AMD Can't Shake Off US Class-Action Lawsuit over Bulldozer "8-Core" Advertising

mope54 said:
That's fine if you can't articulate the fine points to a consumer. But by the same token you can't use vague terms to market those esoteric differences to a consumer.

It's fine to say that 8 cores to 6 cores isn't a meaningful distinction when comparing different architectures. The problem lies in marketing teams who deliberately shifted to these slight to meaningless differentiations in order to sell more product to the masses.
Click to expand...
So you're saying AMD claimed their 8 core bulldozer CPU was better than Intel's 4 core...generally, and not just in whatever benchmarks they used in marketing?
 
mope54 said:
That's fine if you can't articulate the fine points to a consumer. But by the same token you can't use vague terms to market those esoteric differences to a consumer.

It's fine to say that 8 cores to 6 cores isn't a meaningful distinction when comparing different architectures. The problem lies in marketing teams who deliberately shifted to these slight to meaningless differentiations in order to sell more product to the masses.
Click to expand...

And there you go falling right back to your performance argument. What was that about performance not mattering?
 
Nobu said:
So you're saying AMD claimed their 8 core bulldozer CPU was better than Intel's 4 core...generally, and not just in whatever benchmarks they used in marketing?
Click to expand...
I'm saying that AMD listed "8 cores" on their CPU box so now they have to prove in court that it actually has 8 "cores." If they can't definitively state what a core is, then they can't have it as a bullet point in their sales strategy.

Multiple people in this thread have said that cores can't be defined because the definition changes based on the architecture among other reasons. Until an industry standard of what does or doesn't constitute a core is established in law, corporations are vulnerable to their claims as to how many exist in a given product.

If AMD can establish in court that they listed "8 cores" in their marketing strategy and that their claim is firmly within an industry established standard, then they will not be held liable for falsely claiming their chip had 8 cores.

Even if that is the outcome of this case (AMD prevails; they accurately portrayed their CPUs as having 8 "cores"), it still wouldn't render the court case frivolous. Courts are the venue where disputes are heard. Since there currently exists a dispute between the claimants' (people filing) understanding of "core" and the respondent's (AMD) understanding of "core" the court must step in and resolve the conflict between the two parties.
 
mope54 said:
I'm saying that AMD listed "8 cores" on their CPU box so now they have to prove in court that it actually has 8 "cores." If they can't definitively state what a core is, then they can't have it as a bullet point in their sales strategy.

Multiple people in this thread have said that cores can't be defined because the definition changes based on the architecture among other reasons. Until an industry standard of what does or doesn't constitute a core is established in law, corporations are vulnerable to their claims as to how many exist in a given product.

If AMD can establish in court that they listed "8 cores" in their marketing strategy and that their claim is firmly within an industry established standard, then they will not be held liable for falsely claiming their chip had 8 cores.

Even if that is the outcome of this case (AMD prevails; they accurately portrayed their CPUs as having 8 "cores"), it still wouldn't render the court case frivolous. Courts are the venue where disputes are heard. Since there currently exists a dispute between the claimants' (people filing) understanding of "core" and the respondent's (AMD) understanding of "core" the court must step in and resolve the conflict between the two parties.
Click to expand...

Completely disagree. The case is entirely frivolous and will do nothing to settle any dispute. If anything, it's an indicator of America's broken judicial system that allows people to pursue frivolous lawsuits for some small amount of monetary gain while wasting large amounts of corporate money and lining lawyer pockets.
 
mope54 said:
I'm saying that AMD listed "8 cores" on their CPU box so now they have to prove in court that it actually has 8 "cores." If they can't definitively state what a core is, then they can't have it as a bullet point in their sales strategy.

Multiple people in this thread have said that cores can't be defined because the definition changes based on the architecture among other reasons. Until an industry standard of what does or doesn't constitute a core is established in law, corporations are vulnerable to their claims as to how many exist in a given product.

If AMD can establish in court that they listed "8 cores" in their marketing strategy and that their claim is firmly within an industry established standard, then they will not be held liable for falsely claiming their chip had 8 cores.

Even if that is the outcome of this case (AMD prevails; they accurately portrayed their CPUs as having 8 "cores"), it still wouldn't render the court case frivolous. Courts are the venue where disputes are heard. Since there currently exists a dispute between the claimants' (people filing) understanding of "core" and the respondent's (AMD) understanding of "core" the court must step in and resolve the conflict between the two parties.
Click to expand...
I'm afraid you're going to be disappointed then, that's about all I can say. It may not go the way I think, but I highly doubt it'll go the way you suspect.
 
Nobu said:
I'm afraid you're going to be disappointed then, that's about all I can say. It may not go the way I think, but I highly doubt it'll go the way you suspect.
Click to expand...
What do you think I'm going to be disappointed about?

What's the way you think?
What's the way you think I suspect?

Your post reads like you've made a determination about how the case should resolve and you've assumed some way I think it should resolve.

You haven't been tracking the conversation accurately.

I haven't made any prediction about the case. In fact, I'm not even talking about how it should resolve.
 
mope54 said:
What do you think I'm going to be disappointed about?

What's the way you think?
What's the way you think I suspect?

Your post reads like you've made a determination about how the case should resolve and you've assumed some way I think it should resolve.

You haven't been tracking the conversation accurately.

I haven't made any prediction about the case. In fact, I'm not even talking about how it should resolve. Unfortunately, the conversation seems to have gone right over your head.
Click to expand...
No, I'm aware you haven't said what you think, because you've avoided saying it at every opportunity. I don't think it'll go the way you suspect, and that's all I have to say about that.

I think the court will hear the plaintiff's case, and then allow AMD to explain why they advertised the way they did. I don't think anyone will be asked to change their marketing strategy, AMD or otherwise.
 
Why do people keep going with Class Action Lawsuits? At best, it will be settled, lawyers get millions, the rest get pennies and AMD won't admit any wrong doing.
 
This is almost as bad as "Orange man bad" logic.

Bulldozer was rough, but an FX-8x00 certainly runs multiple guest VM's better than a 6 or 4 core, including 4 core Intel parts.
 
thebufenator said:
This is almost as bad as "Orange man bad" logic.

Bulldozer was rough, but an FX-8x00 certainly runs multiple guest VM's better than a 6 or 4 core, including 4 core Intel parts.
Click to expand...
eh, it's performance is not really any better than a 4 core 8 thread Intel chip. It's only better then regular quad core intel chips. A i7 2600k vs a fx-8150 is a pretty rough comparison for AMD... and its only BARELY faster than a 2500k at multi threaded processes...

The argument that it beats a 4 core chip is pointless, you need to compare it to either a 4 core 8 thread chip or a straight up 8 core chip.
 
elite.mafia said:
eh, it's performance is not really any better than a 4 core 8 thread Intel chip. It's only better then regular quad core intel chips. A i7 2600k vs a fx-8150 is a pretty rough comparison for AMD... and its only BARELY faster than a 2500k at multi threaded processes...

The argument that it beats a 4 core chip is pointless, you need to compare it to either a 4 core 8 thread chip or a straight up 8 core chip.
Click to expand...
It doesn't matter how it compares to other chips of other architectures. An 8 core chip will be able to perform 8 operations, one on each core, simultaneously.
 
Nobu said:
It doesn't matter how it compares to other chips of other architectures. An 8 core chip will be able to perform 8 operations, one on each core, simultaneously.
Click to expand...

How is that any different from a 4 core, 8 thread chip?
 
elite.mafia said:
How is that any different from a 4 core, 8 thread chip?
Click to expand...
That chip is able to perform 2 operations per core, simultaneously. Edit: rather, in this post and the last, substitute operations with tasks. It'll be performing multiple ops, but it'll be limited to one or more tasks per core (depending on the architecture).

Further edit: to be clear, a 4c 8t chip will be able to perform up to 8 tasks simultaneously, but some tasks may have to wait if a compute unit (ALU or FPU) is not free. An 8c 8t chip has 8 ALUs (in most/all modern CPUs, at least), so it will never have to wait for those operations, but it may have to wait (as in the case of bulldozer) if there are no available FPUs.
 
Last edited:
Nobu said:
No, I'm aware you haven't said what you think, because you've avoided saying it at every opportunity. I don't think it'll go the way you suspect, and that's all I have to say about that.

I think the court will hear the plaintiff's case, and then allow AMD to explain why they advertised the way they did. I don't think anyone will be asked to change their marketing strategy, AMD or otherwise.
Click to expand...
I haven't made any prediction about the case because it's not relevant to the points I was making.

The problem is that you created a straw man that you've been debating with instead of paying attention to what I've written. In fact, the post you quoted literally has both outcomes and the explanation of their ramifications (if they lose, this happens; if they win, this happens). You quoted that very post of me explaining the two outcomes of a court case and concluding that I'm inclined to believe it should resolve in one of those two ways...a position you wholly made up!

Here are the points listed out for your convenience:

Dickey and Palmer have one definition of a core
AMD has a different definition of a core.

The court room is the venue for resolving these differences of opinion between two disputing parties.
A jury will determine whether one party or the other has a more convincing argument of how a core should be defined.

These types of court processes (discovery, testimony, and jury opinions) will produce a definition of a core that others will rely upon.

Ultimately these kinds of intersections between society, technology, and the law are one of the ways in which *things* in society become defined.

The only subjective conclusion I made was that this function of the courts was necessary and beneficial to society. I suppose you could debate that point, but that is the only subjective position I've taken in this conversation. Anything else you created in your mind and are inferring from my posts.
 
mope54 said:
I haven't made any prediction about the case because it's not relevant to the points I was making.

The problem is that you created a straw man that you've been debating with instead of paying attention to what I've written. Here are the points listed out for your convenience:

Dickey and Palmer have one definition of a core
AMD has a different definition of a core.

The court room is the venue for resolving these differences of opinion between two disputing parties.
A jury will determine whether one party or the other has a more convincing argument of how a core should be defined.

These types of court processes (discovery, testimony, and jury opinions) will produce a definition of a core that others will rely upon.

Ultimately these kinds of intersections between society, technology, and the law are one of the ways in which *things* in society become defined.

The only subjective conclusion I made was that this function of the courts was necessary and beneficial to society. I suppose you could debate that point, but that is the only subjective position I've taken in this conversation. Anything else you created in your mind and are inferring from my posts.
Click to expand...
Alright, let's assume they are irrelevant.

I believe that whatever they decide, AMD will not have to change their marketing at all. That is all I have been saying this whole time. I understand that they still have to go through with this regardless of what I think, that does not preclude my thoughts however.
 
Tsumi said:
Except you know, coupling 4 100 HP V2 engines will give you the equivalent of a single 400 HP V8 engine, assuming identical engine architecture. Increased bulk and weight will have its own performance penalties, but that is irrelevant to this discussion.
Click to expand...

But....it didn't. That was the point. And it was not a V8. This is the crux of the lawsuit.
 
Nobu said:
CPU = car
Core = engine
Piston/cylinder = execution units.

?
Click to expand...
https://www.anandtech.com/show/6355/intels-haswell-architecture/8
So, an intel "core" cpu core (i.e. haswell) would have several pistons with multiple injectors and valves (hyperthreading), whereas bulldozer would have more engines with a single injector and valves supplying the pistons (execution units) – some pistons would share an exhaust port and get conjested in certain workloads...or would it just have fewer pistons? ...I'm not good with car analogies.
 
Last edited:
Nobu said:
https://www.anandtech.com/show/6355/intels-haswell-architecture/8
So, an intel "core" cpu core (i.e. haswell) would have several pistons with multiple injectors and valves (hyperthreading), whereas bulldozer would have more engines with a single injector and valves supplying the pistons (execution units) – some pistons would share an exhaust port and get conjested in certain workloads...or would it just have fewer pistons? ...I'm not good with car analogies.
Click to expand...

Sure, I will respond to that as soon as you tell me how many cores they are claiming there... Also where in that article does Intel claim that a hyper thread equals a core?
 
NoOther said:
Sure, I will respond to that as soon as you tell me how many cores they are claiming there... Also where in that article does Intel claim that a hyper thread equals a core?
Click to expand...
I never said it did. I told you I'm bad at analogies. I'm just trying to get one that works.
 
Nobu said:
I never said it did. I told you I'm bad at analogies. I'm just trying to get one that works.
Click to expand...

You implied it did. The whole legal argument revolves around how many actual cores the CPU has and its performance.
 
NoOther said:
You implied it did. The whole legal argument revolves around how many actual cores the CPU has and its performance.
Click to expand...
Where? Please fix my analogy because I'm sure it's nonsense. That's not what I was trying to say at all.
 
Page 5, my take on it most motherboards had the option to set it to 1m/1c instead of 1m/2c so to fix the 50% hit the cores was not much more then smt/HT but using cut down cores the main stuff was in the shared module witch the 2 cores had to fight for

The cpu should of been called x core cpu(the amount of modules it has) with core assist (something along them lines) or just don't have core count on the box (not sure If Intel was doing it that way for long time)

If you look at the bulldozer silicon you can visually see the 4 modules or cores with the 2 mini cores inside them

Just effectively HT or smt with real cores

the main design flaw with the bulldozer architecture was that they did not Mark the extra cores as smt or HT threads if they had done that that would have made Windows 7 and higher utilise the cores more or less correctly and a lot of the issues would of been Less (but still would of had the 50% loss per thread issue once total cpu load was past 50% on FPU operations)
 
Last edited:
NoOther said:
But....it didn't. That was the point. And it was not a V8. This is the crux of the lawsuit.
Click to expand...

Performance is irrelevant. Besides, in operations that were not heavily dependent on a fast front end, Bulldozer will scale linearly up to 8 threads. Such applications are extremely rare, and the only time Bulldozer is actually limited to 4 execution units is when running 4 256-bit threads. Each 256-bit FPU was perfectly capable of working as two independent 128-bit FPUs.
 
crypto mining scales a hell of a lot better on 4+ threads on the FX arch taht it does a 4c 8t I7.
 
I'm not sure who started it, but i've had to correct many many people who go around on forums spouting off that FX isnt a true 6 or 8 core chip. It is, its just not as good, for technical reasons far beyond what most forum whiners could hope to understand.
Its a pity, cause the 6 core and 8 core dozers were and continue to be decent budget gaming chips. 1080p, 1440p, all current AAA games (as of last fall anyway), just work. Dont expect 144hz, but we're in budget land.
This lawsuit is garbage.
 
travm said:
I'm not sure who started it, but i've had to correct many many people who go around on forums spouting off that FX isnt a true 6 or 8 core chip. It is, its just not as good, for technical reasons far beyond what most forum whiners could hope to understand.
Its a pity, cause the 6 core and 8 core dozers were and continue to be decent budget gaming chips. 1080p, 1440p, all current AAA games (as of last fall anyway), just work. Dont expect 144hz, but we're in budget land.
This lawsuit is garbage.
Click to expand...

I dislike this court case for a great many reasons, but perhaps most for the implication that if a core's design is not competitive with the best in the industry - you might get sued based upon some laypersons flawed interpretation of CPU designs. And further, they may attempt to enforce something they don't understand.

Designers need the flexibility to try different things: provide differing strengths for different markets, different cost points - just like they always have. Not everything can be or should be top-tier desktop per-core performance, at top-tier prices. And of course, the differentiation will not always be simply clock rate. Sometimes it means you save money on cache, or the branch predictor, or god forbid FPU units. It's reality.
 
PhaseNoise said:
I dislike this court case for a great many reasons, but perhaps most for the implication that if a core's design is not competitive with the best in the industry - you might get sued based upon some laypersons flawed interpretation of CPU designs. And further, they may attempt to enforce something they don't understand.

Designers need the flexibility to try different things: provide differing strengths for different markets, different cost points - just like they always have. Not everything can be or should be top-tier desktop per-core performance, at top-tier prices. And of course, the differentiation will not always be simply clock rate. Sometimes it means you save money on cache, or the branch predictor, or god forbid FPU units. It's reality.
Click to expand...
Designers will still be free to do whatever they want in the lab, the difference will be in how marketing teams sell the things that are made in those labs.

The public doesn't have a "flawed" interpretation of CPU designs--they only know what the marketing team tells them is important. That's the point of marketing and the point of laws designed to ensure marketing is fulfilling its legal responsibility to adequately inform consumers about what they are purchasing.

The surprising thing is how many enthusiasts, on a site dedicated to fact-checking the snake oil claims of the industry we play in, are supporting of these marketing teams hoodwinking consumers and selling snake oil for so long. The only thing that makes sense to me is some of you seem to be confusing the case (how marketing teams sell a piece of technology) with what can and can't be designed if this case prevails (research and development, which won't be impacted at all).

We aren't in the 1700s anymore, we don't operate under laissez-faire policies in US consumer law, but the tech industry has been operating as if it's still the wild wild west for the better part of the 20th century. Now that we're in the 21st century it's time to start reigning these industries in before they walk all over us. Junk food and bunk medicine has more oversight in what/how in can and can't market products than the technology sector...and given how ubiquitous and significant technology is now (and "cores" arguably at the center of it) some terms that have been solidified in the market space need to be defined.
 
mope54 said:
I'm saying that AMD listed "8 cores" on their CPU box so now they have to prove in court that it actually has 8 "cores." If they can't definitively state what a core is, then they can't have it as a bullet point in their sales strategy.

Multiple people in this thread have said that cores can't be defined because the definition changes based on the architecture among other reasons. Until an industry standard of what does or doesn't constitute a core is established in law, corporations are vulnerable to their claims as to how many exist in a given product.

If AMD can establish in court that they listed "8 cores" in their marketing strategy and that their claim is firmly within an industry established standard, then they will not be held liable for falsely claiming their chip had 8 cores.

Even if that is the outcome of this case (AMD prevails; they accurately portrayed their CPUs as having 8 "cores"), it still wouldn't render the court case frivolous. Courts are the venue where disputes are heard. Since there currently exists a dispute between the claimants' (people filing) understanding of "core" and the respondent's (AMD) understanding of "core" the court must step in and resolve the conflict between the two parties.
Click to expand...

they did state what a core was.

it was in my post.

https://hardforum.com/threads/amd-c...re-advertising.1976211/page-2#post-1044053993

this is the basis of their lawsuit.

190398_core.png
 
mnewxcv said:
My phone has 8 cores but it's slower than my 6 core laptop. Class action lawsuit!




*not all cores are created equal.
Click to expand...

No, no, no, it only means something if AMD did it. :D o_O:rolleyes: There are some people who are still really upset that AMD did not just die in a fire. (DIAF) :D FX 8*** processors were all 8 core ones but, I never got one that I could overclock to 4.8 GHz or more, but, oh well, they still worked well for what they were.
 
Master_shake_ said:
they did state what a core was.

it was in my post.

https://hardforum.com/threads/amd-c...re-advertising.1976211/page-2#post-1044053993

this is the basis of their lawsuit.
Click to expand...
I'm not sure who you think "they" are, but the portion you quoted is the claimant's definition of a core. The definition is in dispute because there is no codified definition of a core. Until it's legally codified, companies can continue to call anything they want a core and laypersons will continue to believe cores are things that they may or may not be depending on how they're marketed.
 
ManofGod said:
No, no, no, it only means something if AMD did it. :D o_O:rolleyes: There are some people who are still really upset that AMD did not just die in a fire. (DIAF) :D FX 8*** processors were all 8 core ones but, I never got one that I could overclock to 4.8 GHz or more, but, oh well, they still worked well for what they were.
Click to expand...
Intel was also under the gun a little over ten years ago. I'm can't recall specific cases, or if Intel was sued specifically for misleading claims, but PC Mag was sued for their review and only prevailed because they specifically linked the "core" claims to relevant performance within the benchmarks.

So no, it's not just AMD.
 
mope54 said:
I'm not sure who you think "they" are, but the portion you quoted is the claimant's definition of a core. The definition is in dispute because there is no codified definition of a core. Until it's legally codified, companies can continue to call anything they want a core and laypersons will continue to believe cores are things that they may or may not be depending on how they're marketed.
Click to expand...

the claimant doesn't know what a cpu is, what it does, or how it operates.

once amd's lawyers are done running him through the ringer the jury will see how stupid he is and likely rule in favor of amd.

and if we are lucky force the claimant to pay AMD's legal fees

AMD won't have to prove anything.
 
Master_shake_ said:
the claimant doesn't know what a cpu is, what it does, or how it operates.

once amd's lawyers are done running him through the ringer the jury will see how stupid he is and likely rule in favor of amd.

and if we are lucky force the claimant to pay AMD's legal fees

AMD won't have to prove anything.
Click to expand...

I wonder if Intel will help the claimant side ?
 
mope54 said:
I'm not sure who you think "they" are, but the portion you quoted is the claimant's definition of a core. The definition is in dispute because there is no codified definition of a core. Until it's legally codified, companies can continue to call anything they want a core and laypersons will continue to believe cores are things that they may or may not be depending on how they're marketed.
Click to expand...

It's in dispute because the plaintiffs are saying that AMD's cores cannot perform like Intel's cores. Performance is the crux of this argument, whether you want to admit it or not. And once again, for the umpteenth time, PERFORMANCE IS ENTIRELY IRRELEVANT TO WHAT MAKES A CORE. Which, once again, makes this lawsuit entirely frivolous and unproductive, and is merely a cash grab and/or publicity stunt by the plaintiffs.

Once again, the definition of a core cannot be compared to MPG ratings, or HP ratings, or TQ ratings, something you seemingly are entirely incapable of comprehending. MPG, HP, and TQ ratings are most similar to benchmark scores.
 
Master_shake_ said:
the claimant doesn't know what a cpu is, what it does, or how it operates.

once amd's lawyers are done running him through the ringer the jury will see how stupid he is and likely rule in favor of amd.

and if we are lucky force the claimant to pay AMD's legal fees

AMD won't have to prove anything.
Click to expand...
AMD will have to prove that their definition is within industry standards. The standard of proof in a civil case is preponderance of the evidence, which is more likely than not (50.1+ vs. 49.9-).

AMD likely would be able to demonstrate to a jury's satisfaction that they weren't mis-using the term, but in the process (the process being discovery, written testimony, spoken testimony, industry papers, researched opinions, the trial, and eventual judgement one way or the other) a definition of a core will emerge.

The next time a company is sued for using "core" in a way that the consumer doesn't think matches what they believed (and it doesn't much matter the technical bona fides of a the consumer, since the definition of a core comes from the marketing teams primarily of Intel and AMD; that is, Intel and AMD have built up an idea on the marketplace of what a core is, what it does, and what it's relationship to performance is or should be--this claimant didn't just make that definition up out of thin air), the company will look back at v. AMD and cite the evidence and judgment (win or lose, the definition that emerges during this trial will become codified) and that will make a more clear definition of a core. Eventually, "core" will be well-defined, much like those other industry standards like MPG and torque that I used as examples and also had similar class action lawsuits that brought those industry standards into legal standards.

That's the process of law and society. The other way in which things get defined into law would be legislatures, which would be weird and polar opposites of what people are wanting in this thread. It's so bizarre, they just want to argue to argue without understanding our system of law.

In a marketplace, terms have to be defined. Things can't be accurately marketed and consumers don't have any assurances when things aren't defined. Compare this to when "organic" was unregulated and plastered all over every thing just to sell it. Then it was "Natural," then "All natural" which are all marketing phrases that have become regulated over time after politicians or court cases defined them into what they are today.

Maybe you don't care about red dye #2, but that doesn't mean it's not relevant to know when it is or isn't in a product (eventually banned in the US). M&Ms even removed their red ones because of the hyped up threat over #2 (didn't actually have that dye to begin with) and took about a decade to recover.

So either definitions emerge through an iterative process where the industry leaders come into court rooms and show their reasoning for why they call something the way they do, or we have politicians creating rules about labels (and we see how well that works out from food dye to weapons), but under no circumstances do we live in a society where neither of those things occurs and people just sell stuff and call it whatever they want, label it however the best way to move product off the shelf, or whatever kind of "free" market people fantasize we live within.
 
You must log in or register to reply here.
Back
Top