AMD Agrees To Pay Out $35 Per Chip Over FX Marketing Lawsuit

[LAwLz]

51 minutes ago, Jito463 said:

Exactly what did "the law" protect people from?

Deceptive marketing.

 

52 minutes ago, Jito463 said:

People claim AMD used deceptive marketing, but I can't recall any from that time period and no one has been able to actually provide any specific examples.

Here are some direct quotes from AMD:

Quote

Take megatasking abilities to extreme levels with AMD FX-Series FX 8300. It is the first native 8-core desktop processor built with dynamic, tuneable performance to handle multiple intensive apps without breaking a sweat.

Quote

The industry’s first and only native 8-core desktop processor for unmatched multitasking and pure core performance with all-new “Bulldozer” architecture

It's somewhat easy for us on LTT forum to think "oh, well it's not quite what I'd call a native 8-core processor but it does technically have two integer execution units so I guess you could call it an 8 core", but false advertising laws are meant to protect ignorant users from being mislead.

As soon as you have to start justifying a somewhat shady action with "well technically, in our marketing we used a different definition from what might be common, but..." then you are not exactly acting in good faith.

 

AMD knew what they were doing. They deliberately hyped the "8 core" part of the chip knowing that it would make it seem better in marketing material. I don't get why people are so hellbent on justifying what is clearly shitty behavior. They tried to be misleading. They could just as easily have called it a quad core with 8 threads like we see happening today.

I have a feeling that if Intel suddenly started marketing their HT 6-cores as 12 core processors then people wouldn't be taking Intel's side. And that's not too far from what AMD did with Bulldozer. It's basically only an integer execution engine away from being exactly the same.

 

1 hour ago, Jito463 said:

Why isn't it up to AMD to decide?  It's not like there's a set definition of what specifically is included in a CPU core, apart from some very basic elements.  It literally had two physical cores to process data, that they shared components didn't make either of them less of a core.

Well, it has two physical cores to process specific types of data (integer). For other types of data (float point), it physically only had one core to process data (per module).

 

It isn't up to AMD to decide because they are the ones creating the marketing material.

Think of it rationally for a second. If I start a lottery in the US promising 1 000 000 dollars as the price, and get a ton of people buying it. Then when I am suppose to pay the price out, I reveal that it's "LAwLz dollars", my own crypto currency which is worth around 0.001 US dollars.

Would you not say I was using deceptive marketing too? Just because I left some technical details out? I was technically telling the truth!

 

It is not up to the company making the marketing material to decide if something is deceptive or not. If a large portion of people seeing the ad gets the wrong impression, then it is deceptive. That is how it works, and how it should work. You can't make up your own definitions of words and then pray on people who assume the words mean something else.

[leadeater]

1 hour ago, LAwLz said:

It is not up to the company making the marketing material to decide if something is deceptive or not. If a large portion of people seeing the ad gets the wrong impression, then it is deceptive. That is how it works, and how it should work. You can't make up your own definitions of words and then pray on people who assume the words mean something else.

Non understanding isn't the defining metric for deceptive marketing either though, you can have the wrong impression and not have been deceived in any way even if this happened in a large scale.

 

And it largely is up the the manufacturer to define what a core is, their core. Unless someone like the IEEE define the requirements as to what one is then that responsibility does fall back to the designer of the product. In exactly the same way GPU cores are referred to as cores they, what is being talked about, are not cores but elements of such a thing that is able to do work. A CUDA core is not a core, an SM is a closer representation of a 'core' in a classical belief of what cores are but that doesn't stop Nvidia or AMD labeling FPU elements as cores in their block diagrams either.

 

When we have a situation like Bulldozer where the performance is not as good as the customer or the community expects and certain workloads perform significantly less than desired are we in any better position, or worse, than that of the manufacturer to start defining what a core is. It ultimately make no difference if the spec sheet says 4 or 8 when reviews exist that give you direct information about performance and you also have the market align itself to the value of the product, why anyone would expect a $200 CPU to perform better than a $400 because it has 8 cores or the $700 laptop is faster than the $1200 for the same reasoning is lacking a bit of common sense.

 

Does anything actually change if it were marketed in exactly the same manor but under a 4 core banner? Was the marketing itself any different than comparable examples on the market, not limiting this to CPUs or GPUs. Marketing rarely if ever does not come with bold claims, worded just right. Sometimes companies get the wording just that little bit wrong and can get burned but here the majority of the supporting argument around the marketing relies back on this what is and is not a core which honestly is not that relevant and it fundamentally doesn't change anything. 4 or 8 the same spiel would have been rolled out and the performance would have been exactly the same, bad.

 

Products can, and should be allowed to, be worse than another competitors.

[mr moose]

22 minutes ago, leadeater said:

Non understanding isn't the defining metric for deceptive marketing either though, you can have the wrong impression and not have been deceived in any way even if this happened in a large scale.

 

And it largely is up the the manufacturer to define what a core is, their core. Unless someone like the IEEE define the requirements as to what one is then that responsibility does fall back to the designer of the product. In exactly the same way GPU cores are refereed to as cores they, what is being talked about, are not cores but elements of such a thing that is able to do work. A CUDA core is not a core, an SM is a closer representation of a 'core' in a classical belief of what cores are but that doesn't stop Nvidia or AMD labeling FPU elements as cores in their block diagrams either.

 

When we have a situation like Bulldozer where the performance is not as good as the customer or the community expects and certain workloads perform significantly less than desired are we in any better position, or worse, than that of the manufacturer to start defining what a core is. It ultimately make no difference if the spec sheet says 4 or 8 when reviews exist that give you direct information about performance and you also have the market align itself to the value of the product, why anyone would expect a $200 CPU to perform better than a $400 because it has 8 cores or the $700 laptop is faster than the $1200 for the same reasoning is lacking a bit of common sense.

 

Does anything actually change if it were marketed in exactly the same manor but under a 4 core banner? Was the marketing itself any different than comparable examples on the market, not limiting this to CPUs or GPUs. Marketing rarely if ever does not come with bold claims, worded just right. Sometimes companies get the wording just that little bit wrong and can get burned but here the majority of the supporting argument around the marketing relies back on this what is and is not a core which honestly is not that relevant and it fundamentally doesn't change anything. 4 or 8 the same spiel would have been rolled out and the performance would have been exactly the same, bad.

 

Products can, and should be allowed to, be worse than another competitors.

I think what Lawlz meant was that when the company uses marketing knowing full well that in order to be informed it relies on a technicality that majority of consumers don't understand, then that marketing is deceptive.  Allowing the company to be the one to decide what is deceptive would be like allowing companies to decide what  consumer laws mean after paying into the coffers of US senators...  oh wait... 

[leadeater]

23 minutes ago, mr moose said:

I think what Lawlz meant was that when the company uses marketing knowing full well that in order to be informed it relies on a technicality that majority of consumers don't understand, then that marketing is deceptive.  Allowing the company to be the one to decide what is deceptive would be like allowing companies to decide what  consumer laws mean after paying into the coffers of US senators...  oh wait... 

We're not allowing them to decide what is deceptive the problem is we're saying it is deceptive due to reliance on what we define what a core is. What a core is in this case is what AMD said it was and has no effect at all on performance, performance isn't actually a measure of the number of cores. A core's performance can be terrible, is that actually deceptive? The sum of those cores were still overall better than AMD's previous products.

[LAwLz]

3 minutes ago, leadeater said:

We're not allowing them to decide what is deceptive the problem is we're saying it is deceptive due to reliance on what we define what a core is. What a core is in this case is what AMD said it was and has no effect at all on performance, performance isn't actually a measure of the number of cores. A core's performance can be terrible, is that actually deceptive? The sum of those cores were still overall better than AMD's previous products.

I don't think performance matters at all. What matters (at least in my eyes) is if the general consumer got the wrong impression of the product based on the claims made by AMD.

AMD suddenly changing their definition of a "core" and then start advertising it as "the only native 8 core processor" was deceptive.

 

For crying out loud, if they didn't want to be deceptive then they could have called them quad cores, right? The reason why they chose to call them 8 cores wasn't because "it was kind of technically the truth, if we change our definition of a core", it was because calling them 8 cores would hopefully make more people buy them.

 

Performance is completely irrelevant. The relevant things is intention and if the general consumer got the wrong impression of the product based on the marketing material.

[leadeater]

Just now, LAwLz said:

For crying out loud, if they didn't want to be deceptive then they could have called them quad cores, right? The reason why they chose to call them 8 cores wasn't because "it was kind of technically the truth, if we change our definition of a core", it was because calling them 8 cores would hopefully make more people buy them.

Because they weren't 4 cores either and in the same vein as you are using now would be equally deceptive, they are not 4 core CPUs. So if they are not 4 cores and also now not allowed to be called 8 cores then what is it?

 

If changing CPU architecture design is going to lead to litigation you are directly going to restrict progression in the industry. There can be legitimate reason to change things like this, for better or worse. Luckily this does not set any ongoing legal precedent for this, but don't think as a result AMD and Intel will now think twice before investing money in similar efforts.

[mr moose]

43 minutes ago, leadeater said:

We're not allowing them to decide what is deceptive the problem is we're saying it is deceptive due to reliance on what we define what a core is. What a core is in this case is what AMD said it was and has no effect at all on performance, performance isn't actually a measure of the number of cores. A core's performance can be terrible, is that actually deceptive? The sum of those cores were still overall better than AMD's previous products.

 

I still think it boils down to informed versus uninformed,  The average users was not informed of a performance metric (so how it actually performs is largely moot), they were told the product had 8 pure performing cores,  it would be natural for an average consumer to believe this meant it was twice as good as any 4 core processor (including Intel) without the necessary caveats.    AMD failing to give those caveats is the only reason this whole case exists in the first place.

[leadeater]

14 minutes ago, mr moose said:

I still think it boils down to informed versus uninformed,  The average users was not informed of a performance metric (so how it actually performs is largely moot), they were told the product had 8 pure performing cores,  it would be natural for an average consumer to believe this meant it was twice as good as any 4 core processor (including Intel) without the necessary caveats.    AMD failing to give those caveats is the only reason this whole case exists in the first place.

Then all 6 cylinder engines are less performant than 8 cylinder engines then correct? What about displacement? What about naturally aspirated versus forced induction? Like I said non understanding doesn't equal deception and the claims made aren't unique to similar marketing. Nobody caveats in marketing, they show the product in the best possible way they can. Caveats rarely show up and when they do are along the lines of "When combined with a proper diet" or "Consult your doctor if symptoms persist".

 

The performance is not moot, it's literally what is being used to justify that there is a problem in the first place, that consumers were deceived and the cores are not cores, because of said performance.

 

Edit:

Let me ask a simple question, would there have been a lawsuit and would people have complained about Bulldozer if it were exactly as it was architecturally but was twice as fast as Intel CPUs?

[Drak3]

5 hours ago, Jito463 said:

Though I do agree with you about the APU "compute cores".  While it's technically true, it was irritating trying to decipher how many CPU cores vs GPU cores an APU had.

I don’t. Having bought multiple APUs and machines with APUs from the era, the CPU and GPU core counts were clearly listed in the specifications, and I believe that AMD’s actual marketing listed the total compute core count along side the break down of CPU and GPU cores.

[RoseLuck462]

I have the  FX-8120

 

Guess I could use that $35 to buy a silver dollar.

[NunoLava1998]

>buy like 1,000 FX 4100s or something for like 15$

>later

>get 35,000$ in payouts for that

>buying the 1000 cost 15,000$

>net profit: 20k$

 

how 2 make money 101

[Jito463]

5 hours ago, LAwLz said:

Take megatasking abilities to extreme levels with AMD FX-Series FX 8300. It is the first native 8-core desktop processor built with dynamic, tuneable performance to handle multiple intensive apps without breaking a sweat.

I'm still not seeing the deceptive nature you're talking about.  Yes, it's marketing and it painted the chip in the best light possible, but what precisely was deceptive about it?

5 hours ago, LAwLz said:

The industry’s first and only native 8-core desktop processor for unmatched multitasking and pure core performance with all-new “Bulldozer” architecture

The only thing you might have an argument with, is the "pure core performance" line, since single core performance was fairly weak.  Then again, I always take marketing fluff with a grain of salt.

5 hours ago, LAwLz said:

It's somewhat easy for us on LTT forum to think "oh, well it's not quite what I'd call a native 8-core processor but it does technically have two integer execution units so I guess you could call it an 8 core",

Actually, I always understood what it meant and still accepted it as an actual 8 core.

5 hours ago, LAwLz said:

As soon as you have to start justifying a somewhat shady action with "well technically, in our marketing we used a different definition from what might be common, but..." then you are not exactly acting in good faith.

Except there really wasn't a "common" definition for cores, not even today.

5 hours ago, LAwLz said:

AMD knew what they were doing. They deliberately hyped the "8 core" part of the chip knowing that it would make it seem better in marketing material. I don't get why people are so hellbent on justifying what is clearly ***** behavior. They tried to be misleading.

Marketing material for a company painted their product in the best light?  Say it ain't so?!  It's not like every company in history has tried to present their product in the best way possible.  I'm not trying to defend any behavior, I'm just trying to get someone to prove that they actually were misleading.

4 hours ago, LAwLz said:

For crying out loud, if they didn't want to be deceptive then they could have called them quad cores, right?

Except that it's not a "quad core", it's an octocore.  They could have explained the CMT (Clustered Multi Threading) more fully, but that's not intrinsically deceptive.

5 hours ago, LAwLz said:

Well, it has two physical cores to process specific types of data (integer). For other types of data (float point), it physically only had one core to process data (per module).

FP is a unit, not a core itself.  Now you're redefining the definition of a core.

6 hours ago, LAwLz said:

Think of it rationally for a second. If I start a lottery in the US promising 1 000 000 dollars as the price, and get a ton of people buying it. Then when I am suppose to pay the price out, I reveal that it's "LAwLz dollars", my own crypto currency which is worth around 0.001 US dollars.

Would you not say I was using deceptive marketing too? Just because I left some technical details out? I was technically telling the truth!

Sorry, not seeing the analogy.  That might make sense in your own mind, but you haven't proven that's what AMD has done.

[spartaman64]

the amazing amd 0 core cpu

[mr moose]

4 hours ago, leadeater said:

Then all 6 cylinder engines are less performant than 8 cylinder engines then correct? What about displacement? What about naturally aspirated versus forced induction? Like I said non understanding doesn't equal deception and the claims made aren't unique to similar marketing. Nobody caveats in marketing, they show the product in the best possible way they can. Caveats rarely show up and when they do are along the lines of "When combined with a proper diet" or "Consult your doctor if symptoms persist".

 

The performance is not moot, it's literally what is being used to justify that there is a problem in the first place, that consumers were deceived and the cores are not cores, because of said performance.

 

Edit:

Let me ask a simple question, would there have been a lawsuit and would people have complained about Bulldozer if it were exactly as it was architecturally but was twice as fast as Intel CPUs?

 

There is a lawsuit because the product did not fit the physical description it was marketed at.  

 

If you want to use an engine as an analogy, it would be like advertising 8 full performing cylinders, but not explaining those 8 cylinders are each sharing enough valves, intake and fuel injecting for 4 cylinders.   

 

So in the end trying advertising  8 pure cores when in reality it was 8 half performing is deceptive.   Only those with intimate knowledge of the CPU design understood why they claimed 8 cores. 

[Mira Yurizaki]

While there may not be a strict definition of what a "CPU core" is, there is still an expectation of what the public thinks of what a CPU core is. If you want to call out a "core" is whatever the hell a company defines it as, then this can lead to confusing ways to market your product. For example, the Motorola Moto X was said to use an "X8 Mobile Computing System", with the "X8" part meaning it had 8 processing cores. If you didn't know anything else about the product or had a thorough enough grasp on technology, what are you going to immediately default "processing cores" to mean? To stretch the argument even more, if we were to say a "core" is something that processes data, as opposed to storing/managing it, then one could argue that an AMD Ryzen or Intel Core processor is really a 20+ core per "core" processor, because there are that many elements that process data in some form or fashion.

 

Besides that, multicore processors have been around in general consumer space for at least five years before Bulldozer was launched and a cursory Google search tells me the first processor generally accepted as multicore was made by IBM in 2001. So if you want some basis for a "technical" definition, I would take what most people consider to be the first thing of it that existed.

 

On top of that, if we take what AMD called a core for Bulldozer, does this mean I could just split up the back-end of a given processor and call it multicore processor? Could I claim a multi-rail PSU is a redundant PSU? Could I claim a multi-platter hard drive is a RAID system? A 4-cylinder engine is really just a dual-two-cylinder engine system?

[leadeater]

5 hours ago, mr moose said:

So in the end trying advertising  8 pure cores when in reality it was 8 half performing is deceptive.   Only those with intimate knowledge of the CPU design understood why they claimed 8 cores. 

But it wasn't half performing cores and now you're litigating a company for a product having worse performance than a competitor. Like I said companies should be allowed to have inferior products than a competitor, you don't have to buy them and if you don't do some very basic logic checks like "it costs less so it should perform less" then you were not deceived you just failed common sense.

[leadeater]

3 hours ago, Mira Yurizaki said:

To stretch the argument even more, if we were to say a "core" is something that processes data, as opposed to storing/managing it, then one could argue that an AMD Ryzen or Intel Core processor is really a 20+ core per "core" processor, because there are that many elements that process data in some form or fashion.

Those back end elements don't have any capability to process the data, they move data around and store it. Block diagrams label cores as cores or execution units because that is where the processing happens and each of of them can be given a task of work, which is the difference between SMT because the tasks share the same execution units. Things like shared L2 cache isn't new either, Core 2 was built around shared L2 cache and had no L3 cache at all.

 

We even have Intel processors on the market right now that have different amount of FP units per core. Xeon Gold 5000 and below have a single AVX-512 unit and products above those have 2 AVX-512 units.

 

There is merit in an argument around the shared fetch and decode of bulldozer but two units of work can still be assigned to two different physical execution units, totaling 8, as long as you are doing INT. Each INT unit has it's own scheduler as well as the FP unit. 

 

Dare I bring up ARM big.LITTLE and how the products were marketed with the total number of cores but only half could be active, either the low power cores or the high power cores not both. Were the general buyers any more informed about what was going on there? No. Maybe we should sue for that as well then.

 

How dare a company try something new.

[mr moose]

23 minutes ago, leadeater said:

But it wasn't half performing cores and now you're litigating a company for a product having worse performance than a competitor.

No, they are litigating because what the company claimed was wrong.   The product was sold on the marketing that they natively had 8 pure performing cores (when they did not), and something they specifically claim the competition dosen't.    Therefore it is reasonable for the average consumer to believe they are better than anything with only 4 cores.

 

23 minutes ago, leadeater said:

Like I said companies should be allowed to have inferior products than a competitor, you don't have to buy them and if you don't do some very basic logic checks like "it costs less so it should perform less" then you were not deceived you just failed common sense.

They are allowed to have inferior products.  They are not allowed to market them as having something they don't which specifically insinuates them to be superior.

 

The claim that they have 8 pure cores is wrong.  they aren't "pure" cores by any stretch of the imagination, so claiming they are the only ones with this feature is wrong.

 

That's all it takes for false advertising to be a thing.  Everything else is moot.

 

 

[mr moose]

3 hours ago, Mira Yurizaki said:

While there may not be a strict definition of what a "CPU core" is, there is still an expectation of what the public thinks of what a CPU core is.

 

A very good point, and when AMD claim they are the only ones with 8 pure cores, this is AMD telling consumers that whatever you consider a pure core to be this has 8 of them.

Sadly when lawyers found out they bent the truth a little to make their product look better that landed them in trouble (if you can call this trouble, I think it'll wash away and be of no consequence to AMD).

 

 

[Blademaster91]

41 minutes ago, leadeater said:

But it wasn't half performing cores and now you're litigating a company for a product having worse performance than a competitor. Like I said companies should be allowed to have inferior products than a competitor, you don't have to buy them and if you don't do some very basic logic checks like "it costs less so it should perform less" then you were not deceived you just failed common sense.

It was half performing cores in quite a few use cases, the FX 8 cores were beaten by dual core i3's and locked quad core i5's, except for some apps that could see the 8 cores or were optimized for how the bulldozer cores worked. I agree with you, a company should be allowed to have inferior products, except AMD was marketing their product as being superior compared to the competitor yet for the tasks most people bought them for the performance was worse. It's like buying a V8 pickup truck but you can only use 4 of the cylinders unless you're towing a large trailer.

[leadeater]

25 minutes ago, mr moose said:

No, they are litigating because what the company claimed was wrong.   The product was sold on the marketing that they natively had 8 pure performing cores (when they did not), and something they specifically claim the competition dosen't.    Therefore it is reasonable for the average consumer to believe they are better than anything with only 4 cores.

Well they were better than 4 cores, better than AMD's previous generation of 4 core products. What is pure performing cores? Pure performing what? That specific claim didn't actually claim the performance was greater.

 

And no it is not reasonable for the average consumer to believe they are better than anything with only 4 cores. Seriously no. There are 4 core products on the market that performed differently, AMD had 6 core products on the market that performed worse than competing  4 core products. Bigger isn't better is literally a common saying, so is buyer beware. The was existing precedent for this assumption to not be true.

 

Intel did not have the ability to execute 8 INT tasks and AMD did, Intel's 4 were still better. AMD could execute 4 FP tasks, Intel could execute 4 FP tasks.

 

Again the argument falls back to what is a core and without an industry or legal definition of what that is we are left with opinion and only opinion, I will always defer to the manufacture in such an instance unless there is very clear and explicit grounds for not doing so. There are 8 INT schedulers, 8 INT execution units, 8 L1 Data Caches, 8 Load/Stores, 8 L1 DTLBs. Where is your specific point where it is not a core? What specifically are you using as ground to define where the line is for what is and is not a core.

[mr moose]

Just now, leadeater said:

Well there were better than 4 cores, better than AMD's previous generation of 4 core products. What is pure performing cores? Pure performing what? That specific claim did't actually claim the performance was greater.

 

And no it is not reasonable for the average consumer to believe they are better than anything with only 4 cores. Seriously no. There are 4 core products on the market that performed differently, AMD had 6 core products on the market that performed worse than competing  4 core products. Bigger isn't better is literally a common saying, so is buyer beware. The was existing precedent for this assumption to not be true.

 

Intel did not have the ability to execute 8 INT tasks and AMD did, Intel's 4 were still better. AMD could execute 4 FP tasks, Intel could execute 4 FP tasks.

 

Again the argument falls back to what is a core and without an industry or legal definition of what that is we are left with opinion and only opinion, I will always defer to the manufacture in such an instance unless there is very clear and explicit grounds for not doing so. There are 8 INT schedulers, 8 INT execution units, 8 L1 Data Caches, 8 Load/Stores, 8 L1 DTLBs. Where is your specific point where it is not a core? What specifically are you using as ground to define where the line is for what is and is not a core.

You don't need a legal definition for a core.    AMD clearly claimed they were the only ones with this therefore anyone who bought it thinking it would be better because of that claim and discovered it wasn't was subject to false advertising.

[leadeater]

6 minutes ago, mr moose said:

You don't need a legal definition for a core.    AMD clearly claimed they were the only ones with this therefore anyone who bought it thinking it would be better because of that claim and discovered it wasn't was subject to false advertising.

Wrong, they claimed they had something the competitor did not and that was true. Assuming more is better is on the buyer, AMD already had more and performed worse, the expectation was already there, if you ignored it then it's on you.

 

You're still going through litigation because the product was inferior to a competitors, which is perfectly acceptable situation.

 

I get that this is around the marketing but this is all falling back on consumer assumptions, ones that were already not true.

 

Quote

"AMD is pleased to have reached a settlement of this lawsuit. While we believe the allegations are without merit, we also believe that eliminating the distraction and settling the litigation is in our best interest."

Without Zen AMD would fight this, right now it is not worth the reputation damage to fight it so made it go away. AMD has not conceded anything here they are just doing what is in their current best interest. 

[mr moose]

6 minutes ago, leadeater said:

Wrong, they claimed they had something the competitor did not and that was true. Assuming more is better is on the buyer, AMD already had more and performed worse, the expectation was already there, if you ignored it then it's on you.

 

You're still going through litigation because the product was inferior to a competitors, which is perfectly acceptable situation.

 

I get that this is around the marketing but this is all falling back on consumer assumptions, ones that were already not true.

 

I think we are going to have just settle as not agreeing on this.  

 

6 minutes ago, leadeater said:

Without Zen AMD would fight this, right now it is not worth the reputation damage to fight it so made it go away. AMD has not conceded anything here they are just doing what is in their current best interest. 

I really don't care what their public reasoning is,  Nvidia lost in similar circumstances,  Coles lost a false advertising case about their "daily baked" bread.  More like daily warmed.  Redbull made the same PR claim when they were sued and settled for claims regarding performance, concentration and reaction times.   False advertising rarely needs a legally defined condition beyond the claims being misleading.  

[leadeater]

1 minute ago, mr moose said:

I really don't care what their public reasoning is,  Nvidia lost in similar circumstances,  Coles lost a false advertising case about their "daily baked" bread.  More like daily warmed.  Redbull made the same PR claim when they were sued and settled for claims regarding performance, concentration and reaction times.   False advertising rarely needs a legally defined condition beyond the claims being misleading.  

Those examples all have specifics where you could measure the claim as true or not. Baking and heating are actually different, you can qualitatively and quantitatively measure that. You can also measure human performance and claims of improvements. Nvidia settled because they advertised the product as having 4GB of high speed GDDR5 which you could physically and measurably disprove.

 

The Nvidia case is the most similar and similarly Nvidia decided to settle. Both that case and AMD Bulldozer heavily rely on technical aspects to be able to actually show if the advertising was deceptive.

 

Similar is only similar, that doesn't actually mean the same.