INTEL MAKES THE TACTICAL DECISION TO SWITCH TO SOLDER ON THEIR IHS
wccftech.com
pcbuildersclub.com
This will result in much better thermal performance and much better power consumption, how will AMD possibly recover from this devastating blow?
The upcoming eight core 8700K is going to hit 5GHz out of the box and is going to rape the fuck out of every AMD CPU on the market.
Other urls found in this thread:
learnabout-electronics.org
twitter.com
The only reason why I bought a Ryzen was because it was not a housefire and didn't have all the vulnerabilities, more cores for less.
Glad they are making this decision, will wait for the 8 core variant if it has decent TDP.
CANT WAIT FOR THE CRACKED DIES FROM THEIR SHITTY THIN CPUS
>eight core
>5GHz
It will require an industrial refrigerator just like their rigged demo
>5ghz eight core
>95w
>$559 - $600 retail for same old architecture
Rubbing hand intensify
Looking forward to the 9900K taking a big diarrea shit all over AMDs Cinebench scores, your fanboy tears will be delicious.
You really thought Intel was just going to let Lisa Su parade her dogshit half-baked CPUs out with no counterattack?
>Intel has been a spectacular foundry developer and that's a space that is super complicated and you know, they led the world for 20 years and somebody has a hiccup and everybody is like ‘oh woe is me, what's going to happen next’. Personally I'm thinking I know what's going to happen next: fix it and keep going.
It's over. Jim joined them
>intel needed jim to tell them to use solder instead of paste
How dumb were these niggers exactly?
You should be thanking AMD for forcing intel's hand.
There's something wrong with their build. Ryzen is is known to be a beast specifically at GTA V minimums.
Where's the 10nm :^)
SHUT UP SHUT UP SHUT UP SHUT UP
NOOOOO MY HOUSE
>CPUs are slooowly getting better and better
>RAM is still the true bottleneck
>cache development is still stale
>there is no real benefit from using more registers in IA64 because it results in larger executables, RAM usage and more cache misses
>OOOE is being disabled because spectre and meltdown
Enjoy your CPOOU fights, faggots.
>creates a new cpu uarch that doesn't even guarantee x86 backwards compatibility
can't wait lul
|
|>
|
|3
|
You're not funny.
>OOOE is being disabled because spectre and meltdown
Yeah if you're using pootel kek
Spectre-class CPU vulnerabilities are a natural byproduct of out of order execution (or otherwise known as speculative execution). If you execute code on branching paths based on if conditions, some of those if conditions will be cryptologically important. Or important in other ways for access control and data security.
Meltdown was just dumb. Intel cut corners and just stored speculative execution results in the cache and just kept it there, letting any process read it.
h-h-haha the pajeets say intel is finally deigning to stop cumming over all our cpus bros, a-amd b-b-btfo haha
Intel won't be shitting on anyone but themselves until they get rid of their monolithic dies.
They're stuck in 14nm.
Man it must suck for Intlel to be completely and utterly BTFO by some "half-baked" CPUs, their processors are literally aborted xeon garbage.
>you are now aware that the reason they didn't solder their 14nm++ 8th gen CPUs was to make their 9th gen 14nm++ look better when in fact they are practically identical to what came before.
>all those pajeets
>all those kikes
>all those signs in hebrew
TOP
LEL
>much better power consuption
enlighten me on how switching the tim will affect power consuption. i'm waiting.
>comparing TR, FX, and R3 to i7
>no R7s or R5s
Gaymer nexus showed 10 c drops in using liquid metal over intel mayonnaise
That doesn't mean it produces less heat, only that it gets dissipated better.
and how exactly does that affects power usage?
You're not funny.
Stop with these threads, both of you.
Also, how will that change anything when their chips are fucking housefires regardless? Solder will bring temps down, but you would be delusional to think they could deliver a 5GHz 8 core chip without some serious thermal issues.
>how much power will my cpu use if I take off my heatsink?
Die retard
Imagine the shitters ruining their brand new CPUs trying to delid it out of habit and breaking the die.
Has anyone noticed that he looks like the Hercules dude?
Also, AMD has a win for the next few years but we'll see what Jim and Raja cook up now that the chief retard is gone and they're big honchos with a lot of freedom.
still can't touch the price to performance ratio of amd cpus. why would i want to spend $600~ for an intel cpu (in my local currency) to play children's games and browse this bangladeshi kite building forum. i don't understand why some shills insist on intel for day to day needs. literally 90% of this board could get away with having a sub $200 cpu and not notice the difference
As much as it can before it starts thermal throttling.
>The upcoming eight core 8700K is going to hit 5GHz out of the box
> no 8c/8t i7s
REEEEEE
that should be easy with the soldered IHS.
Power consumption will increase since it won't nail 95C in 27ms and thermal throttle like a pimp choking a bitch ass ho.
5.0+ GHz out of factory(meaning each die needs substantially more voltage than hand overclocked chips) will end up drawing 250W+ from the CPU alone, making most your Z boards lit up in flame.
> - 8700K - 5.1GHz, 1.375V, delidded, no offset
>- ASRock Extreme4
>- Small FFT AVX - 290W
>-90C with Noctua NH-D15 (max RPM), closed case with 6 140mm, low RPM fans
That's 6 cores, 8 cores will be nuts.
No it wouldn't
you're forgetting Intel's arch scales terribly once they pass 6 cores.
You're not funny.
>what is resistance/power consumption increasing with heat
They're in Haifa
Fucking
>Core i3-3240
Fucking
>Ivy Bridge - when they started using dog shit for TIM
Fucking
>stock clock
Fucking
>stock heatsink
>inb4 "there's your problem" - I
Fucking
>know
Fucking
>load its shitty twin cores with something remotely difficult
Fucking
>watch the thing shoot up to 90C
Fucking
>watch it emit flames fans spin to
Fucking
>max and then bluescreen
>too
Fucking
>poor to replace with a Ryzen
That intel garbage is going to cost almost $600, with 2700x being $300 I highly doubt this thing is going to be twice as fast
Imagine people saying that Intel needs to go back to TIM because their solder is somehow even worse.
based intel
intel should just stick to the shitty thermal paste. at least it allows you the option to delid your cpu.
based jewish people, based feminism frequency and based diversity!!!
The 1200, 8370, and especially 1950x shit on the pentium on multicore.
lol
Multi-core doesn't matter :)
c-c-calm down /v/
If it's sub $500 I'll finally upgrade from 4770k
who cares about AMD shit?
>If it's sub $500
AG\HAHAHAHAHAHAHAHAAHHAAHAHAHAH
Non-jewish people.
So how exactly was AMD btfo? AMD soldered hi-end chips anyway
>will wait for the 8 core variant if it has decent TDP.
It will still have same vulnerabilities, intel doesn't plan to release non-shit cpu. Don't pay for faulty products. If you do, it will only get worse.
It will pull 200W instead of current 140W when boosting.
I am AMD fan and I'm glad. It won't affect intel's sales because intelfags don't care about quality and would buy intel regardless. But the competition will prevent AMD from holding features and force them to release their best lineup sooner.
>The absolute state of underage shintelkikes
The power your cpu draws isn't due to it's heat. The heat is due to the power drawn and the stress on your cpu.
jew spotted
>will wait for the 8 core variant if it has decent TDP
If they make the base clock around the vacinity of 2.5ghz, I'm sure they'll advertise a "tdp" of 60w. Would that count?
read
again
>TIM boiled and evaporated under such heat
>TACTICAL DECISION
L M A O
>If I put a tiny fridge inside a large fridge, the small fridge will work without drawing any power because it's cold and there's no heat! ha-haH!
Read a physics book other than your middle-school one, rajesh shekelberg.
Your analogy is fucked up.
I cannot help to wonder if you are baiting or if you truly feel so strongly about two companies that are completely unaware of your sorry existence
>The upcoming
JUST
.....wait
T
M
Will these CPU's still made on their archaic 14nm finfet lithography ?!
Have they fixed Meltdown without breaking computers ?
Why do i feel like that these amd vs intel threads are filled with bots?
You're dumb.
That's that impression I always got from their absolute refusal to solder. They were saving it for the final refresh of the refresh of the refresh. Until security holes and AMD forced their hands to take one final bullshit cash-grab. As if they coulden't have been soldering them all along.
The solder wont even be able to offset the additional 2 cores, at 5ghz this cpu might be the worlds first fusion reactor.
Is it coming on 14nm++ or 14nm+++?
Just retarded shills from reddit.
>dude locks his CPU to a given vcore and frequency
>fucks with heatsink
>CPU draws between 20 and 30 more watts at higher temps
There's a name to the phenomenon of electrons can move through insulators, but can't recall the name of it.
electron migration?
you mean leakage?
christ allmighty you really are dense
let me point you to this post
OMFG
learnabout-electronics.org
The conducting part of the cpu is increasing its resistance when in high temps.
The insulating part of the cpu is decreasing its resistance when in high temps.
So, in high temps you need more voltage to push enough electrons through the same gates
And your walls seperating those wires are getting weaker and let electrons pass through them.
That's 1st semester physics
I want the /v/ children to leave.
oyy veeey
you're one of 'em, aren't you?
please stop trying so hard to "fit in" on an anonymous imageboard
krauts ruining everything as always.
>10°
Nice meme, it's more and liquid metal also cools better than indium.
Pretty insignificant over the sorts of temperatures we're talking about.
It is completely fake
Intel isn't going back to solder at this stage of the game.
The real reason is because of material and thermal stresses with post-32nm massive monolith silicon. The bean counters loved since it lowered production costs.
If Intel does happen to go back, then enjoy "sudden" death syndrome chips when the solder/thermal cycling breaks down the substrate cover of the silicon.
Categorically wrong. The temperature coefficient of resistance for copper is .393% per degree Celsius. From 20° to 80°, that makes 23.6% increased resistance. Power transmission and distribution conductors are derated all the time due to ambient temperatures, where the delta isn't even close to 60°. In a microprocessor, switching frequency comes into play as well as other factors. Why comment on something you don't know the first thing about?
When will ryzen 7nm get released?
april at the earliest
this is the stupidest shit i ever read
saged
threadripper is for actual work not gaming
the ryzen r3 1200 on that chart should say enough.
too far