d1e08d No.14086612
tl;dr every Intel processor made within the past decade has a severe design flaw that could grant hackers complete control over your computer no matter what OS you're using, and the fix rolling out soon costs 5-30% of performance.
Enjoy your framerate hit, Intelniggers.
https://archive.is/RaNZI
4f2f8d No.14086617
>second image
That looks like System Shock.
48e260 No.14086627
The performance hit only affects system calls, right? It might actually not be that detrimental to the majority of code execution.
2aa5db No.14086640
5a3bd2 No.14086647
>>14086612
>a severe design flaw that could grant hackers complete control over your computer
You mean intentional backdoors for the CIA that have been compromised?
d1e08d No.14086655
>>14086627
This shit affects pretty much everything.
>It is understood the bug is present in modern Intel processors produced in the past decade. It allows normal user programs – from database applications to JavaScript in web browsers – to discern to some extent the contents of protected kernel memory.
>The fix is to separate the kernel's memory completely from user processes using what's called Kernel Page Table Isolation, or KPTI. At one point, Forcefully Unmap Complete Kernel With Interrupt Trampolines, aka FUCKWIT, was mulled by the Linux kernel team, giving you an idea of how annoying this has been for the developers.
>Whenever a running program needs to do anything useful – such as write to a file or open a network connection – it has to temporarily hand control of the processor to the kernel to carry out the job. To make the transition from user mode to kernel mode and back to user mode as fast and efficient as possible, the kernel is present in all processes' virtual memory address spaces, although it is invisible to these programs. When the kernel is needed, the program makes a system call, the processor switches to kernel mode and enters the kernel. When it is done, the CPU is told to switch back to user mode, and reenter the process. While in user mode, the kernel's code and data remains out of sight but present in the process's page tables.
>Think of the kernel as God sitting on a cloud, looking down on Earth. It's there, and no normal being can see it, yet they can pray to it.
>These KPTI patches move the kernel into a completely separate address space, so it's not just invisible to a running process, it's not even there at all. Really, this shouldn't be needed, but clearly there is a flaw in Intel's silicon that allows kernel access protections to be bypassed in some way.
>The downside to this separation is that it is relatively expensive, time wise, to keep switching between two separate address spaces for every system call and for every interrupt from the hardware. This adds an extra overhead, and slows down the computer.
147d74 No.14086668
>>14086612
How long until we see class action?
f44728 No.14086670
>>14086668
i found the jew
2aa5db No.14086672
48e260 No.14086681
>>14086670
>Oy vey, Intel did nothing wrong goyim
f44728 No.14086694
>>14086681
im just hecking you mang
48e260 No.14086828
>>14086647
To be fair, AMD would have it too if it was a backdoor.
7f19e5 No.14086836
>>14086612
>allowing ANY outgoing connections you don’t know about
lol
80489d No.14086851
85e392 No.14086868
>>14086828
nah, intel's got far more nefarious things going on.
https://please use archive.is.com (use archive.is)/2016/06/15/intel-x86-processors-ship-with.html
there really is no defending Intel, and doing so just makes you look like an agent of ZOG.
85e392 No.14086873
43e374 No.14086882
>>14086612
>Enjoy your framerate hit, Intelniggers.
>implying CPUs are the bottleneck in framerate
f7c12c No.14086891
>>14086882
>not bottlenecking your CPU
GPUlet detected
48e260 No.14086892
>>14086868
AMD has their own version of that. Look up AMD PSP. That's why I say that AMD would have it too if it was an intentional backdoor.
e2ed17 No.14086900
>>14086892
Didn't they start up a petition for removing PSP and then pretended it never happened?
2812cd No.14086908
>>14086873
>>14086868
AMD and ARM CPUs also have an "hidden" CPU with higher privileges level than the main cores.
85e392 No.14086928
>>14086908
>>14086892
fair enough
>>14086900
they would be wise too. the whole thing reeks.
ddd6c2 No.14086935
>>14086908
ARM cpus don't. They have TrustZone instead.
2812cd No.14086945
>>14086935
>ARM cpus don't. They have TrustZone instead.
Wait, trustzone isn't an extra core that can disallow execution or force execution of certain things on the main cores?
ddd6c2 No.14086952
>>14086945
TrustZone is a cryptography processing engine. It isn't like ME.
ddd6c2 No.14086967
>The code they submited to the Linux kernel was purposefully obfusicated to hide what it was patching
Defend this GPLfags.
48e260 No.14086980
>>14086967
Ready to join the FreeBSD master race?
f3fc32 No.14086981
>>14086892
So it's not a faulty NSA backdoor, just shit design on Intel part. Well I never doubted they've been using substandard pay undergrads to design their CPUs, there's no way actual professional engineers would've produced piece of shit so inefficient it casually runs 40 degrees hotter than the next AMD processor.
72d7b5 No.14086986
>>14086612
>PC vidya
>Intel, AMD, Nvidia all dirty jews
>Console vidya
>Sony, Nintendo, and MS is more severe jews
What do, /v/? What should I do?
ddd6c2 No.14086991
>>14086986
Program and play on Amiga/Acorn/early IBM PC compatables.
d1e08d No.14086995
>>14086986
Pray for POWER9's price to come down.
48e260 No.14087010
>>14086995
Imagine telling someone in the 80s that IBM would one day be the least jewish of their contemporaries.
48e260 No.14087026
>>14086986
The Sega Dreamcast homebrew scene is still alive :^)
2a3856 No.14087040
>>14086986
https://www.crowdsupply.com/eoma68/micro-desktop
Or other selfmade solutions, maybe you could even find small time silicon producers that would be willing to jump on the market searching for security by freedom
98da72 No.14087042
Good thing I only use a computer to watch anime and play VNs.
f3fc32 No.14087044
>>14087042
>watch anime and VNs
FTFY
535083 No.14087051
>>14086986
play Grimoire on Temple OS on a PC you've built from scratch in a faraday cage.
429b91 No.14087056
The worst part about losing control over my system would be if the guy deleted all of my porn and video games.
No script will keep me safe though, r-right guys?
eb4962 No.14087062
>>14087056
>noscript
>not µmatrix
48e260 No.14087064
>>14087056
Just update man. You know it's bad when they work so shadily and hastily to patch it.
>>14087051
Grimoire isn't FLOSS you CIA nigger.
535083 No.14087074
>>14087064
>Just update man
>we've patched all of our flaws, backdoors and exploits, and replaced them with new ones!
5837f6 No.14087076
As far as the security issue goes, does it require connecting to Notmalicious.network.biz or some shit?
7f6130 No.14087079
So which Intel processors are affected?
2812cd No.14087082
>>14087079
All the relevant ones.
429b91 No.14087084
>>14087064
I'm already using a 2010 quad core, any more slowdowns and I won't be able to run Warcraft 3, which has, ironically, only gotten more bloated system requirements over the years.
Thanks Blizzard.
>>14087062
Is it better?
I'll check it out.
48e260 No.14087087
YouTube embed. Click thumbnail to play.
>>14087074
Then you shouldn't be using windows at all. Probably not even Linux to be honest.
I feel now is as good a time as any to remind everyone that, even in a best case scenerio (pre-PSP AMD), x86 is shady as fuck. See embed related if you have the time.
535083 No.14087094
>>14087087
I wish self driving car would just fucking kill me already.
2fe3c3 No.14087101
>>14086612
I'm too tired to read thoughtfully so excuse me, but
If this forces a kernel redesign for operating systems, doesn't it affect all the processors the systems are going to run on?
The performance hit, if it comes from OS design, shouldn't it be the same for both intel and AMD silicon?
fc161a No.14087104
>>14087101
The kernel has a flag that checks if you're running Intel before activating the solution.
Since AMD is made by the less retarded variant of Pajeets, apparently.
cb64b0 No.14087105
>>14086668
Sorry anon, if past Intel class actions lawsuits are anything to go by, the settlement will be for $20 a person and Intel will laugh all the way as they made more money by convincing you to buy an Intel that was 15% faster than an AMD.
Intel are evil kikes, they will find some way to blame this on someone else and fuck over consumers. When they got in trouble with cheating in SYSMARK and lost that class action lawsuit about that, they only paid out like $10 to people.
One of the biggest douchebag things Intel did, was when the courts ordered them to display text that Intel basically cheats in benchmarks, they released the text as an image so you can't find it easily via searching.
Intel is pure evil and when all of you Intel fanboys lose 30% of your CPU performance or give Pajeet Javascript access to kernel memory, maybe now you will see masturbating over +/- 5% in benchmarks is a waste of time.
2fe3c3 No.14087125
>>14087104
That makes sense. Thanks.
5837f6 No.14087126
>>14087076
Hello Faggots! I have an actual question
9d2ad9 No.14087134
>>14087105
All corporations are evil, all are in on it. You might as well accept the fact that if the powers that be want to spy on you they will. The solution, naturally, involves a degree of violence the majority of the populace is not willing to accept, so here we are, stuck in this hell and waiting to see when it all comes crashing down.
f3fc32 No.14087137
>>14087105
This time it would have hurt pretty big companies so expect lawsuits for ridiculous compensations that they will not be willing to settle out of court.
80e981 No.14087148
>>14086866
>fam
Get out nigger.
816388 No.14087151
>>14086640
>>14087093
>tfw my 6 year old AMD CPU keeps getting faster and faster due to better industry-wide multicore support
>tfw the last time I had to upgrade to a new socket type was over half a decade ago
Going red felt kinda risky at the time, but good god has it proven to be a real long-term investment.
80e981 No.14087179
>>14086612
>At one point, Forcefully Unmap Complete Kernel With Interrupt Trampolines, aka FUCKWIT
>FUCKWIT
Are we being trolled?
fcdf0e No.14087185
>>14086986
Abandon electronics
9d2ad9 No.14087186
>>14087151
You do understand AMD processors have the same kind of built backdoor, right?
a66495 No.14087196
>>14087186
there is no escape from this hell.
005a79 No.14087200
If you want to call me an intel shill or whatever (Even though my two main work PC's are intel) there is literally no evidence your PC will be slowed down at all as normal programs never talk to the kernal anyways except if its to do something like open a device like a USB drive (You know this as the USB handshake request) or get access to read/write a file, opening a device driver file etc
99.99% of the shit you do daily is done in user space, not kernal space thus wouldn't be slowed down by these changes anyways, shit you do in userspace includes:
-Games
-Music
-Rendering software
-Web Browsing
etc etc etc
If anything the only place this could cause performance issues is massive read/write database files such as server farms etc, but even to you the home user nothing should get slowed down as no user runs their CPU at 100% load 24/7 and having a huge amount of read/write calls will only bump your CPU use up from .4% to .7%
5837f6 No.14087211
535083 No.14087212
>>14087196
There is, on this very site!
>>>/suicide/
f3fc32 No.14087214
>>14087126
Literally any program running on your Intel CPU can do this. All it needs to do is to find the kernel and then access some of its memory. The kernel memory is normally protected. The CPU will preload the code to access this memory in advance but shit Intel chips will not check whether or not the program had privilege to access this memory until after it runs, and thus will happily run it.
80e981 No.14087215
>>14087200
>read/write is slowed down but it will never affect normal users
Literally EVERYTHING you do on a computer uses read/write. Me typing and submitting this reply uses read/write.
75e9aa No.14087222
>OSwide updates have to be rolled out
>won't happen for any windjews OS before 10
Looks like giving that shit out for free wasn't enough, now they have to fuck anyone not running it.
5837f6 No.14087224
>>14087214
So this doesn't really mean anything unless I run some "This is totally not a virus trust me.exe" or some shit. Right?
f3fc32 No.14087227
>>14087179
No, Linux devs legitimately think that letting any program run without checking its access level is pants on the head retarded.
ddd6c2 No.14087228
>>14087200
>Games
Have to access graphics through the kernel.
>Music
Have to access stereo through the kernel.
>Rendering software
Have to access graphics through the kernel.
>Web Browsing
Have to access LAN through the kernel.
This effects everything you do.
48e260 No.14087232
It's too bad TempleOS only works on x86_64. I guess God didn't account for (((them))) in his master plan.
>>14087200
Hardware interrupts are affected. I hope you like stuttering every time you press a key, intel shill faggot.
>>14087222
source?
80e981 No.14087241
>>14087227
Aren't they correct though?
d1e08d No.14087245
>>14087179
The name is a giant middle finger to Intel for being retarded enough to make implementing FUCKWIT necessary.
>>14087224
Supposedly it's possible to trigger the bug through browser exploits or any networked program getting exploited, so unless you disconnect from the web or install a performance-degrading security update you're basically fucked.
005a79 No.14087247
>>14087215
No it doesn't, did you save the file?
Your web page cache requires kernal mode access in order to write temp files to your hard drive, but shit such as all the text on the screen/graphics are in userspace
>>14087228
GPU's do not constantly call the kernal, they only call the kernal once when you load a device driver
>>14087232
Keyboard hardware interrupts are part of the keyboards on board system, a hardware interrupt for it as a device which then sends out a signal to your PC that X key has been pushed, this is handled by your keyboard, not by your computer
Unless you're unplugging/replugging in your keyboard thousands of times a second nothing will change performance wise. Mouse movement is also 100% in user space
Do any of you know what the fuck you're talking about?
ddd6c2 No.14087249
>>14087232
Someone should create a microkernel for POWER that acts as a backend for an x86_64 emulator that runs TempleOS.
dc19d3 No.14087251
Looks like someone in the AA got mad with someone with a computer.
I thought they would only use those things to cripple a nation, not as niggerish things to be done on a random day. First the backdoors and overclock lock, now this.
f3fc32 No.14087253
>>14087247
>GPU's do not constantly call the kernal
Except they do. Thousands of times per frame. In fact most games are bottlenecked by draw calls, not actual rendering payload.
8ea134 No.14087255
>>14086900
Some BIOS now have an option to "disable" PSP, but it's not sure if it actually turns it off, or it just disables communication between PSP and other parts of the CPU and/or chipset.
5837f6 No.14087263
>>14087245
Realistically couldn't that be just any security compromise in a browser or whatever?
I mean it sucks dick because it'd basically mean any of the most popular software is going to have aggressive attempts to take advantage of this, but still
85e392 No.14087270
>>14087247
>Do any of you know what the fuck you're talking about?
he doesn't.
005a79 No.14087272
>>14087253
Unless you're running DX9 games under 100% CPU load you shouldn't be affected, and I can't think of a DX9 game made that would kill a modern processor as DX11-12 all were designed around improving API and Direct Draw calls by at times almost a thousand fold
d1e08d No.14087276
>>14087247
>doesn't know how to spell kernel
>reddit spacing
485b46 No.14087280
>>14086612
I know its the OS's that need to do the patching right now but post yfw processors in the future need firmware. Which will most likely lead to requiring they be (((always online))).
d1e08d No.14087290
>>14087272
>as DX11-12 all were designed around improving API and Direct Draw calls by at times almost a thousand fold
>doesn't know the difference between drawcalls and DirectDraw
Newfag detected.
>>14087280
Processors already have updatable microcode, but this flaw is so deeply ingrained in the processor that no mere microcode tweak can fix it.
a30390 No.14087297
And to think, Brian Krzanich (Intel CEO) sold most of his shares just over a month ago (keeping only the bare minimum he is required to hold). I somehow doubt he didn't know about this when he cashed out (Microsoft at least knew about it as far back as at least November, so Intel likely knew about it even earlier)
f3fc32 No.14087298
>>14087272
Dogshit API not relevant here, the relevant part is the context switch, which is expensive as fuck already and it's going to become WAY more expensive yet. With DX11 you're looking at 100 000 ish kernel calls per second, with DX12 that's 1 000 000 ish.
ae617f No.14087299
>>14087292
How much you pay?
d1e08d No.14087311
Anyhow, 485b46 is retarded because many modern games rely heavily on texture and level streaming and many implementations of Denuvo are constantly doing fucky shit with the hard drive. Biggest vidya performance impact will probably be on open world games or Denuvoshit.
4cd5db No.14087317
>>14087299
I bought it at release, so $320.
005a79 No.14087321
>>14087298
Looks like programmers will have to learn how to pack calls again instead of brute force executing code won't they?
That said DX11/DX12 most of those calls are handled by the GPU itself and very little is done by the CPU, which is why microsoft etc has been trying to get the industry to switch for almost a decade
Vulkan is an example of almost all draw calls being handled by the GPU
79c39b No.14087324
Unless this new code is being forced down my throat via some kind of telemetry, then I elect to refrain from accepting it. What's the worst that could happen?
ddd6c2 No.14087329
>>14087324
Depends on which OS. You will be hit with severe performance drops.
4cd5db No.14087334
>>14087321
Except Physics and DX11 Multithreadding (That thing that makes Nvidia own AMD in DX11)
48e260 No.14087336
>>14087329
He's asking about if he DOESN'T update.
6872f6 No.14087342
f3fc32 No.14087353
>>14087321
What part of a "kernel call" do you not understand? Everything that's not resides entirely within your own program gonna involve calling a kernel. And unless your game carries its own GPU driver, it's gonna call kernel every single time it needs to contact the GPU, for any reason.
4cd5db No.14087354
"Kernel Call" sounds like anything that would use winapi.
6872f6 No.14087358
>>14087324
>worst that could happen
Someone pwns your bios and something along the lines of your cpu microcode's microcode. Which is to say spying and stealing of your information.
afd6ca No.14087371
>>14087241
absolutely, what makes you think they aren't
f3fc32 No.14087373
>>14087324
You will be mining coins for someone else 24/7, the computer will be powering itself back on if you power it off, and wiping the hard drive won't and doing clean install won't solve the problem.
f3fc32 No.14087375
>>14087373
>won't do shit to stop it
f3fc32 No.14087379
>>14087369
Literally everything uses virtual memory you fucking dickweed.
4cd5db No.14087381
>>14087369
I would like to hear more about this bug.
If it is something exclusive to VT-D then not "every" Intel processor is vulnerable.
If it is something specific to memory and using a certain series of assembly instructions causes escalated code execution then it affects everyone and would be very dangerous.
4354f0 No.14087395
>>14086612
Good for me. I basically gave up on security, and I don't update shit. Still running Windows 7 with no updates. Because of this shit, at least they might give up on the old backdoors, and might therefore benefit me even if I don't update.
>>14086986
I honestly kinda regret spending this much money on my PC. Just get a used console for a good price when you can. And use older, cheaper computers for games that you can run pretty easily. Gaming on PCs is pretty annoying. A lot more issues than any console would ever have, and it forces me to use Windows, and Windows is fucking garbage. I don't play very many recent games, and they are all on PS3, and I own one anyway, for the exclusives. It would have been smarter to buy an old, durable machine, for lighter emulation and internet. Maybe a used PS4 when it's cheap enough.
Overall, fuck post-modern technology.
a5e0db No.14087417
d1e08d No.14087423
b999de No.14087427
9c3a49 No.14087432
>>14086612
>bought a AMD processor at low price
>always regretted it
>this many years later it pays off
WITH JEWS YOU LOSE
4cd5db No.14087448
0518ce No.14087456
>when companies aren't purposefully fucking you in the ass they're letting someone else do it instead by being incompetent
Holy shit, I just want to play videogames and shitpost, why can't they let me do that in peace? If this shit really does have a big performance impact then I ain't getting it, and if that permanently ruins my computer then I'll go find a new hobby, because this one isn't worth the trouble.
d1e08d No.14087472
>>14087448
>IO heavy workloads
Like hosting an imageboard. I bet Jim feels real happy about his Intel servers now.
5837f6 No.14087479
>>14087472
Wait, Jim Jim or different Jim?
e9f9fc No.14087482
>5-30 percent
>numbers are dependant and vary considerably by workload
>that fucking range
>still no actual benchmarks released
>just speculation and rhetoric by fanboys
So in other words I can expect to see little to no performance penalty
Thanks for letting me know fam
4cd5db No.14087483
>>14087472
You're not wrong.
7f19e5 No.14087490
>>14087472
Why wouldn’t he? His purpose is to destroy the site and its userbase. We’re already looking at a quarter of what we had before he took power. He’s probably ecstatic about it.
>>14087479
Who the fuck do you think.
5837f6 No.14087492
378244 No.14087500
>>14087490
>Why wouldn’t he? His purpose is to destroy the site and its userbase.
Codemonkey has been keeping this site running a lot fucking better than fyretyres ever did
d1e08d No.14087508
>>14087490
Endchan and Meguca exist, but I wouldn't be surprised if their servers use Intel CPUs too.
2812cd No.14087515
>>14087500
Still hasn't fixed the biggest bug the site has since 2014
b999de No.14087517
>>14087500
>Codemonkey has been keeping this site running a lot fucking better than fyretyres ever did
Hey anon, do me a favor and enlarge your reply box and try to make it smaller again.
e9f9fc No.14087521
>>14087487
Why does the performance impact for Skylake look so much more negligible compared to Coffee Lake? Not complaining since I use Skylake but does this bug simply not effect Skylake as badly?
4354f0 No.14087523
>>14087490
>His purpose is to destroy the site
I hope he does. This place is shit. So bad that this is actually the best board. Where's my 16chan?
378244 No.14087524
>>14087515
>Still hasn't fixed the biggest bug the site has since 2014
Which one?
>>14087517
At least it runs decently fast and is relatively stable. But yeah I didn't even know the reply box was fucked up
b999de No.14087539
>>14087524
>At least it runs decently fast and is relatively stable
Well we keep being okay with it because we constantly compare this site to how bad of a state cuckchan is in. Especially now since it's literal malware. But we should start looking up and not so far down since it's not that hard being better than a giant fucking turd.
I know Chodemonkey is doing a somewhat dandy job, but for some reason whenever he touches anything something else breaks.
Almost like a certain pig phobic who fucks shit up so he stays employed to "fix" it again.
7d6e09 No.14087548
The real question is if they will provide the update to windows XP, Vista, 7 and 8/.1 like they probably will or if they will actually make good on their claim of no longer providing support to old OS like they never have.
378244 No.14087553
>>14087539
>Well we keep being okay with it because we constantly compare this site to how bad of a state cuckchan is in
I was comparing it to how the site ran before and during the infinity next debacle. I haven't been back to halfchan since I left in 2013 so I don't have anything to compare this to other than that.
>Almost like a certain pig phobic who fucks shit up so he stays employed to "fix" it again.
Ron is Jim's son, I think he would keep him employed either way. I'm more inclined to believe that he's more focused on Jim's other sites and 8chan is just a sideproject for them so he puts minimal effort into it. Yeah my standards are pretty low but I really like this place. I never understand when people complain about how shit it is, there's a few downsides but for the most part its good
75e9aa No.14087558
>>14087548
>XP
No way in hell.
>vista
Likewise.
>7
Probably not.
>8.1
Maybe, since it still uses their awful appstore and such.
5837f6 No.14087560
>>14087548
They already said no and IIRC they never even officially supported x86 in anything but 8 and 10
2812cd No.14087561
>>14087521
The first few benches are not CPU limited on the 6800k while they are on the 8700k from what I gather (file I/O bench, NVMe SSD on one side SATA 3 on the other)
>>14087524
>Which one?
The post inlining memory leak
Inline a post in the last post of a thread, when a post is made it won't display after the page autorefresh fetches it but it will still count as there being a new post, next refresh will recount that post (and if other post have been made they will be added to that counter as well) as a new post and since the auto refresh shortens on new post it will add those to the new post counter endlessly every 5 seconds, now that's the visible side of it I dunno what happens under the hood but in a couple hour you will crash pretty much any browser except FF quantum for some reason
b999de No.14087565
>>14087553
>there's a few downsides but for the most part its good
I'd like to believe people because they care.
Because they know it could be a little better, but I'm not gonna pretend I'd know how to do a better job, I'm just kinda raising my concerns as user.
b999de No.14087567
>>14087565
*like to believe people bitch because they care
95a4ee No.14087575
>>14087548
>use win 10 or this russian we didnt hire will use you like a truck stop whore
oh shid
e9f9fc No.14087584
>>14087561
Why did Skylake see an improvement in h264 encoding fps? Also why didn't they bench Skylake with Nvme/PCI mode SSDs when Skylake supports it?
378244 No.14087585
>>14087561
Oh the memory leak, has it really been in since 2014? I never noticed it until the last few months.
>>14087565
Yeah, being content with everything just leads to complacency that's true. I'm done with the metaposting though since it's shitting up this thread
48e260 No.14087592
>>14087548
Earlier this year (or rather last year) they backported a patch to fix an exploit in the Samba protocol (the thing that caused that huge ransomware debacle. It's worth noting that the tools to make use of this exploit were LITERALLY programmed by the CIA) all the way back to XP, even though XP is no longer officially supported.
So yes, probably at least 7 will get a patch depending on just how severe this bug is.
2812cd No.14087607
>>14087584
>Why did Skylake see an improvement in h264 encoding fps?
>242.6 vs 243.2
Probably just margin of error stuff, not very relevant.
>Also why didn't they bench Skylake with Nvme/PCI mode SSDs when Skylake supports it?
Good question, dunno the answer.
>>14087585
> has it really been in since 2014?
Early on it just broke the last post and shoved every new post in that post while breaking the formatting in various ways, then that got fixed late 2014 but instead it caused a memory leak.
7d6e09 No.14087609
>>14087560
>>14087558
Microsoft has been saying they wont update XP since 2014 and yet they updated it just last year to patch a security hole as >>14087592 mentioned. Hell all the cash registers at my job run on XP my money says we are going to have a fun time with the update even though we probably don't even have effected processors
3127e5 No.14087615
>>14087292
>>14087317
>320
What's the point of a budget AMD CPU if you're going over 300 bucks? The goal is to try to save money. I bought an R5 1600X for 215 in November and from what I've seen the price went up a bit. I put it under water with a cheap Corsair H50 which was on sale and the Tdie gets pretty low with how cold it's been here.
I still can't seem to get a stable 4.0GHz on all cores, but maybe I just suck at overclocking or lost the silicon lottery. 3.9GHz stable at best on my R5 1600X. Meanwhile, my i7 5930K on a custom loop does a stable 4.7GHz from a base of 3.5GHz.
7f19e5 No.14087627
>>14087500
You’re mentally defective.
2812cd No.14087629
>>14087615
>What's the point of a budget AMD CPU if you're going over 300 bucks?
Higher MTP at release than the best regular i7 which was pricier than that,
>I still can't seem to get a stable 4.0GHz on all cores
Yeah that's about right for Ryzen, hopefully the next rev fixes that shit.
378244 No.14087644
>>14087627
Nice one faggot, I'm right though. Hotwheels was cool but he was shit at programming
7f19e5 No.14087652
>>14087644
>lol i right i say so i win
You’re. Mentally. Defective.
378244 No.14087655
7f19e5 No.14087659
>>14087655
>lol ha ha i have no argument but you have no argument lol i win you lose
Reported, then.
378244 No.14087664
828e79 No.14087694
Are Ryzen and Threadripper prices going to skyrocket now just like Vega GPUs?
Combined with the current skyhigh DDR4 prices, a new build will be too costly for the next year at least.
Very annoying.
758970 No.14087708
>>14087321
>Looks like programmers will have to learn how to pack calls again instead of brute force executing code won't they?
They don't hire programmers anymore, you incredible dunce. They hire script-kiddies with neon hair and diversity hire qualifications.
What has happened in the past ten to twenty years that makes you think anyone in the game dev environment is competent or willing to become such? Look at /agdg/, for fuck's sake, look at their OP spiels. Look at how they discourage learning to program and look at how it's mostly about pre-made engines.
The pros have always been far lazier than the hobbyist developers, no matter how highly you think of them or their abilities. If the hobbyist isn't interested in doing it the right way, what makes you think the pros are going to do it?
f67d14 No.14087712
>>14087263
No, those compromises should not affect ring0. Also, this is different in the sense that it's not something that you can simply patch or mitigate.
5837f6 No.14087718
>>14087694
>There was a sale on Ryzen processors for like 30% off
>Didn't take advantage of it because brit bongo friend said I didn't need two computers
GOD DAMN IT
On the bright side at least I have a Case Powersupply, storage and a Mobo already
d1e08d No.14087750
>>14087708
>Look at /agdg/, for fuck's sake, look at their OP spiels. Look at how they discourage learning to program and look at how it's mostly about pre-made engines
8/agdg/ has several enginedevs and three of five games on http://8agdg.wikidot.com/general:finishedprojects were built using custom engines, including Speebot.
526e87 No.14087751
>>14087694
DDR4 doesn't give you any improvement in gaming. If your building your rig just for gaming, I'd stick with DDR3.
5b654a No.14087761
>>14086873
>complains about kike BO about not being able to direct-link
O U T K I K E D
535083 No.14087794
Fuck, i need to buy a new motherboard an a CPU since i have an asus z 87K
543618 No.14087811
Way to go for forced obsolescence.
Intel, the Microsoft of hardware.
2812cd No.14087814
>Have to wait till April for a potentially not shit CPU for emulation
>>14087751
Yeah anon that's a really smart idea, or it would be if the platform could use DDR3 at all which it can't.
48e260 No.14087817
>>14087811
>microsoft
>not apple
You're confusing your kikes, mate.
758970 No.14087834
>>14087750
>OP
>spiels
Reading comprehension.
cdb0b2 No.14087850
6112a7 No.14087868
>>14087105
>15% faster
Due to software compiled with Intel compilers that purposely choose unoptimized code paths for AMD and VIA processors, even if those processors are capable of running the optimized instructions. Spoof the system to make a Ryzen look like an Intel, and that performance improvement disappears. Buying Intel means buying into market dominance, not buying better hardware.
5eaf90 No.14087872
>>14087247
>GPU's do not constantly call the kernal, they only call the kernal once when you load a device driver
Part of the graphics driver is inside of the kernel you fucking imbecile.
>>14087272
>Direct Draw calls
>kernal
Anon please stop pretending you know what you're talking about.
>>14087297
Doesn't this count as insider trading?
d31704 No.14087875
Seems like videogames are fine but other workloads are more affected.
https://archive.fo/f7MSz
f5bdc6 No.14087892
>>14087212
You'd literally go to real hell, dumbass.
336062 No.14087924
>>14087913
>choosing to download that software.
the problem intel is happening due to hardware you stupid shill.
2812cd No.14087935
>>14087924
>inb4 he throws the TLB bug your way
336062 No.14087937
>>14087935
oh god that shit is ancient.
f5bdc6 No.14087968
>>14087937
but not forgotten!!!
a1e909 No.14087971
>>14086882
>He doesn't know a weak CPU can bottleneck your GPU
e9f9fc No.14087994
Supposedly BSD actually does not have this issue because they have never implemented Page Table Splitting to begin with and the OS is simply more secure by design so they have no need to isolate Kernel and Userspace
It sounds like the issue can be mitigated with smarter kernel and userspace abstraction. You can push more shit to userspace to mitigate the need to have it call to kernel memory OR simply have a more optimized memory management subsystem. Its likely Microsoft has their own solution but went have to wait 7 days before the patches roll out and we can see if there's any strong impact on their side. But overall the issue is more or less a kernel design problem and it's likely this will impact Linux (being a monolithic kernel) far worse than Windows or OSX
3b6844 No.14088055
>>14087872
>>14087297
>Doesn't this count as insider trading?
Pure coincidence I'm sure :^)
>Intel's CEO Just Sold a Lot of Stock
http://archive.fo/AQvyZ
>Since Krzanich was appointed Intel CEO in May of 2013, he'll need to have 250,000 shares by May 2018 – or about five months from now.
>What's interesting, then, is that before Krzanich made any of the transactions that he reported in his most recently filed Form 4, he held 495,743 shares.
>After the options exercises and subsequent sales (which left Krzanich's position unchanged at 495,743 shares), Krzanich then made two more transactions: a sale of 242,830 shares and a sale of 2,913 shares, with each transaction happening at an average price of $44.555, per the filing.
>Those two transactions left Krzanich with exactly 250,000 shares – the bare minimum that he's required to hold as CEO.
>What does it mean?
>Perhaps Krzanich sold those 245,743 shares valued at nearly $11 million at the time of the transactions to pay for a new house or buy a rare piece of artwork.
>However, I find those explanations tenuous at best, particularly in light of the big windfall that he received when he exercised and sold all those stock options.
>Instead, given that Krzanich seems to have sold all the shares he could save for those he is required by Intel's corporate bylaws to hold, the impression that I get is that Krzanich doesn't have a ton of faith in the potential for Intel stock to appreciate, perhaps driven by a lukewarm (or potentially even negative) view of the company's near- to medium-term business prospects.
df219e No.14088059
>>14087324
I don't think you know what the word "telemetry" means.
>>14087487
Oh shit, this is huge. The PostgreSQL and Redis losses alone will make sysadmins shit their pants, and I'm sure they'll be mirrored with other similar software as well.
5837f6 No.14088095
>>14088084
Look intel's already got the faggot quota filled don't make a surplus
17f1b1 No.14088098
>>14088084
>Buying any kind of hardware for video games at all
t. cuck that loves getting fucked in the ass
4ab639 No.14088103
5837f6 No.14088107
>>14088103
Give me like 2 weeks.
4ab639 No.14088119
>>14088107
In 2 weeks you'll be able to own Intel itself for the change in your pocket. And you'd still be better off with the change.
a5c119 No.14088120
>just built a new PC a few months ago
>wanted Ryzen but decided to go with Intel so I could play with Hackintosh
I made a huge mistake
d453fd No.14088128
>>14088120
Look at it this way, anon: you learned an important lesson that will serve you well throughout your life.
828e79 No.14088131
>>14088120
It's not a big deal. Just don't run any software that makes system calls.
75e9aa No.14088132
>>14088120
>sitting on a four year old EOL i5 and shitty compact mobo
>have had 0 need to upgrade since everything plays just fine
>this bullshit hits
I guess I know what I'm doing with my next paycheck.
e9f9fc No.14088147
>>14088131
Its not syscalls per se, just usermode memory accessing kernel-mode memory
If the kernel API calls you want to access are exposed in usermode, or if you run a program that runs 100 percent in either kernel mode or usermode, this will not effect you. The biggest impacts of the bug are database software that requires constant access to kernel mode while abstracting it from usermode for security reasons. Its the reason why people are saying only server admins need to worry. Things like video games and video decoding shouldn't be affected
2812cd No.14088173
>>14088147
>. Things like video games
Someone posted a few benches that agree with that but no word on emulation and similar stuff.
3a921b No.14088185
Is this the hack someone posted in another thread that was a 45 minute dissertation on the -2 level permission? Basically some deprecated patch for 90s processors that they never took out for current designs despite current not needing to remap the BUS manually anymore?
671298 No.14088192
>tfw you don't know shit about computers
>build PC
>buy AMD because it's cheaper
Dodged a bullet.
I feel bad for people who had a PC running this given to them for Christmas.
e9f9fc No.14088195
>>14088173
There is no reason for any emulator to want to access kernel memory so I doubt they will be affected either
Also again, it's to be expected that this will impact Linux far worse than Windows since Linux uses much more aggressive kernel to usermode abstraction. BSD isn't even bothering to implement PTI for example
75e9aa No.14088199
>>14088185
>implying it wasn't intentional
Knowing intel, it was probably designed for the NSA or some shit, but someone spilled the beans.
e9f9fc No.14088213
>>14088195
And before any autist wants to correct me, I of course only mead vidya gaem emulators, not hypervisors or hardware assisted emulation like KVMs
9981c1 No.14088214
YouTube embed. Click thumbnail to play.
For anyone genuinely unaware of Intel's long history of massive kikery, here is a fantastic summary you can watch. It's 34 minutes long but it is presented as a slide show, so I could probably make a watchable WebM from it, but that will have to wait an hour or so as I'm not at my PC at the moment to run it through ffmpeg.
The short version: Intel loves to back out of licensing agreements and pay huge piles of cash to companies for them to not purchase AMD hardware.
3a921b No.14088285
>>14088199
I saw someone posted the video in this thread too. I'd suggest people just watch it because it's actually pretty neat how the guy found and shows you how to perform the hack. If you're on Linux please don't think you're safe from this
48e260 No.14088287
>>14088214
>I could probably make a watchable WebM from it
Good think ffmpeg doesn't seem to be affected :^)
48e260 No.14088314
YouTube embed. Click thumbnail to play.
>>14088285
Actually I posted the *other* talk about x86 fuckery by the same guy. They're both worth watching. However, I don't think this is the same bug we're talking about in this thread.
d3c5af No.14088320
>>14087186
There's a lot of rumors that the FX series don't have them since they're 2012 architecture processors extended to around 2014. If that's the case I really dodged a rainstorm of bullets.
2812cd No.14088323
>>14088314
>this is the same bug we're talking about in this thread.
Intel has had to fix at-least 3 major bugs (maybe more iME has been fucked to death) in the last 6 month all having major security implications on top of that.
dc19d3 No.14088328
Now tell me, lad, why is the american army trying to force the people to buy new hardware?
First it was the samsung shit that was burning while recharging, now apple did the same, they claimed that something was making the hardware shitty because of a software upgrade, and now theres some wierd bug on these intel hardware, wich by the way is very wierd. Do you remember that accident that happened with that Malasya Airlines plane?
https://www.google.com/patents/US8650327
This was the patent worth killing. Just read the ending of the page.
>US9417801 * Mar 27, 2014 Aug 16, 2016 Intel Corporation Virtual general-purpose I/O controller
>referenced by
>intel corporation
https://www.linkedin.com/pulse/coincidence-rothschild-inherits-freescale-patent-bryan
There's clearly some sort of connection.
71ac45 No.14088338
So… If im on Windows 7 is this update forced?
d31704 No.14088355
>>14088287
ffmpeg is affected.
697759 No.14088362
Is there any actual proof that there is an issue and that this is not just a scam to gimp older hardware so people buy new stuff? Intel are turbo kikes so it would be entirely in character for them to try something like that.
75e9aa No.14088363
>>14088338
Who knows. Apparently 7 has two more years of support, but they're going all out on W10. If I was a kike, I wouldn't roll this out on anything but W10, and force everyone to upgrade or get fucked.
a36011 No.14088368
>>14087592
I'm under impression that this happened because people stubbornly still use XP and Microsoft didn't want their machines hijacked to spread malware that could affect newer machines
6872f6 No.14088373
cc728f No.14088389
>>14088376
So it got faster after the patch that supposedly slows it down 30%? What the fuck.
d31704 No.14088394
>>14088376
>>14088389
Whoops, my mistake I thought I/O shit being affected would include ffmpeg, it still affects SQL Databases depending on what it's storing and how many transactions it needs to process are also affected by a large amount.
a2874f No.14088402
>>14088389
"what is margin of error"
2812cd No.14088403
>>14088389
Some things are unnafected, and the difference is so minimal it's not worth talking about (4.19 to 4.17 and 6.04 to 6.11
70945a No.14088405
>>14088394
>it still affects SQL Databases depending on what it's storing and how many transactions it needs to process are also affected by a large amount.
So what you're saying is Windows' update deployments are going to get even slower?
6b7deb No.14088407
>>14086655
This is brilliant. Now every software process needs a clearance from the hardware to even run. Self built unregistered computer? Too bad, nothing runs on your rig, and when something runs it's doorkicker time for breaking some arbitrary law somewhere. And "life" in prison is not what it used to be. I love our dystopian horror timeline. It's neat.
48e260 No.14088409
>>14088389
That's just a random discrepancy. There's no way a program will execute twice taking exactly the same amount of time. It's more or less the same.
48e260 No.14088416
>>14088407
Anon, that's not at all what's going on. However, one day that'll probably be the case.
d31704 No.14088422
>>14088405
No, you know this website you're using? It uses a database management system and contains records like images, videos, text, boards, bans, logs and ips the last time I recalled. But the servers that use databases would have to update in order to get that performance hit, and usually servers use 'outdated' but stable software. It's unlikely.
48e260 No.14088427
>>14088422
Don't you think a major security bug would be a high priority update?
70e3f0 No.14088432
Damn it, I was already having issues with Krita slowing to a halt, I don't need this shit too.
d31704 No.14088438
>>14088427
Most of select outdated software used for stability will have 'updates' to fix security bugs. Whoever is maintaining the server (the developers who set everything up will be gone to other projects) is going to update whenever the support for the outdated software ends. These outdated software get life support for more than 5 years, so don't expect it soon.
6b7deb No.14088441
>>14088416
I remember the Intel-internet-tuned "fast" chips+Mbs that never really slept. Secretly always on, for officially no reason. No living person ever found out why. Posting for a paranoid friend.
507433 No.14088470
>>14087134
Yeah, fuck Chick-Fil-A and their involvement in this scandal!
45f6a9 No.14088534
>>14087324
you will get fucked in the ass
3127e5 No.14088539
>>14087629
Ryzen+ needs higher clocks or at least greater overclocking potential.
70945a No.14088545
>>14086672
Reminds me of Microsoft's Critical Update Notification Tool
3127e5 No.14088567
>>14088147
>>14088195
>>14088213
So it's just another instance of fanboys of X brand making a big stink over Y brand. They get something legitimate to complain about and blow it out of proportion until most of the complaints become downright illegitimate lies. When autists are dedicated enough they can drag out disinformation and exaggerations for ages. For instance, the Nvidia house fire shit during Fermi still isn't dead even after AMD's Tahiti surpassed Fermi in power consumption and heat.
e9f9fc No.14088576
>>14088362
Since the vast majority of consumer software is run entirely in user space, this will have virtually no effect on 99.9 percent of users. And the little use cases this does affect can be mitigated with better kernel space abstraction optimization. As long as you don't need to access kernel memory from userspace millions of times a day you shouldn't give a fuck about this. Like every time this happens you tend to run into a lot of fanboys making it a bigger deal than it is. If you want to upgrade to a newer CPU at this point though Ryzen is certainly looking to be the better choice but if you are currently on the latest Intel shit its not a big deal
a2874f No.14088578
>>14088539
Ryzen+ is just a process change but it looks like a pretty good one, 15% improvement seems likely.
It's Ryzen 2 that's going to be a big deal because Samsung's 7nm process is supposedly as dense as Intel's, so this will be the first time in years that Intel doesn't have a manufacturing advantage.
662cbe No.14088586
>>14088567
I haven't seen a housefires joke for ages, not even on /tech/.
5837f6 No.14088596
>>14088578
>Samsung
Please tell me that's a joke.
098818 No.14088599
>>14088567
This is a huge windfall for AMD, and a lot of sysadmins are going to be losing sleep and their fucking minds trying to deal with the fallout. Gays like you are not the only market out there.
Just be glad this shit is getting fixed before an exploit came out in the wild, are you would have been affected.
62865e No.14088604
c7c57f No.14088607
>>14088586
WOOD SCREWS
"muh board culture" killed /g/
a2874f No.14088609
>>14088596
It was a collaboration with IBM, if that makes you feel better.
3e5258 No.14088611
well Ryzen and threadripper are pretty good alts to intel right now.
5837f6 No.14088614
>>14088609
So 1500x 1600x or 1700x?
12f6de No.14088618
>>14086828
No, because it would be too obvious when it inevitably leaked.
6872f6 No.14088623
>>14088609
>samsung AND IBM
a2874f No.14088625
>>14088614
Whatever makes your heart gladdest, but you should look at DDR4 prices before you put too much thought into it.
5837f6 No.14088629
>>14088625
I got 16gb of DDR4 that's getting pulled out of my intel machine when it gets here.
e9f9fc No.14088707
Good news lads!
OUR VIDEO GAMES ARE SAFE!
https://www.phoronix.com/scan.php?page=news_item&px=x86-PTI-Initial-Gaming-Tests
My intuition was correct, usermode applications that do not rely directly on kernel mode memory access are not impacted
fd44aa No.14088710
>>14088214
Alright here's the WebM version, had to split into two parts.
e9f9fc No.14088728
>>14088725
Fuck me I clicked it
How do I know if there's a Javascript memory sniffer running in the background?
fd44aa No.14088733
>>14088728
Bro, you gotta learn to think before you click.
Also why don't you have NoScript installed?
e9f9fc No.14088739
>>14088733
>Also why don't you have NoScript installed?
Laziness, although I might install it now
It appears to just be a harmless spambot though who wants to shill some ad-filled blog si I think I'm safe
48304f No.14088740
>>14088728
>not having Norton
fd44aa No.14088743
>>14088739
You at least have uBlock Origin installed, right anon?
fd44aa No.14088744
>>14088740
The only 7chan meme anyone remembers.
6872f6 No.14088747
>>14088728
>not already having turned pajeetscript off
First you turn off your computer and remove the battery, the AC power, and the CMOS battery. Then you desolder your MMU and CMOS/BIOS logic controller chip. Then you dump the last state of the MMU powered on and make a backup flash of the CMOS's bios portion and comapre those to a known good backup you have a backup right?. If those are good and the same and you aren't using windows as a OS you should be fine. Unless you want to get extremely autistic in checking more things.
e9f9fc No.14088748
>>14088743
M-maybe Windows defender will catch it?
6872f6 No.14088753
>>14088748
Not a snowflake's chance in hell.
6e2232 No.14088756
TFW this would be less of a problem if programers could optimize their code
JUST
fd44aa No.14088763
>>14088753
To be fair, I have seen Microsoft Security Essentials successfully block the execution of a Java application that was part of a drive-by download.
a7c419 No.14088765
Same question as this anon >>14088614. Been meaning to upgrade for awhile now. Seems to be a big price jump from 1600X to 1700X, benchmarks aren't that different granted those are lies. Fucking hell I didn't want to deal with this.
>>14088748
And here I thought I was retarded.
fd44aa No.14088767
>>14088765
The non-X Ryzen chips come with decent coolers, the X chips expect you to bring your own cooling. All models can be overclocked, so there is really no point in paying for the extra clock speeds when you can just apply them yourself and save the cash.
The AM4 socket will be supported for at least 4 years, so you will be able to drop in the next few Ryzen generations easily.
7f8c98 No.14088768
>>14088120
Why would you want a hackintosh anyway?
You can do everything on a mac on windows.
6872f6 No.14088773
>>14088768
>hackintoch
Well the older mac laptops have powerpc proccessors so they wouldn't be affected by this bullshit. Just put a GPU in a usb port and slap linux on it. Now you have emulators and FOSS games with web browsing. But there goes the entire x86 library that isn't FOSS.
5837f6 No.14088774
>>14088768
What about PocketTanks, isn't that MAC only?
a7c419 No.14088784
>>14088767
I already have a custom cooler I was kinda hoping I could reuse but some searching about is indicating otherwise.
Christ, if it wasn't for Win10 I'd just buy a new PC.
e9f9fc No.14088786
>>14088753
>>14088763
Running a full offline scan, there's no way I'm installing extra anti-virus shit that fights malware by becoming malware itself and I don't want to use GNU/Linux since this is my main media PC and I already have a dedicated Linux machine
>>14088773
Macfags will likely not notice a difference since OSX already uses Page Table Isolation
5582ff No.14088810
>>14086627
from what I've read/heard the impact to everyday applications like games is probably going to be about 5%. Not huge but it'll be noticeable.
The real issue at hand is how this will affect enterprise applications. I've heard numbers from 10 to 30 percent, which is fucking huge. Intel shit the bed big time and they're going to have a bitch of a time recovering unfortunately all the kickbacks they get from the intelligence agencies for putting hardware backdoors in all their shit will likely keep them afloat
e9f9fc No.14088829
>>14088810
This will have zero impact on regular usermode applications. 99.9 percent of users will not notice a difference, see
>>14088707
>>14088576
>>14088376
>>14088147
The biggest impacts of this bug more of an oversight really but still will be database applications like SQL that need to make millions of calls to kernel space from userspace a day. That's why Google and Amazon and Microsoft are all collectively shitting themselves. You really shouldn't care all that much though, if you're getting a new CPU though you'd be retarded not to go with AMD/Ryzen however
447145 No.14088842
>>14087105
>>14087134
>All corporations are evil, all are in on it.
But with hardware you always lose. AMD, Intel, nVidia, they are all total cancer.
a89cf7 No.14088864
>tfw still using a shitty ancient athlon II X4
And here I thought I was cursed
70fb54 No.14088874
>>14088774
I play PocketTanks on Windows
5582ff No.14088915
>>14088864
even if you upgrade to ryzen you should hang on to your athlon build. that's one of the last generations of chips without hardware backdoors
5bc9f4 No.14088993
>>14087247
>tries to sound smart
>can't spell kernel
12f6de No.14089009
>>14088915
Not to mention those are some of the last quality cpus.
5ba0f0 No.14089052
>>14086612
Some reports say it might even hit performance by 50%. The jew has outjewed itself.
e9f9fc No.14089105
>>14089052
The vast majority of programs do not need to access kernel mode memory and are seeing a 0 percent performance impact though
5ba0f0 No.14089111
>>14089105
Good to know shills are covering this up. But you're wrong, it affecting everything.
5ff151 No.14089116
mouthbreather here, what does this mean for Mac users?
e9f9fc No.14089118
e9f9fc No.14089121
>>14089116
OSX already uses PTI so you will certainly see no performance impact either
5ba0f0 No.14089123
>>14089118
Yes you are
>>14088728
>How do I know if there's a Javascript memory sniffer running in the background?
You mongoloids glow in the dark.
>>14089116
Same shit, don't trust this kike though >>14089121
2242a2 No.14089124
>>14089118
You are one dumb motherfucker.
>>14089121
>mouthbreather telling other mouthbreather
Like poetry.
e9f9fc No.14089130
>>14089123
>>14089124
I understand reality is difficult to understand but it's okay because gaslighting someone on an anonymous imageboard is easier than using braincells
5ba0f0 No.14089142
>>14089130
>oy vey youre the ones gaslighting
You don't even know what gaslighting means you kike. Fuck off.
f4948d No.14089144
>>14087395
>Fuck post modern technology
>Fuck Intel and AMD
>PC doesn't even feel good to play on
>Buy a PS3/PS4
e9f9fc No.14089148
>>14089142
I done really understand the animosity. You should be happy people didn't completely waste their hard earned money over some retarded companies oversight. But you don't really care about that and are more interested in being a brainless fanboy
12f6de No.14089149
>>14087395
This is the most jewish thing I saw today, and I stalk our board owner.
5ba0f0 No.14089151
>>14089148
>oooooooy veeeeeeey
991140 No.14089161
>>14089148
Yeah, this is getting asinine. Answer me something on-topic - I built a PC recently with fx-8350, 16GB RAM, Gigabyte MOBO. Should I upgrade to Ryzen or will my build last me another 4-5 years?
991140 No.14089162
>>14089161
*and a MSI R9 390 GPU.
e9f9fc No.14089165
>>14089161
I would say definitely get Ryzen because the massive jump in even single-core performance is worth it. The only reason I'd say to hold off would be because you'll obviously also need to get DDR4 memory (not sure how Ryzen plays with DDR3) and there's a big DDR4 shortage at the moment
1e609a No.14089167
>>14089161
Upgrade to Ryzen
FX8350 is pretty outdated
ddd6c2 No.14089169
>>14089161
Don't. PSP has yet to be open sourced and is present on Ryzen.
e9f9fc No.14089181
>>14089169
This is a board about video games, I done think he really fucking cares about wither something is FOSS or not, this isn't /tech/ he's not going to share your autistic sentiments. I don't really think PSP is on the same level as ME anyhow
ea4900 No.14089188
>>14089165
>DDR4 shortage at the moment
nah, it's price fixing by the kike-gooks
hope they restore the original price on ram modules now that the holiday season is over
991140 No.14089191
>>14089165
>there's a big DDR4 shortage at the moment
Yeah, you weren't kidding. 160 USD for 16 gigs of RAM? Almost double compared to DDR3.
My main issue with upgrading is price - 300 USD worth of hardware swapped out (won't sell for much either) for 400 USD CPU, MOBO and overpriced RAM that may not provide an appreciable performance increase for the cost.
662cbe No.14089195
>>14089181
Coreboot rates it about the same on the danger scale, as does Libreboot. The whole issue is it's a secondary CPU overseeing the main one with no idea or control of what it's doing, perfect for a backdoor. The correct way is to have the security coprocessor under user control.
e9f9fc No.14089196
>>14089191
>MOBO and overpriced RAM that may not provide an appreciable performance increase for the cost.
You will be makimg a pretty big performance leap though by moving to Ryzen. Right now memory prices are the big thing holding back builds. Hopefully prices go back down soon
ac3e37 No.14089204
>>14086655
>when you just bought into the Ryzen meme two days ago
1e609a No.14089210
>>14089204
>X meme
Stop posting like a redditor
f3a39a No.14089219
>>14086647
You betcha and it gets hilarious. I can't remember if it was last year or the year before, but other nations had been using the backdoors to spy on Government computers.
They apparently got a lot of stuff before being caught and the government flipped a lid over how companies had these NSA backdoors that other nations were using against the US.
b855d6 No.14089232
>Unplug computer from the internet
>Doesn't matter, update still makes it slower
BRAVO
R
A
V
O
ARE YOU SAFE YET
YOU FUCKING IN TEL AVIV NIGGERS
t. AMD FAG
b77133 No.14089234
>>14086986
Stop playing vidya, go outside, make friends, get laid, raise a family. Save yourself.
991140 No.14089237
>>14089196
>You will be makimg a pretty big performance leap though by moving to Ryzen
600-650 USD worth of performance leap? That is the opp. cost, even after RAM prices drop. I'll just overclock a bit more till I can't run the majority of games on ultra at 60fps.
b93480 No.14089238
>every Intel processor made within the past decade has a severe design flaw that could grant hackers complete control over your computer no matter what OS you're using, and the fix rolling out soon costs 5-30% of performance.
good thing I'm still using my pentium then
2242a2 No.14089250
>>14089130
>oy veeeeeeeeeeeeey
>>14089148
>Oy vaaaaaaaaaaaaaaaaay
Top kek lad.
e9f9fc No.14089274
>>14089238
This bug likely also effects Pentium 4s as well
>>14089250
Seek mental help
fd44aa No.14089275
>>14089165
>(not sure how Ryzen plays with DDR3)
It doesn't.
As for price fixing accusations, the main issue seems to be there are only a few manufacturers of RAM and they are in the process of transitioning to a smaller manufacturing process, which hurts yields. iToys, game consoles, and GPUs are gobbling up RAM these days, so you have low supply and high demand, meaning high prices. Also one of the manufacturing plants had some kind of accident in the summer that had them shutting down production fir a while so they could fix it.
85e392 No.14089279
>>14086986
>amd/intel compromise your privacy on a device you use for everything
>consoles are still worse!
>amd/intel spying on everything you do, providing backdoor access for the US government and israeli government and whoever else they want
>consoles are still worse!
>the device you file taxes on, manage order shipments, reach out to family, friends, and others
>consoles are still worse
m8 you might be a bit brainwashed. sony/microsoft/nintendo no doubt spy on you by collecting statistics but what information are you putting on those systems? Hours played in a game? Better that than the whereabouts of your fucking extended family.
e9f9fc No.14089284
>>14089275
A lot of embedded applications like iToys as well as game consoles still use DDR3 so that doesn't really explain it. Even GDDR5 used for GPUs is based on DDR3. The only thing I can imagine is Smartphone but AFAIK only the top-end flagships use DDR4 while the low-end is taking forever to play catchup
e70ea0 No.14089286
>>14089144
Actually, pirating a PS3 costs far less than pirating on pc at the moment. Any random fat model can be softmodded, so 50 bucks and you're set. How many pc components can you buy for 50 bucks, an internal hd maybe?
fd44aa No.14089290
>>14089284
iPhone 8 and X both use DDR4, their production gobbled up a ton of the supply.
e9f9fc No.14089291
>>14089279
You're right anon, consoles are better, I can't imagine why people don't file their taxes on game consoles
12f6de No.14089309
>>14089210
Stop making posts with the redditor meme.
2242a2 No.14089312
>>14089274
Take your own advice lmao.
b93480 No.14089313
>>14089274
is this a pentium 4
why did I even come into this thread when a 10 year old child is more proficient with computers than I am
12f6de No.14089318
>>14089313
Your cpu is affected by the bug.
And it's recent, should be 2 or 3 years old.
700a58 No.14089322
>have i5 3570k
I'm in the clear I think
a2874f No.14089327
>>14089322
Why would you think that?
d4a82c No.14089328
>>14089313
Look at the launch date.
5ba0f0 No.14089329
>>14089327
Because some CPUs are not affected.
4b2e23 No.14089334
>use AMD almost always
>decide to try out intel just to see the difference
>this happens
1e609a No.14089395
>>14089286
>Pirating on outdated locked down hardware is cheaper than modern hardware
>You can't build a PC for 50 bucks
>Second hand PS3 for 50 bucks
You can't even find a PS2 with the gamepads for that price
You're full of shit
>>14089309
>t.Reddit
a623f3 No.14089454
>>14089334
>switching to Intel right when AMD got their shit back together
429b91 No.14089493
>>14089454
>was planning on getting an AMD rig because it's cheaper next year after almost a decade of not upgrading
This fiasco and the fact that jewvidia uses new drivers to gimp the performance of their older cards cements it.
Although I really wish we'd get a non-jewish third GPU/CPU manufacturer in here.
f3fc32 No.14089503
>>14087432
Regret what? Greener grass on the other side of the fence?
>>14089493
>big biz
>non jewish
You know full well that won't happen until after the final solution comes about.
f3fc32 No.14089513
>>14089454
>applying ice to the case
>to the cold side of the case
>not to the radiators
I could forgive noodle limbs and a missing pair of ice crystals, but that's just fucking retarded.
2930ac No.14089542
>>14087105
>20$ per intel processor
in the last 4 years they have sold 400 million processors
= 8 000 000 000$
i dont think they will "laugh all the way"
e51a1d No.14089570
>>14089237
if you can run the majority of games on ultra with your current build, then the performance leap is not worth it. I assume you know what you are doing with overclocking.
662cbe No.14089589
>>14089542
More important than that is that some major players are gonna be very pissed off at a fault like this, and the necessary performance reductions to fix it. There could be payouts for them. At the very least less will want to consider Intel in future.
662cbe No.14089596
>>14089589
Never mind, found it.
e70ea0 No.14089669
>>14089395
>Pirating on outdated locked down hardware is cheaper than modern hardware
Notice how he purposedly mixes up an obviety with the lie is trying to pass: that PS3 is outdated (obviously) and that it is more pricey to pirate than PC (false, since good shit doesn't run on toasters).
>You can build a PC for 50 bucks
>Not get a second hand PS3 for 50 bucks
Laughing track 01.flac
dcf725 No.14089745
>>14086980
>FreeBSD
>developer code of conduct
<master race
f6182d No.14089824
>>14089669
>you can get a second-hand PS3 for $50
By any chance, are you from Israel?
2812cd No.14089833
>>14089824
>bid
>bid
>partially working
>$40-$75 shipping
Are you by any chance retarded?
ce9674 No.14089858
>>14089824
>all that dust on the 2nd one
005a79 No.14089870
>>14089111
Having spend a good two hours reading about this last night and reading benchmarks with theses new changes added to Linux, stuff like encoding/gaming/general day to day shit you and I do as users is literally uneffected by this change
So no, gaming etc is uneffected
However like I pointed out yesterday shit like access to files etc is massively effected, meaning intel systems running servers etc will see a massive decrease in performance
5582ff No.14090021
>>14089313
>>14089328
you'll probably be seeing an OS-level fix coming down the pipes soon. no reason to throw it out.
197e32 No.14090066
>>14087850
1600x with 3200Mhz 14-14-14-34 FlareX motherfucker. That Infinity Fabric is running hot as a motherfucker B)
ccb07f No.14090079
Long story short, in Layman terms, tell a complete PC retard, should I just update my Windows 7 up to date? I haven't installed jack shit since they began pushing that Windows version as an update.
ce9674 No.14090153
>>14090079
I don't think they've patched it yet. I think it's supposed to be tuesday. And I'm pretty sure the updates that are listed as critical are safe, it's only the optional updates that have telemetry and forced upgrades.
197e32 No.14090307
>>14090079
I'd say don't do it yet, and manually install only the specific patch that it's included in.
Also just bought AMD stock to hopefully ride this wave B)
64900d No.14090486
>>14089238
Is the fix an automatic update?
48e260 No.14090603
HAHAHAHAHAHAHA
THE KIKES KNOW NO BOUNDS
https://archive.is/kFte4
83f138 No.14090729
>>14086655
Linus rant when?
83f138 No.14090734
>>14086967
>>The code they submited to the Linux kernel was purposefully obfusicated to hide what it was patching
>Defend this GPLfags.
Don't want crackers to notice it and exploit it?
429b91 No.14090736
>>14090603
There is no oven with enough fire in the world for these creatures.
Only God's holy wrath will do.
48e260 No.14090814
>>14090736
I'm just wondering if Windows will go the same route. Probably, considering that if Intel is paying (((free and open source))) software developers to hide the fact that their CPUs are now far worse than AMD's, they'll obviously pay MS too. At least with Linux you can just fix the kike code and compile the kernel yourself, but Windows users are probably up shit creek without a paddle.
429b91 No.14090891
>>14090814
I don't think Intel has enough clout to bribe MS. MS is on a completely different level to the point that petty hardware squabbles do not concern them.
They'll patch what they have to, but they'll do Intel no favours unless it turns out that they're too incompetent to make the patch Intel-only which would not surprise me in the least.
Wangblows is shit, but at least I can choose not to update it.
48e260 No.14090906
>>14090891
>unless it turns out that they're too incompetent to make the patch Intel-only which would not surprise me in the least.
Me neither.
d1e08d No.14090930
>>14090891
>MS is on a completely different level to the point that petty hardware squabbles do not concern them.
Nigger, Microsoft is so dependent on x86 that they'd probably do it for free just to stay in Intel's good graces.
2812cd No.14090988
>>14090930
They could also use it as leverage against intel regarding their attempts at making x86 emulation on ARM more of a thing.
3127e5 No.14091436
>>14086612
>>14090603
>'LOL ENJOY YOUR 30% PERFORMANCE HIT INTELNIGGERS'
>turns out AMD is also affected
Could have avoided this by acknowledging that it's not 30% in all cases and mostly affects servers. Now "le 30%" meme applies to AMD as well. Guess you should have fucked off with your disinformation and exaggerations because now you've just convinced every naive tard of a 30% performance hit which hits AMD too.
d1e08d No.14091470
>>14091436
AMD isn't affected, the kikes just applied the performance-draining fix to all x86 processors and ignored a patch which limited it to affected Intel processors.
48e260 No.14091479
>>14091436
Fuck off with your brand war faggotry, intel kike.
da6140 No.14091550
>>14091436
AMD confirmed they are not affected, fuck off shill
2812cd No.14091590
bb13bc No.14091595
Can someone give me a short rundown of the situation?
Does it effect only x86(32-bit) CPUs or what?
What the fuck am I suppose to even update? A Windows update? Does it even exist for W7?
I am on a Ryzen chip, but my family is not, so I will have to eventually fix their shit.
48e260 No.14091606
>>14091595
>Does it effect only x86(32-bit) CPUs or what?
It AFFECTS (learn the difference between verbs and nouns, faggot) both 32-bit and 64-bit Intel CPUs.
>What the fuck am I suppose to even update? A Windows update? Does it even exist for W7?
It will most likely be pushed out next tuesday. Windows 7 still receives updates until 2020.
dd0573 No.14091815
>>14091015
>grsecurity
>talking shit about kikes
>when they stopped providing community security patches for the kernel
Really gets the noggin joggin.
3127e5 No.14091842
>>14091470
The patch was applied to AMD. Whether it was mistaken or not is irrelevant as AMD was affected.
2812cd No.14092016
Windows fix benches are up
http://archive.is/ml8aK
df219e No.14092064
>>14092016
Wow, it's fucking nothing.
1bd551 No.14092123
>they don't upgrade every 6 months
This 30% you cucks lost I'll earn back soon.
2812cd No.14092241
And would you look at that we're all fucked anyways
http://archive.is/Jj6lf
tl;dr
<other exploit that achieves the same thing as the intel specific one
<affects everyone
Yes that means you too ryzenfags / ARMtards
<impossible to fix without new redisigned hardware
>harder to exploit (no indication of what that means in practical terms though)
d1e08d No.14092258
>>14092241
There's two separate bugs: Meltdown and Spectre. Meltdown only affects Intel hardware and is what these patches for, while Spectre affects basically all modern processors and will be much harder to fix (while also being harder to exploit).
5582ff No.14092364
>>14092241
>trusting NYTimes
>trusting a woman when it comes to cybersecurity
a2874f No.14092793
>>14092258
There's two variants of Spectre. The less malicious one affects everyone, even ARM, the more malicious one only affects Ryzen if you use a non-default kernel configuration.
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
So far it's looking like the impact of the Meltdown patch on gaming is pretty minimal. Nobody's tested multiplayer games or the more Denuvoed titles yet though.
314fe4 No.14092870
>>14092793
>Nobody's tested multiplayer games or the more Denuvoed titles yet though.
Computerbase.de has tested AC:O (Denuvo 4.7+Vmprotect) the impact is present but under 5%.
a2874f No.14092984
>>14092870
That's disappointing, I was hoping for Armageddon.
5582ff No.14093068
>>14092984
It's not much in the gaming sphere but you can expect Intel's stock prices to start shitting the bed which will probably affect us in the long run
ea9a20 No.14093414
>>14087561
Shit, that's why the GG and other high post threads crash my toaster?
1e609a No.14093497
Hooktube embed. Click on thumbnail to play.
>>14089669
>Good shit doesn't run on toasters
Every old game ever
>You can't get old parts and make a computer for 50 bucks
Embed related
>You can get a PS3 for 50 bucks
Where?
I wouldn't even make that price for a friend since I could just sell it for 100-150 to collectable faggots.
Get fucked
1aedb8 No.14097600
What kind of new opportunities for gaming consoles hacking could this provide?
d1e08d No.14097629
>>14097600
Meltdown gives almost none, while Spectre could be a fucking goldmine.