Intel's 9th Gen Core Gaming Benchmarks Flawed and Misleading

[lexluthermiester]

actually that never happen.. thank you fr playing

Actually it did, and those "leaks" got a lot of people excited about Ryzen.

 

[iO]

Click on the source link at the end of the article.

Or here directly to the PDF

 

[ShurikN]

Or here directly to the PDF

I forgot to mention this in my first post. The reason why TPU didn't post pictures of those shady results is because that would probably breach the NDA that they signed (otherwise they wouldn't care). Steve didn't sign an NDA (bought the review sample) therefore he can post whatever he can

 

[laszlo]

they get paid and they delivered the stuff as requested...

question is how many other reviewers do the same....

unfortunately uneducated buyer will buy everything no matter if is worthy, needed or not, nothing new under the sun

 

[JalleR]

I forgot to mention this in my first post. The reason why TPU didn't post pictures of those shady results is because that would probably breach the NDA that they signed (otherwise they wouldn't care). Steve didn't sign an NDA (bought the review sample) therefore he can post whatever he can

Oh... Yes that makes sense.

 

[Dave65]

About what id expect out of Intel!
Intel, you're dead to me

 

[R0H1T]

So Intel's worse than the cheating scums we all thought knew they are

 

[Robcostyle]

Well, If people pre-order something, it's entirely their fault. People should never pre-order something without first knowing all the facts.

There's thousands of them - I'm taling about people preordering stuff, willing to pay ahead, or pay for whatever made "wow" effect on'em, even without knowing what that thing is for. Hundreds of pumpkinheads braged about preordering 2080\(Ti).
And that makes their "faults" being our problems - for instance, the price rise we see today. Market is oriented to the majority - and the majority isn't used to use their head before paying a bill.
And that's sad.

 

[bug]

So Intel's worse than the cheating scums we all thought knew they are

Only if you enjoy jumping to conclusions.
Sure, the setup was suboptimal. But at this point we don't know how much it actually influenced the results. For all we know, the results may not change much if you tighten the timings for AMD, because you'd be running the memory at lower speeds and IF doesn't really like that anyway. And it's not like running with 4 memory banks populated is an invalid scenario in the real world.

 

[mcraygsx]

Let us not forget we are comparing a $275 CPU against a CPU that cost more then $550.99. These results are not very far off from each other anyways and consumers will likely run into GPU bottleneck at higher resolutions. Not everyone spends thousands of dollars to play AAA titles at 1080p

 

[medi01]

i hope AMD builds a better chip.

AMD's CPUs perform on par with Intels, sometimes 10% behind (normally single threaded stuff or stuff that depends on nvidia's Intel optimizations in drivers) sometimes quite ahead, especially in multithreaded apps, even with core for core . AMD's power consumption is on par or better, despite inferior fab process.


I hope people get real about AMD/Intel/NGreedia oferrings.

Sure, the setup was suboptimal. But at this point we don't know how much it actually influenced the results.

Ah. Because... Erm... Because if "not much" it's only 'not much cheating" I guess.
Logical.

 

[bug]

Ah. Because... Erm... Because if "not much" it's only 'not much cheating" I guess.
Logical.

No. It's because even with optimal settings, the outcome may still be the same.
But when it comes to faulting Intel/Nvidia, we don't need numbers any longer, just an accusation will do. Nevermind that we're already discarding reviewer negligence/incompetence and are already assuming Intel asked them to set the benchmark up as they did.

 

[R0H1T]

No. It's because even with optimal settings, the outcome may still be the same.
But when it comes to faulting Intel/Nvidia, we don't need numbers any longer, just an accusation will do. Nevermind that we're already discarding reviewer negligence/incompetence and are already assuming Intel asked them to set the benchmark up as they did.

What about the accusation that some(?) of the tests were run with 2700x as a quad core, did hardware unboxed make that up?

Now that you mention it, lemme see ~ after spectre/meltdown, shouting ~ hey they were affected as well. Except meltdown is not a Zen flaw.

Amdflaws.com ~ did anyone get to the bottom of that?

Faking 28c 5GHz record breaking demo, which was then rebadged as a Xeon (on socket P) & will be released shortly, not as HEDT though!

Nope, Intel's always guilty in my eyes. I don't like them & I don't like giving them free passes every time they do "whoops, we did it again"

 

[ironwolf]

Update from Hardware Unboxed...

The benchmarks carried out by Principled Technologies are even more bogus than we first thought. A few viewers pointed out that the Ryzen 7 2700X was listed as tested in the “Game Mode” within the Ryzen Master software and I foolishly thought they might have just made a simple copy and paste error in their document as they would have used this mode for the 2950X. This does explain why the Threadripper CPUs were faster than the 2700X in every test.

What this means is a CCX module in the 2700X was completely disabled, essentially turning it into a quad-core. I’ve gone ahead and re-run the XMP 2933 test with Game Mode enabled and now I’m getting results that are within the margin of error to those published by Principled Technologies.

Rest of the update: https://www.patreon.com/posts/21950120

 

[R0H1T]

DP

 

[Vayra86]

Why does AMD need to build a better chip

It has be proven with even previous gen that intel on average is barely 10% faster than AMDs offering while having much higher clocks

Are people that naive to rave about a measly performance gain but a bigger hand in your pockets

10%? Source pls. I csn show you sources where Intel extracts FPS precisely relative to its clockspeed advantage and its no secret either...

There is a large gap but you only notice it with high refresh rates. For 60 fps most CPUs are fine.

That proof you speak of only exists in a performance summary. When you look at scenarios where there is no restricting factor such as a GPU, the single thread gap is still quite big and its almost entirely due to clocks.

Regardless, Intel should be burned for this and they are. Good riddance

 

[londiste]

Game Mode is the dick move.

Memory speeds are the usual manufacturer's benchmark thing where they find the weakest link in configuration that is technically legit. Ryzen spec for memory speed support says 2933 only for two modules. For 4 modules, the supported speed drops to 2133 and even 1866 if the modules are dual-rank.

 

[Vayra86]

Let us not forget we are comparing a $275 CPU against a CPU that cost more then $550.99. These results are not very far off from each other anyways and consumers will likely run into GPU bottleneck at higher resolutions. Not everyone spends thousands of dollars to play AAA titles at 1080p

Not everyone but still a vast majority really does especially with GPU priced through the roof. I reckon its at least as much as people on 4K, likely more.

 

[bug]

Nope, Intel's always guilty in my eyes. I don't like them & I don't like giving them free passes every time they do "whoops, we did it again"

That is your right and I won't dispute that.
But calling them "scum" just because you choose to assume the worst about them is something else.

 

[Salty_sandwich]

I don't care who's got fastest CPU my system spec says it all lol
just as long as this little PC does what I need it to do, gone are the days I want the best or even need it lol
fun to read all the banter people have to go through just to feel the best

 

[Xaled]

Intel for long time were the dirtiest tech company on earth, before nVidia take the the crown. Just remember the recent fake israeli cyber security company thing.. nVidia now is the devil and they are about to kill the whole pc gaming industry..

 

[R0H1T]

That is your right and I won't dispute that.
But calling them "scum" just because you choose to assume the worst about them is something else.

That's because even at their (innovative) best, core 2 era, they've generally been anti consumer & certainly anti competition. There's documented history about their misdeeds, which I'm sure you'd be well aware of. So if not scum ~ sure you can call them whatever you'd like to, but the essence remains same AFAIK & semantics, in this case, don't matter.

 

[bug]

That's because even at their (innovative) best, core 2 era, they've generally been anti consumer & certainly anti competition. There's documented history about their misdeeds, which I'm sure you'd be well aware of. So if not scum ~ sure you can call them whatever you'd like to, but the essence remains same AFAIK & semantics, in this case, don't matter.

They matter in the context of a civil conversation.

 

[iO]

The main thing you can take away from their "tests" is that the 9900K only achieved 5.5% higher FPS on average than the 8700K.

 

[Supercrit]

Didn't Intel uses a Noctua NH14S cooler while AMD had to content with stock? Not that stock AMD cooler is bad, but compare the free cooler with a premium one is a dick move.