PLOT TWIST:

I don't think so. If I have a dual core CPU with 4 threads, then what the fuck is going on there? 1 core with 4 threads or what? This theory doesn't make a whole lot of sense to me. Maybe Intel's general design is just fucking garbage and depends on hyperthreading to work somehow. x86 is a trashfire ISA, but that's only what's exposed to the user. The actual cores are a lot different internally. I've heard they're pretty much a RISC type design. Not even god knows what the fuck happens in there or how it really works.

>being retarded the post
same as above. what the fuck are you talking about? if you have a dual core with 4 threads, you still have a dual core with 4 threads, but with huge performance loss because it's intel. op is specifically talking about the 8400, a 6 core cpu. the joke here is the idea that it's actually a 3-core 6 threads and loses all those performances because the hyperthreading is pozzed.

Attached: 9nohQh4.jpg (478x361, 14K)

>hyperthreads have hyperthreading
That is like hyper-ception. OP is on to something.

>ahah yes, intel is btfo, but the screenshot on the older cpu to make a comparison with shit hardware is on an older version of the software, so your point it's completely invalid even if both the tests on 8400 are on the same release and clearly show a >50% performance loss

Attached: images.jpg (293x172, 16K)

>(4-1)*2=12
>unironically being this dumb

Not shilling for Intlel, nigger, get the fuck out of here with your autistic ramblings.
Off yourself.

>Benchmark version 17.01.64

Attached: 4L_iTH4RT0X.jpg (550x543, 83K)

I got curious and did some quick fuckery with my CPU.
This performance disparity took me less than a minute to do.
Imagine what one could accomplish if they put in any amount of effort.
What did I change?

Attached: nigger.png (803x400, 22K)

version 1.85 vs 1.88 was within the margin of error for me.

Attached: versiontest.png (813x404, 23K)

you disabled boost?

OP might not be a faggot for once.

Anyone got a FPGA and logic analyzer they can hook the CPU up to?

>9MB cache
>divisible by 3
>intel quads usually have 12MB of cache
oh shit

nice outrage intlet

Attached: 186-1869318_67059574-pink-wojak-clipart.jpg (880x815, 252K)

ples sir, do the needful: release the false-flag shill right now

Attached: images (1).png (216x233, 13K)

is this real?
>There are 4 actual cores.
>Then there's a fifth one that doesn't work, but still physically present.
>There's HT secretly enabled for that 5-th "unused core".
>Diagnostic software and benchmarks see this POS as a "6 core".
>It's a literal scam.

Attached: 56d.gif (1189x1189, 2.11M)

IT ALL MAKES SENSE NOW WHAT THE FUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUCK

Attached: wojak-had-enough-of-intels-bullshit.jpg (512x599, 131K)

Which quads? 7600 had 6MB

why do the 8400 and 9400 outperform the 7700k though?

8400 has 9 (3+3+3)
7600 has 6 (3+3)
9400 has 9 (3+3+3)
6600 has 6 (3+3)
THE WHOLE 14NM SEGMENTATION ACTUALLY HAS 3MB PER CORE
HOLY FUCKING SHIT IT'S REAL
MAKE IT STOP
AAAAAAAAAAAAAAAAAAAAAAAAAA

Attached: 155149961993.gif (220x219, 38K)

Ah, so you mean you can divide by two all CPU without HT since Skylake (or Broadwell)?
Ok, do modern motherboards still have function of disabling cores? Can we compare 4C4T i5s with 2c4t i3s? Compare 2c4t i3s with 2c2t Pentiums (or these are single core? hmmmm)

What's so hard to believe about each core having 1.5MB in that situation?
>Intlel

DELID THIS ANTISEMITIC THREAD NOW!!!

Attached: 1515175316605.jpg (691x771, 64K)

Has any tech site done some testing about this?

Wouldn't be surprised if Intel pulled some Jewish tricks on their CPU.

it's just /v/edditors being retarded again

>tfw use 2600k
>have to disable errything if i wand mah ganes

Well then a 2 core 4 thread i3 will perform the same as a 4 core i5.
Apply yourself.

cpu-z thred?

Attached: poasCPU-ZScoras.png (403x402, 14K)

no thanks
make a seperate thread for all the other insecure faggots

What clock?

Attached: 432.png (405x403, 14K)

>i7 2600K matches a Ryzen 1400
God, Ryzen is such a fucking joke.

seething FX fag, eh?

4.5 but have a lot of shit running so guess its closer performing to 4.3 - 4.2

>317$ processor barely matches a 169$ processor
God, Intel is such a fucking joke.

If you arent a poorfag you dont mind spending $500 on a processor every 1 or 2 years.

>spending 500$ on the same performance hardware every year or two
God, Intcels are such rubes

>make 200k a year
>Can't spend $500 a year to maintain a tool you use every day
Poorfag pls.

>unironically using AMD

Attached: (((((mascots))))).png (1600x1223, 1.1M)

>needlessly spending money on hardware that doesn't change year over year
What's it like to have 80 IQ?

Attached: 1545772870574.png (549x413, 90K)

incel inside

Attached: 1502737044765.png (882x624, 39K)

Don't need to be butthurt

OGRE.

Attached: INTURDBTFODED.png (391x117, 360K)

amazing considering it's another 16core part

the actual author of the "benchmarks" here. I've just capped the cpu to 70% of it's frequency via the windows power options
why are Jow Forums posters such fucking brainlets jesus christ

Attached: 1.png (992x935, 122K)

nice shop and try, kid.

>The put the computer to sleep region is CROPPED
BTFO'd, wasted trips

I think a lot o people missed the fact that the i5-8400 is a 6C/6T; not 6c/12,t it shouldn't even have HT to start with.
en.wikichip.org/wiki/intel/core_i5/i5-8400

...supposedly.

See .

Yes user, I understood the point of this thread. What I said makes it even funnier.

pic related was interasdin, with this in mind

Attached: 6823c9beb8.png (754x316, 39K)

there you go, AMIndia bros
valid.x86.fr/77pzs3
valid.x86.fr/626cby

You've said it's a 6/6. It's not.

here's mine! ^_^

Attached: ryzen central processing unit.png (806x804, 37K)

I'm not sure if you're doubling down on the joke ITT or what.
if what, please expound.
if the joke, kudos.

Attached: 7e850a8760.png (286x645, 31K)

>he's still squirming
Just let it go already, Intbecile. You work here was done quite a while ago and now you're not usable anymore.

So it says the OS protection isn't enabled, but how do I enable it...?

I installed the update.

Attached: 2019-05-22 12_08_21.png (1072x831, 56K)

>he thinks OP or I were joking

Attached: 56fh7r686.jpg (741x487, 53K)

so the 8400 is 4 real cores + 1 fake core and 4 real threads + 2 fake threads?

intredasting......

There's two """""(((updates)))""""". One is mere simple software patch via OS update. The second, and most important one, is HT-disabling microcode (((fix))) via BIOS. Only and ONLY when BOTH are applied, it's considered a "full mitigation". See here:

Attached: 1558209503053.png (600x389, 24K)

Exactly. It's a fucking Ponzi scheme. Remember 3.5GB? Yep.

My CPU has no hyper threading

I installed all the latest windows updates though and even the most recent BIOS which included a new CPU microcode update from less than 2 months ago.

Looks like for 9th gen, it's not a big deal.

Attached: 9700k MDS.jpg (787x823, 133K)

>he hasn't posted any actual proof...

well if you want to convince people... maybe some evidence would help

Only way to know for sure is to destroy one and look at the bare die.
youtu.be/PEoJtluFroA

If you can't count at least 6 clear die areas, then probably safe to say they're doing something funny.

How much is hyperthreading supposed to scale in CPU-Z's benchmark anyway?

This is my 4670k @4.2Ghz. If you scale this up with the multi ratio given in the OP, it would be roughly the same ballpark - and both CPUs are without hyperscaling.

Attached: Benchmark.jpg (406x404, 38K)

based on cache level, multiples of 3, and the halving of the score, it seems to me that the 8400 is actually a 3c6t with 3 fake cores to make it look like a 6c6t.

a test where you fill sequentially use 1 core after another with mitigations should easily show how many physical cores there are.
If once you move from 3 to 4 cores you see performance degradation it means some form of ht kicks in, if this occurs after moving from the 4th to the 5th core, it solidifies your 4+1 theory, if it goes steady until the 6th core then it's real.

there is smoke here.

There's a confirmed performance hit even if you already have HT disabled or don't have it at all. -8~10% on Inturd and even on the absolutely secure Zen it was confirmed to have an up to -3% so far.

Well just comparing CPU-Z scores doesn't reflect this at all, so not sure what you want me to say.

>How much is hyperthreading supposed to scale in CPU-Z
Linearly.

how would you be able to tell when a core is deactivated

>just comparing CPU-Z scores doesn't reflect this at all
It actually does and pretty clearly, too. What's most hilarious about this, is that ALL Inturd's garbage is affected ALL THE WAY BACK TO FUCKING BENTIUM 4. Just fucking L00K @ 'DIS ДOOD (without any "mitigations" and with "full"):

Attached: 1558468805279.png (822x413, 25K)

en.wikichip.org/w/images/thumb/c/cc/coffee_lake_die_(hexa_core).png/650px-coffee_lake_die_(hexa_core).png

That's what Judes try to make you to believe.

I just posted my 9700k before and after the patch you retard.

How the fuck can you tell me my CPU is affected when it obviously saw literally ~1.5% difference for single thread and ~2.2% difference on the multi-thread.

That's close enough to just be a difference in the idle CPU usage.

Attached: 56n3uh64395h6y45.png (500x396, 97K)

>the jews surely didn't glue fake dies on the cpu
>250% the price
>this bench
pathetic

Attached: this_kills_the_intlet.png (403x402, 13K)

ahem, 1,71%

i know, those dies do not look similar AT ALL.

Can't you into reading comprehension, Intbecile? Yet again, you ARE an Intbecile that unironically bought an 9xxx POS, so...no surprise there. Either way, go back and re-read the "8~10%" remark. It's dependant on IPC, too. The less IPC an Inturd CPoo has, the more it loses with """""(((full mitigations)))""""". You didn't lose a lot because, according to Jewtel themselves, "8-th generation and newer" models suffer less from this than older CPoos of theirs. BUT the FACT that you STILL lose it, even without HT (and with HT you'd literally be in a -25%-or-more ballpark), is what matters. At the very least 5%. And CPU-Z clearly shows it.

so if i never installed the microcode bios updates but I did upgrade to the latest win10 (1903) can I be fully free of jewish tricks or will I have a performance hit?

--> ~ ~

Attached: INTURD IS GARBAGE.jpg (1014x1680, 736K)

>At the very least 5%. And CPU-Z clearly shows it
Neither the single core nor multi-core results are above about 2-3% difference, so it isn't at the very least 5%, and CPU-Z CLEARLY doesn't show that.

Attached: test.png (403x402, 15K)

You can't count, eh?

My deepest condolences (not).

Werks fine for any gaming/workload I have in mind.

>315
>gaming

Attached: Intbecile.png (400x400, 1.22M)

If it's a legit hexacore you should see at least 6 identical structures. If they were selling hyperthreaded tricores as hexacores I wouldn't expect there to be 3 dead dies left over. If they only enabled hyperthreading on a some faulty dies and some were legit hexacores they would perform extremely inconsistently.

All we'd need to check is if there are 6 cores or less to prove our suspicions.

8400 is actually 2 cores 6 threads

brainlet detected

someone do this with a i5-8400 plis
youtube.com/watch?v=tSW8dZeE1yM

it is the needful