"Intel gate" - better performance after the patch

guru3d.com/articles_pages/windows_vulnerability_cpu_meltdown_patch_benchmarked,1.html

Other urls found in this thread:

security-center.intel.com/advisory.aspx?intelid=INTEL-SA-00088&languageid=en-fr
myredditvideos.com/
twitter.com/NSFWRedditGif

The benchmark must be false, then.

AMD niggers be begging for this patch now, lmao. Intel wins again

benchmarks are always bullshit
you can't take something with as many variables as a modern processor and say "they're exactly this good :)"
normies are stupid as fuck though and need an arbitrary number to tack onto their products to compare them so it's always gonna be this way.

Don't fall for Intel and Sup Forumskiddies memes, they are deeply compromised.

>damage control

BASED INTEL

Intel is strong, my friends. This recent 'incident' has allowed us to not only increase out chips security. Now, we have increased their maximum performance. This we provide to our loyal customers free of charge.

>dx12
The whole point of it is to reduce syscalls and CPU load in general, obviously it isn't affected by the patch. Fucking brainlets.

IT'S FAAAAAKE FUCKING REEEEE

Would you be saying the exact same thing if the benchmark favored AMD? Give it up, your shilling is showing.
You're confusing it with Vulkan. DX12 is a pile of shit.

...

AMDfan are so pathetic

This is how my haswell laptop faired.

DELID DIS

420 means you should burn it?

how bored do you have to be to engage in processor wars?

That's a big fucking drop for Intel.

AMD fags have nothing to do while they wait.

I'm just staying offline forever then.

>19% drop

how bored do you have to be to be engaged in the happening of your hobby?

and public school is still out for another week

Sorry for you loss.

i7fag here is my shit kill?

>LE summerfags meme xDXD inb4 summerfags need to go :DDDD

There is no perf loss this is DISINFO I repeat this is fake news propagated by AMD shills don't let them bait you

>tfw was saving up for months and bought a new i5 laptop this christmas

don't forget to turn on the mitigation

...

this is my cheap 1600 overclocked with stock cooler running as silent as [spoiler]intel's CEO while selling stocks[/spoiler]

rip

if there are people older than 18 posting these shitty meme images, I feel sad for our future

did any of you retarded niggers even read what the vulnerability/patch was about?
why would a normal benchmark be affected by the patch?

This is the most serious thing to happen in computing since they were invented. Literally every Intel chip since the first fucking Pentium is fucked.

>DAMAGE CONTROL
OY VEY SHUT IT DOWN

what went right?

I'll take that as a "no."

is your reference pre-patch?

have the same, can you deliver some benchmarks ?

How are the Haswell desktops fairing?

that has nothing to do with posting retarded amd hat wojaks

huh guess I should reapply the patch to slow it down :^)

...

the reference seems to be CPU-Z default reference for that processor

I did, you jackass. Did you?
It's an all around shitshow for consumers and professionals alike. There's literally no gain or bright side for Intel from this situation.

At least enable the security fix.

Who needs internet anyway?

But what if our current speeds are fake and the should actually be 25% slower to be safe?

Are we going back or are we just going back to our proper path?

No performance hit they said

...

what if

just bear with me here, everyone

what if Intel have known about the security flaw for years and there are some elements in the more recent CPUs that have always been planned to be turned on only by a patch released after shit hit the fan?

Haswell is the first to have PCID so the hit is less compared to older gen.
Just expect more CPU usage percentage by about 10 to 20% in task manager when doing CPU heavy tasks.

doesn't the patch affect the speed of syscalls and that kinda shit?
how the fuck does one know whether their benchmark is syscall-intensive?
what exactly your benchmarks measuring?

isn't more of a graphics benchmark?

how about and ??
is it just me or they lack, dunno, measurement units?

the only one that seems useful is since file operations rely heavily on syscalls

and don't get me wrong, I'm no shill, but the experiments you guys seem to be running really lack rigor and details

lol
lol

pretty new i7's in benchmarks that say 'don't worry, it's nothing'. Yet most people have laptops and desktops that aren't that.

>Haswell is the first to have PCID so the hit is less compared to older gen.
My librebooted Core 2 Duo machines are fucked, aren't they?

IT WAS SUPPOSED TO BE OUR TIME BROS!

>tfw i5 8600k
Am i crashing down with no survivors?

XDDD

i would like to see the source code of these (((benchmark))) programs.

apply for a job at futuremark if you're so good lol

>54% drop

>how about and ??
>is it just me or they lack, dunno, measurement units?
CPU-Z has a computational benchmark.

shit

i will never buy inlel again, fucking ever. zen+ next year, then hopefully arm on desktop or risc V

>brand new top of the line 8700K gets rekt by an almost year old Ryzen
the absolute state of Intel Aviv, and we are only 2 months away from 12nm Ryzen refresh

does not answer my question: what kind of tests are being run?

because if that program tests, let's say, arithmetic or memory-intensive tasks, sorry, but it does NOT measure the impact of the patch.

the patch mainly impacts the time it takes to perform a syscall (interacting with files, devices, hardware, other processes or itself, etc)

which pretty much explains the impact on file operations as shown by , which makes it much more believable than CPU-Z or whatever for now.

Consumers aren't going to be significantly impacted by this.
It certainly will be a shitshow for anyone who runs a server, but that isn't most people.

this is the new intlel shill meme.

What programs are consumers running that make heavy use of I/O like big server owners are?

could you actually imagine bragging about something you'll never own?

do you also like to talk about how the Ferrari is faster than the Lamborghini at a cheaper price point?

>Consumers aren't going to be significantly impacted by this.
Call bullshit. There are lots of people out there that run their cpus to the max, for all types of shit. Music production is a good example. Then there are lots of software apps out there for home networks, streaming, copying files, torrenting etc... were your computers are basically servers.

>consumers can't access their online services properly
>slowdown of up to 60% in certain operations locally
>but they're not affected by this

kike

>uses flstudio
>has 50 projects
>each running at 90% cpu
>when reaches 90%, bounces audio down to disk
>adds more tracks til cpu is 90%
>uses max cpu
>uses max ram
>uses max hd
Won't be affected because you never reach over 60% on your cpu and you are everyone.

Rev up those class action lawsuits

except everyone depends on servers, you braindead dolt.
wanna see a matchmaking cluster or even a game server itself losing a third of its CPU performance? me neither.

Of course they knew about it. They designed it that way to make it faster at the expense of security, knowing AMD would never do something so incompetent or devious because they'd go out of business if they did.

If you ever wondered how Intel went from getting their netburst ass handed to them by Athlon to leading by a mile, now you know. Literally the Lance Armstrong of CPU design.

So theres nothing i can do against meltdown and spectre as a intel user?

patch and accept shit performance

You can either go offline and airgap your machine without patching, patch and take a large performance penalty but stay online, or don't patch and don't take the performance penalty but then get fucking owned 100 times in the next 100 days.

Meltdown is being patched. Some spectre attacks are being patched but only as they get discovered if that.

Fix is easy, throw that intel shit in the trash and buy an AMD Ryzen.

I was looking to spec put a sandy bridge box for my wife to play some Vista era games, but I feel I would just be kneejerking to by Ryzen right now. Wait a couple weeks maybe but before absolutely frickin everyone buys them out.

Important question since you are using a laptop were you performing the cinebench testing with or without the laptop being plugged in?

I ask because your score does change and if it was plugged in during the testing your laptop may in fact have worse performance once unplugged.

I got an 8700k before all this, How fucked am I.

The amount of cock you suck must be astounding.

margin of error and this mongoloid forgot the registry

this

you need to activate the patch

so according to security-center.intel.com/advisory.aspx?intelid=INTEL-SA-00088&languageid=en-fr my core 2 duo and g3258 are unaffected by this

No, all Intel (and AMD, VIA, ARM, MIPS, etc) CPUs that uses OoOE is vulnerable. Intel introduced OoOE with the Pentium Pro all the way back in 1995 and haven't looked back (apart from Itanium and the first generation Atom)

Pajeet here, we fixed some slowdown errors in the windows kernel with the update so it now runs faster. Linux is btfo though.

I did and still don't notice shit

thing is, amd getting better and doing better means more capital, and more money into r&d.

Amd doing better than intel in a space where people aren't bound to use only intel or only amd, means good things for the future.

Will I ever own a threadripper? yea, especially if I get annoyed enough by render times I will, 1000$ for a cpu isn't to much to ask for me considering what you get but right now Im ok where I am at.

They said normal users that just browse the web and do some light work won't notice a difference, but if you're running a database then good luck

>He doesnt know how to use benchmarks
Fuck off and die please

Kek. You son of a gun

Later on, reports by normal fags saying "the removal of Net Neutrality is taking effect!"

What software is this?

wew lad

A benchmark tool, its cool because it can tell me how fast my CPU can run :^)