[Intel, come on!] Intel FUDing Ryzen 4000 Performance on Battery

[sounds]

Summary

Intel is using sneaky MBA practices to try to get some wins, instead of innovating.

 

In other news, Lenovo set their Ryzen 4000 laptop performance to be really poor, which AMD has to allow them to do.

 

Quotes

Quote

Intel’s explanation for why AMD CPUs lose so much performance
on battery is that the systems wait for 7-10 seconds before
engaging turbo mode

... the settings that control the amount of time before turbo
modes engage and the overall performance delta between AC and
DC power are settings that the OEM controls, not AMD.

... Intel's five comparison systems for itself came from MSI,
Lenovo, Intel itself (in the form of a laptop kit), and two
from HP. Intel is drawing on a much wider range of manufacturers
for its own systems. I don’t know anything about the NUC laptop
kit - haven't had the opportunity to test one - but I would have
preferred the fifth system be a standard commercial comparison,
and the AMD systems should have been drawn from an equally diverse
pool of hardware as the Intel ones were. There are four
manufacturers represented for Intel, and two for AMD.

...

When Intel gave its presentation, it made a point of calling
out the fact that Cinebench R20 doesn’t show the same behavior
as the other benchmarks it had chosen to highlight.

The "oddly" is straight-up FUD. Cinebench R23 also doesn’t
show the 30-48 percent pattern of decline Intel claims. Neither
does Corona Render. Neither does Handbrake. Neither does
JetStream 2. Neither does Blender 2.90. Neither does the
Blender 1.0Beta2 benchmark (not shown, but I ran it).

 

My thoughts

I'm just going to put my palm over my face on this one. I thought after the Toms Hardware "Intel "'Forgot' to Mention 28-Core, 5-GHz CPU Demo Was Overclocked to such an extreme that it required a one-horsepower industrial water chiller" and the $1.25 Billion (to AMD) plus $1.45 Billion (to the EU) over antitrust, maybe Intel would change their ways...

 

Sources

https://www.extremetech.com/extreme/317657-intel-is-spreading-fud-about-supposedly-huge-ryzen-4000-performance-drops-on-battery

[Fasauceome]

Intel:

"People shouldn't rely on benchmarks so much to decide on products"

 

Also Intel:

"Well the benchmarks show..."

[Techstorm970]

Ah, back to their old tricks, I see!

Spoiler

Time for me to go all AMD!

'Twas my plan anyway.

 

[leadeater]

 

To me it looks like this is actually true..... for Intel systems LOL

[manikyath]

isnt this intended behavior when you set your laptop to "conserve battery life"?

 

this is literally how i ran my 2nd gen core i7 laptop for 8 years...

[Jet_ski]

2 hours ago, Fasauceome said:

Intel:

"People shouldn't rely on benchmarks so much to decide on products"

 

Also Intel:

"Well the benchmarks show..."

*rigged benchmarks show

[sounds]

It's on Ars Technica too - FUD machine going full force here.

[Dabombinable]

My sister is getting a laptop with an R5 4350U regardless. Going from an i5 2350M (laptop) and Core M3-6y30 (tablet) it is a gigantic leap forward.
 

The laptop barely runs Minecraft, and the tablet can't even manage that. Plus the tablet nearly always sits at 100% doing anything.

[Arika]

Its almost like marketing is designed to paint your product in the best light. This is not new, nor is it exclusive to Intel.

[Dabombinable]

4 minutes ago, Arika S said:

Its almost like marketing is designed to paint your product in the best light. This is not new, nor is it exclusive to Intel.

The levels of bullshit are always the problem though.

[Arika]

2 minutes ago, Dabombinable said:

The levels of bullshit are always the problem though.

Agreed. Misleading marketing should be treated the same is false advertising. What ever stupid loopholes that allow the degree of cherry picking we've seen lately need to be fixed. Until they are, no first party benchmarks or performance charts should ever be trusted.

[terminalinfinity]

8 hours ago, Techstorm970 said:

Ah, back to their old tricks, I see!

  Reveal hidden contents

Time for me to go all AMD!

'Twas my plan anyway.

 

Anyone who didn't expect this when things got competitive (again) with AMD doesn't know the corporate culture of intel.

 

I'm sure they're already on the horn with Michael Dell to see what his price is this time to be their "friend".

I'm only half joking about the second part. 

[D13H4RD]

At this point, Tiger Lake is going to be a case of a potentially interesting product smeared by an exceptionally questionable (at best) marketing strategy. 

 

From the entire Tiger Lake launch presentation to this recent one, I really feel like Intel's current marketing's strategy is essentially summed up as "mimic what our fanboys have been doing on YouTube comments". That is to say, find strawman arguments to depict competitors' offerings as being inferior rather than trying to convince people that our products are superior. 

 

And that's exactly what Tiger Lake's marketing has been, even from launch. Just a whole bunch of insecure marketers trying to find and pick apart strawman arguments in an attempt to convince people that offerings from their competitors are inferior whilst simultaneously failing to adequately market and bring forth the reasons on why people should consider a Tiger Lake machine. 

 

And honestly, AMD's strategy of limiting peak performance in short bursty workloads on battery power is nothing new. It is a common method even used on Intel machines to conserve battery when peak performance is not required. 

[LAwLz]

I feel like people are on a witch-hunt for Intel.

Of course their marketing is misleading and show their products in the best light, and competing products in the worst light. That is how all marketing works.

AMD does it too, and so does Nvidia.

But for some reason people keep pointing out only when Intel does it. 

 

The simple solution to all of this is, stop listening or caring about first party benchmarks. They are universally shit and can not be trusted, no matter which company puts them out.

[Blademaster91]

The same company that says benchmarks don't matter, puts out bullcrap trying to say AMD is worse at performance.

Intel is getting quite desperate with Tiger Lake, and their disaster of a launch is really evident of them panicking at how good Ryzen 4000 mobile is.

[Master Disaster]

42 minutes ago, LAwLz said:

I feel like people are on a witch-hunt for Intel.

Of course their marketing is misleading and show their products in the best light, and competing products in the worst light. That is how all marketing works.

AMD does it too, and so does Nvidia.

But for some reason people keep pointing out only when Intel does it. 

 

The simple solution to all of this is, stop listening or caring about first party benchmarks. They are universally shit and can not be trusted, no matter which company puts them out.

To be fair to AMD, pretty much all claims they made during the Zen 3 & RDNA 2 launches have been proven to be true but you're still 100% correct.

 

No company really cares about its customers, they care about the amount of money the customer has in their wallet. They're not your friend and they owe you nothing.

 

I think the problem people have with Intel is their recent shift towards being even more anti competition than usual. Things like claiming benchmarks don't matter when they lose while still using them as evidence when they win or spending years telling consumers quad cores are all they need while there's no competition only to reverse that as soon as competition emerges. Its very clear the bean counters are in control at Intel HQ and the engineers have no say in any real decisions.

[LAwLz]

29 minutes ago, Master Disaster said:

To be fair to AMD, pretty much all claims they made during the Zen 3 & RDNA 2 launches have been proven to be true but you're still 100% correct.

Everything Intel says in the graphs above seems to be true as well. Something can be true and misleading.

For example AMD did not mention Raytracing performance because they knew it wasn't as good as Nvidia's performance. Highlighting your strengths and not mentioning your weakness, or highlight your competitors weakness is exactly what companies do in their benchmarks all the time, including AMD, and including in recent times.

 

31 minutes ago, Master Disaster said:

I think the problem people have with Intel is their recent shift towards being even more anti competition than usual. Things like claiming benchmarks don't matter when they lose while still using them as evidence when they win or spending years telling consumers quad cores are all they need while there's no competition only to reverse that as soon as competition emerges. Its very clear the bean counters are in control at Intel HQ and the engineers have no say in any real decisions.

They never claimed "benchmarks don't matter".

I don't get why people keep repeating this blatant misquote or sentence taken out of context. There have been like 4 people in this thread alone who claims Intel said "benchmarks don't matter".

What Bob Swan said was:

Quote

The VOCID-19 pandemic is reshaping business and personal priorities. The world, as we know it, is changing and creating new challenges for technology including supporting the new normal at work, keeping people connected, and using the explosion of data more effectively to solve big global challenges. We should see this moment as an opportunity to shift our focus as an industry from benchmarks to the benefits and impacts of the technology that we create. The pandemic is underscored the need for technology to be purpose-built so it can meet these evolving business and consumer needs. And this requires a customer-obsessed mindset to say close, anticipate those needs, and develop solutions. In this mindset the goal is to ensure we are optimizing for a stronger impact that will support and accelerate positive business and societal benefits around the globe. 

 

This is a mission our collective global industry can rally around across the ecosystem and now more than ever, we'll all have an increasing sense of responsibility, not just for the products we make but the role we play in the world and our ability to make a difference.

We need to create a more inclusive tech industry that is laser-focused on building technologies that advance progress in critical areas like health, safety and carbon neutral computing.

 

Intel has already begun this work by utilizing our size, scale and scope across the ecosystem to enable impact that spans nearly every industry. We aim to help grow and advance the technology business in order to join hands with the broader industry and help drive true impact to people around the world.

 

For example, earlier this year, we introduced our 5G portfolio to help industry-wide network transformation. In a few weeks time, we'll be annoucni9ng a new data center product and platform with Cooper Lake that will further our leadership in AI. Also, we continue to drive our strategy to lead in the projected $65 billion intelligent edge opportunity. And later this summer, we'll introduce Tiger Lake and cement our position as the undisputed leader in mobile computing and PC innovation.

 

This is a moment for our industry to come together, creating technology that enriches lives, that creates value and supports and accelerates positive business and societal benefits should be our collective goal.

 

In closing, I encourage you all to make the most of this virtual event and continue the tradition of COMPUTEX for driving future innovation. We look forward to more collaboration with our Taiwan ecosystem partners and TAITRA in the future. Stay safe and well, and thank you.

 

The feeling I get from the announcement is that it is not aimed at your average LTT "gamer bro" who wants the fastest gaming PC. I mean, that crowd of people aren't really interested in data center products or what collaborations Intel does with health services or the likes.

 

A lot of the talk is dedicated to the current pandemic and the impression I get is that this is Bob saying "hey, right now there is a global pandemic and hundreds of thousands of people are dying. I think the industry should focus on helping people who are currently struggling rather than try and compete on benchmark graphs". 

Of course, there is a lot of marketing talk in that speech as well, I don't really get the feeling at all that Intel are sore losers that don't want people looking at benchmarks anymore. I get the feeling that Intel's CEO, when seeing the death toll rise went "I think we have lost focus on what is really important here".

 

If AMD had made the same statement then I think it would have been received very differently. They would have looked humble and caring because they are in the lead. Since Intel are behind it makes them look bitter. 

 

 

But "Intel wants tech companies to collaborate and help people affected by COVID-19" doesn't get as many clicks as "Intel are bitter they are losing in benchmarks".

[RejZoR]

I don't quite get how Intel is nerfing AMD here. Is it like Intel extort Lenovo to purposely cripple Ryzen chips so that Intel looks better?

[leadeater]

48 minutes ago, LAwLz said:

Everything Intel says in the graphs above seems to be true as well. Something can be true and misleading.

Well the other side of that is it's only true if the laptop is configured that way and it's both an OEM choice and user power profile choice. It's the classic "it depends" issue, not exactly a bedrock to stand on and claim about your competitors when it is actually not a trait of the CPU at all but rather a configuration choice.

 

Like I can go in to Windows power settings and cap an Intel CPU to 50% clocks on battery then do a bunch of testing and publish an article, it's founded on bullshit but still "true".

 

On desktop so battery option doesn't show but you get the point:

 

So while not directly the same as what Intel is talking about in regards to AMD CPUs it's still founded on bullshit, and from memory you can configure similar behavior on Intel too.

[Blademaster91]

9 hours ago, RejZoR said:

I don't quite get how Intel is nerfing AMD here. Is it like Intel extort Lenovo to purposely cripple Ryzen chips so that Intel looks better?

I wouldn't be surprised if Intel was still getting companies to make the Ryzen CPU's slower on purpose. Asus made performance worse with the TUF gaming 15 laptop with the ryzen cpu.

This is just Intel throwing false claims at AMD, even though in benchmarks AMD is still faster.

 

[leadeater]

17 minutes ago, RejZoR said:

I don't quite get how Intel is nerfing AMD here. Is it like Intel extort Lenovo to purposely cripple Ryzen chips so that Intel looks better?

No it's a case of AMD allowing an option to change boost behavior for their CPUs and OEM can choose how to use that. So for more power and portability focused laptop designs it's common to configure a boost delay to conserve power and extend run time. On more performance focused laptop designs this tends to not be configured this way.

 

Intel is pointing to this to show how their CPUs are better, which ignores the variability of the issue and they also neglect to mention that the Intel systems have worse on battery run times unless configured equivalently or more conservatively, which is possible.

 

It's a bit like filling two buckets with water then putting a hole in each, of different sizes, then proclaiming your hole is better because the water drained faster. Well sure if that is the only goal it is possible to also put a bigger on in the other bucket to. And thus the whole thing is a pointless comparison.

[samcool55]

 

R23 4800U AC 8143, on battery 8333.

 

That result shows something is clearly not right. The battery result should either be the same or lower, but not higher. That doesn't make any sense.

[LAwLz]

42 minutes ago, leadeater said:

Well the other side of that is it's only true if the laptop is configured that way and it's both an OEM choice and user power profile choice. It's the classic "it depends" issue, not exactly a bedrock to stand on and claim about your competitors when it is actually not a trait of the CPU at all but rather a configuration choice.

 

Like I can go in to Windows power settings and cap an Intel CPU to 50% clocks on battery then do a bunch of testing and publish an article, it's founded on bullshit but still "true".

 

On desktop so battery option doesn't show but you get the point:

-picture-

 

So while not directly the same as what Intel is talking about in regards to AMD CPUs it's still founded on bullshit, and from memory you can configure similar behavior on Intel too.

Well aren't all benchmarks "it depends"? 

The benchmarks AMD showed on their launch event for example are only true if you use exactly the same components, exactly the same settings, exactly the same software and exactly the same part of the games they used to benchmark.

For example AMD showed a lot of benchmarks where their 6000 cards competed with Nvidia's, but they made sure ray tracing were disabled before doing their tests.

You have to use very skeptical eyes and mindset when evaluating any first party results. It is very, VERY rare that they give anything close to the whole picture and even if they are technically true, they are typically only true in some specifically manufactured scenario that shows that companies products in the best light possible. 

 

But I don't think you can compare what Intel did here to someone going in and changing a bunch of settings in the power manager in Windows. As the ExtremeTech article explains, the settings dictating these specific things are set by the manufacturer, not AMD nor the end user. It seems like Intel's tests are not manufactured by them deliberately changing a bunch of settings to make AMD look bad, but rather them deliberately picking certain AMD products that behave this way out of the box (such as the Lenovo laptops they showed).

 

 

Anyway, first party benchmarks are shit and can't be trusted. More news at 11.

[leadeater]

21 minutes ago, LAwLz said:

It seems like Intel's tests are not manufactured by them deliberately changing a bunch of settings to make AMD look bad, but rather them deliberately picking certain AMD products that behave this way out of the box (such as the Lenovo laptops they showed).

But presented as if this is a trait of the CPU themselves which is not the case. So me going in to power management and setting 50% maximum clocks is completely as fair is what Intel is portraying here. Neither are a trait of the CPU, it is nothing more than a possible setting.

 

It is completely fair to compare two laptops against each other and talk about this type of difference, as it is a trait of the laptop not every single Ryzen Mobile CPU found in every single Ryzen Mobile laptop.

 

21 minutes ago, LAwLz said:

For example AMD showed a lot of benchmarks where their 6000 cards competed with Nvidia's, but they made sure ray tracing were disabled before doing their tests.

You have to use very skeptical eyes and mindset when evaluating any first party results.

Yes but they and neither anyone else has tried to characterize the GPU rasterization (RT off) performance of RX 6800 as being that of the Ray Tracing enabled tests. Even when comparisons are done between products with RT On and Off, comparing all combinations, it's signified as to each what they are.

 

And that's what Intel could have done, compared a laptop with is boost conservation on and off on products for both vendors CPUs. Because the full truth of the matter is Intel boost parameters are configurable too.

 

To be clear I don't actually care that much about this, I find it amusing if anything. What I find most amusing is tech publications failing to do due diligence and fact check Intel's claims before publishing. It's their face they are risking for egg to be on it, but I am keen to throw the eggs lol.

[Arika]

21 minutes ago, samcool55 said:

 

R23 4800U AC 8143, on battery 8333.

 

That result shows something is clearly not right. The battery result should either be the same or lower, but not higher. That doesn't make any sense.

depends how many times you run it. being that close is within margin of error. you can run the same test multiple times and never get the same score.