If you overclock, you have to much time on your hands for negligible gains...

I just OC'd my ECC RAM from 2400 MHz to 2933
CPU speed is fine for now, not worth the increased power consumption as it runs 24/7

>computer bsod randomly

Great gains!

>>hurr what is safe overclock

If it was safe the manufacturer would clock it to the safest possible speed.

>safest possible
it's like saying that if your computer eats up 500W in the worst possible case, you buy a 550W power supply instead of a 700 or 800W one. ever heard of safe overhead?

Back in the Core 2 Duo/Core 2 Quad and even early Core i days you could get around 1ghz overclocks sometimes even more, now a days, a good portion of 2700x can't even all core 4.3ghz which is it's default single core turbo speed.

Overclocking is dying unless you have a lot of cash to blow.

It's not really dying, they just figured out how to automate overclocking in the chip itself so that it always basically maxes out its performance for the given power/cooling constraints.

Overclocking didn't die, it just got automated.

And the result is he's depressed

look he chanel

Attached: 1541533664325.png (1000x1000, 386K)

>If you OC you better replace your CPU every year
OC'ed since day 1 still runs like a charm.

Attached: every year goyim.png (458x691, 47K)

Acutally, it is pretty much dead.

CPU vendors are releasing high-end SKUs they are already close to their clockspeed/voltage celling. They also lock down the lower-end SKUs.

There's no more Celeron 300A + 440BXs or E6600 + i965/Q6600 + P35/P45. Those days are long gone.

Attached: 7c9.gif (320x240, 970K)

Depends. If you take modern intel hardware - it is pushed to the limits at stock settings and is pointless to overclock. It already runs as hot as if it was overclocked. Older intel hardware could be easily overclocked with close to no effort for very high gains. With modern AMD hardware overclocking memory provides reasonable gains, overclocking cores pretty much as pointless as intel.

Intel's stock clocks are what, 2.8 GHz? And it boosts to 5.2 or something. That guy is right, overclock is automated.

You made the exact same thread a few days ago you ballbag.

They don't boost across all cores though. Neither do AMD. If you want your overclock to be across all cores, you need to do it manually.

My 4790k boosted to 4.4GHz on two cores. I clocked it to boost to 4.8GHz across all four @ 1.275v.

My FX-8320E boosted to 3.5GHz across all cores (full load). It boosted to 4.0GHz across 4 cores (half load). Manually I clocked the chip to hit 4.6GHz across all cores @ 1.404v.

The i5-8600k will boost to 4.3 on one core, 4.2 on 2-4 cores and 4.1 on all six cores.
The i7-8086k will boost to 5.0 on one core, 4.6 on 2 cores, 4.5 on 3 cores, 4.4 on 4-5 cores and 4.3 on all six cores.

If you want the 8086k to boost to 5GHz across all 6 cores, you're going to have to overclock manually. The reason the boost clocks drop the more cores are boosted is voltage regulation. When you start playing with the voltages yourself manually, you quickly find that to get all cores stable at high clocks requires bringing the chip close to the recommended maximum voltage for chip longevity, if not going beyond it.

Take for example. The absolute most recommended voltage for Sandy Bridge was 1.38v. You could get away with more however if you had a meaty cooler or some kind of water cooling solution, or you didn't particularly care about the longevity of the chip. This was well known at the time. I had a Q6600 (max 1.5v) and Q8400 (max 1.375v) and I remember people killing the 45nm generation of chips because they were still pumping the same sort of voltages through them that the 65nm chips were capable of handling. People did the same thing with Sandy Bridge and I remember the Overclockers UK forum making a pinned post about it after communicating with Intel and testing some chips to arrive at a safe voltage.

Most of the blame for all that can be put on Intel, since they removed the recommended VID information from their ark pages. Westmere was the last generation where they had a specific VID information, with the max being 1.3v.

>They don't boost across all cores though. Neither do AMD.
They do. Both Intel and AMD. Can you be any more clueless?

>They do. Both Intel and AMD. Can you be any more clueless?
No they don't.

en.wikipedia.org/wiki/Coffee_Lake#Desktop_processors

And they never have. I've been overclocking as long as I've been building PCs. And that was back in the late 90s.

>Hurr I'm retarded.
Pic related.

>with the max being 1.3v.
And the reason for that is to keep chips below 1.3v. Hence why when you stress an "unoverclocked" K chip, it'll boost to it's default levels across it's given cores, and never pull more than 1.3v. It'll do everything it can, not to cross that threshold unless you clock it manually and regulate the voltage yourself manually. Intel's built in default since Westmere has been to never exceed 1.3v.

It's what happens as silicon shrinks and getting rid of that excess heat becomes more of a challenge.

Attached: Turbo-Freq-3.png (1088x787, 120K)

using aftermarket coolers and pushing the voltage is a waste of time. my ryzen 1700 is OC'd to 3.5 on stock voltage with the stock cooler. super stable with great temps.

It is what it is. The only difference between the 1700 and 1700x is binning. You've just got a decently binned 1700 chip that is capable of running at the 1700x default across all cores.

takes like 1min to OC my GPU and i get 15-20% better performance at negligable power consumption and temperature increase
you have to be retarded for it to take you MUCH TIME to do it
i fucking hate that i replied to this bait thread from a retard pimping videos from a boomer making videos on absolute non issues

it doesn't matter if my 1700 is at 3.2, 3.5 or 3.7. you missed the point. overclocking is good, up to a point. that point is tweaking the voltage and buying aftermarket coolers.

that duron 600 at 900mhz made a difference tho.