Trying to overclock an 8700k. Computer wont even boot at 4.6ghz @ 1.4v

dont worry, I have similar experience with ryzen

this

Attached: 2019-02-15 21_04_05.png (661x209, 9K)

OP here. I read this as well. I'm not sure what I could be doing wrong.

I'm literally only changing clockspeed and vcore

You dont OC a ryzen CPU though. Just buy fast RAM and good cooling and let it do the rest by itself.

Pumping more than 1.3v through your chip isn't safe unless you're on monster cooling.

LLC?

could be motherboard VRM, or cooling, or terrible powersuppy, if its neither of those then I really dont know

fuck you piece of shit, RAM OC depends on the memory controller in the CPU chip as well, so even though I have premium b-die ram, it doesn't OC that well and pretty much doesn't boot, regardless of settings or voltage below a certain latency.
The chip itself doesn't OC that well either.
Honeslty this whole OC business if a fucking scam, you can lure customers into buying your shit and at the same time you aren't guaranteeing them jack shit, diddly squat! so if it OC like utter garbage you can say "ah but it fulfills it's advertized base clocks"

Try resetting CMOS

imagine the joys of this baby

Attached: 2600.jpg (1564x2148, 953K)

Get a better motherboard and faster ram you fucking retard. Holy fuck.

Im saying that Ryzen sucks at manual OC but precision boost does its job and you'd get pretty much the same performance if your RAM and cooling are adequate. But they certainly wont help you getting better OC and I agree it's a waste of time on Ryzen.

OP are you using Load Line Calibration? I need it for mines to be stable or else O get vdroop and it crashes.

>Im saying that Ryzen sucks at manual OC
and I am saying is that 2600 is a overhyped piece of shit, it's literall bottom binned shit people who expect easy overclocks while at low voltage are getting scammed.

> but precision boost does its job and you'd get pretty much the same performance
precision boost is very retarded and voltage happy, you absolutely need voltage offset to even make it decent.

>and I am saying is that 2600 is a overhyped piece of shit, it's literall bottom binned shit people who expect easy overclocks while at low voltage are getting scammed.
Not that guy, but...

That's what the X chips are for. It's no different than Intel having K chips. But unlike Intel, AMD keeps regular chips unlocked so you can attempt at least some kind of overclocking on them.

I haven't had an AMD chip in a while. But the last one I had was an 8320e which was the lower TDP version of the 8320. My particular chip I was able to get 4.4GHz stable across all cores, while others on the OCUK forums I use were unable to get the original or E chip to similar clocks, even pumping more voltage through them. It's luck of the draw on each individual chip. The 4790k I got after I was able to get 4.8GHz across all four cores at 1.275v, less than the chips own recommended 1.3v max for longevity. Again, I got a good chip. Others needed more voltage for similar clocks or less clocks.

My motherboard is great, and my ram is 3200. Are you ok, user?
Hmm, it could be my powersupply? It is pretty old, HX650

> But unlike Intel, AMD keeps regular chips unlocked so you can attempt at least some kind of overclocking on them.
Selling meme dreams with no promises.
Gee sure is nice for if the customer buys on the vague idea of something.

I mention powersuppy because they are responsible for the initial stability of the voltage delivered, VRM just cleans it up as best it can.
Again not to imply that is the reason for your troubles, just that a garbage powersupply will produce probems if the voltage ripple is too high.

THIS IS WHY YOU BUY AN 8086K
>tfw 5ghz on all cores

Attached: 1550242723984.jpg (1100x1500, 96K)

AMD Ryzen 2600 = 3.4 - 3.9 = £144
AMD Ryzen 2600X = 3.6 - 4.25 = £184

The 2600 has a lower TDP than the X too. So save yourself £40 and perform a small clock and you've got a less power hungry and cheaper X chip. If you get more than 4.25 out of the standard 2600, then everything else is a bonus.

AMD has always been good to overclockers. When yields for Phenom were shit, they simply disabled problematic cores and then provided the option to re-enable them should you be so inclined. That's why we ended up with 3 core Phenoms that could be unlocked to 4 cores and still overclocked.

Intel on the other hand decided to lock their chips to prevent people doing what they were doing with AMD chips and then hiked the price up on their "specially binned" K series for overclockers.

Thanks anons. Didn't know what this was, and I'm now looking into it.

I had 5ghz all cores on my 8700k. Would expect more than that from an 8086k...

>The 2600 has a lower TDP than the X too.
Because its lower clocked.
>So save yourself £40 and perform a small clock and you've got a less power hungry and cheaper X chip.
This is highly optimistic and as I said almost scam tier, in getting customers buying the bottom binned garbage chips in vague hopes that it's as good as better binned ones.
>If you get more than 4.25 out of the standard 2600, then everything else is a bonus.
Most people should not expect more than 3.9-4.0GHZ on the 2600, anything beyond that is very lucky.

>This is highly optimistic
Not really. Only a moron wouldn't be able to get 4.25GHz out of a 2600.

Looks like you got a shit chip.

>Not really. Only a moron wouldn't be able to get 4.25GHz out of a 2600.
Ah I see, so you are covering up a scam with "dear customer it must be your fault, clearly you don't know what you are doing".
As I said 3.9-4.0GHZ is what people should realistically expect from 2600, anything beyond that is highly optimistic considering you can end up with chips that literally wont boot past 3.9 even with 1.42v

I know......but turns out most of them are like this, this ins't the "exception' but the majority.

My 2600 is about the same as this guy's, needed 1.4v for 4GHz, fell for the "2600(nonX) OCs to 4.2GHz ~1.3V" meme. Posted about it here and got screamed out

Exactly. Even this drunk Irishman can do it.

youtube.com/watch?v=B9V4qMsMDrc

>Posted about it here and got screamed out
Not surprising, since Jow Forums has quite a shill force working here at certain times of the year.
/v/ is for video game shilling
Jow Forums is for hardware shilling
if you disagree with the narrative or bring up any critique to the product being pushed you'll get shitposted to death

>fell for the "2600(nonX) OCs to 4.2GHz ~1.3V" meme.
ahem it's actually
2600 to 4.2GHz at 1.2V and sub 50C temps using the stock cooler while doing prime95 or aida64

So I've cranked LLC to level 1. Now I notice in HWmonitor the Vcore is actually close to what I set it in the bios (1.375). Before the monitor would shot like 1.1v or something low (I assumed it was just incorrect).

Currently running prime95 blend tests at 4.8ghz (-1 AVX offset, so actual clock is 4.7 it seems?), at 1.375v.

Well OP can't.

I got a 2500k @ 4.6ghz OP. Need LLC and a motherboard with good VRMs.

show*

HWMonitor reports vcore at 1.328v (bios 1.375) with an LLC at level 1.

please dont use hdmonitor, instead use HWiNFO64.

Alright, will get it now.

HWinfo reports 1.439v vcore.

Is this accurate? I set vcore to 1.375 with level 1 LLC.

Setting LLC too high, can OVERvolt.

Do not just set it to the highest.
If at idle your voltage is more than what you put in bios then you are overvolting with LLC.
Set it much lower and work your way up.

Keep in mind that the higher you go in voltage, the less aggressive LLC you want to use if you value your CPU life.
Because the way LLC works is that it attempts to compensate for voltage droop when a load hits.
When that load disappears and it goes back to idle, the droop has the opposite effect and voltage shoots up into a spike.

So let say you set your voltage to 1.3 and your Vdroop is 0.05V so it goes form 1.3 to 1.25V under load, as you leave load, it will spike and go some where around 1.31 and then stabilize at 1.3

With flatline LLC for instance
you have 1.3 idle, then 1.3 under load but when going back to idle, your voltage will spike to 1.36 and then stabilize back to 1.3

My point here is that you don't want your spike to be in the dangerous region if you value long term life.

Generally speaking you want some voltage droop, as it's impossible for the VRM to compensate for doops and spikes, so you have to roll with it.
If you are in 1.3 voltages then you have more leeway and can use LLC to stabilize it around there, rather than raising voltage by much.

Take away: Your LLC should NEVER push your voltage beyond a flatline, it should not overvolt. And for most users, some Vdroop is nessesary and "healthy" for long term component life.

Thanks chaps. Although if its overvolting to 1.439v and still crashing (it just did), I don't like my hopes of any success at lower levels. Will need to downclock to 4.6ghz and try again.

too high voltage can cause instability as well, though massive heat spikes man. Set the voltage to something sane like 1.3 or 1.35 and make sure it doesn't overvolt from what you state in bios. from there you can actually try shit