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

The objective and tangible targets came about because of the lawsuits. It's very strange how you continue to post about a thing that exists as if it always existed without acknowledging how it came to be.

I don't know which country you are in, but in the United States every single car is sold with it's HP and Torque claims loudly and proudly marketed. In the past car companies used to claim ridiculous HP and Torque until they were sued and now have to publish *when* that power occurs (lb-ft@rpms).

The same way with response time. Companies used to shift all kinds of claims about what mattered when it came to response time in a monitor until they had to step into court and codify what they meant. That led to the now standard of reporting "Grey to Grey." Industry didn't just decide to be less vague and more accurate with their marketing claims because people wanted them to, they did it because people made them through the court system.

Whether this particular case prevails isn't particularly relevant to the process I'm describing to you. The conversations that occur in the court will become part of the body of our law that will direct future court proceedings. Eventually that body of law will build up into a formal understanding of the thing ("core" in this case). That's different from informal understandings of a thing, which is where were are now.

/facepalm

Yes, they publish peak HP and torque numbers, but they do not publish HP and torque curves.

What you're saying is akin to consumers assuming performance based on number of cylinders and displacement, and then wondering why a DOHC direct inject engine performs better than a pushrod carburetor engine of the same cylinder count and displacement. Number of cores is the number of cylinders. Clock speed is the displacement. Architecture is your intake, exhaust, cam, fuel delivery, etc. Car companies publish peak HP and torque numbers. They don't publish HP and torque curves, which determines how the engine performs throughout the RPM range, and they don't publish numbers on how the transmission performs. Similarly, AMD publishes and advertises performance in certain benchmarks, but they do not publish numbers on a wide range of software.

In case you still don't get it, the technical existence of 8 cores is in no way similar to MPG, HP, torque, or any other specification. The closest match will be number of cylinders. Claiming a Bulldozer core can't be considered as a full core is most similar to saying that a pushrod cylinder cannot be considered a full cylinder because it doesn't perform as well as a DOHC cylinder.
 
/facepalm

Yes, they publish peak HP and torque numbers, but they do not publish HP and torque curves.

What you're saying is akin to consumers assuming performance based on number of cylinders and displacement, and then wondering why a DOHC direct inject engine performs better than a pushrod carburetor engine of the same cylinder count and displacement. Number of cores is the number of cylinders. Clock speed is the displacement. Architecture is your intake, exhaust, cam, fuel delivery, etc. Car companies publish peak HP and torque numbers. They don't publish HP and torque curves, which determines how the engine performs throughout the RPM range, and they don't publish numbers on how the transmission performs. Similarly, AMD publishes and advertises performance in certain benchmarks, but they do not publish numbers on a wide range of software.

In case you still don't get it, the technical existence of 8 cores is in no way similar to MPG, HP, torque, or any other specification. The closest match will be number of cylinders. Claiming a Bulldozer core can't be considered as a full core is most similar to saying that a pushrod cylinder cannot be considered a full cylinder because it doesn't perform as well as a DOHC cylinder.
WAIT!
What about manifold pressure!
Is it still a cylinder if the manifold pressure is 10atm? Thats not fair. I want a free turbo.
 
this reaks of "I didn't know coffee was hot befor I spilled it on me"

edit:

and also people who don't know shit about technology, being in a position to make judgements about said technology.....

but hey, congrats to everyone who is a part of this class action, I hope you spend your 8 dollars well when you get it....
meanwhile the lawyers will be riding around in new bentlys

The McDonald's hot coffee lawsuit was entirely meritorious and you should stop trying to use it as an example as it's obvious that you don't know the truth of it. It produced third degree burns that required skin grafts, do you have any idea how hot it had to be to do that? And do you know how many safety reports McDonald's had ignored about the heat of the coffee from accidents?, hint, it was not trivial at all.
 
this reaks of "I didn't know coffee was hot befor I spilled it on me"

edit:

and also people who don't know shit about technology, being in a position to make judgements about said technology.....

but hey, congrats to everyone who is a part of this class action, I hope you spend your 8 dollars well when you get it....
meanwhile the lawyers will be riding around in new bentlys
Gonna go off track here, but anytime I see people mention the hot coffee thing, I like to bring up the case of Stella Liebeck, who its famous for.
Some have seen the netflix or HBO shows on it. Lots of people like to laugh about it....but it was horrific.
Coffee was so hot, it nearly melted the womans genitals off.
https://travis.pflanz.me/hot-coffee-2011/
https://travis.pflanz.me/wp-content/uploads/stella_liebeck_burned_by_mcdonalds_coffee.jpg
 
The McDonald's hot coffee lawsuit was entirely meritorious and you should stop trying to use it as an example as it's obvious that you don't know the truth of it. It produced third degree burns that required skin grafts, do you have any idea how hot it had to be to do that? And do you know how many safety reports McDonald's had ignored about the heat of the coffee from accidents?, hint, it was not trivial at all.

I expect any coffee I buy to be at or near boiling temperature.

McDonalds isn't able to super-heat water at atmospheric pressure.
 
No, no coffee should be served at our near boiling, at that setting it isn't fit for human consumption, and that was part of the problem, coffee can be served up to around 60 Celsius, they were serving theirs at 94-97, at those temperatures it produces third degree burns in 3 to 7 seconds, 3rd degree is the worst type of burnt.

Also they had had over 700 similar cases and notes regarding it being a safety hazard openly ignored and admitted so during the trial.

It was most certainly a meritorious lawsuit.
 
The objective and tangible targets came about because of the lawsuits. It's very strange how you continue to post about a thing that exists as if it always existed without acknowledging how it came to be.

I don't know which country you are in, but in the United States every single car is sold with it's HP and Torque claims loudly and proudly marketed. In the past car companies used to claim ridiculous HP and Torque until they were sued and now have to publish *when* that power occurs (lb-ft@rpms).

The same way with response time. Companies used to shift all kinds of claims about what mattered when it came to response time in a monitor until they had to step into court and codify what they meant. That led to the now standard of reporting "Grey to Grey." Industry didn't just decide to be less vague and more accurate with their marketing claims because people wanted them to, they did it because people made them through the court system.

Whether this particular case prevails isn't particularly relevant to the process I'm describing to you. The conversations that occur in the court will become part of the body of our law that will direct future court proceedings. Eventually that body of law will build up into a formal understanding of the thing ("core" in this case). That's different from informal understandings of a thing, which is where were are now.
You realize you just proved everyone else's point right? hp and torque are the engine output, the actual power put down to the road is less... with your argument, car manufacturers are being deceptive and we should be able to sue them.

Continuing with the crappy car analogies, an 8 cylinder engine can have a single cam all the valves run off, or have the valves run off multiple cam shafts, is the push rod engine not an 8 cylinder since it only has one cam?
 
You realize you just proved everyone else's point right? hp and torque are the engine output, the actual power put down to the road is less... with your argument, car manufacturers are being deceptive and we should be able to sue them.

Continuing with the crappy car analogies, an 8 cylinder engine can have a single cam all the valves run off, or have the valves run off multiple cam shafts, is the push rod engine not an 8 cylinder since it only has one cam?
They were already sued and now they have to specify whether they are citing power to the wheels or not.

It's not my argument; it already happened. I was using historical analogues but apparently I was explaining history to some of you at the same time.
 
Even windows 10 lists fx cpus as 4 core 8 logical just to make sure different workload we're not put on like the same to cores on the same module as there was a 50% performance penalty for doing it

The problem is that when Bulldozer was released, no OS on the market understood how to use a module with 2 x integer + 1 x FPU and everything ran like crap. It was essentially a throwback to the problems with Pentium 4 HT where no applications (ESPECIALLY games) could take advantage of HT and would effectively run like it was on a CPU of half the speed. I remember the bemoaning of many a gamer with shiny new Pentium 4 HT rigs only to be horrified that it was even slower than their older Pentium III or Athlon rigs. I managed to convince several to go into the BIOS and turn HT off, and to their amazement, their games ran almost exactly twice as fast since the full CPU could be utilized. There were only a tiny number of applications that ever supported HT in the Pentium 4s lifespan, most related to content creation that the general public hardly ever used.

Microsoft had to re-write the thread dispatcher for Windows 7 and 8 just for the Bulldozer architecture to keep resource intensive threads on different modules and backfill the second thread with lower priority tasks as much as possible. The Linux kernel also had to have the same treatment, and you can use the top monitor in bash to see the behavior in action on Bulldozer based CPUs, but not on Intel based CPUs.

The fact that operating systems had to rewrite how they dispatched processes meant that there was a standard core definition, at least in x86 land by that point that a "core" is one integer unit and one floating point unit, with the option of SMT. Bulldozer fell outside of that definition and was a train wreck that thankfully AMD moved on from. Though it is disappointing that they're still releasing new (mostly entry level and mobile) parts based on the old bulldozer architecture smurfing as Ryzen parts.
 
Wrong. It was settled and consumers got $30 per card they bought Some stores separately gave credits. I got $60 from the settlement and 20pct from my vendor for my pair of gtx 970 cards.

Really? I signed up for the class action for my old pair of 970s but never heard back after the original emails. Guess that's my fault for not keeping up with it.
 
Wrong. It was settled and consumers got $30 per card they bought Some stores separately gave credits. I got $60 from the settlement and 20pct from my vendor for my pair of gtx 970 cards.

I thought that whole lawsuit was kind of silly.

The boards technically did have 4GB of RAM. Yes the last 500MB was much slower than the rest, but it was still 4GB.

Anyone who bought one read the launch reviews and knew exactly how it performed it titles, so they got exactly what they paid for.

I thought that everyone crying about it after the fact was just lame.
 
Really? I signed up for the class action for my old pair of 970s but never heard back after the original emails. Guess that's my fault for not keeping up with it.
I got some money back from the suit for one card and Amazon honored the complaint and tossed me a couple bucks on the other.
 
I thought that whole lawsuit was kind of silly.

The boards technically did have 4GB of RAM. Yes the last 500MB was much slower than the rest, but it was still 4GB.

Anyone who bought one read the launch reviews and knew exactly how it performed it titles, so they got exactly what they paid for.

I thought that everyone crying about it after the fact was just lame.

To be honest, I never really had any issues with the 970 except one of my Zotacs bit the dust and Zotac replaced it. The last 500 megs being 32 bit while the rest of the RAM was 256 bit was what actually bit Nvidia in the ass on that one. If I remember right the card was advertised as a 4GB card at 256 Bit. I still don't know why they did this ...
 
To be honest, I never really had any issues with the 970 except one of my Zotacs bit the dust and Zotac replaced it. The last 500 megs being 32 bit while the rest of the RAM was 256 bit was what actually bit Nvidia in the ass on that one. If I remember right the card was advertised as a 4GB card at 256 Bit. I still don't know why they did this ...

If I recall it had to do with some sort of even multiplier problem. With the buss width of the RAM, the multiplier worked out to be 3.5, but I think they waynted to boost it up to 4 for marketing purposes. I can't remember exactly how this stuff works anymore.
 
If I recall it had to do with some sort of even multiplier problem. With the buss width of the RAM, the multiplier worked out to be 3.5, but I think they waynted to boost it up to 4 for marketing purposes. I can't remember exactly how this stuff works anymore.
I think you're right and I don't know how the heck it works either ;)
 
If I recall it had to do with some sort of even multiplier problem. With the buss width of the RAM, the multiplier worked out to be 3.5, but I think they waynted to boost it up to 4 for marketing purposes. I can't remember exactly how this stuff works anymore.

It was that they disabled some of the ROPs on the 970, which disabled the 8th L2 cache block. This meant that the 7th L2 cache block had to service two 32-bit memory controllers for the last gigabyte of VRAM. As long as you stayed below 3.5GB, that 7th block behaved just like the first 6. Once you went over, every access to a 3.5GB-4.0GB address invalidated cached data from the 3.0-3.5GB address space, and vice versa. This was causing cache misses nearly every single time your game accessed VRAM.
 
It was that they disabled some of the ROPs on the 970, which disabled the 8th L2 cache block. This meant that the 7th L2 cache block had to service two 32-bit memory controllers for the last gigabyte of VRAM. As long as you stayed below 3.5GB, that 7th block behaved just like the first 6. Once you went over, every access to a 3.5GB-4.0GB address invalidated cached data from the 3.0-3.5GB address space, and vice versa. This was causing cache misses nearly every single time your game accessed VRAM.

Thanks for the refresher!
 
No, no coffee should be served at our near boiling, at that setting it isn't fit for human consumption, and that was part of the problem, coffee can be served up to around 60 Celsius, they were serving theirs at 94-97, at those temperatures it produces third degree burns in 3 to 7 seconds, 3rd degree is the worst type of burnt.

Also they had had over 700 similar cases and notes regarding it being a safety hazard openly ignored and admitted so during the trial.

It was most certainly a meritorious lawsuit.

To be fair, with all the details people put in this thread, yeah pay for the burns. My grippe is with the "content is hot" band-aid solution if I may say.
The fact that the coffee was too hot, won't miraculously not be too hot with that sentence and will still be wayyy too hot even with it. I fail to see why that sentence alone protect mcDo from stupid practice ?

Again, I didn't read much of the lawsuit and I'm only discussing about it because of that stupid sentence. Now if you tell me that sentence won't protect them from other suits if it ever happens again with the same pre-requisites, well wtf is that sentence doing there ?
Seems like they wrote that on the cup to protect from future lawsuits if any were to happen but that would also lines with the fact that some pointed out they played the media and this seems to do it.
 
Ohh they had to change the temperature besides the warning (referring to McDonald's), so yeah in general during the lawsuit it was obvious to the jury that besides the life altering burns, McDonald's had to pay extra punitive damages for the gross disregard of customer safety. They really weren't giving any damn about anyone and stupidly made it very clear.

The sentence was added because even now it's served just on the fringe of viable human consumption. If they are ever found to knowingly tamper with the settings again it won't save them from another lawsuit with even higher punitive damages tho.
 
you don't get to define what the designers determine is a core, amd took a gamble and said 'this will be good' turns out it wasn't exactly all that good, it wasn't because they claimed it was a 8 core but infact it wasn't, it was because it was just a poor design, combined with lack of proper software, everything was being designed according to what intel determined was a proper cpu, it was encoded to work well on intel cpus...

again they are saying because of how amd designed it, it wasn't a 8 core....

I mean, I am not exactly educated on how all this goes, but from my understanding they are basicly claiming, a V8 engine isn't a v8 engine, if it can be beat by a 4 cylinder, or 6 cylinder....



da fuck?

oh and don't get me wrong, im not supporting the bulldozer, imo it was a absolute piece of shit..... I stayed clear away from amd since the duron/Athlon/athlonxp days...…. I only recently came back with the 2000 series ryzen cpus..


but im also not sitting here claiming it wasn't a 8 core cpu….

No, what they are claiming is 4 V2 engines do not make a V8 engine. And the performance degradation shows that.
 
No, what they are claiming is 4 V2 engines do not make a V8 engine. And the performance degradation shows that.

But that's not a valid comparison.

Sure, the FX solution is like having 4 2-cyl engines. But by comparison, every single other multi-core chip ever made is like having 4 (or 6, or 8) 1 cylinder engines.

The fact that AMD grouped their cores is not why the FX sucked performance wise. It is the fact that their cores sucked that was the problem.

Sure, there was a marginal impact from the CPU caches not being able to keep up with the 2 core module design, but that only had a marginal impact on performance.

In the end it is completely moot though. AMD did not trick these people into buying a slower CPU than they thought they were getting. Bulldozer launch day reviews were everywhere. You could easily google and find a benchmark to see how it would perform in your workload (or at least one close enough that you could make an educated guess)

Once you have actual performance figures, the architecture itself is completely irrelevant.

If these people bought the CPU and were surprised by its lack of performance, then quite frankly they have no one to blame but themselves.

Each FX CPU marketed as an 8 core has 8 integer cores. They may be grouped two by two, but that doesn't change the fact that they are complete integer cores.

This lawsuit is COMPLETELY without merit being pursued by a bunch of whiny crybabies and opportunists who think they are going to cash in, even though in reality - even if this thing succeeds - they are only going to get a 58 cent check with all the millions going to the lawyers.

The only thing this will do is harm AMD's chances at focusing on their comeback. I wouldn't be surprised if Intel somehow were behind this. It sounds like something scummy they would do.
 
Last edited:
No, what they are claiming is 4 V2 engines do not make a V8 engine. And the performance degradation shows that.

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.
 
AMD did not trick these people into buying a slower CPU than they thought they were getting. Bulldozer launch day reviews were everywhere. You could easily google and find a benchmark to see how it would perform in your workload (or at least one close enough that you could make an educated guess)

Once you have actual performance figures, the architecture itself is completely irrelevant.

If these people bought the CPU and were surprised by its lack of performance, then quite frankly they have no one to blame but themselves.
The expectation that more "cores" equals more power was built up by AMD and Intel themselves. You have the flow of information reversed: the corporations trumpeted the value of "cores" during their marketing campaigns to try and one-up one another during the core wars. People old enough will remember this playbook from the Gigahertz wars (where customers were first told that more gigs are better and then told that gigs weren't relevant "depending on architecture").

I'm not sure why so many people in this thread are suffering from Stockholm syndrome, but it's not acceptable in any other industry to make vague marketing claims that are only true when the company says they are true and then a 3rd party industry that somehow removes responsibility from the marketing team to be honest with what they are selling and moves it to the consumer. That's never acceptable and it's ridiculous to be hashed out in these forums whose very existence was formed out of these tech companies making claims about their products that weren't accurate.

Who cares who could have googled what because it literally doesn't matter; the only thing that matters is whether what's inside the box matches the claims on the outside of the box. It doesn't matter what reviewers say, it doesn't matter what CEOs say, it doesn't even matter what 1st party press releases say. If the product description on the box doesn't match what exists under the hood then it's false/misleading advertising.

Bottom line: IF you can't define a core, you CAN NOT market a core.

It seems like that should be common sense but for some reason a few people are having difficulty understanding that basic principle of, not only law, but also logic. You can't describe accurately that which you don't understand. You can't market or sell that which you can't define.

AMD can't have it both ways--can't eat their cake and have it, too. Either cores matter and all the advertising about cores being a big deal and how they have more cores than their competitors is important, or cores don't matter and all of that marketing was bullshit. Either cores are something that can be described accurately and therefore it's ok to put, "My CPU contains X distinct cores" or they aren't well-defined and you can not put any number of cores on your box because you can't definitively say what does or does not constitute a core.

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.
You and Zarathustra keep trying to hinge this debate on performance when it's completely irrelevant. It doesn't matter which chip performs better than the other; all that matters is whether the marketing matched what's going on under the hood. AMD built up a definition in consumers' minds about what does or does not make a "core" and now they have to live with that definition. They can't waltz into court now saying well cores don't really matter...doesn't really matter how we count em up.
 
The expectation that more "cores" equals more power was built up by AMD and Intel themselves. You have the flow of information reversed: the corporations trumpeted the value of "cores" during their marketing campaigns to try and one-up one another during the core wars. People old enough will remember this playbook from the Gigahertz wars (where customers were first told that more gigs are better and then told that gigs weren't relevant "depending on architecture").

Who cares who could have googled what because it literally doesn't matter; the only thing that matters is whether what's inside the box matches the claims on the outside of the box. It doesn't matter what reviewers say, it doesn't matter what CEOs say, it doesn't even matter what 1st party press releases say. If the product description on the box doesn't match what exists under the hood then it's false/misleading advertising.

And it does. What you are saying makes absolutely no sense.

Does a 8 core Bulldozer not perform better in threaded loads than a 4 core or 6 core bulldozer?

People DID get what it said on the box.

There is no Stockholm syndrome here. No one has been mislead, everyone got exactly what they were promised, and the lawsuit is COMPLETELY without merit. Not even an ounce of it in there.
 
And it does. What you are saying makes absolutely no sense.

Does a 8 core Bulldozer not perform better in threaded loads than a 4 core or 6 core bulldozer?

People DID get what it said on the box.

There is no Stockholm syndrome here. No one has been mislead, everyone got exactly what they were promised, and the lawsuit is COMPLETELY without merit. Not even an ounce of it in there.
You are exhibiting stockholm syndrome.

AMD is claiming they can't define a core while continuing to sell cores to you.
You continue to buy cores even though you readily type in here that cores can't be defined.

A marketing team has sold you on a bill of goods that don't approximate the technology and you are more than willing to lap it up, defend them, and then point to 3rd party industry to "save" you from the misinformation out there instead of holding the marketing teams of the corps they represent accountable.

You can't say that people got what it said on the box because earlier in the thread you said that cores couldn't be defined. So explain how, if something can't be defined, people definitely got what they bought when they picked up an "8 core" CPU?
 
The expectation that more "cores" equals more power was built up by AMD and Intel themselves. You have the flow of information reversed: the corporations trumpeted the value of "cores" during their marketing campaigns to try and one-up one another during the core wars. People old enough will remember this playbook from the Gigahertz wars (where customers were first told that more gigs are better and then told that gigs weren't relevant "depending on architecture").

I'm not sure why so many people in this thread are suffering from Stockholm syndrome, but it's not acceptable in any other industry to make vague marketing claims that are only true when the company says they are true and then a 3rd party industry that somehow removes responsibility from the marketing team to be honest with what they are selling and moves it to the consumer. That's never acceptable and it's ridiculous to be hashed out in these forums whose very existence was formed out of these tech companies making claims about their products that weren't accurate.

Who cares who could have googled what because it literally doesn't matter; the only thing that matters is whether what's inside the box matches the claims on the outside of the box. It doesn't matter what reviewers say, it doesn't matter what CEOs say, it doesn't even matter what 1st party press releases say. If the product description on the box doesn't match what exists under the hood then it's false/misleading advertising.

Bottom line: IF you can't define a core, you CAN NOT market a core.

It seems like that should be common sense but for some reason a few people are having difficulty understanding that basic principle of, not only law, but also logic. You can't describe accurately that which you don't understand. You can't market or sell that which you can't define.

AMD can't have it both ways--can't eat their cake and have it, too. Either cores matter and all the advertising about cores being a big deal and how they have more cores than their competitors is important, or cores don't matter and all of that marketing was bullshit. Either cores are something that can be described accurately and therefore it's ok to put, "My CPU contains X distinct cores" or they aren't well-defined and you can not put any number of cores on your box because you can't definitively say what does or does not constitute a core.


You and Zarathustra keep trying to hinge this debate on performance when it's completely irrelevant. It doesn't matter which chip performs better than the other; all that matters is whether the marketing matched what's going on under the hood. AMD built up a definition in consumers' minds about what does or does not make a "core" and now they have to live with that definition. They can't waltz into court now saying well cores don't really matter...doesn't really matter how we count em up.

/facepalm

You are so full of contradictions. Yes, more cores equals more performance in multithreaded tasks... compared to fewer cores of the same architecture. How it compares across different architectures is entirely irrelevant.

AMD is claiming they can't define a core while continuing to sell cores to you.
You continue to buy cores even though you readily type in here that cores can't be defined.

WTF are you even talking about?
 
You are exhibiting stockholm syndrome.

AMD is claiming they can't define a core while continuing to sell cores to you.
You continue to buy cores even though you readily type in here that cores can't be defined.

A marketing team has sold you on a bill of goods that don't approximate the technology and you are more than willing to lap it up, defend them, and then point to 3rd party industry to "save" you from the misinformation out there instead of holding the marketing teams of the corps they represent accountable.

You can't say that people got what it said on the box because earlier in the thread you said that cores couldn't be defined. So explain how, if something can't be defined, people definitely got what they bought when they picked up an "8 core" CPU?


I can very clearly define a core. I never said I couldn't. You must be thinking of someone else. It is the integer portion of a traditional CPU. It does not include any caches, or FPU's.

An CPU marketed as an 8 core Bulldozer has 8 integer cores. End of story.

And you know what? I agree with you. The selling cores is a little nuts. I'd much rather have fewer faster cores, than more slower ones. But that doesn't mean that these FX chips didn't have 8 cores, and if people are silly enough to over-prioritize high core count, AMD is perfectly within their rights to sell it to them.

It is easy to see them on a chip diagram. If you can count to 8, you realize this is a frivolous lawsuit.
 
I can very clearly define a core. It is the integer portion of a traditional CPU. It does not include any caches, or FPU's.

An CPU marketed as an 8 core Bulldozer has 8 integer cores. End of story.

And you know what? I agree with you. The selling cores is a little nuts. I'd much rather have fewer faster cores, than more slower ones. But that doesn't mean that these FX chips didn't have 8 cores, and if people are silly enough to over-prioritize high core count, AMD is perfectly within their rights to sell it to them.

It is easy to see them on a chip diagram. If you can count to 8, you realize this is a frivolous lawsuit.
The problem arises because the industry doesn't define cores that way. Sometimes they do, because they can be defined that way, but sometimes they can defined in another way. And therein lies the conundrum: AMD and Intel have advertised their CPUs as having more cores than the competition, and what those higher core counts means, and they do it by attaching varying degrees of importance to different aspects of the technology that lay-persons (and even techies) don't have information about.

We can look at a CPU diagram to see what the legend label as a "core" but that's not enough since the company selling the CPU is the one drawing the diagram...and the one telling us what is or isn't an important part of the CPU.

That's my bad if I said you were claiming cores weren't well-defined. Other people in this thread will disagree with your definition, however, as will some people in this lawsuit. The industry has an operational definition, I agree with that point, but it's not codified and it should be now that it's become a salient marketing point.

I don't agree this lawsuit is frivolous. It wouldn't have made it past the first goal post. Regardless of the technical merits of the claimant's case, this court case serves a vital purpose of codifying what does or doesn't constitute a discreet core according to our current technological understanding. That's a separate issue from whether AMD should lose or not, which is what I think you mean when you label it frivolous (that AMD should prevail).
 
/facepalm

You are so full of contradictions. Yes, more cores equals more performance in multithreaded tasks... compared to fewer cores of the same architecture. How it compares across different architectures is entirely irrelevant.



WTF are you even talking about?
Pretty straightforward.

It's even obvious to you, but since you want to try and "win" this argument rather than learn something about product law and understand the principles I'm laying out relatively in clear terms, you continue to load your statements up with caveats.

You know it's not enough to simply claim, "8 cores equals more performance" yet that's what AMD and Intel both did.

There is no contradiction in what I'm typing. The confusion comes from you refusing to see what's plainly in front of your face. You want to argue with me, basically simply because it's me, but in order to do so you have to argue against common sense and what everyone in the US knows about buying things as consumer already understands.
 
Pretty straightforward.

It's even obvious to you, but since you want to try and "win" this argument rather than learn something about product law and understand the principles I'm laying out relatively in clear terms, you continue to load your statements up with caveats.

You know it's not enough to simply claim, "8 cores equals more performance" yet that's what AMD and Intel both did.

There is no contradiction in what I'm typing. The confusion comes from you refusing to see what's plainly in front of your face. You want to argue with me, basically simply because it's me, but in order to do so you have to argue against common sense and what everyone in the US knows about buying things as consumer already understands.

So, exactly and expertly define a core, right here and right now, I will wait.
 
So, exactly and expertly define a core, right here and right now, I will wait.
Why would you ask me that when I've been explaining that cores aren't well-defined, yet?

The fact that you asked that question of me indicates you haven't read and comprehended the argument I laid out.

If a core isn't well-defined, it can't be marketed.
You are implying that a core isn't well-defined (you're going to wait for a definition from me), so therefore if you understood my argument you would be agreeing with me that AMD shouldn't be marketing a term that is not well-defined.

Otherwise you are arguing for an environment where I have some concept of a core in my mind, you have some concept of core in your mind, Intel uses it in some particular way, and AMD is free to interpret it in some other way, and all of these companies can simply use various features of a "core" to try and upsell their products to the public when in fact there is no standard definition yet.

Why would any consumer argue for a context within which that can occur? That's just opportunity for industry to make up definitions along the way and sell us whatever they say matters at the time and then switch it when it no longer servers their purpose, which is how things have been so far with predictably disastrous results.
 
Pretty straightforward.

It's even obvious to you, but since you want to try and "win" this argument rather than learn something about product law and understand the principles I'm laying out relatively in clear terms, you continue to load your statements up with caveats.

You know it's not enough to simply claim, "8 cores equals more performance" yet that's what AMD and Intel both did.

There is no contradiction in what I'm typing. The confusion comes from you refusing to see what's plainly in front of your face. You want to argue with me, basically simply because it's me, but in order to do so you have to argue against common sense and what everyone in the US knows about buying things as consumer already understands.

AMD did not claim that their 8 cores beat Intel's 8 cores. They simply stated 8 cores equal more performance. They were not specific what more performance was relative to outside of a few canned marketing benchmarks, so you have no grounds to assume they meant compared to Intel quad cores in all scenarios, or whatever you want to assume. If you assume something and bought wrong, the onus is on you for being a dumb idiot and not bothering to do your due diligence.
 
Why would you ask me that when I've been explaining that cores aren't well-defined, yet?

The fact that you asked that question of me indicates you haven't read and comprehended the argument I laid out.

If a core isn't well-defined, it can't be marketed.
You are implying that a core isn't well-defined (you're going to wait for a definition from me), so therefore if you understood my argument you would be agreeing with me that AMD shouldn't be marketing a term that is not well-defined.

Otherwise you are arguing for an environment where I have some concept of a core in my mind, you have some concept of core in your mind, Intel uses it in some particular way, and AMD is free to interpret it in some other way, and all of these companies can simply use various features of a "core" to try and upsell their products to the public when in fact there is no standard definition yet.

Why would any consumer argue for a context within which that can occur? That's just opportunity for industry to make up definitions along the way and sell us whatever they say matters at the time and then switch it when it no longer servers their purpose, which is how things have been so far with predictably disastrous results.

I asked you to straight up define it and you attempted to claim you cannot do so. Therefore, there has never been any cores, by your definition, not from Intel, not from AMD, not from IBM, not from whatever is in our phones, nothing. Then if, as you claim, a core is then defined, every single core that does not exactly match up with that in the past will need to be sued and that company will have to give money back to all purchasers. :D ;)
 
That's not at all what I wrote; but your post does fit the description of someone who wants to argue without engaging in a good faith discussion.
 
AMD did not claim that their 8 cores beat Intel's 8 cores. They simply stated 8 cores equal more performance. They were not specific what more performance was relative to outside of a few canned marketing benchmarks, so you have no grounds to assume they meant compared to Intel quad cores in all scenarios, or whatever you want to assume. If you assume something and bought wrong, the onus is on you for being a dumb idiot and not bothering to do your due diligence.
I'm not assuming anything.

People in this thread are arguing that one can not clearly define a "core."
If that is true, when you put "8 cores" on the box you are opening yourself to a lawsuit.

This doesn't have anything to do with relative performance (I'm certain I already wrote this); the only relevance performance has in this discussion is if the consumer wasn't harmed there would have been no standing in court. Beyond that, however, the relative performance of 8 core AMD chips to 4 core Intel chips is irrelevant. The real meat is going to be how AMD defines a core and whether that lines up with what the industry has been using as an operational definition for long enough.

If AMD has been using "core" in line with an informal industry standard, then AMD will prevail in court. If they have not, then AMD will lose in court. Regardless of whether they win or lose, however, that doesn't render the case frivolous. There is still a debate in our culture and law as to what does or does not define a "core" and this definition needs to be hammered out over time. That's the part that seems to be flying over your head in your rapid defense of AMD. It's still a necessary case. We can't live, purchase, or develop technology in a world where the terms aren't defined.
 
So, exactly and expertly define a core, right here and right now, I will wait.
Zarathustra already provided a definition a post or two above your question. Did you miss it? Here it is:
I can very clearly define a core[...] It is the integer portion of a traditional CPU. It does not include any caches, or FPU's.

The question is not my personal definition of a core. The question is whether you agree with his definition or whether you think that definition is in dispute. So far, numerous people in this thread have claimed that a core is not defined, can't be defined, and/or shouldn't be defined.

I don't hold that opinion, but given that opinion exists means there is confusion or debate over the salient aspects of a core. As long as there are still factors in dispute, then they must be defined. Until they can be defined, they can not be the basis of one's marketing.
 
I'm not assuming anything.

People in this thread are arguing that one can not clearly define a "core."
If that is true, when you put "8 cores" on the box you are opening yourself to a lawsuit.

This doesn't have anything to do with relative performance (I'm certain I already wrote this); the only relevance performance has in this discussion is if the consumer wasn't harmed there would have been no standing in court. Beyond that, however, the relative performance of 8 core AMD chips to 4 core Intel chips is irrelevant. The real meat is going to be how AMD defines a core and whether that lines up with what the industry has been using as an operational definition for long enough.

If AMD has been using "core" in line with an informal industry standard, then AMD will prevail in court. If they have not, then AMD will lose in court. Regardless of whether they win or lose, however, that doesn't render the case frivolous. There is still a debate in our culture and law as to what does or does not define a "core" and this definition needs to be hammered out over time. That's the part that seems to be flying over your head in your rapid defense of AMD. It's still a necessary case. We can't live, purchase, or develop technology in a world where the terms aren't defined.

You keep claiming that the issue is that "core" isn't defined, and then you keep falling back to performance as to why it needs to be defined. And yet you also claim performance doesn't matter. So which is it? Get your argument straight.

And yes, this lawsuit is frivolous. The only consumers that were harmed by the 8-core marketing were the ones that deserved to be harmed by it. There is absolutely no merit to having "core" defined the way the plaintiffs want it, or the way you want it for that matter.
 
Zarathustra already provided a definition a post or two above your question. Did you miss it? Here it is:


The question is not my personal definition of a core. The question is whether you agree with his definition or whether you think that definition is in dispute. So far, numerous people in this thread have claimed that a core is not defined, can't be defined, and/or shouldn't be defined.

I don't hold that opinion, but given that opinion exists means there is confusion or debate over the salient aspects of a core. As long as there are still factors in dispute, then they must be defined. Until they can be defined, they can not be the basis of one's marketing.

It is possible to define a given architecture's core. The point is that the definition is not going to match in any meaningful way from one architecture to the next, which is what you're saying must exist.

I have worked on products which used a 6502 (embedded controller in a coffeemaker). That's a core. As is the Zen. A couple control systems used 386s.
What possible good could come of trying to extract some definition from these of those beyond "a set of execution units capable of processing an instruction stream"?
How would any definition you derive actually serve the customer? I don't know how to explain what a TLB is to Joe Public, or what cache associativity means, or pipeline stages, or numbers of ALUs in a compute cluster, etc.

A core is defined by a great many sub-components, all of which can and must vary to work in multiple segments. The core in my phone is different than my desktop. Different than the one in the microwave. By necessity.

What a customer cares about is what performance they will get, and they can absolutely obtain that information.
 
It is possible to define a given architecture's core. The point is that the definition is not going to match in any meaningful way from one architecture to the next, which is what you're saying must exist.

I have worked on products which used a 6502 (embedded controller in a coffeemaker). That's a core. As is the Zen. A couple control systems used 386s.
What possible good could come of trying to extract some definition from these of those beyond "a set of execution units capable of processing an instruction stream"?
How would any definition you derive actually serve the customer? I don't know how to explain what a TLB is to Joe Public, or what cache associativity means, or pipeline stages, or numbers of ALUs in a compute cluster, etc.

A core is defined by a great many sub-components, all of which can and must vary to work in multiple segments. The core in my phone is different than my desktop. Different than the one in the microwave. By necessity.

What a customer cares about is what performance they will get, and they can absolutely obtain that information.
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.
 
Back
Top