Intel is at it again

>falling for the delidding meme
Enjoy your voided warranty

>>58020879
Apparently you can just glue it back for RMA and they are fine with it.

Intel and Apple are basically family in the sense that all their products are memes, are overproduced, and are highly overpriced, yet people still buy them
if only AMD wasn't fucking retarded, then we'd actually be getting decent CPU improvements every year
hopefully Ryzen is as good as it seems

>>58020919
Nah, only Apple.

Intel is a monopoly, not a meme.

WHEN'S THE PRICEDROP INTEL.

Kaby lake is Skylake. Ryzen is ready to fuck you in the ass. Drop the goddamn prices already so I can upgrade.

>amd
>wanting a house fire
its like you enjoy bdsm or something

>>58020954
I like how you are still calling AMD the housefire, when it is Intel who purposefully gimps their shit so it runs at 99 DEGREES CELSIUS.

It isn't even the first time, they've been at it since Ivy Bridge.

>>58020852
>falling for the (((intel))) meme when based shit wrecker is a month away

>>58020971
No wonder ryzen was able to run at a lower temp despite their fab process being slightly worse.

>>58020852
The TIM intel uses isn't great, but it isn't the cause of the problem. Its shim height. For whatever reason intel simply doesn't have the IHS pressed down far enough to make good contact with the die. Its unknown why this is, but it is known that it is explicitly intentional, and they have maintained this over several generations of chips now.

File: le-happy-merchant-coke.jpg (93KB, 650x762px) Image search: [Google]

93KB, 650x762px

>>58021035
That's even more crooked

>>58021063
It is.
Some people have speculated that its to reduce cracked dies during packaging. I doubt this excuse, but at the same time it is the only logical explanation.
Sabotaging your own product, including mobile parts, makes no sense. Though intel matter of factly is doing just that. Theres no way to explain this away.

>>58021035
It's not shim height. It's the fact that they are purposefully NOT using solder on the K chips. They use soldered IHS on every other one of their chips, including the Extreme lines.

The K line chips are just purposefully gimped so they can't overclock as high.

>>58020971
>>58021001
Ryzen is matching Kaby right now right? Wonder if that changes with better TIM.

>>58021104
i7E parts can't overclock as high no matter what, the increasing power draw, and thermal limitations are far greater.
A non soldered IHS isn't an issue if done right, the difference in thermal transfer is paltry.

File: Teddy_Lawl.jpg (15KB, 374x250px) Image search: [Google]

15KB, 374x250px

>>58020879
>K part
>warranty
You have to buy a warranty for K parts

>>58020852
Is there any reason why I can't just throw the lid to the trash and directly cool that surface with proper thermal paste? Genuinely curious, it just doesn't make sense to me.

Damage prevention I assume.

But being careful it should work right?

>>58020852
>delidding
never got this meme

are you supposed to do lapping till you don't need the thermal paste? do you just discard the lid and jam the cooler directly?

pls explain

File: 1481670182199.png (673KB, 1520x1460px) Image search: [Google]

673KB, 1520x1460px

beep beep beep beeeeeeeeeeeeeeeeeeeeee

>>58021180
>>>/leave/

Man shit like this doesn't surprise me anymore. That's why I'm waiting for zen instead.

File: 1429555438686.png (486KB, 526x700px) Image search: [Google]

486KB, 526x700px

>>58021192
rude

>>58020852
They've been doing this shit since Ivy, I'm not sure how exactly it's still a surprise for some people.

In any case buying a CPU now is stupid, wait for Ryzen and see what it brings to the table, then decide what to buy.

>>58021234
You don't even need to lap you retard. Just clean off the thermal paste after deliding and apply new one.

It's pretty embarrassing you even have to do this in tye first place desu.

Hopefully Zen won't be a complete farce like kabby lake has been.

If zen is actually good intel might try to make a successor to sandy bridge.

wow it runs hot at 5.1ghz

File: sunny displeased.gif (2MB, 700x394px) Image search: [Google]

2MB, 700x394px

>>58020852
>2x more for the performance you could get free from overclocking.
Nope.

>>58021128
>A non soldered IHS isn't an issue if done right, the difference in thermal transfer is paltry.

99C with default TIM and 66C with liquid metal TIM is a paltry difference to you? With solder, the temps would be even lower...

>>58021160
>Is there any reason why I can't just throw the lid to the trash and directly cool that surface with proper thermal paste? Genuinely curious, it just doesn't make sense to me.

- coolers would then not reach the die properly,
- you'd have a high chance of chipping of the die corner

But with liquid cooling and a custom shim, you probably could it.

>>58021460
Op is saying Intel wants you to pay 2x more for 3-5% performance increase which you could easily get from delid and overclocking.
That's what Intel has done since the bridge.

>>58021382
>You don't even need to lap you retard. Just clean off the thermal paste after deliding and apply new one.

Nah, you need liquid metal TIM.
Simply using different/better thermal paste will only give you minimal decrease in temps. I've been there.

>>58020879
Void warranty in the US is illegal and they know it, if they pull that shit you can sue and win.

>>58020852
Wow fuck that, going straight for Ryzen then.

>>58020852
Just use new thermal paste

>>58021666
>can't wait to buy new CPU to delid it!!

>>58020954
My AMD FX-8350 runs at about 60C at its hottest. It's overheating if it gets hotter, and can only handle up to 72C max.

These Intel CPUs run at stock hotter than mine can safely run period.

>>58021489
Again, this is from shim height of the IHS. It isn't making proper contact with the die.
The difference between TIM when properly applied and solder is paltry.

https://forums.anandtech.com/threads/lapped-my-i7-3770k-and-there-was-zero-improvement-in-operating-temperatures.2261219/


>>58021681
Reminder: AMD CPUs and APUs use thermal margin instead of a temperature in degrees.

>>58021678
I'm not the retard what blew up a perfectly good i7 core processor because he was too retarded to overclock it safely now am I?

>>58021681
You forgot to mention how your 8350 is also 33% worse at both single and multi thread tasks than the Intel CPU, and that yours is hooked up to a dual fan water cooling unit while the Intel ones are using stock air coolers. Faggot.

>>58021742
I have a hyper 212 evo and I know it's a shit CPU.

It's not a housefire though.

>>58021711
Wrong thread

https://forums.anandtech.com/threads/delidded-my-i7-3770k-loaded-temperatures-drop-by-20%C2%B0c-at-4-7ghz.2261855/

>>58021711
>Again, this is from shim height of the IHS. It isn't making proper contact with the die.

That's why there needs to be some inbetween material, in this case thermal paste. Except that the thermal paste has shit thermal conductivity, hence why Intel K processors ever since Ivy Bridge have shit temps and cannot overclock due to hitting thermal limits.

The Extreme platform cpus all use solder and have better temps, even if you delid your K cpu and use better TIM paste.

>The difference between TIM when properly applied and solder is paltry.

No. Solder has better thermal conductivity than anything except maybe Liquid Pro.

>https://forums.anandtech.com/threads/lapped-my-i7-3770k-and-there-was-zero-improvement-in-operating-temperatures.2261219/

That one talks about lapping the IHS, and no one here is talking about that.

File: i7-3770KtemperatureswhenshimmedwithNTH1.png (44KB, 480x485px) Image search: [Google]

44KB, 480x485px

>>58021873
Just stop talking out of your ass already.

Conclusion: The Intel stock CPU TIM is not the reason Ivy Bridge's run hot, and replacing the Intel stock CPU TIM is not the reason a delidded Ivy Bridge runs so much cooler - the benefits of delidding are entirely due to the resultant reduction in gap height between the CPU silicon die and the underside of the IHS.

>>58020852
Wow! That's almost 1337 Kelvin!

>>58021715
>I'm not the retard what blew up

lol

>>58021897
Source: your ass

>>58021897
The point is that if they soldered the IHS on, you wouldn't have them running at fucking 99°C, because the solder would fill up that empty space with better thermal conductivity than any TIM.

In fact the reason why the space is there is because the material they use as solder is thicker before soldering.

>>58022036
The point is that shim height is responsible for the high temps, not the chosen TIM. TIM is not inherently bad in any way. If this height is reduced there is no significant difference between using TIM, and putting a water block directly on a bare die.
This is too well documented to refute, you're arguing that the sky isn't blue.

>>58022084
>The point is that shim height is responsible for the high temps, not the chosen TIM.

I'm not arguing the quality of their thermal paste, I'm saying that they knowingly gimped their CPUs by using thermal paste in the first place, instead of using the soldered IHS they use for damn near every other chip in production, including the Extreme line ones.

>>58020852
>7700k
>less power efficient than 6700k
>5% less performance than 6700k
>less overclock than 6700k
>costs $100 more and requires a new motherboard
>poo tier thermal interface
>$459

2017 will be the like a time travel back to when amd64 released and intel went on LITERAL suicide watch


im removing my intel sticker from my case, whos with me?

did you tards ever bother to think the reason IHSes exist it is distribute the load of the heatsink over the interposer rather than place it solely on the die. when things get hot they expand and there is going to be a lot more stress on the interface between the die and interposer without an IHS.

>>58022308
>package substrate
>interposer
Try again.

FYI all pressure of the heat sink is used to ensure good contact with the IHS. If there is no IHS present the clamping pressure required is only a fraction, and the mountain hardware takes all the force.

The IHS exists to protect fragile dies from cracking in consumer hands. There is no other reason for them.

>>58022308
You asseating tardfag did you even read the thread?
Nobody wants to advocate removal of the ihs the issue is that intel uses bio degradable shit tier consumer grade thermal paste instead of soldering it. Intel advocates the ihs removal by saving pennies.

>>58020927
This

>>58022346
>The IHS exists to protect fragile dies from cracking in consumer hands

/thread.
jewtel uses cheap TIM because they rightfully view it disdainfully as just part of an anti-idiot shield system.

File: Screen Shot 2016-12-15 at 9.27.52 PM.png (141KB, 1416x762px) Image search: [Google]

141KB, 1416x762px

>>58022346
>Try again.
stay retarded anon

>>58022368
>Nobody wants to advocate removal of the ihs the issue
literally in OP's post he is advocating delidding

>>58022546
Keep on trying.
Not all substrate packages are interposers, though they're sometimes used interchangeably they are not the same class of packaging.


http://www.intel.com/content/www/us/en/support/processors/000006761.html
http://eesemi.com/substrates.htm

File: richland_mobile_5300M.jpg (262KB, 1803x1200px) Image search: [Google]

262KB, 1803x1200px

>>58022546
holy fuck you're actually retarded
cpus dont have dies upside down with bondwires covered in resin

your image is completely irrelevant to the discussion choke on a cod u cunt

>>58020954
>calling AMD a housefire
>when intel's been fucking up thermal paste for a while
My sides

>>58022616
>http://www.intel.com/content/www/us/en/support/processors/000006761.html
>mobile processors

>>58022694
>grasping at straws when caught talking out of your ass like the clueless kid you are

File: img_192134.png (1MB, 1207x624px) Image search: [Google]

1MB, 1207x624px

uhmm... thanks alot guys

File: 1377997944924.jpg (29KB, 493x387px) Image search: [Google]

29KB, 493x387px

>>58022736
YOU DID IT :D

>>58022736
Where the fuck is the actual silicon that you see in all the pictures?

File: 33187.gif (500KB, 155x116px) Image search: [Google]

500KB, 155x116px

>>58022736
>uhmm... thanks alot guys
YOU DID IT

>>58020919
>if only AMD wasn't fucking retarded
I've been waiting since I upgraded from my Athlon XP all those years ago, so I wouldn't hold your breath.

>>58023057
Firmly attached to that heat spreader.

>>58022275
I remember when people were saying the 6700k was a bad CPU, kek.

I've been bretty happy with mine for the past year. Not too upset about Crappy Lake since I wasn't planning on upgrading.

And quite frankly I haven't thought to myself, "Damn, my computer would run so much better if my CPU were faster" since my C2Q Q9450. The only reason I've upgraded the past few years is because:
A) previous build just started breaking down (legit the mobo started crapping out and I'm sure as fuck not going to buy old ass parts)
B) shits and giggles
C) I like to pretend I still play video games

When does the 7700k come out?

>>58023591
Next month supposedly.

As badly as I need to upgrade though, Ryzen is looking like the better buy.

>>58023684
fark I'm buying a new computer very soon, is it worth me waiting for the 7700k or should I just get the 6600k?

>>58023723
Wait for Ryzen, AMD will probably be offering 8 cores with 16 threads at the price of an i7 7700K

>>58023723
6700k is good to go right now. 7700k is like a 1% improvement, not worth waiting for.

Ryzen is worth waiting for though. Wait till that's out and see how it holds up then decide.

>>58023723
Wait for zen then buy whatever you want, however the 7700K has zero improvements if you aren't looking into OC'ing, and even then it's pretty bad at that thanks to the shit TIM/IHS they are using to deliberately gimp their non extreme CPU's
Ryzen will bring prices down like it has never seen in x86 land

>>58023756
7700k will be worth it for the higher clock rates and thus higher overclcocks compared to a 6700k.
Compared to a Ryzen it will only be worth it if you highly value single thread pref.

>>58023768
>Compared to a Ryzen it will only be worth it if you highly value single thread pref.

Depends how close Ryzen matches it on IPC. If it's close, it'll be a no brainer really.

>>58023760
Quite the upgrade from my i5 2500k lol.

>>58020879
cuck.

>>58021604
this meme needs to end, it's a limited warranty

>>58021960
Hes British

>>58024112
Like the pajeets in customer service knows the difference.

>>58020879
>voiding warranty on a CPU
boohoo, it's only the single most resistant and durable component of a modern PC

>>58020919
>>58020927
The difference between Apple and Intel is that Apple sells substandard products for outrageous prices by marketing them aggressively towards idiots

Meanwhile, Intel simply sells the best products available in their class, so they get to charge outrageous premium prices for them.

Waiting for poocessor from AMD
Not buying 6700k on black friday for 250$
You fucked up kiddo

>>58020879
leave dirty normie

>>58022084
>The point is that shim height is responsible for the high temps, not the chosen TIM.
Question: Would a soldered IHS with the same shim height run just as hot?

>>58025021
It is until you start cutting into the heatspreader

>>58022523
>jewtel uses cheap TIM because they rightfully view it disdainfully as just part of an anti-idiot shield system.
Then why do they solder all of their xeon and extreme series chips?

>>58022996
>>58023057
>>58023081
This is a 2500k posted in a thread from 2007 on [H]ardforum

File: 1461769811159.png (43KB, 481x500px) Image search: [Google]

43KB, 481x500px

>>58020852
Why don't they just solder the die, I just don't get it. Surely the thermal intermediate is worse than using die to IHS soldering and more expensive to boot then a tiddlywink of solder.

There any actual reason to why they do this, or do they like getting RMAs as a fetish?

>>58026264
To prevent overclocking and to sell +1% kaby lakes with moar megahurtz.

>>58026264
>Why don't they just solder the die, I just don't get it.

Because then you can overclock a i5 to so high speeds that there is no point in buying the more expensive Extreme platform cpus, or even as much as an i7.

So by gimping overclocking, they give more incentive for people to buy the more expensive chips.

>>58025216
same, but enterprise customers will bitch more for conspicuously cut corners

>>58028619
How much more does it cost intel to solder the IHS instead of using their shit TIM+shim like they do on the K series?

>>58020852
Did performance/W even improve with Kaby Lake?

Seems like it stayed the same, aka 7700k performs like a slightly overclocked 6700k (higher speed, but also higher power use).

>>58028673
something on the order of like a fifth of a penny.

>>58028673
>How much more does it cost intel to solder the IHS instead of using their shit TIM+shim like they do on the K series?
Not a lot, the same reasons why Xeon E3s dont cost that much more than a comparable i7. Its just that all the butthurt fags here didnt want to spend an extra $20 on their CPU.

>>58028883
So how does that justify using a shitty TIM that seems to have no benefit except for artificially making the cheaper CPUs even worse?

You said it's just done with disdain, but I don't buy it. Are you really suggesting they're artificially fucking over their customers out of pure malice, with no ulterior motive? No company could possibly hate their source of income that much. If they wanted to fuck over their consumers, they'd just increase the price tag. (Heck, they do)

If it's as cheap as you make it seem, then it's literally not even cutting corners, there has to be some deeper reason for wanting to gimp their CPUs. I mean they could just artificially gimp the chips themselves if that's their only motive.

>>58027441
Like they didn't want you to dual celeron A's back in the BP6 days.

>>58029049
>out of pure malice, with no ulterior motive?
Money is their motive you retard. See >>58027441

>>58027441
But why gimp overclocking by using shitty TIM instead of just.. gimping overclocking?

They literally control the hardware, why would they have to fuck around with the heat spreader. They can lock the multiplier or clock range.

>>58022084
>>58021897
>>58021711
>>58021128
>>58021035
Anon I would stop. These children parrot what ecelbs say. Right now the big ecelebs say solder is better so to them it's better no matter what. They need an eceleb to parrot and follow.

They can't comprehend that proper pressure greatly effects TIM performance.

File: 1463122623561.png (50KB, 426x364px) Image search: [Google]

50KB, 426x364px

>The 2500K was so good OC'd it still competes today
>Intel are now afraid to make another good processor and shoot themselves in the foot

>wah why doesnt intel use solder it only costs a few cents more
>wah why doesnt intel just give me a faster clock version of the chip, the silicon is all the same
/g/, full of crybaby poorfags

>>58021517
Yes I meant with a custom cooler or something moded to make it fit.
Thanks

>>58029269
>But why gimp overclocking by using shitty TIM instead of just.. gimping overclocking?

Because this way you have the illusion of having a freely overclockable chip.

File: 04509977.jpg (25KB, 211x193px) Image search: [Google]

25KB, 211x193px

>>58025372
>2500k
>2007

>>58029279
>They can't comprehend that proper pressure greatly effects TIM performance.

And you sound like you can't comprehend what solder does.

The purpose of TIM is to fill up the space between two objects, in order to provide better thermal conduction than the air inbetween would (air is a thermal insulator). Pressure does matter in this case, because more pressure means the objects in question are closer, and you need less paste to fill up the space between them. Paste will always transfer heat worse than actual physical connection between the objects. This is also why people do lapping, since that will make the surface of the objects completely smooth, again leaving less space for air to be trapped and insulate the thermals.

But soldering will have the two object physically joined. There is no empty space inbetween. The two objects (in this case the die and the IHS) are bonded together. Pressure does not matter in this case because we have solid objects, not a paste between two metals.

File: 1465853199399.jpg (62KB, 925x960px) Image search: [Google]

62KB, 925x960px

>>58029634
>Pressure does matter in this case
>more pressure means the objects in question are closer
Obongocore really failed this generation by dropping basic physics.

>>58029634
>>58029675
Pressure aids thermal contact conductance by increasing the shared surface area between two imperfect surfaces.

TIM/paste/whatever is intended to bridge the air gaps through a moldable material with higher conductivity than air if not metal.

With a TIM, you can still get slightly higher surface contact between surfaces with higher pressure, and you can squeeze out some of the TIM sitting between parts of the surface that would otherwise be in contact.

Solder is basically the same idea except the higher thermal conductivity means less of a penalty where excess material sits between the surfaces, and you can't really squeeze out excess after it freezes.

>>58029841
>solder
>freezes

>"tim"
>higher surface contact than solder
>solder literally only bonds through perfect surface contact

kys fuckin retard

>>58030432
not saying that TIM contact surface greater than solder you faggot, just that pressure helps TIM in a similar but distinct way from unaugmented contact.

>>58021092
>Sabotaging your own product, including mobile parts,

mobile chips ship without heatspreaders, it's up to OEMs to choose TIM for those.

File: 1478646830753.png (173KB, 3000x3000px) Image search: [Google]

173KB, 3000x3000px

>>58025372
>2500k
>2007

File: question.png (21KB, 900x900px) Image search: [Google]

21KB, 900x900px

>>58025372
>sandy bridge i5
>2007

>>58029397
>>58030585
>>58030801
>he doesn't have a dad who works at Intel and gets early copies of the next Pokemon game

>>58027441
What a load of shit.

The consumer-grade chips have the same, or superior, single-thread performance than the enthusiast/professional chips. It would make no sense for a gaymer, or casual user, to spend 2-3x the amount on an enthusiast chip.

Intel don't solder their consumer chips because it cuts costs, not to "trick" people into buying enthusiast hardware.

>>58021104
They haven't soldered anything since ivy bridge?

>>58025372

r u a warlock?

>>58020852
>free from overclocking.
and yet they overcharge for unlocked multipliers..
somehow this is free to you? rellying on expensive ass motherboards for FSB overclocks, is aids.

>>58030890
This
If solder is so much better than paste, where are the gains?

>have 6700k
>OC'ed to 4.5GHz
>never hit 99°
>never hit 70°

>>58031040
>have 3770K
>OC'd to 4.6GHz
>almost hit thermal threshold until I discovered delidding
>now it hovers at 70°
I should probably do the same with my 5775C. I only just discovered that it uses TIM as well.

You who are arguing forget that Intels paste dries over time. I'm on i3-4170 atm waiting for Zen to explode the used components market and I'm pissed the fucking thing heats like crazy so its cooler (random 180W tdp zalman) gets loud. The i3 is seriously hotter than my old i5-4690k. Way too lazy to delid before upgrading back into something proper.

>>58023809
If you're using a 2500K you sure can wait a couple more months for actual Ryzen benchmarks and reviews. Hell, you could wait a year or two more and you still wouldn't be outdated.

zen will only beat xeons
The core line will remain overpriced and undisputed king for gaymen

>>58023760
>Ryzen will bring prices down like it has never seen in x86 land
>believing fantasies
AMD has had intel beat in price:performance since forever now, intel doesn't care and just keeps releasing skyhigh priced bullshit.

>>58022275
I was about to put mine on next to the MSI sticker thanks

>>58031509
Kill my self

At least I got my 6700k for $260

You use a fancy thermal paste between the CPU and the heatsink and use water cooling etc only to find that you are limited by the shit under the CPU lid.

That's it. I'm done with Jewtel.

>>58031149
So it uses TIM
What are you going to use instead?
>inb4 Arctic silver

>>58030890
>Intel don't solder their consumer chips because it cuts costs, not to "trick" people into buying enthusiast hardware.

Then how come they are soldering every other chip except the K ones?

Please AMD, don't fuck it up. Fuck Intel's shit up so I can finally ditch Fiery Bridge. Summit Ridge should already outperform it straight out of the box, all that's left is for it to not suck on single threaded shit.

>>58030432
Wow, you sure did refute all those points.

>>58031280
>i3-4170

That's one of the unsoldered chips, so yeah, it'll have shit temps no matter what cooler you use.

>4960k works fine, never above 80°
>lower temps sounds nice
>delid as per /g/ instructions
My new 6600k works great btw, it's much cooler now, thanks for telling me to delid my 4960k

>>58020954
Stop buying the garbage fx9k series

File: 15192581_199795247140043_396068067912833421_n.jpg (38KB, 675x547px) Image search: [Google]

38KB, 675x547px

>>58020852
i7-E soldered IHS mustardrace reporting in

>mfw the cpu alone uses 200w when oc'd
n-no bully pls

>>58021035
>>58021092
Seems logical to me, if the part is intended to be run at a specific standard clockspeed and the raised shim still keeps it within required bounds that seems like the right business choice to make if lowering it puts the entire die at risk. Better a customer receive the product as described than them receiving a broken one.

More importantly if de-lidding is effective for enthusiast then there doesn't seem to be any issue, you risk instability when messing with the clock, you risk damage when messing with the voltage, de-lidding seems like just another simple step for hobbyist who care to maximize the device and they can actually do it, there's nothing artificially holding them back, it's a physical problem with a valid business reason.

>>58032930
Liquid metal pastes exists for this reason

File: image.jpg (42KB, 735x541px) Image search: [Google]

42KB, 735x541px

>>58033822
Do they really exist for this reason

>>58033723
Soldered i7-2600k mustard race here :^). At least we're not gimped OC air gap peasants.

>>58033851
No, but it's awfully convenient for delidding
>near solder thermal performance
>solidifies into a near-unbreakable bonding layer
The only downside is that it's conductive and hard to apply on silicon.

>>58021681
How you liking watchdogs 2?

>>58020919
>hopefully Ryzen is as good as it seems
>Bulldozer will be the return of AMD

>>58030432
Yes, solder freezes when its too cold to be a liquid. It's called the freezing point, and it exists for every element.

>>58020954

Someone didn't see the demo

File: 1010618_1024126484297550_4100109023308938178_n.jpg (46KB, 960x862px) Image search: [Google]

46KB, 960x862px

>>58023760
>Ryzen will bring prices down like it has never seen in x86 land
AMD did that already. Look at the existing <$200 octo cores. Intel isn't going to price match shit. They've never price matched anything. They're going to do their usual thing of bribing OEMs and shilling the fuck out of social media to try and keep Zen down. If you haven't noticed, it's been getting more and more intense as we near Ryzen's release. They hit /g/ hard, just like always.
>Buy this Intel processor now before Ryzen comes out!
>Look, we dropped the price of the i7 at some obscure Microcenter that 12 people can actually buy from but it's still worth full price so buy it!
>Ryzen? Looks like a ring of poop around it! BTFO!! Buy Intel instead!
>AMD's stock dropped 1/4% during lunch today! AMD dead in the water! Buy Intel!
>Look an Indian! LOL I shooped an AMD logo near him! Don't buy from Pajeet! Buy Intel!
etc. etc.

>>58034540
Intel shilling exists because they're objectively better and have been since 64
AMD also has had inconsistent claims about zen for the last few months

>>58034578
Can you point to some?

>>58034578
Great fucking job anon. You're going to turn "objective" into a soft and abstract term like you faggots did with "literally". No, they are not "objectively" better. Concepts like value, longevity, multithreaded processing, and quality of TIM are all objective components that Intel fails at. By definition they can only be subjectively better. Now go the fuck away.

>>58034578
>AMD also has had inconsistent claims about zen for the last few months
Such as?

File: 1479075781356.jpg (138KB, 640x480px) Image search: [Google]

138KB, 640x480px

>>58020928
Ryzen octacores will demolish Intel LGA1151 in overall MT performance. That is almost guaranteed.

But you know what? People will still whine about single-thread perfomrance being somewhat lower and/or come up with some lame excuses, like the "but Athlon 64 is less stable" in 2004/2005. And will still buy Skylakes/Kaby Lakes. Or at least tell other people to buy them.

File: lemon angel.png (363KB, 640x480px) Image search: [Google]

363KB, 640x480px

>>58021897
>The Intel stock CPU TIM is not the reason Ivy Bridge's run hot, and replacing the Intel stock CPU TIM is not the reason a delidded Ivy Bridge runs so much cooler - the benefits of delidding are entirely due to the resultant reduction in gap height between the CPU silicon die and the underside of the IHS.

Utter bullshit. If that was true, then how would delidding and returning the heatspreader back just with another thermal grease improve temps?

And it indeed does - significantly. Most people actually don't run completely bare because that is impractical and risky.

>>58037650
>If that was true, then how would delidding and returning the heatspreader back just with another thermal grease improve temps?

Because most people also scrub down the black glue holding the IHS in place, which makes the IHS sit lower.

>>58037974
let's put artwork "quality" aside, can you make a point without dragging /pol/-tier crap into it?

This way people are just gonna think you are a falseflagging troll or genuinely retarded, even if the point was sound.

>>58037913
I kinda doubt that's the secret. I would probably hear about it by now, since I get to talk/work with a professional reviewer that has tested maybe 100 CPU coolers and did delid tests too. If it was just a matter of gaps and pressure, he would be the one I would expect to come with this enlightement, and not /g/.

>>58038319
Pathetic appeals to authority only highlight your clueless state. That chart and the accompanied text came from Anandtech user "idontcare." Its been a well known and multiply repeated fact for 4 years. Its not up for debate.

If you want to lie just say your dad works for Nintendo, kid.

>>58020852
I really, really hope Ryzen isn't going to be as big of a disappointment as I'm expecting it to be. I'm so tired of being bent over and gaped by Intel.

>>58038307
they can't stop, it defines them. Their entire lives are ruled by their distaste for jews.

They have nothing left to cling to in their meaningless lives so hatred is all they have left.

>>58021035
>>58021104
>>58021711
>>58021873
>>58022036
>>58022084
>>58026264
>>58030899
>>58033298
The reason why Intel uses TIM is due to the small and long die since Ivy Bridge, when solder is used the die actually cracks after some use. However that being said the 7700K does not seem to have a small enough die for that to occur. Hence, I would argue that the 7700K should use solder instead of TIM under the IHS.
>>58025216
They don't. Xeon E3 also uses TIM. Only E5 and up uses solder.

>>58038645
I can smell them though the computer screen. the stale eight memes, the sardonic superiority-complex. They stink of immaturity and autism.

i picked up a 6700k, msi z170a m5 gaming mobo, 2x8gb 3000mhz ddr4 evga ram, and a couple odd and ends from microcenter today for $550. how'd i do? btw upgrading from a x5660 @ 4.2ghz, asus sabertooth x58, 12gb ddr3 1600mhz ram.

File: 1463266354228.png (150KB, 500x205px) Image search: [Google]

150KB, 500x205px

>>58038647
If the die cracks after some use wouldn't that further justify them using solder, gotta make those shekles somewhere by giving their chips a usage life.

If they wanted to avoid the possibility of the lower end chips being overclocked to match the performance of higher end ones. Then I don't know why they don't negate the ability to overclock it in the first place.

>>58038720
If they reduce their warranty to 30 days that would probably work out. They will fail like really quickly, way before 3 years.

>>58038647
Well then maybe Intel shouldn't have made Ivy Bridge so damn small. My 3570K temps are hard to drop during summer yet doesn't help to heat up my room during winter.

>>58038741
30 days? Holy fuck

Are there any examples of this you could provide occurring in solder vs. TIM I can look at. 30 days seems awfully short for a chip failure.

>>58038784
Well afaik they want to use the same die for both PGA laptop CPUs and Socket H type CPUs to cut costs, hence the die is small

>>58038717
Doesn't seem like a very cost efficient upgrade imo.
probably like 30% better single core performance and roughly the same multi core

>>58037974
watched the wan show an hour ago, he spent about a minute crying about the name, about 30 seconds discussing the actual performance (using the leaked slides from videocardz and didnt even mention anything about the horizon event) and spend like the next 10 minutes on some boring irrelevant tangent about thunderbolt compatibility.

>>58038800
That was something arbitrary I wrote, please don't mind that. However the problem of solder/die failure is real,especially for LN2 users:

"The micro cracks will also decrease the thermal conductivity but will especially increase the thermal resistance at the corner of the DIE. ...delamination of the solder preform would occur after few thermal cycles. Micro cracks occur after about 200 to 300 thermal cycles...The micro cracks will grow over time and can damage the CPU permanently if the thermal resistance increases too much or the solder preform cracks completely."

File: 1376618725388.gif (1MB, 499x341px) Image search: [Google]

1MB, 499x341px

>>58038857
Thanks for clearing that up friendo, appreciate it

>>58038810

i will be selling the x5660 and motherboard and ram. feel like i can get $200-250 for all. plus there's a $20 rebate on the z170a motherboard. so after all is said and done, $300 or so upgrade from a 6 year old platform. i feel like it's worth it. plus the sabertooth x58 is garbage for SLI (the pci slots are on top of each other, literally 0 room for the card to breath). now i can actually drop in a second 980 ti if i feel like it - already have a 850w psu.

>>58025212
You'd have to be a retard to fuck it up.

>>58038810
30% better single core score is, like, 15 new generations of chips.

>>58020852
>2016
>buying jewtel

>>58038647
>The reason why Intel uses TIM is due to the small and long die since Ivy Bridge, when solder is used the die actually cracks after some use.
And what about the xeons?

>>58039775
Xeon E3 uses the same die as Core i series desktop CPUs and PGA type Core i series laptop CPUs. Xeon E5 uses the same die as Socket R i7(actually the other way round).

File: 1469256129238.jpg (17KB, 237x223px) Image search: [Google]

17KB, 237x223px

>yfw RyZen will have the same TIM/gap-issues because AMD saw Intel getting away with it

>>58039791
The Xeon die is also colossal.

File: LOL WUT.jpg (42KB, 600x669px) Image search: [Google]

42KB, 600x669px

>>58022275
What's the point if it's less power efficient AND has less performance?

>>58040117
Well only for the E5 and E7 series. E3 also uses Socket H, and E3 mobile are...mobile.

>>58021742
>>58021802

>8350
>shit CPU

>trying this hard to shill

>>58040173
Better iGPU, goyim.

>>58022275
i thought they worked on z1070 mobos?