Intel removes non-K overclocking after the fact, pushes blue-screening updates to punish overclockers through Windows Update
https://forum.teksyndicate.com/t/rant-30-intel-removes-non-k-overclocking-after-the-fact-tek-syndicate/96946
Who the fuck buys a non K and expects a decent overclock anyway, yeah only morons.
>>52983171
/thread
>>52983171
The fact is, people should be able to do whatever the fuck they want with their hardware. Sure, intel is free to create cpus meant to overclock, but that does not mean they should limit other cpus.
>>52983245
>The fact is, people should be able to do whatever the fuck they want with their hardware.
Wrong.
>>52983255
This logo sincerely disgusts me.
It feels me with its repugnancy the second I look into it.
>>52983245
>buy hardware
>"I should be able to hack into NSA and FBI servers cause I paid for the hardware."
>>52983171
The people who bought non-K were not expecting to over cock it. Then they realized they could and it worked. Then Intel pushed and update to remove it and it broke peoples systems.
Its not as simple as people buying non-k processors and not being able to overclock.
It being possible and intel pushing code onto your computer to remove features is so fucking jewish.
>>52983245
>Sure, intel is free to create cpus meant to overclock, but that does not mean they should limit other cpus.
Then they'd have a lineup of like 10 CPUs instead of 88.
>>52983245
>buy non-K CPU knowing very well it's not designed to be overclocked
>people put out bios patches to overclock them anyways
>intel reverts it
How is this any different from
>AMD releases 6950s which are gimped 6970s
>People put VBIOS flashes on them to turn them into 6970s
>AMD reverts this
>>52983314
As far as I'm concerned, the government's hardware is not mine, and as such, I'm not free to access it without due permission. That's not to say that the opposite is true, as much as I'd like it to be.
>Intel is not letting me overclock the $300 CPU I paid for
God dammit AMD this is what happens when you go to shit
>>52983073
I suggest you buy a real processor.
Cough cough AMD
>>52983334
It's not. I think artificially limiting anything you buy is plain stupid, and any company that does that doesn't deserve my money.
That being said, that gpu scandal is already 6 years old. I'd prefer something newer in terms of comparison.
>>52983245
I agree, butttttt I'm pretty sure the locked versions are sold that way for a reason. T
They don't just slap a K on the end and higher the price, each chip is unique you know this. It makes sense to sell lower quality units locked and the ones that test higher unlocked. Your warranty isnt voided by overclocking on the K series, not sure how they would know but still.
People who overclock locked chips always get pathetic overclocks anyway, "look im 400MHz higher!!" yet they've disabled the turbo boost feature which is usually higher than their overclock.
Point being, nobody seriously buys a locked processor and expects a high overclock, correct me if I'm wrong, it's rare these chips even hit 4GHz safely.
what about the pentium g3258?
>>52983358
but the govs hardware was payed by your taxes.
>>52983453
no AVX, no buy.
That's right, the AVX instructions and registers are perfectly functional on those CPUs, just disabled to jew you.
>>52983453
>>52983542
...And not JUST AVX
>>52983446
I don't know enough about silicon and lithography to debate the chip quality argument. That aside, if you buy something like an airsoft or paintball gun, you get it with the expectation that you can take it apart (Whilst voiding the warranty) and do whatever the hell you want with it.
The same should apply with cpus in my opinion. Like with many android smartphones, using fastboot oem unlock will set the state from locked to unlocked. While using oem lock will set it to relocked. Point being it isn't that hard to create a three/two state register that is set off the moment the clock settings are touched, allowing people to buy chips to fuck with them, while still protecting the manufacture from user stupidity.
>>52983073
>teksyndicate
>>52983557
does that even matter for dolphin?
looking to get one to make a cheap emu/htpc for my living room
>>52983478
Taxes paid by: my neighbors, my colleagues, my friends, my family, people who live in my town, people who live in my county, people who live in my state, people who live in my region, people who live in my country, and by myself.
If the majority of those people disagree with the usage of that hardware and give me explicit permission to touch it, I'd give no craps about what the government says, granted that majority would be willing to back me up. Otherwise, as much as we might despise the government, the majority of the population still believes that it acts in their favour, and as such meddling with such affairs would be pointless.
>>52983654
Dolphin the emulator?
Look at their hardware recommendations. They have a list on their website.
Also it might be worth waiting till skyfail passes. As long as intel's shit is together for their next release, you *should* be good. Although dolphin benefits from strong overclocks, so I expect if you want a machine that can keep up with most things, you're going to have to drop a pretty penny on intel's K chips. You could also wait to see if AMD's Zen is any good with dolphin, although if current comparisons are true, you might be better off with the intel chips due to faster single threads.
(Note that dolphin can use up to three threads, so you'll want to get a quadcore cpu atleast, and bump it to real time, while setting the affinity to not use the hyperthreads).
SHILL 30
H
I
L
L
3
0
>>52983327
Who actually updates their BIOS anyway?
>>52983813
People that have hardware issues or want to hack/modify their bios for various reasons.
>>52983358
what a fucking stupid analogy
https://dolphin-emu.org/docs/faq/
>
AMD: Due to very low IPC, AMD CPUs are poor choices for Dolphin. The FX8000 line of processors is the best they have, and even then most of those processors will be around half the speed of the 4670K’s performance with Dolphin.
AMDPOORFAGS BTFO
B
T
F
O
>MUH CORES
>>52983844
Maybe I'm just lucky, but I never had a hardware issue that could be attributed to a BIOS problem. Also, If you are hacking/modifying BIOS, then the lock is a moot point as it can be hacked/modified out.
>>52983898
>analogy
a·nal·o·gy
/əˈnaləjē/
Noun
noun: analogy; plural noun: analogies
a comparison between two things, typically on the basis of their structure and for the purpose of explanation or clarification.
Where was the analogy in >>52983358 ?
>>52983595
The only thing I can imagine is it's just intel covering their own arses. With overclocking being available from the OS itself these days, the average incompetent user can try it being told it can speed up his old shit. Even though clearly going against everything intel says, you know the media shit storm that hits when enough idiots get together with the papers. Still they shouldn't have killed others previous clocks.
>>52983933
Yeah I've rarely had hardware issues, but my current laptop need a BIOS update before I could use stock nvidia drivers rather than ones modified by HP. (Which didn't exist when I got the machine, so that was annoying).
>>52983813
Its being pushed though windows update.
>>52983955
I'm not a lawyer, but I don't think its that hard to state that overclocking reduces product lifespan, and may cause adverse side-effects to the functioning of the computer, none of which intel is responsible for. Then just use a single bit register that sates whether the cpu clock has been touched.
To prevent any idiots from bypassing such paper notifications (and hence lawyering up), you just have the mobo chipset check the cpu type and display $non-k warning before any settings are touched related to the clock, even after the register has been changed.
>>52983073
ITT we blame Intel for MS shit.
>>52983372
>im too stupid to pay the extra 10 dollars for unlocked
>>52984054
I didn't get my point across too well, blame the alcohol. What I'm trying to say is, even though Intel would be perfectly safe from a legal point of view, that all goes out of the window when a media shit storm hits.
I know it's different in America, but here if a shitty story picks up traction - even when the party on the offensive side is a fucking moron who is 100% to blame, it can even be thrown out of court or they could have already lost, but as soon as there's any traction behind them in the public, it becomes a fucking fucking spiralling nightmare and ultimately costs money.
Still getting my shit across badly but I can see it from a business point of view.
>>52984260
No, you're definitely right, PR is expensive as shit. While a court case can be thrown out, PR takes lots to build, quickly destroyed and unfortunately as in the case of AMD can suffer due to the illegal practices of competitors.
>>52984224
$50 difference in most popular CPU choice.
>>52984355
>im too much of a retard to pay the extra 30-40 dollars for some potential extra performance
You should just go back to your fx 9590 that performs worse than a 3570k with a mild overclock.
>>52983784
totally the wrong direction when it comes to price i can get a wii and soft mod it and get a stick computer for cheaper and do just about the same thing
>>52984420
>implying I'm not using VIA cpus.
To be honest though, I'm all in favor of using the best tool for the project, granted that tool doesn't break the bank. I have a laptop with a quad i7, great machine in general. However, I disagree with the practices of intel. Thats why I like to buy varied parts when the best tool is just a generic x86_64 cpu, and usually AMD is much more fairer about keeping their features on their chips unlocked. This meaning that both companies have and do do unfair things to the consumers. Although in the end I see AMD as slightly better. (Although getting rid of PSP would be nice so libreboot can be a thing).
>>52984553
I disagree with how intel played this memelake shit out, but a lot of the fault is with the motherboard manufacturers in this case.
>through Windows Update
Prime fucking reason to turn that shit off.
>>52984630
Seriously? I haven't heard that before. Got any sauce?
>>52984423
Yup. Unless you want HD GCN/Wii. Still love my soft-modded wii. Have one of the older ones with the fully hackable boot loader. :)
>>52984651
Some of these motherboard makers were advertising it as a main feature of their boards. Asrock being one of the more common ones.
>>52984729
I thought you meant the total memelake failure in general, not how stupidly broken memelake is.
That being said, does Asrock not have the ability to make hardware function in a specific way, even if the manufacturer did not intend it?
I don't want to waste a lot of time, can someone summarize this in greater detail than one line but readable in 10 seconds?
>>52984787
Skylake locked processors can be overclocked via bios and motherboard trickery, intel doesnt like it and rolls out an update to remove it with a microcode fix for freezing, people dont update, intel makes windows release an update that blue screens you unless you have the first update.
Something like that its stupid overall m8 memelake is dildos
>>52984294
I can literally seeing it come down to "well.. you shouldn't have let me do it!" style defence. It's shitty they reverted previous stable clocks - which I didn't know they did, I was under the impression they just actually full on locked down the new non K chips from now on.
I'm sure there'll be a way around it, but I still can't see any one buying a locked chip expecting to OC at any point, it's always people that have bought it then look into overclocking a long time after purchase, and they never get a decent clock anyway, but still shouldnt have reverted previous stable clocks like I said.
>>52984707
i'd probably just borrow my brothers modded wii to play the 3 games i want to play on it, the metroid prime trilogy, every thing else is all on the snes/gba/ds/genesis/psone, which i'm sure a little compute stick can easily emulate, and playing some 720p movies and shows, i'd have to try 1080p
WHAT THE FUCK DO YOU MEAN I CAN'T OVERCLOCK MY NON OVERCLOCKABLE CPU NOW
>>52984855
Don't cite me on this, because I'm a bit hazy here (and could be completely wrong.).
This method of overclocking (The transport bus?) has been around for a while and publicized in previous generations on YT and smallerish blogs.
It seems for some reason though, only with skylake has this method gone mainstream enough for motherboard manufactures to incorporate one-click solutions to use this method. Which subsequently made it a big enough issue for intel to block. Probably not to compete with their k chips which have the main clock multiplier unlocked, but still a little pointless imo.
As you said, there will probably be a way around the locked clocks anyways, and this'll be a big issue in a few generations again.
>>52984958
WHAT THE FUCK DO YOU MEAN I CAN'T USE THE THING I PAYED FOR THE WAY I WANT TO NOW
>>52985022
I PAYED FOR IT I SHOULD BE ABLE TO FUCK IT UP AS MUCH AS I WANT NOW AND THEN RUIN IT AND SEND IT FOR RMA GIVING YOU BAD REVIEWS AND FUCKING UP YOUR IMAGE ALL BECAUSE OF MY FAILURES
So lemme get this straight, you bought a non overclockable CPU from intel and think it's your right to overclock it, grow up you entitled millennial.
>>52984988
The reason its popular now is because before it was only stable up to about 105mhz on a great chip, now it can go much higher even on a shitty chip.
>>52983245
The difference between k, and non-k is a microcode. That's it. Educate yourself kiddo.
>>52985084
actually no not at all
"hur dur I watched a teksyndicate video now I understand everything"
boy all this freedom is great
>>52985106
>implying AMD doesn't sell locked processors
>>52985084
lol no
>>52983073
>jewtel
And I didn't care one bit.
>>52985052
Doesn't turboboost let you overclock anyways?
Also you seem a bit salty.
Once again, intel could implement a single bit register that is swapped when clocks are modified.
I'm also more annoyed over intel artificially limiting the features of their cpus than anything else. (Think instruction sets, for example in the pentium g3258, as noted by >>52983542 and >>52983557 or virtualization in Xeon chips or hyperthreading in chips like the g3258).
In addition, as >>52983446 pointed out, K series despite overclock abuse can be sent in for warranty service; so I doubt if the cpus are just the same minus firmware differences, it really doesn't makes sense that the same CPU would have a generally different failure rate. This assuming we're comparing a 6600k vs 6600, or something with a similar numbering pattern for model number.
>>52985084
Like I mentioned above, microcode only differences are only guaranteed in in XXXK vs XXX. Other models might be modified to have lower power consumption, although this is in theory also possible by microcode. Cores could also be disabled by microcode, but this is a bit backwards. However, bulk manufacturing might make this cheaper than producing multiple chips. Also laptop chips are also going to be different due to potential size/heat limitations, and the rapidly increasing usage of bga sockets.
>>52985101
Okay, find me a source that says other.
>>52984988
Overclocking by the FSB has been around for along time yea, it was how I did my first overclocks, but if I'm not mistaken they were overclocking these non K "locked" chips with the Multiplier still. They could get to like x38 Mutlipliers even though they shouldn't have been able to. Overclocking by the BLCK isn't really valid these days, and infact it's a sure fire way to fuck your shit up when idiots just start tinkering with it.
>>52985451
I'm not sure what you mean by BLCK (and a quick google/wikipedia has failed me). :/
>>52983073
Don't overclock, don't care.
>>52985538
Solved. BCLK not BLCK.
Seems like skylake non-k overclocks use a combination of the bclk and transport bus for overclocking.
>>52985371
How about you find me a source that backs up your claim.
>>52985371
So I can claim I have 20 inch dick when hard and unless someone finds a source that says otherwise my claim stands? What a time to be alive.
>people rather want machine-lockups due to the AVX-bug than lose the miniscule overclocking on their otherwise non-overclockable poorfag-CPUs
lel, never change /g/
>>52983073
>pushes blue-screening updates to punish overclockers through Windows Update
Anti consumer class-action suit inbound folks.
>>52988270
oy vey
|
|>
|
|3
|
>>52983073
the multiplayer is shit on non-k, are people retarded
>>52988270
sooo if I don't use windblows I don't get this stupid microcode update
>>52988455
And what about the thousands of people who are effected by this?
Hell, AsRock are going to look like the good guys here standing up and saying "We enabled this feature for the good of the customer!"
Hiring more women won't get Intel out of this one if it does end up in the courts.
>microcode-updates are flagged as non-free in leenox-distros
enjoy your unfixed cpu-erratas lyl
>>52984958
>>52985022
>>52985052
I thought OC'ing was a breach of the guarantee provided when you buy the product? Like when you open the case of a prebuilt PC.
>>52988455
>Overclock a processor
>For a non windows OS
Wait, you could overclock non-K?
In any case, like everyone said in this thread, why would you?
>>52985371
>Make claim
>Don't back it up
>Request source for the contrary
>>52983595
There's no way to tell if a CPU was OCed. Easy to claim "It just died" after a failed OC.
>>52983415
Nvidia says they're selling 4gb 970, 3.5 instead
>>52983073
> AMDsyndicate
Fuck off OP. Reported for spam.
>>52988831
It's just an overblown media bullshit again.
This shit affected a few hundred people at most, but in the age of the internet, everyone is loud and offended at the simplest things.
>>52989201
Why are you getting so defensive? Are they paying you in liters of precious cum or what?
>buy car
>add accessories
>car salesman comes and break my engine with a hammer
How is it allowed?
>>52989226
This only made YOU look like a paid poster.
I don't care about megacorporations, I am just mad that people defend them, attack them, whatever.
WHO CARES?
It's not like AMD will get any better if you shill for them all day.
It's not like Intel is going to go bankrupt if you make threads from your parent's basement each day. And so on.
>>52989242
uhm... see Volkswagen scandal?
> ITT people who don't read news
Intel sure does have a lucrative business model
>i5's could have 4 cores but they don't
>i7's are just i5's with hyperthreading enabled
>xeons are just higher binned i5/i7 with ECC
>>52989458
So... same as AMD?
>>52989458
Did you know that the higher end chips are more difficult to fabricate? When they manufacture the chips, they run all kinds of tests on what they have and when certain tests fail for the chip, they disable that part of the chip and only claim that the parts that passed are reliable. If you try to re-enable those features on the lower models, you should not expect reliable performance from your chip because they didn't guarantee that kind of operation when they sold it to you.
>>52989458
woah woah woah...i5's don't have 4 cores?
>>52983073
wow pretty rude desu
>>52989676
They do tho
>>52983334
> >AMD reverts this
They didn't revert it, they used lasers to physically cut unused blocks in later shipments.
>>52989191
>biased little bitch
Fuck off. Reported for monopoly.
>>52989458
Uh, are you talking about laptop i5's? Because I have an i5 in my desktop right now and it's a quad core.
>>52983415
>that gpu scandal is already 6 years old
There was no scandal, just the second run of GPUs made being put together differently to prevent people from unlocking them.
I'm not even sure if it's a case of artificially limiting, back in the day there was a lot of talk about stability problems on unlocked cards and we never got to hear if they were stable in the long run.
>>52983813
I usually do for whatever fixes or additional functionality it might have. Why run with some old as shit BIOS?
>>52985258
>Doesn't turboboost let you overclock anyways?
No, you need to go look up the definition of overclocking.
>microcode only differences are only guaranteed in in XXXK vs XXX
Well, it's a little broader than that. I think it's reasonably safe to say the only difference between all current gen desktop i5s are the same die but differently binned. The discrepancy shouldn't be enough to make a fuss about. The difference between an i5 and an i7 is larger though, with bits being cut away by laser. I've heard a rumour that the i3, Pentium and Celeron line is a different die but no source to back that up.
>>52989458
This is pretty standard across the industry. Go compare Opterons to their FX counterparts or NVidias workstation cards to their desktop cards or even laptop cards. Heck, AMD used an underclocked HD 7870 as their flagship laptop card for the 7970m, 8970m and m290x.
>>52983638
who is this semen demon?
>>52996016
Linus Sebastian
He's not as cute as Luke though.
>>52984754
Intel released a press release about this. To me it sounded like them admitting they have a gun to the mobo mfgrs head
>>52984958
If it wasn't overclockable, people wouldn't have been able to overclock it.
