Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Maintenance is complete! We got more disk space.
Become a Patron!

/g/ - Technology


View post   

[ Toggle deleted replies ]
File: 73 KB, 700x565, 1558545774096.jpg [View same] [iqdb] [saucenao] [google] [report]
71376569 No.71376569 [Reply] [Original] [archived.moe] [rbt]

https://youtu.be/vzV24ZqrJ4s?t=394

>Oh and AMD also said that they ran all of their Intel system tests that they shared today without software and firmware mitigations in place for these security flaws, which gives Intel a much better shot at 'Best-Case Scenario' performance.

It wasn't even their final form. They're hiding their power level.

>> No.71376602

>>71376569
holy fuck it's like Lisa has constructed every part of this to destroy Intel

>> No.71376611

>>71376569
I mean it barely matters when 16 cores is $800 from amd but $2000 for 18 cores from intel, not counting the other costs

>> No.71376641
File: 1.51 MB, 1002x1198, 1558925788675.png [View same] [iqdb] [saucenao] [google] [report]
71376641

ALSO, they didn't use the May Windows update, which also would give a boost to AMD.

>> No.71376752

>>71376641
>>71376569
Sandbagging or just too lazy to redo the benchmarks?

>> No.71376765

>>71376641
>May Windows update
WTF is that supposed to mean?

>> No.71376778

>>71376752
If you know Lisa Su, you know she's doing this to btfo Intel on launch day.

I wonder what memory speeds they ran the benches on too

>> No.71376840

>>71376765
The May Windows update gives AMD an additional ~15% boost in performance.
We're potentially looking at even crazier performance numbers from AMD once it's tested with updated systems.

>> No.71376847

>>71376569
NOOOOOOOOOOOOOOOO GOYIM

>> No.71376855
File: 383 KB, 1773x998, 1551133938209.png [View same] [iqdb] [saucenao] [google] [report]
71376855

>>71376569
OHNONONONO

>> No.71376867

>>71376840
15% even on single thread stuff?

>> No.71376887

>>71376855
I don't remember seeing that slide in the conference. It could be numbers for the APUS.

>> No.71376897

>>71376840
>The May Windows update gives AMD an additional ~15% boost in performance.
How the f-...

>> No.71376911
File: 57 KB, 329x329, 1560208885772.jpg [View same] [iqdb] [saucenao] [google] [report]
71376911

>>71376897
>>71376855
>>71376569
I... I don't feel to good. It's... over.

>> No.71376923

>>71376867
>>71376897
Only multithreaded stuff, it's scheduler is ccx aware now.

>> No.71376932

>>71376855
will this benefit rpcs3? it was already manually limiting itself to the second CCX only.

obviously crossing CCX would probably still be off the table for old gens.

>> No.71376933

>>71376897
I don't know why exactly, but from my understanding the current Windows 10 build isn't optimised for AMD's CCX design because Intel co-delevoped parts of it years ago to work well on their Xeons.
The may update will rework scheduling to give AMD an advantage.

>> No.71376961

>>71376569
But they didn't need to, because the performance fuckups were Intel exclusive?

>> No.71376964

>>71376933
The May update is already out though as 1903?

>> No.71376978
File: 65 KB, 1200x661, ryzen_3000_windows_optimizations-100798898-large.jpg [View same] [iqdb] [saucenao] [google] [report]
71376978

https://www.pcworld.com/article/3401086/amds-ryzen-9-3950x-is-a-16-core-cpu-aiming-to-topple-intels-gaming-dominance.html

This article talks about it a bit.
This was interesting.

>With previous builds of Windows, AMD said it could take around 30 milliseconds for the CPU to ramp up to higher frequencies.
>As of the update (and with a new chipset driver) it’ll take just 1 to 2 milliseconds for the chip to reach its top speed.
>These fixes give some games a boost of 15 percent, while the faster clock ramping can yield 6-percent improvements.

That's one hell of an improvement in how quickly the clocks ramp up after that upgrade. That's like ~20x faster.

>> No.71377003

>>71376569
>>71376641
Did Intel decline her job application in the past or something? This shit is some sadistic as fuck humiliation that has to be personal.

>> No.71377008
File: 257 KB, 1280x720, 1531001043667.jpg [View same] [iqdb] [saucenao] [google] [report]
71377008

Mommy is too powerful, she must be stopped

>> No.71377076

>>71376569

Intel have kinda shot themselves in the foot. And the head.

>> No.71377115
File: 54 KB, 293x399, 1287252235931.jpg [View same] [iqdb] [saucenao] [google] [report]
71377115

big, swinging dick move

>> No.71377119
File: 10 KB, 251x242, 1440258438184.jpg [View same] [iqdb] [saucenao] [google] [report]
71377119

Meanwhile:

>Intel: Come challenge us in real world gaming
>Lisa: Okay.

https://www.pcgamesn.com/intel/worlds-best-gaming-processor-challenge-amd-ryzen-3000

>> No.71377194
File: 428 KB, 1442x810, 1541107623591.jpg [View same] [iqdb] [saucenao] [google] [report]
71377194

How can one company be this BASED?!?!!

>> No.71377230

>>71376933
>The may update will
it's been out for a while now

>> No.71377247

>>71377003
Who knows what's behind it.
I'd wager Intel folks have been giving her shit every time they've seen AMD people in press conferences and other occasions. That must have gotten a bit annoying after a decade of hearing their jeers.
AMD finally got fed up and decided to destroy the fuck out of them out of spite. It's going to get even worse when Threadrippers come out and completely obliterate the competition in the HEDT field.

>> No.71377274
File: 320 KB, 2560x1440, 1553020194709.png [View same] [iqdb] [saucenao] [google] [report]
71377274

>le epic edgy meme from pol

this is what you trying to hard AMD shills look like right now

>> No.71377285

>>71377274
>edgy
>/pol/
Zoomers, please.

>> No.71377295

>>71377274
>what you trying to hard AMD shills
can you reformulate this in english you brown tranny?

>> No.71377297

>>71377274
>get rekt
>cry /pol/
Every god damned time.

>> No.71377307

>>71377247
Feels like AMD is that pathetic fat kid everyone made fun off and then it learned martial arts over the summer.

>> No.71377327
File: 378 KB, 816x1356, pol.jpg [View same] [iqdb] [saucenao] [google] [report]
71377327

>>71377274
why not go to another website you schizo?
you literally have all social media to yourself
are you a masochist

>> No.71377378

>>71377285
>>71377295
>>71377297
>>71377327
this rapid response cause Im right. step the fuck up shills.

>> No.71377407

>>71377378
>shit up thread with schizo tranny nigger shit
>ahaha people getting pissed means I'm RIGHT(in broken english)

>> No.71377430
File: 2.00 MB, 250x158, 1316647238711.gif [View same] [iqdb] [saucenao] [google] [report]
71377430

>mfw I almost bought a 9900k a few months back

>> No.71377474

>>71377407
>t. thinks its broken english because he is ESL

>> No.71377500

>>71377003
Intel has done some major scumbag shit against AMD in the past. Good on her for getting payback.

>> No.71377510

>>71377474
>t. is a nigger who thinks ebonics is correct english

>> No.71377760

holy fuck intel is dead

>> No.71378133

>>71377760
And that's a good thing.

>> No.71378222

>>71376933
It will not give AMD an advantage but will make sure scheduling with AMD actually works properly now. I mean the end result is that it does improve performance with AMD but it will not affect intel performance negatively.

>> No.71378420

>>71377274
Congrats on the (You)s you fucking attention whore

>> No.71378580

>>71376569
Based Mommy Su

>> No.71378700

>>71376569
it wouldn't surprise me to hear they purposely used slow RAM with the Ryzen to sandbag further. They as a company know that their benchmarks, regardless of their testing methodology will be seen as biased. So they're gonna resort to waiting on 3rd party benches who WILL be testing with proper 3200MHz RAM. These reviews will speak for themselves.

>> No.71378736

>>71378700
Third parties will also most likely have Intel mitigations applied, this is gonna be interesting

>> No.71378741

>>71378700
Does the sandbagging theory really hold up? Have they done this in the past?

>> No.71378833

>>71378736
that's very true. All the handi-caps that AMD may have offered Intel in their E3 showing will be missing.
>>71378741
AMD hasn't been in this position of owning Intel this hard since their Original Athlon 64 days. Now that AMD's Ryzen architecture is in full swing with proper optimizations on the OS level as well (Windows 1903 update), everything is swinging in their favor. I wouldn't be surprised if AMD is sandbagging, at least a little bit, to REALLY twist the fucking knife and break the blade off in Intel for all the years of AMD being dominated.

This was a long time coming and with Su as the helm, they're a force to be reckoned with.

>> No.71379044

>>71376752
Dude they sandbagged Zen performance numbers for like 2 years before first gen Ryzen to lull Intel into a false sense of security

>> No.71379183

>>71376569
ARE THERE ANY REAL WORLD BENCHMARKS OUT YET?! DON'T GIVE ME FUCKING RELATIVE FPS! I WANT 99%s, 1%s and 0.1%s IN DEMANDING AAA TITLES, NOT SYNTHETIC BENCHMARKS!

>> No.71379297

>>71379183
Wait for the embargo to lift on 7/7. Pre-order now though.

>> No.71379299

why can't women be at least 5% as logical, focused, modest, calm and based as mommy su?

>> No.71379400

>>71377119
kek

>> No.71379422

>>71376932
as far as I know rpcs3 just implemented their own thread scheduler because the windows one sucked, so I don't think this will help

>> No.71379546

1903-based LTSD build when?

>> No.71379590

>>71379546
there will be no 1903 based build
LTSC has a 3 year release cycle

>> No.71379931

>>71379590
https://en.wikipedia.org/wiki/Windows_10_editions
>There are currently 3 releases of LTSC: one in 2015 (version 1507), one in 2016 (version 1607) and one in 2018 (version 1809).
So we're long for a new one, then?

>> No.71380011

>>71379931
yep

>> No.71380670

>>71377008
Why contain her?

>> No.71381014

>>71376569
Yep, my predictions were right, if 2950X was already capable of boosting to 4.4GHz, I expected Zen2 to reach 4.7GHz on high end chips.

>> No.71382007
File: 24 KB, 1712x284, 1547405897008.png [View same] [iqdb] [saucenao] [google] [report]
71382007

>>71377194

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action