems, there is no damage to the mainboard, the socket or the processor. This also proves once more that Intel probably didn’t allow compatibility with the old motherboards for sales reasons.
Click to expand...
 
You must log in or register to reply here.
Back
Top