9th gen i7 WONT be hyperthreaded

Removing features is still very kikeish

k

How do you feel about that?

This is what happens if you hire poo.

Attached: 1531831307453.jpg (262x192, 10K)

lol no

>threads matter more than architecture
Sage

You’re acting like it’s possible this WON’T happen

only problem is its the 4th skylake rebrand featuring more housefires

in general architecture > clock > cores > threads for single-threaded workloads i.e. most of the shit you do

Attached: 1528445007753.jpg (691x771, 112K)

>single-thread perf is what matters
>so what if they dont have hyperthreading anymore just buy an 8 core anyway goy

Attached: YourAsian_xyPKOMBg2REgngF0KdsN.jpg (625x781, 70K)

Is this the power of 14nm+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++?

Attached: 1521984254812.jpg (777x704, 57K)

>single-thread perf is what matters
>We'll just bribe software developers to stop using more than one thread.

Attached: 1531079876768.jpg (751x720, 54K)

this is not really news, since the last couple of hardware exploits ייntel cant fix that without a complete redesign of the arch. so avoiding hyperthreading on the next lake rebrands was a given

i wonder what investors are going to think about this once they notice that the new rebranded cpus cant keep up with the old ones

Attached: 1515023946073.jpg (638x599, 123K)

What I hyper threading?
Is it like having a CPU core pretend it is actually two cores?
How the hell does it make it faster in some tasks?

>i13 comes with a black crucifix heat spreader glued upside down

>t. AMD shill who has no idea how CPUs work

AHAHAHAHAHAHAHAHAHAHAHAHAHAHAAHAHAHHAHAAHAHHHAHAHAH

Attached: 1532328170498.jpg (250x247, 12K)

That's unironically true. I've been working on some demanding simulations using HFSS and they actually recommended turning off hyperthreading because of performance.

actually it's theoretically the same. Use the GMP formula without clock speeds:
2C + T
6c/12t = 24.0 GMP
8c/8t = 24.0 GMP

So yeah... clock by clock it will deliver about the same multicore performance as the 8700K. Since it boosts a little bit higher, it will be a little bit faster.

The 2700X will still destroy it on multicore tho...

Here

>Multi-core doesn't matter!
>Productivity doesn't matter!
>Price/performance doesn't matter!
>Performance per watt doesn't matter!
>Power usage doesn't matter!
>Temperatures don't matter!
>Soldered dies don't matter!
>Stutters don't matter!
>Streaming doesn't matter!
>Data centers don't matter!
>Locked CPUs don't matter!
>OEMs don't matter!
>Hyperscalers don't matter!
>Upgradeability doesn't matter!
>Anti-competitive business practices don't matter!
>Locked platform features don't matter!
>Synthetic loads don't matter!
>PCI-e lanes don't matter!
>Burnt pins don't matter!
>ECC support doesn't matter!
>*NEW* 7nm doesn't matter!
>*NEW* StoreMI doesn't matter!
>*NEW* Security doesn't matter!
>*NEW* Stock coolers don't matter!
>*NEW* Games don't always matter!
>*NEW* Hyperthreading doesn't matter!

Attached: 1502432194480.png (350x314, 6K)

hi brian, how are you holding up? is your wife still with you after getting fired over that extra-conjugal deal inside the corporate premises?

You forgot
>Backdoors don’t matter!

and with that intel concedes muticore performance to amd. sure would be a shame if ryzen 7nm beat intel in single core performance too.

Underrated

Hyperthreading is a hardware feature. It detects when a CPU core is idle, and uses it to process another instruction. It basically makes the CPU more efficient.

Just follow the nvidia method and push driver and firmware 'security' updates to gimp the older processors. Bingo! happy investors

nvidia also tries to block vt-d strangely on their consumer tier cards so people had to spoof it's ID as a AMD card

Just like Intel making you pay for a fucking Raid key on their coffeelake + motherboards

Fucking tired of intel and Nvidia kiking me around.

and cucked consumers

Nope.
8c/8t < 6c/12 < 4c/16t

POWER9, fuck yeah!

Basically in a CPU, there's some parts that prepare calculations/operations, some parts that do the calculations, and some parts that handle the results of the calculations. Hyperthreading doubles down on the preparation and result handling while sharing the actual core (that does the calculations). To the operating system, this appears as two cores, but sometimes operations will have to wait for the core to be available.

As a rough rule of thumb, hyperthreading adds about 50% more performance when doing something involving two threads.

>tfw papers on new side channel attacks will continue to surface until processor manufacturers are forced to remove SMT, speculative execution, multi-cores, caches, pipelining, everything architectural advancement since the 8086

I'm more interested in the 4.7ghz all 8 cores. These things are going to be benchmark monsters and when put under home cooling will never reach that. I bet there's not a single review that has less than a 280mm rad despite being only 95w parts for both the i7 and i9.

No just intel

poo in the joo

Both AMD and Intel have backdoors with PSP/ASP and ME.
Not to mention a bunch of other blobs, of varying replace-ability
coreboot.org/Binary_situation

25%

>14nm++
OH NO NO NO

Do they have a specific reason that's being defending blindly or is it genuinely just for jewing

>Meltdown doesn’t matter!

Fuck off back to Jow Forums.

Did you even bother to read the link?

Fuck off back to cripplechan

I asked why its being defending
lmao shut up pussy

They give you the same chip different socket. In a way I admire these kikes. The purity of their greed is a wonder of the modern world.

Attached: Holy+i+didnt+have+that+one++_27f4ea5616d2dbe9b98e1e04f0cfb675.png (500x500, 9K)

8c/16t > 8c/8t

The most greedy deceiving manipulating corrupt semiconductor company to ever exist on this planet.

Not necessarily

A 9700k with hyperthreading would beat one without it, that's the point I was trying to make.

Attached: 1523063695818.jpg (700x700, 58K)

HT is part of architecture. The fact they're including it in their high end chips shows it is something they believe people will miss when they remove it from i7s.

Attached: 1531946551286.png (426x510, 154K)

>95 watts
I smell bullshit

Attached: 1527824489442.webm (1280x720, 1.01M)

Intel? More like Incel OH NO NO NO NO

Core i9 9900k is 8 core/16tr with a 95w tdp and 4.7ghz boost on all cores. For less than 500 dollars. It will rape amd.

And I guarantee the i5 and i7 without hyperthreading will still rape amd. Can't wait to post benchmarks and laugh at you soon.

hyperthreading doesn't matter!

Attached: 1502213714326.png (575x641, 13K)

they finally got to 10nm

I-intel will w-w-win

Attached: 1489421658644.png (653x726, 84K)

Yep, and they are winning now. Go look at the market share and stock value.

No they didn't

how else do you explain 95W TDP then :^)

>wccftech.com
I recommend the use of the Options field.

That was the joke dingbat

INTEL IS FUCKING FINALLY FINISHED
(((THEY))) WILL NEVER RECOVER
INTELSHILLFAGS BURIED AND DEAD

jews did i9/i11

Why fuck they don't remove that fucker igpu it's not like anyone buying CPUs need integrated shit.
I know Intel will bin i3 crap from it but come-on, even amd ditched shit igpu

HFSS is extremely optimized and tight code written by real programmers in real programming languages. Ask yourself how often is software optimized to this degree in the wild. You have shit programmers churning out shit code in languages that don't even compile a rule now days.

HT earns its keep when a thread stalls or the CPU can't figure out a way to run multiple instructions at once. I did a comparison on a HFSS sim I ran and got a 2% reduction in run time. Not worth it because letting those fake cores stay free increases responsiveness of the machine while the sim is in progress. I get about 15% reduction in compile times. Stuff like python probably gets a better boost than this.

Meanwhile we are living in a world where 2-5% performance per generation is normal. You are throwing out 3-5 generations of performance by cutting out a feature that doesn't cost a lot of die space and has already been figured out.

Attached: 1532025054280.png (500x400, 34K)

Hyperthreading doesn't matter

Oy vey you tell them Chaim!

Wow its fucking nothing.

The i9 is the new mainstream flagship aka i7, the i7 the gaymer i5 without hyperthreading and the i5 is the budget i3. The i3 is removed from the lineup. In other words everything is exactly the same as before. Sage for AMDrones currytech clickbait.

>threadlet seething

if the i9 is$450 does it even matter the i7 doesn't have hyperthreading

>clocklet poozen coping

Can you quit shilling your crappy faux metric please?

holy shit!

Guess i'll keep using my 3570k for another couple years...

>architecture > clock > cores > threads
Clocks trump everything.

t. netburst engineer

They won't even fix it with this generation, will they? How are they getting away with this?

Everything else being equal, duh.

Its just the naming scheme that changed to one up AMDs lineup because AMD copied their naming scheme with ryzen and the motherboard chipsets.

fake and gay. none of those turbo speeds are even correct

The post you responded to was pointing out that architecture (i.e. IPC) matters first and that then clocks are king, so I believe you just made his point.

No, improvements in clock generally respond better in every workload given the same architecture, whereas improvements in architecture respond less well across different workloads given the same frequency.

I had a hard time (possible mental breakdown) when I found my B960 had no hyperthreading. That's why 2 years ago I went team AMD.

YOU NEVER NEEDED IT ANYWAY GOY
4 CORES IS ALL YOUUUUU NEEEED

Attached: 1447161347847.jpg (1846x1923, 607K)

And let me emphasize doubly how true this is these days, since architectures have reached a point where they tend to extract almost all ILP that is available in a program at all. We can't really hope for much further improvements in architecture without leaving von-Neumann architectures altogether. Whereas improvements in frequency scale very well.

>currytech

Attached: 1526078534530.gif (500x493, 631K)

Who /2500k/ here?
We still king

>What is marketing
If this is anything like the 7980xe ordeal (and considering Intel's track record,it is), the chances are it's either incorrect info or it's 100 Extra watts of TDP for every 1000th of a volt they add to vcore. No offense but the leaks we have put Intel's 10 nm at 2020 or very,VERY late 2019.

When are they going to release a CPU without MELTDOWN? It's supposedly been a year since the MELTDOWN fault was shared under embargo, although it's likely that Intel knew about it longer than that and designed it that way for performance.

Intel literally had nothing in the pipeline, certainly no response to Ryzen.