AMD's take on benchmarks.

[cageymaru]

The best benchmark is you.
http://www.techradar.com/us/news/computing-components/processors/the-best-benchmark-is-you-1296238

Agree? Not agree? When DX12 comes out how important are those single core tests going to be really? Interesting read though.

 

[durquavian]

I have debated this to death. There are two sides, both credible and both with positives and negatives. Unfortunately all benchmarks in reviews tend to favor one type of throughput = single thread and none speak to user's everyday use. So consumers get skewed outlooks on what they should expect.

DX12 will change the gaming benchmarks as games utilizing it surface, but for at least a short time we will still see DX11 and those single core games flooding the suites. Never really did see the point of flashing 800x600 benches to show CPU use when the 1440/4K hell even 1080p numbers were lower than the lowest on the CPUs. Look at any thread on games and see that eyecandy is the highest metric and that alone diminishes the CPUs involvement to max FPS.

 

[Quix]

This reads like the sort of marketing fluff piece you write when your product isn't competitive. DirectX 12 will mean that CPU performance as a whole will be much less important. The biggest feature DirectX 12 brings is not better multi-thread performance but much less overhead overall.

Will FX series processors be more competitive in games under DirectX 12? Sure, but so will all lower-end processors, at least until the game developers start using that extra CPU power that's been freed up.

AMD made the decision to give up on bulldozer years ago and hasn't released a new performance CPU design since Vishera in 2012. They decided to sink all their development money into the replacement "Zen" core. AMD understands what their problem is and are working hard to solve it, this is just them trying to hold the line until their real solution is released.

P.S. Most non-gaming benchmarks are extremely multi-threaded and even AMD's top-end chips are still not competitive with Intel's mid-range i5s so we can't expect a miracle.

 

[durquavian]

This reads like the sort of marketing fluff piece you write when your product isn't competitive. DirectX 12 will mean that CPU performance as a whole will be much less important. The biggest feature DirectX 12 brings is not better multi-thread performance but much less overhead overall.

Will FX series processors be more competitive in games under DirectX 12? Sure, but so will all lower-end processors, at least until the game developers start using that extra CPU power that's been freed up.

AMD made the decision to give up on bulldozer years ago and hasn't released a new performance CPU design since Vishera in 2012. They decided to sink all their development money into the replacement "Zen" core. AMD understands what their problem is and are working hard to solve it, this is just them trying to hold the line until their real solution is released.

P.S. Most non-gaming benchmarks are extremely multi-threaded and even AMD's top-end chips are still not competitive with Intel's mid-range i5s so we can't expect a miracle.

Read more. 1st vishera is piledriver, we now have excavator Carrizo that has SB on die so there have been updates. Second there are no real multi thread benchmarks, well maybe a scarce few, but rather a plethora of single thread duplicated on multiple threads. Besides that the article was written by a member of AMD so there is a slight slant. But it doesn't change the message. Read it again and this time with an open mind and devoid of you apparent disdain.

 

[Araxie]

Second there are no real multi thread benchmarks, well maybe a scarce few, but rather a plethora of single thread duplicated on multiple threads. .

not really.. I think its all the contrary.. in the past few years we have more and more multi thread benchmarks.. not gaming but for all of things of real world.. Asus RealBench I think its the best exponent... but also, Frybench, Cinebench, and even real apps used as benchmark like POV Ray, Handbrake, as just examples for both sides.. 3DMark Physic test its also heavy in Multi-Thread... with the arrival of DX12 multi threaded bench will be just more popular..

 

[pxc]

Sounds like AMD doesn't expect its products to be performance competitive in the way they're being evaluated now and is trying to replace shift it to APU-centric benchmarks, particularly the relative strengths of the GPU, which may or may not be important for certain customers. Throw in buzzwords and that's pretty much the message.

The problem with the strategy is that it's not going to help in the largest segments of the x86 market (business/enterprise PC sales, which don't care about iGPU performance or HSA), or the most profitable (commodity x86 servers). There's definitely a segment for APUs to excel, but it's in low margin products, which is the same problem AMD has faced since Llano was introduced.

If AMD's power targets are met with new products, that solves only half its problem: performance is still an issue. Given that Zen is supposed to come out next year, it's not very reassuring that AMD is shying away from competing on CPU performance.

 

[CaptNumbNutz]

Will FX series processors be more competitive in games under DirectX 12? Sure, but so will all lower-end processors, at least until the game developers start using that extra CPU power that's been freed up.

Exactly. DX12 will also affect Intel CPU's, a fact that desperate AMD fanboys keep ignoring. DX12 is not a saving grace for AMD, not to mention it isn't even released yet and there are very few DX12 games being released in the next year. Providing a competitive product for OEM's to consider in more machines, producing enough to satisfy OEM needs, and decent marketing to sway average joe's opinion will be their saving grace.

When's the last time you have seen an AMD TV commercial, or any commercial for a PC with an AMD logo present? I don't think I ever have, even in their heyday of K7-K8. They lost the high end enthusiast and server markets, and thus are losing word of mouth through their most prominent supporters. Every person I ever talk to that has little knowledge of computers knows Intel, but not AMD. The few that do bitch about the shitty laptop they had with an AMD chip, and thus associate shit quality with AMD and the OEM (HP, Dell, etc.) instead of just the OEM.

When you start seeing Dell TV commercials for decent looking high end desktop and laptops while prominently seeing AMD logos on screen is the day that AMD starts turning major profits. Those high end machines have more value as marketing tools and opinion swayers than anything else. Average joe may not be able to afford them, but they see high end and see quality, much like how a few Corvettes can help sell a ton of Chevy Impalas through brand recognition.

This piece is more bullshit from AMD, nothing more. It sickens me they can't figure out better marketing. They desperately need to. If Zen becomes a killer product but they can't market it, they will still utterly fail.

 

[Pieter3dnow]

Sounds like AMD doesn't expect its products to be performance competitive in the way they're being evaluated now and is trying to replace shift it to APU-centric benchmarks, particularly the relative strengths of the GPU, which may or may not be important for certain customers. Throw in buzzwords and that's pretty much the message.

The problem with the strategy is that it's not going to help in the largest segments of the x86 market (business/enterprise PC sales, which don't care about iGPU performance or HSA), or the most profitable (commodity x86 servers). There's definitely a segment for APUs to excel, but it's in low margin products, which is the same problem AMD has faced since Llano was introduced.

If AMD's power targets are met with new products, that solves only half its problem: performance is still an issue. Given that Zen is supposed to come out next year, it's not very reassuring that AMD is shying away from competing on CPU performance.

It depends on how you define computing all I know is PCMark 8 v2 or any other benchmark is just that it is no tool to purchase computers by if you are knowledgeable you never look at these.
AMD's problem is that it is to far ahead and the marketing side of things have trouble getting this across. AMD does try to show in several benchmarks that it is the champion and running several demanding tasks concurrently it beats the competition hands down.

AMD maybe a little tired of getting hit with the same stick and the article shows that.

Creating high-quality benchmarks that truly represent today's workloads and provide an accurate picture to users' needs should be inclusive of the whole industry. When the industry does not work together, it can result in benchmarks that are not representative of real-world tasks and can be skewed to favor one hardware vendor over another. This has happened before and while one hardware vendor may benefit, the real loser is the consumer who is presented with skewed performance figures and may pay for a perceived performance benefit.

How are you going to explain Mantle/DX12 to a new audience where AMD cpu that were "slow" are now matching or just better at high batch count scenarios.

Computing is changing and something that are better handled on the gpu should run there and if benchmark software does not reflect this then it is another reason not to use it.

 

[Pieter3dnow]

Exactly. DX12 will also affect Intel CPU's, a fact that desperate AMD fanboys keep ignoring. DX12 is not a saving grace for AMD, not to mention it isn't even released yet and there are very few DX12 games being released in the next year.

You do know that most engines which support Mantle will be updated for DX12 (not sure if that would mean that current catalogue would get a patch as well).

And you are right it is not saving AMD it is showing that when you have an API that is not crippling multi core it actually makes a huge difference.

Btw that Dell part was that on purpose the company which will not buy AMD even if Intel has no chips available ?

 

[Nenu]

You can benchmark with the actual application you wish to use, exactly like [H] video card and game reviews.
AMD didnt make any mention of this, it looks like they would rather side step the results of close inspection.

 

[Bowman15]

I agree with everything he says in the article. The problem is users know Intel is doing everything better with what he talks about. At some point if developers and programmers don't get on board with your architecture then its up to you too adjust. AMD is running out of excuses if not money.

 

[CaptNumbNutz]

You do know that most engines which support Mantle will be updated for DX12 (not sure if that would mean that current catalogue would get a patch as well).

And you are right it is not saving AMD it is showing that when you have an API that is not crippling multi core it actually makes a huge difference.

Games and game engines supporting Mantle now getting updated to support DX12 has no bearing on this at all. Please reference what Quix said above and what I quoted in my previous post.


DX12 will benefit Intel and AMD equally, with the end result being the same status quo as it has been for the past few years: AMD will still be behind on performance (at least until they put out a new architecture). This performance differential means jack shit to you and me who already know how much cpu power is really needed (an AMD cpu is good for most situations already). The difference is the minimum will be raised for both CPU's since DX12 is more about removing the CPU workload from the equation. Removing CPU workload affects ALL CPU's equally.

People are putting way too much faith into DX12. It isn't going to magically make that FX 8 core be utilized to its full potential and make its performance equal to every Intel chip. Even if it did, single Intel cores get nearly twice the workload done per single AMD core and you are right back to comparing an i5 quad core against an FX 8 core as has been the case for 5 years now. The only thing that is going to change this status quo is new CPU architectures such as Zen. This takes me back to the point of my post: AMD needs to get some higher performing parts out along with competent marketing to start swaying public opinion. DX12 is not going to change this.

Btw that Dell part was that on purpose the company which will not buy AMD even if Intel has no chips available ?

You are right. Dell may not have been the best choice of examples since they have been historically Intel only. My point still stands though. You get the OEM's putting out high end laptops/desktops with AMD in them and advertising them on tv clearly saying and or showing AMD in the commercial, they will start to take off just like the sports cars and Car manufacturers. The real term for this is "Halo Effect". People see quality elsewhere in the product line, they will start to assume quality across the board. AMD cpu's/apu's in only low end desktop's and laptop's means people are only associating AMD with the low end. That's how you get people like my naive father who won't buy a computer that has an AMD chip in it.

Get the OEM's to make high end products with AMD and market them as such, AMD will thrive by increased sales in all price brackets. The problem is convincing the OEM's who will need some solid evidence AMD chips outperform the Intel chips in power savings and performance. AMD will have to market themselves better. This is primetime for them to do it, especially since the Intel Anti-Trust situation is over. Putting out bullshit articles like the one in the OP is making things worse, not better.

 

[durquavian]

Sorry Capt but you like so many others are taking DX12 performance out of context. No one, well any one with a bit of intelligence, is stating DX12 Makes AMD on par with Intel, but rather alleviates any need, well most of it, for highend CPUs to run at performance levels. This means that even $100 CPU/APUs can produce equal or nearly equal results using the same GPU as >$300 varieties. This also means the GPU purchase is the bigger deciding factor almost to the point as the sole deciding factor when building a gaming machine.

It is even possible that my 8350 will out perform, in as little context as can be had, any i5 and even be on par with most i7s simply because I have the 8 cores. Or a better way to look at it is that Intel SB and up as well as AMD FX will now have even longer life than what we were used to because of DX12.

And as far as Putting out bullshit articles like the one in the OP is making things worse, not better. . Well you are what is making it worse. His points are quite valid and if one were to have an open mind they would see his point of view. Like I stated in my first post there are equally compelling arguments on both sides. His point is not to enthusiasts but to the general public. That is the first fact you have to consider, then after that everything else follows and in respectful order. Consider that any normal purchaser, talking the 95% or more of the purchasing public here, would not be able to tell the difference in use between an Intel and AMD. We might because we know what to look for, even then some who claim they can probably couldn't in reality. The authors point in that respect is quite true, for the majority of the public most of these benchmarks mean total jack. Cinebench does not say how ones experience logging into facebook and playing CandyCrush Saga will go. SiSaSoft doesn't speak to how Youtube will play.

Honestly I have never been able to get the same results for anything thus far that a reviewer got, mine was always better and a great deal of the time WAY BETTER.

As far as laptops I know a number of people personally that went from Intel to AMD on their laptops and like the AMD way more. Of course the quality of the setup could be the reason more than which would tech be better but it doesn't change their experience or what they concluded.

 

[jamesgalb]

i7s will benefit more than FX chips...

it will be interesting to see if FX benefits more than i5s...

The difference maker in the Intel vs FX improvements will be the iGP on Intels chips, which will be usable for post-processing effects and such.

All said, everyone is about to get a huge upgrade towards next generation games. Save your current systems when you upgrade next time, as the current generation will last a long time as its hardware gets fully utilized.

 

[CaptNumbNutz]

Sorry Capt but you like so many others are taking DX12 performance out of context. No one, well any one with a bit of intelligence, is stating DX12 Makes AMD on par with Intel, but rather alleviates any need, well most of it, for highend CPUs to run at performance levels. This means that even $100 CPU/APUs can produce equal or nearly equal results using the same GPU as >$300 varieties.

Hmmm. Let's try this again.

4 thread $130 AMD A10-7800 3.5ghz performs on par with 4 thread $130 Intel i3-4330 3.5ghz using same GPU (GTX 770). Fucking brilliant AMD. That's what we want to see. Now just scale this up to your high end and MAKE SURE AVERAGE JOE KNOWS THROUGH MARKETING.

Please oh please show me where a $100 AMD CPU/APU will perform on par with a $300 Intel Core i7. I don't doubt that it can happen but I have yet to see any legitimate site prove it. Currently the closest comparison might be an FX-6300 at $110 versus a $300 Core i7-4770k.
But that is still beside the point. I'm not here to argue DX12 and what its capable of doing.

Let's examine again what I said. I was agreeing with Quix. Let me quote things again:

Will FX series processors be more competitive in games under DirectX 12? Sure, but so will all lower-end processors, at least until the game developers start using that extra CPU power that's been freed up.

Exactly. DX12 will also affect Intel CPU's, a fact that desperate AMD fanboys keep ignoring. DX12 is not a saving grace for AMD, not to mention it isn't even released yet** and there are very few DX12 games being released in the next year. Providing a competitive product for OEM's to consider in more machines, producing enough to satisfy OEM needs, and decent marketing to sway average joe's opinion will be their saving grace.

**In the context of that sentence I meant "released to the masses via Windows 10 retail release." I know full well you can experience DX12 now. Don't pick the fly shit out of the pepper.

Are you understanding my point yet? You are quoting only one of my posts without acknowledging the previous one where I gave the context behind what I said. I am not arguing whether or not DX12 will help AMD's benchmarks, to which AMD themselves obviously want everyone to ignore anyways per OP article. I am arguing it won't help their bottom line. I even agreed that DX12 will help them somewhat and even make an FX chip perform on par... at least until what Quix stated starts happening...developers start using that extra CPU power that's been freed up. When they do, those cases where FX is on par with an i7 will start to disappear again. That will be true until AMD can get faster parts out the door such as Zen. Quarterly loss after quarterly loss with so few periods of profit. They are already on the razor's edge. They need something big, and they need it now. Marketing is it.

And as far as Putting out bullshit articles like the one in the OP is making things worse, not better. . Well you are what is making it worse. His points are quite valid and if one were to have an open mind they would see his point of view. Like I stated in my first post there are equally compelling arguments on both sides. His point is not to enthusiasts but to the general public. That is the first fact you have to consider, then after that everything else follows and in respectful order. Consider that any normal purchaser, talking the 95% or more of the purchasing public here, would not be able to tell the difference in use between an Intel and AMD. We might because we know what to look for, even then some who claim they can probably couldn't in reality. The authors point in that respect is quite true, for the majority of the public most of these benchmarks mean total jack. Cinebench does not say how ones experience logging into facebook and playing CandyCrush Saga will go. SiSaSoft doesn't speak to how Youtube will play.

I am not the one making it worse. AMD does that well enough on their own. My mind is very open to the possibilities.

And once again, you completely missed my point. I already acknowledged that average joe can't determine the performance difference on your average pc or laptop. I was pointing out the Halo effect that AMD is missing out on. This attitude that AMD presents of "concede all other performance segments except low-end" IS what's hurting them. They don't market themselves enough.

I will say it again. HALO EFFECT. AMD needs it bad. They never run TV commercials and general populace knows fuck all about AMD other than "I saw that logo once on a sticker that I peeled off". You know all those Microsoft Surface and Intel 2 in 1 Laptop commercials? Lets just say hypothetically AMD got the design win for Surface and all those Ultrabooks. You get the slick HP, Dell, Lenovo, or even Samsung TV commercial saying "with the amazingly powered all in one AMD APU processors...blah blah blah AMD... blah AMD...[the word AMD stated another 10+ times]" end of commercial AMD logo prominently displayed and you get an AMD guitar riff or something...anything to stick in the mind of the consumer.

THAT'S WHAT THEY NEED. Not DX12. DX12 won't save them. They need to start making products that get high end design wins with the OEM's, then let the OEM's start making some of these commercials for them.

Stating benchmarks suck because average joe knows fuckall won't save them. They are right, benchmarks mean nothing to average joe, they make that abundantly clear. My point, which you still haven't got, is for them to present the article in the OP is terrible marketing. You and I know where AMD's performance really is real world, and that is why I call this a bullshit fluff piece of an article that doesn't help them. They are telling enthusiasts who visit an enthusiast website how the best benchmark is you. Well no shit AMD, us enthusiasts already fucking knew that. Stop preaching to the choir and tell everyone else that through mainstream media and not tech enthusiast websites. That's why I called it a bullshit article.

Sounds like AMD doesn't expect its products to be performance competitive in the way they're being evaluated now and is trying to replace shift it to APU-centric benchmarks, particularly the relative strengths of the GPU, which may or may not be important for certain customers. Throw in buzzwords and that's pretty much the message.

The problem with the strategy is that it's not going to help in the largest segments of the x86 market (business/enterprise PC sales, which don't care about iGPU performance or HSA), or the most profitable (commodity x86 servers). There's definitely a segment for APUs to excel, but it's in low margin products, which is the same problem AMD has faced since Llano was introduced.

If AMD's power targets are met with new products, that solves only half its problem: performance is still an issue. Given that Zen is supposed to come out next year, it's not very reassuring that AMD is shying away from competing on CPU performance.

This man gets it.

 

[durquavian]

Sorry Capt but you show your ignorance so well. Again you take points out of context. Not the least to mention STARSWARM isn't a Gaming benchmark but just a batch count benchmark. And by the way even it doesn't change what I said. Is the problem here that you just cant comprehend what I stated or are you too blind to the truth?

Looks more to me like you are scared of the truth or you refuse to see it. With my current setup I play Witcher3 full Ultra, VSR @ 3200/1800, HW on but using AMD CCC tweak, FPS capped at 60 and it plays wonderfully, only hitches in cutscenes with HW but only slightly. Now how is Intel going to change that? It wont. Performance for my Monitor is already maxed (60hz) so by your logic Intel should increase my performance but how is that possible? Car analogy: one 4 cylinder one 8 cylinder and both limited to 60 mph. So how is the 8 cylinder which obviously has more power going to change the experience at 60 mph? It wont seeing both can do it. Same case here seeing how the Majority of users are using 1080p on Desktop. Your elitist view doesn't span the whole of users.

Like I said you HAVE to read the article with an open mind and understand his point of view. Obviously you are having issues doing this as at no point to you give credence to his argument. You don't have to agree, but that doesn't make his point less valid. You act like they are saying there is a conspiracy and this is somehow a marketing ploy, and to a small degree it is. But more so this is a thought opening post that hopefully will make some who weren't more aware now there is more than meets the eye when it comes to benches.

Like CINEBENCH is not a Multi-thread processing bench in the true meaning of multithread. It is a bench of multiple instances of single thread work. Each thread operation is independent of the other hence not true multithread. True multi-thread would require 2 or more thread for a single operation thereby cutting the time taken for said task.

Maybe if you stop looking at this from the doom and gloom perspective you might just understand it better. I KNOW that most cant tell the difference between an Intel and AMD system in the mainstream. Many a test with Games were done and a majority actually preferred the AMD setup thinking it might have been the Intel. These were not extreme setups where it might have been far easier to tell but it doesnt change the facts.

Nor was the posters point about Halo products or how to sell more, it was a point in fact of the industry today and the many cases where some manufacturers cheated to show they had superior equipment. Where even some tried and true benches are not so agnostic whether by design or simply a sign of the times. Just because AMD doesn't lead the score doesn't mean the bench is unfair, this isn't the point of the article. But rather a word of caution and a recommendation. Is your assertion here that the bench suite he mentions not as agnostic as he claims?

At any rate you seem to be doing what most fanboys do, whether you intend it or not. You are sidestepping the point and true nature of the message, taking it out of context and adding points to it by inference when at best it is a long shot. The simple message, nothing else, is that with the market changing and so many new software changes imminent ie: W10, HSA the tried and true software bench suites will not adequately depict the true nature of the field. He asks that one look into the bench suites that do for better or worse. Simply it, that is all.

 

[Michaelius]

There's nothing difficult here to understand - if AMD chips were topping performance benchmarks this article wouldn't exist.

It's simple damage control PR

Texts likes this are a reason why we have hundreds of webpages dedicated to testing pc hardware product instead of relying on company marketoids.

 

[babadook]

CINEBENCH is not a Multi-thread processing bench in the true meaning of multithread. It is a bench of multiple instances of single thread work. Each thread operation is independent of the other hence not true multithread. True multi-thread would require 2 or more thread for a single operation thereby cutting the time taken for said task.

What apps do you have in mind? Where does AMD shine and it isn't reviewed properly?

http://www.phoronix.com/scan.php?page=article&item=amd-a10-7870k&num=4
Some Linux app testing with fairly unusual apps. As always, the A10-7870K trades blows with the Core i3, while being outclassed by the i5.

Same in gaming benchmarks. I've never seen one where AMD doesn't lose by a fairly wide margin to any i5. And once I'm buying a midrange (~250$) GPU, I probably won't care if the i5 adds another Benjamin to my bill.

I've grown up on AMD CPUs. Even when Core 2 Duo swept the world, I held out. But at some point even I had to admit they dropped the ball and Sandy Bridge won me over.

And I was a Radeon guy. Ever since the 9500 Pro could be pencil-modded to a 9700 Pro they had my heart. Last card was the 5870. Man, what a babe. But then they started relabelling like crazy and delaying needed refreshs. So at some point I had to admit they dropped the ball and got myself a GeForce 680.

If AMD manages to make Zen at least somewhat competitive, I'll buy it. But the whole R9 300 disaster, where they'll rebadge the 7970 AGAIN, doesn't give me any hope.

DX12 won't help them one bit either. Game devs will just use the freed up performance for AI and physics.

 

[xTrident]

Car analogy: one 4 cylinder one 8 cylinder and both limited to 60 mph. So how is the 8 cylinder which obviously has more power going to change the experience at 60 mph? It wont seeing both can do it. Same case here seeing how the Majority of users are using 1080p on Desktop. Your elitist view doesn't span the whole of users.

When all is equal and 60 mph is the limit, sure. But put those two engines under stress. Example would be a trailer with weight. All things continue to be equal where the 4 cylinder starts to struggle at a given weight and not maintain 60 mph the 8 cylinder will maintain said speed and continue to carry more weight at that speed.

You mention your machine playing Witcher3 just fine. What about the next game, or game engine? I think Capt's point is Intel is still going to have more headroom and that DX12 is just a temporary quick fix. At least based the comments I've read that's what I'm getting out of it.

 

[drescherjm]

Car analogy: one 4 cylinder one 8 cylinder and both limited to 60 mph. So how is the 8 cylinder which obviously has more power going to change the experience at 60 mph?

Not in the case of AMD versus Intel. More accurately the 8 cylinder has around 1/2 the horsepower per cylinder ( Intel's wide versus AMD's narrow cores) in the case of AMD so in a best case scenario the 8 cylinder matches the 4 cylinder.

 

[Michaelius]

Cylinders in this example weren't analogy to number of cores

 

[durquavian]

Boy you guys just keep grasping at straws. So tell you what, trying to stay on original topic, I will explain what I think his intent was.

Lets start with Mantle. Original i am sure AMD took this on as a possible sales technique but as it went on realized that they could force earlier adoption of dx12 thereby making far greater utilization of their architecture. This turned out to be a great strategy. And honestly the first time in 10 years that the software market changed so dramatically.

Now AMD wants to change the benchmark space because as we see with HWbot alot of them are either still using archaic code or optimized heavily in favor of one manufacturer or another. In most cases AMD strengths are not given any light (having 10% of the market doesn't help with this either). So what do you do if you are AMD? You reach out to a benchmark provider and try to get some benches that accurate illustrates what you bring to market.

In this particular case I am certain he is directly referencing HSA. Do you blame him? If you have something that makes even your most paltry processor seem elite, would you not want the world to see it? This was the point of the whole article. He could have blamed Intel or the benches, whining and such but he didn't. He articulated calmly and professionally. The true intent was to change the criteria of benchmarks, get some buzz if only to question what was going on.

No part of that post was flaming marketing. There is a counter argument to his point that this far not one of you has mentioned, I inferred it earlier in this post. Griping about sales and smoke and mirrors shows either you didn't really read the article or lack the understanding of his point.

 

[CaptNumbNutz]

Sorry Capt but you show your ignorance so well. Again you take points out of context. Not the least to mention STARSWARM isn't a Gaming benchmark but just a batch count benchmark. And by the way even it doesn't change what I said. Is the problem here that you just cant comprehend what I stated or are you too blind to the truth?

Looks more to me like you are scared of the truth or you refuse to see it. With my current setup I play Witcher3 full Ultra, VSR @ 3200/1800, HW on but using AMD CCC tweak, FPS capped at 60 and it plays wonderfully, only hitches in cutscenes with HW but only slightly. Now how is Intel going to change that? It wont. Performance for my Monitor is already maxed (60hz) so by your logic Intel should increase my performance but how is that possible? Car analogy: one 4 cylinder one 8 cylinder and both limited to 60 mph. So how is the 8 cylinder which obviously has more power going to change the experience at 60 mph? It wont seeing both can do it. Same case here seeing how the Majority of users are using 1080p on Desktop. Your elitist view doesn't span the whole of users.

Like I said you HAVE to read the article with an open mind and understand his point of view. Obviously you are having issues doing this as at no point to you give credence to his argument. You don't have to agree, but that doesn't make his point less valid. You act like they are saying there is a conspiracy and this is somehow a marketing ploy, and to a small degree it is. But more so this is a thought opening post that hopefully will make some who weren't more aware now there is more than meets the eye when it comes to benches.

Like CINEBENCH is not a Multi-thread processing bench in the true meaning of multithread. It is a bench of multiple instances of single thread work. Each thread operation is independent of the other hence not true multithread. True multi-thread would require 2 or more thread for a single operation thereby cutting the time taken for said task.

Maybe if you stop looking at this from the doom and gloom perspective you might just understand it better. I KNOW that most cant tell the difference between an Intel and AMD system in the mainstream. Many a test with Games were done and a majority actually preferred the AMD setup thinking it might have been the Intel. These were not extreme setups where it might have been far easier to tell but it doesnt change the facts.

Nor was the posters point about Halo products or how to sell more, it was a point in fact of the industry today and the many cases where some manufacturers cheated to show they had superior equipment. Where even some tried and true benches are not so agnostic whether by design or simply a sign of the times. Just because AMD doesn't lead the score doesn't mean the bench is unfair, this isn't the point of the article. But rather a word of caution and a recommendation. Is your assertion here that the bench suite he mentions not as agnostic as he claims?

At any rate you seem to be doing what most fanboys do, whether you intend it or not. You are sidestepping the point and true nature of the message, taking it out of context and adding points to it by inference when at best it is a long shot. The simple message, nothing else, is that with the market changing and so many new software changes imminent ie: W10, HSA the tried and true software bench suites will not adequately depict the true nature of the field. He asks that one look into the bench suites that do for better or worse. Simply it, that is all.

Boy are you dense. Read much? You have horrible reading comprehension. I even bolded, underlined, and even increased the size of my lettering in hopes that you would get past this benchmark bullshit argument which I already agreed with you several times and am not arguing to begin with.

I pointed out StarSwarm to congratulate AMD as a sign of what was possible. Thus agreeing with you. I even asked for a comparison of an FX versus an i7 because I would like to see some progress but haven't seen any yet. But that was beside the point. I immediately stated, underlined, then bolded where I was going with the discussion and how it was sidetracking my point.

I'm not scared of any truth. You keep bringing the DX12 argument back into play like it will save AMD. I DON'T FUCKING CARE ABOUT THAT ARGUMENT AS I AGREED WITH YOU MULTIPLE TIMES IN AN ATTEMPT TO GET PAST IT AND GET BACK TO MY POINT.I shouldn't have to yell in a forum. I even pointed out that DX12 wasn't even the point of the OP article. I fucking get it. I stated before the article makes great points but in the end is pointless bullshit because they are stating the obvious to the enthusiasts that already know this information. THIS IS MY POINT. You even have the gall to say I'm sidestepping the argument, saying people are grasping at straws when you are the one beating your chest with your personal Witcher3 benchmarks that no one fucking cares about and has nothing to do with the OP article.

In the end it won't make the difference in AMD's bottom line. I would love nothing more than AMD to create hardware that stomps Intel, but they did that before with K7 and K8 and still got nowhere. Anti-Trust issues are gone. Its time for AMD to start marketing heavily and getting in the public mindset.

I'm done discussing this with you because you obviously won't even try to understand where I'm going with this.

 

[Ramses]

I've no doubt Intel has "better" CPU's in a number of ways and has had for awhile, hell I even have an i7 laptop, but I really think there ought to have been a place in the world for "good enough", which is exactly what AMD cpus have been in my experience.

I went from an 8350 to a 9590 to a 4790K, there is jack-all difference in even my relatively heavy use.
I enjoy a good benchmark and game some but nothing crazy. I'm (fairly) sure the 4790K will be relevant longer, but as of now and a year ago and quite possibly a year from now at the rate things are going, they were sidegrades for all intents and purposes. Nor did my utility bill drop on a PC I use for working from home and personal use that is on for 15 hours a day or better. I was happy and vocal about AMD and put my money where my mouth was for an Intel setup (with faster ram and ssd at the same time no less), and all it's done is prove to me that performance has plateaued in a practical sense for most users. People ask me what CPU or such to buy and I tell them whatever's cheaper with four cores unless they are doing something fairly specific that needs otherwise. Or you want to run a lot of benchmarks.

 

[xTrident]

I've no doubt Intel has "better" CPU's in a number of ways and has had for awhile, hell I even have an i7 laptop, but I really think there ought to have been a place in the world for "good enough", which is exactly what AMD cpus have been in my experience.

I went from an 8350 to a 9590 to a 4790K, there is jack-all difference in even my relatively heavy use.
I enjoy a good benchmark and game some but nothing crazy. I'm (fairly) sure the 4790K will be relevant longer, but as of now and a year ago and quite possibly a year from now at the rate things are going, they were sidegrades for all intents and purposes. Nor did my utility bill drop on a PC I use for working from home and personal use that is on for 15 hours a day or better. I was happy and vocal about AMD and put my money where my mouth was for an Intel setup (with faster ram and ssd at the same time no less), and all it's done is prove to me that performance has plateaued in a practical sense for most users. People ask me what CPU or such to buy and I tell them whatever's cheaper with four cores unless they are doing something fairly specific that needs otherwise. Or you want to run a lot of benchmarks.

I think there has been, for a really long time. And that's basically the 95% of consumers not enthusiasts being referenced in this thread. If the machine gets the job done, what do they care?

And that's what Capt' is talking about when referencing AMD's marketing. So DX12 could vastly increase AMD's performance compared to Intel... What does it matter if AMD doesn't tell the world? Instead us enthusiasts continue to see it, read about it etc, and also just as Capt' said we already know AMD CPU's are plenty good enough. But we're not the people that need to know...

 

[CaptNumbNutz]

And that's what Capt' is talking about when referencing AMD's marketing. So DX12 could vastly increase AMD's performance compared to Intel... What does it matter if AMD doesn't tell the world? Instead us enthusiasts continue to see it, read about it etc, and also just as Capt' said we already know AMD CPU's are plenty good enough. But we're not the people that need to know...

THANK YOU.

Hopefully others read what you just typed. I was beginning to think my point was flying over everyone's heads.

 

[Ramses]

I think there has been, for a really long time. And that's basically the 95% of consumers not enthusiasts being referenced in this thread. If the machine gets the job done, what do they care?

And that's what Capt' is talking about when referencing AMD's marketing. So DX12 could vastly increase AMD's performance compared to Intel... What does it matter if AMD doesn't tell the world? Instead us enthusiasts continue to see it, read about it etc, and also just as Capt' said we already know AMD CPU's are plenty good enough. But we're not the people that need to know...

I had an experience that was sorta sad too with my i7 laptop, as an AMD fan. I actually bought an AMD laptop, I forget what it was but I was perfectly happy with the APU in it, games a little, ran everything I'll ever run on a laptop fine, $400 or so, trouble was it was a steaming pile of trash surrounding a perfectly OK APU. Case, keyboard, trackpad, etc, all junk. I don't buy laptops new sight-unseen so I went to the box stores and such and played with some more, and all the AMD models were pretty much junk. I ended up spending twice as much for an HP with an i7, and promptly but an SSD and lots of RAM in it just like I did with the AMD example. I can understand a $400 laptop being kinda crappy (even though my last $400 tk-57 amd lappy was fine), and I can understand a grand for an i7 HP(sorta), what I don't understand is why I couldn't spend say, $600-$700 on an AMD based piece that had a better chassis/kb/tpad/etc. I wouldnt' be surprised to find that they exist somewhere, but they were not to be had in any of the two decent sized cities I looked at here in central CA. I can't help but think that sort of thing is hurting AMD when they ought to be competitive. The average bulk consumer barely even has a chance to buy an AMD based anything.

 

[durquavian]

Boy are you dense. Read much? You have horrible reading comprehension. I even bolded, underlined, and even increased the size of my lettering in hopes that you would get past this benchmark bullshit argument which I already agreed with you several times and am not arguing to begin with.

I pointed out StarSwarm to congratulate AMD as a sign of what was possible. Thus agreeing with you. I even asked for a comparison of an FX versus an i7 because I would like to see some progress but haven't seen any yet. But that was beside the point. I immediately stated, underlined, then bolded where I was going with the discussion and how it was sidetracking my point.

I'm not scared of any truth. You keep bringing the DX12 argument back into play like it will save AMD. I DON'T FUCKING CARE ABOUT THAT ARGUMENT AS I AGREED WITH YOU MULTIPLE TIMES IN AN ATTEMPT TO GET PAST IT AND GET BACK TO MY POINT.I shouldn't have to yell in a forum. I even pointed out that DX12 wasn't even the point of the OP article. I fucking get it. I stated before the article makes great points but in the end is pointless bullshit because they are stating the obvious to the enthusiasts that already know this information. THIS IS MY POINT. You even have the gall to say I'm sidestepping the argument, saying people are grasping at straws when you are the one beating your chest with your personal Witcher3 benchmarks that no one fucking cares about and has nothing to do with the OP article.

In the end it won't make the difference in AMD's bottom line. I would love nothing more than AMD to create hardware that stomps Intel, but they did that before with K7 and K8 and still got nowhere. Anti-Trust issues are gone. Its time for AMD to start marketing heavily and getting in the public mindset.

I'm done discussing this with you because you obviously won't even try to understand where I'm going with this.

Actually went back and read what you wrote ALL OF IT and read it again. Sorry but your communication skills are terrible. This is what you wrote first that I responded to:

Exactly. DX12 will also affect Intel CPU's, a fact that desperate AMD fanboys keep ignoring. DX12 is not a saving grace for AMD, not to mention it isn't even released yet and there are very few DX12 games being released in the next year. Providing a competitive product for OEM's to consider in more machines, producing enough to satisfy OEM needs, and decent marketing to sway average joe's opinion will be their saving grace.

My response was to the fact that you seem to allude someone made the assertion that DX12 made AMD on par with Intel and no one stated that. Hell the person that you responded to for that comment was responding to me. Of course then you move to sales and how to get them which again as I have been trying to do for most all of this thread was point out that is not the PRIMARY point or intent of the article.

And honestly your demeanor makes it quite hard to tell what your stance really is:

Please oh please show me where a $100 AMD CPU/APU will perform on par with a $300 Intel Core i7


Now after you stated you were agreeing with me I went back and read your posts again but as you can see in the above underlined you don't sound like you are agreeing with anyone but in conflict with others points, or in this case since you were responding to me, conflicting with my view.

At any rate you really never commented on the article or his point as I have numerous times. Sales or how to attain them is not really on topic. try responding to this instead: (my post)

Lets start with Mantle. Original i am sure AMD took this on as a possible sales technique but as it went on realized that they could force earlier adoption of dx12 thereby making far greater utilization of their architecture. This turned out to be a great strategy. And honestly the first time in 10 years that the software market changed so dramatically.

Now AMD wants to change the benchmark space because as we see with HWbot alot of them are either still using archaic code or optimized heavily in favor of one manufacturer or another. In most cases AMD strengths are not given any light (having 10% of the market doesn't help with this either). So what do you do if you are AMD? You reach out to a benchmark provider and try to get some benches that accurate illustrates what you bring to market.

In this particular case I am certain he is directly referencing HSA. Do you blame him? If you have something that makes even your most paltry processor seem elite, would you not want the world to see it? This was the point of the whole article. He could have blamed Intel or the benches, whining and such but he didn't. He articulated calmly and professionally. The true intent was to change the criteria of benchmarks, get some buzz if only to question what was going on.

No part of that post was flaming marketing. There is a counter argument to his point that this far not one of you has mentioned, I inferred it earlier in this post. Griping about sales and smoke and mirrors shows either you didn't really read the article or lack the understanding of his point.

 

[CaptNumbNutz]

Actually went back and read what you wrote ALL OF IT and read it again.

At any rate you really never commented on the article or his point as I have numerous times. Sales or how to attain them is not really on topic.

You sure you went back and read what I wrote? I don't think you did. I commented on the article 4 separate times. Sales and Marketing is completely on topic, because that is exactly what this article is. It is public relations, it is marketing, it is an article pandering to the enthusiast community. If you can't see that, you are blind.

I will quote it all again because your reading comprehension sucks.

Post#7

This piece is more bullshit from AMD, nothing more. It sickens me they can't figure out better marketing. They desperately need to. If Zen becomes a killer product but they can't market it, they will still utterly fail.

Post#12

Putting out bullshit articles like the one in the OP is making things worse, not better.

Post#15

My point, which you still haven't got, is for them to present the article in the OP is terrible marketing. You and I know where AMD's performance really is real world, and that is why I call this a bullshit fluff piece of an article that doesn't help them. They are telling enthusiasts who visit an enthusiast website how the best benchmark is you. Well no shit AMD, us enthusiasts already fucking knew that. Stop preaching to the choir and tell everyone else that through mainstream media and not tech enthusiast websites. That's why I called it a bullshit article.

Post#23

I stated before the article makes great points but in the end is pointless bullshit because they are stating the obvious to the enthusiasts that already know this information. THIS IS MY POINT.

As you can see. I have quite clearly commented on the article. The article is bullshit PR damage control. It is definitely beautifully written with wonderful points to be made, but ultimately a fruitless endeavor because it was written for enthusiasts and posted on an enthusiast tech website. I don't need to breakdown all the points in the article, as I agree with most of them already and do not care to address them. Why can't you understand that I can read the article, fully comprehend the article, agree with the article, and at the same time call a spade a spade? I'm calling it out as a bullshit PR article, nothing more. I'm allowed to have an opinion on the article, and my opinion has been stated.

You have called me ignorant no less than twice already, and have called me a fanboy. I am neither. My only goal from the get go when I replied to Quix was state my opinion on this piece.

try responding to this instead: (my post)

I don't need to reply to your other arguments, because I have either agreed with them already on some level by not arguing them at all, or simply no longer give a shit.

If you think my communication skills are terrible, fine. I don't know how else to clearly convey my point to you especially when other's in this thread already stated they understand. Just delete everything I have typed from that head of yours and insert this man's point instead. Admittedly, he put it much clearer than I did. He said my viewpoint in one post.

There's nothing difficult here to understand - if AMD chips were topping performance benchmarks this article wouldn't exist.

It's simple damage control PR

Texts likes this are a reason why we have hundreds of webpages dedicated to testing pc hardware product instead of relying on company marketoids.

 

[durquavian]

Ok then try reading your own objectively, even these posts of yours you quoted. They have angry overtones and never speak directly to any point he made. So no you didn't comment on the article. Obviously you didn't understand his point or you didn't comprehend. You speak as if his point was marketing as to sell AMD hardware. Granted that would be a secondary point or effect of what his topic was. Hence why I used the Mantle example. Its primary objective was to get the software to change and then ultimately sell more GPUs and possibly more CPUs (when price becomes the bigger factor because performance is less an issue gaming anyway).

He speaks of benchmarks. So this is where you directly reference his point about benchmarks, as I did. You took the opportunity to thread crap on a point that at best was secondary to his point. If he really wanted to sell more CPUs/GPUs then he could have referenced the HP or Dell Laptop comparison between Intel and AMD where the performance for the real world task ( granted a bit specific a task so more akin to cherry picking) favored AMD slightly and was $300 cheaper. Or even the 9590 with 295X2 against the 6Core Intel and 295X2 where both gaming had very similar results to the point of being equal. Those two examples were not being discussed and had they been then you could talk Marketing toward sales all you want. But in this case it is a stretch.

As far as Michaelius his point is almost as bad as yours. The article shows absolutely no sign of inherent bias but rather a plea to pay attention to what Information you are getting. Just because an AMD employee is making the plea makes it no less valid. If they were on top of benchmarks they wouldn't write the article, well my god that is outstanding and so insightful (heavy sarcasm) or as I like to put it NO SH**. Stating the obvious doesn't make an insightful post. And keep in mind how long now, over 10 years, that AMD hasn't really been leading the benchmarks, a lot because of Intel immoral and illegal activity in the beginning. The writer posted this now so likely it has little to do with not leading the benchmarks as much as it has to do with new criteria and metrics that coming that aren't being utilized in a lot of the common benchmarks.

Now if you have issue with his stance on the current state of benchmarks, then now would be the time to counter argue his point. Like I said there is a good argument against that thus far none has brought up. Instead you and others have decided to bring hostile and overtly negative points that have very little to do with the original authors point. Take this moment to learn how to debate and stay on topic rather than look for opportunities to air your grievances.

 

[xTrident]

No way is David Bennett's primary point not to sell or an attempt to sell more "AMD" products. David hides said point well through the angle he takes giving his opinion on benchmarks. But I think you're buying into his sugar coating of wanting the industry to work together. I'm not saying he's wrong, because that would be great for the consumer but it's also unrealistic as he's talking about competing companies.

And the point stands that if AMD was winning the benchmark wars, and had been the past ten years - what would they care what the benchmarks measured? Their market share would probably be much larger if such had been happening. Hence the obvious point being made that such an article wouldn't exist and why Capt' references it as a bullshit piece. No way would AMD want the benchmarks adjusted to help their competition, and rightly so.

I don't have an issue with David's stance on benchmarks as I agree. But I think you're mistaken if you don't think David's underlying point was to put a positive spin on AMD's products in an effort to sell.

Anyway, I hadn't read the article before I was just going by what the people commenting in here had said. Now that I read the piece, that's the way I took it.

 

[Pieter3dnow]

DX12 will benefit Intel and AMD equally, with the end result being the same status quo as it has been for the past few years: AMD will still be behind on performance (at least until they put out a new architecture). This performance differential means jack shit to you and me who already know how much cpu power is really needed (an AMD cpu is good for most situations already). The difference is the minimum will be raised for both CPU's since DX12 is more about removing the CPU workload from the equation. Removing CPU workload affects ALL CPU's equally.

You don't get it the multi core addressing GPU is important where you had _only_ 1 core talking to the GPU single thread performance was king....

The part where it relies on 1 core now is gone, AMD was always considered as having weak cores and this is where it will shine if you want to see some stats on this you can visit old AMD GPU14 day (oxide presentation) and it has some batch count numbers of certain cpu running star swarm under Mantle.

And to a degree you are right but on higher batch counts where AMD multiple cores are "better" suited for this so with DX12 the cores pretty much scale where Intel hyperthreading just is not good enough where it previously was considered "good".

That the graphics card does all the heavy lifting now everyone already knew but the part you are forgetting is rather important.....

 

[durquavian]

Not saying the end game would not be sales but the primary point here is benchmarks. Seriously you guys are trying hard to reach here beyond the scope of what he said. Reading far too much into it. FIRST he is trying to change the image of benchmarks or in the least draw a few references to the possibility of skewed results because of optimization based on particular hardware. SECOND he is recommending a benchmark suite that may more accurately detail the now or real world results (of course this benchmark could be the very same for his side as the others are against). THIRD assuming others reading take his point to heart and we get this shift then HERE AND ONLY HERE can we start talking sales.

As I stated earlier I think this benchmark has HSA hence why he would recommend. And yes at some point this could correlate to sales but again you guys keep putting the cart before the horse. I think he was professional, far more than some of you in response. There was an air of slant evident early in his article but it was slight and far less by a mile than a lot of review sites than many claim to be fair. The topic here is benchmark neutrality.

So did he lie at any point or make outrageous claims? Was he wrong in any way for attempting to shine a far more fair light on his companies product, whilst acting in the best interest of said company? Did he cross a moral line? These questions have largely gone unanswered. If an Nvidia or say Intel employee had written the article would your disdain still exist? Or do you think AMD should say nothing, do nothing to air their concerns even when done as calmly and professionally as possible?

 

[xTrident]

Not saying the end game would not be sales but the primary point here is benchmarks. Seriously you guys are trying hard to reach here beyond the scope of what he said. Reading far too much into it. FIRST he is trying to change the image of benchmarks or in the least draw a few references to the possibility of skewed results because of optimization based on particular hardware. SECOND he is recommending a benchmark suite that may more accurately detail the now or real world results (of course this benchmark could be the very same for his side as the others are against). THIRD assuming others reading take his point to heart and we get this shift then HERE AND ONLY HERE can we start talking sales.

As I stated earlier I think this benchmark has HSA hence why he would recommend. And yes at some point this could correlate to sales but again you guys keep putting the cart before the horse. I think he was professional, far more than some of you in response. There was an air of slant evident early in his article but it was slight and far less by a mile than a lot of review sites than many claim to be fair. The topic here is benchmark neutrality.

So did he lie at any point or make outrageous claims? Was he wrong in any way for attempting to shine a far more fair light on his companies product, whilst acting in the best interest of said company? Did he cross a moral line? These questions have largely gone unanswered. If an Nvidia or say Intel employee had written the article would your disdain still exist? Or do you think AMD should say nothing, do nothing to air their concerns even when done as calmly and professionally as possible?

I don't feel as though I'm reaching at all with my claims, because literally the first thing that popped into my head putting the pieces together (AMD representative talking about the bias in benchmarks) was that this was an article with a bottom line of making AMD look good/better than they currently do. I can't tell you how many times I've seen the same kind of statement made by enthusiasts in this forum and others. As for your three reasons - like I said before, I think you're buying into his sugar coating, which means David has done his job.

You agree that this could correlate to sales, which means you understand my point. Yes, I understand your stance that you think all he's getting at is benchmark neutrality. And I stated in my previous comment that I think that would be great, but unrealistic. Also, I don't think his only point is benchmarks. I think at face value David wants the reader to assume that is all he's talking about. I believe there is more to it.

As for the questions you ask at the end, I find them irrelevant if you're asking me as I'm not arguing/debating any of those. Was he wrong in any way for attempting to shine a far more fair light on his companies product? Absolutely not, but I ask what's the purpose of doing such? Why make claims that your product is better? So consumers choose your product over the competition, I.e. sales.

Finally, as for the statement of whether nVidia or Intel did the same, I was actually going to ask you same thing if you'd take the same stance. I don't have disdain for what David Bennett is saying. I can only say I agree with his premise so many times, but I believe his point in the end is to benefit AMD in an effort to increase sales. Whether that happens or not remains to be seen. Yes, he was professional and the article read that way. But when bias in benchmarks come up I almost immediately think of AMD vs. Intel, and we all know who typically wins in benchmarks.

Like I said... as enthusiasts we all know about this, and have for quite some time and yet AMD still continues to struggle. And that's where the conversation of DX12 increasing AMD's performance significantly needs to be marketed comes from.

 

[Sycraft]

Well I agree with the overall thought: The only benchmark that matters to me is wall time, as in how long does it take my computer to do what I want it to do. The problem for AMD is the answer with all the stuff I run is "Intel is way better." The reasons behind that can be argued: Maybe the program isn't multi-threaded enough, maybe it is only optimized for Intel CPUs, maybe it isn't well coded. Whatever the case I don't care, what matters to me is what is going to get me the best performance.

I find that AMD processors actually look the best in some synthetic tests. You do some various media encoding benchmarks or the like and the results are good. However in the program I use, which are mostly video games and pro audio software, Intel pulls way ahead. The reasons really aren't relevant since I have to deal with what it is, not what I wish it was.

 

[CaptNumbNutz]

Ok then try reading your own objectively, even these posts of yours you quoted. They have angry overtones...

When I have to reply to you multiple times, then that's what happens.

Everyone else in this thread understands what I have typed. Why can't you? That means your utter refusal to try and understand after my last post means you you are trollbaiting me to do something stupid.

Other people that disagreed with my post at least didn't state "I showed my ignorance" as a passive aggressive personal attack. This forced me to reply in kind.

...and never speak directly to any point he made. So no you didn't comment on the article.

I read the article. I commented on the article, multiple times. I said I agreed with his opinions and gave my own opinion that it was a bullshit pr writeup. Those are comments. They don't have to address specific things in the article to be comments. I can comment and provide an opinion on the article as a whole. If you don't know what a comment is, then I will provide the definition:

com·ment
/käment/
noun
1. a verbal or written remark expressing an opinion or reaction.
"you asked for comments on the new proposals"
synonyms: remark, observation, statement, utterance

I also stated why I didn't comment directly to any point he made: 1. I agreed with most or the points made in the article, to which I told you multiple times. 2. I'm not required to analyze any of what is in the article. You aren't the comment police. 3. I have no desire to breakdown my analysis of the article. 4. Even if I did have the desire to break down the article at some point, I no longer do because of your frustratingly stupid responses.

Stating the obvious doesn't make an insightful post.

I am not required to make an insightful post. I can post whatever opinion I feel like. If you don't understand it or simply don't like it, find someone else to talk to.

Now if you have issue with his stance on the current state of benchmarks, then now would be the time to counter argue his point.

I don't because I no longer care. How many times do I have to repeat myself? I told you this in my last post.

Please do me a favor. Add me to your ignore list and don't reply to me again.

 

[Ramses]

Well I agree with the overall thought: The only benchmark that matters to me is wall time, as in how long does it take my computer to do what I want it to do. The problem for AMD is the answer with all the stuff I run is "Intel is way better." The reasons behind that can be argued: Maybe the program isn't multi-threaded enough, maybe it is only optimized for Intel CPUs, maybe it isn't well coded. Whatever the case I don't care, what matters to me is what is going to get me the best performance.

I find that AMD processors actually look the best in some synthetic tests. You do some various media encoding benchmarks or the like and the results are good. However in the program I use, which are mostly video games and pro audio software, Intel pulls way ahead. The reasons really aren't relevant since I have to deal with what it is, not what I wish it was.

The "fastest is always best" is interesting in a philosophical sense. Apple for instance isn't fastest very often, they so seldom update their hardware, but the whole scene and approach makes up for it since the hardware is "good enough" until it's crazy out of date.
Obviously AMD hasn't made any headway really with "fast enough" and cheaper, but you'd have thought they would. I think that's the root of so much Intel bashing over their business policies(which I have no idea if it's valid or not), it really seems like AMD should have done OK, but they just didn't. There are a lot of things in life I don't do the easiest or fastest way because some other slower older way is more satisfying or better for whatever reason, you see that everywhere, except computers. It's interesting. There are people that are specifically into vintage computing, and I'm one of them in a very minor way, but it's a crazy small minority.

 

[CaptNumbNutz]

No way is David Bennett's primary point not to sell or an attempt to sell more "AMD" products. David hides said point well through the angle he takes giving his opinion on benchmarks.

Exactly. It's one of many reasons I and others in this thread call it PR talk. David Bennett is Corporate Vice President, Worldwide Commercial Sales at AMD. He is one of their public facing representatives of the company. You have to look at the motives behind the article with a healthy dose of skepticism.

 

[durquavian]

When I have to reply to you multiple times, then that's what happens.

Everyone else in this thread understands what I have typed. Why can't you? That means your utter refusal to try and understand after my last post means you you are trollbaiting me to do something stupid.

Other people that disagreed with my post at least didn't state "I showed my ignorance" as a passive aggressive personal attack. This forced me to reply in kind.
I read the article. I commented on the article, multiple times. I said I agreed with his opinions and gave my own opinion that it was a bullshit pr writeup. Those are comments. They don't have to address specific things in the article to be comments. I can comment and provide an opinion on the article as a whole. If you don't know what a comment is, then I will provide the definition:

com·ment
/käment/
noun
1. a verbal or written remark expressing an opinion or reaction.
"you asked for comments on the new proposals"
synonyms: remark, observation, statement, utterance

I also stated why I didn't comment directly to any point he made: 1. I agreed with most or the points made in the article, to which I told you multiple times. 2. I'm not required to analyze any of what is in the article. You aren't the comment police. 3. I have no desire to breakdown my analysis of the article. 4. Even if I did have the desire to break down the article at some point, I no longer do because of your frustratingly stupid responses.

I am not required to make an insightful post. I can post whatever opinion I feel like. If you don't understand it or simply don't like it, find someone else to talk to.

I don't because I no longer care. How many times do I have to repeat myself? I told you this in my last post.

Please do me a favor. Add me to your ignore list and don't reply to me again.

This is me defying you! You don't even know your own posts do you. That first one I Quoted was your first post in this thread.

Exactly. DX12 will also affect Intel CPU's, a fact that desperate AMD fanboys keep ignoring. DX12 is not a saving grace for AMD, not to mention it isn't even released yet and there are very few DX12 games being released in the next year. Providing a competitive product for OEM's to consider in more machines, producing enough to satisfy OEM needs, and decent marketing to sway average joe's opinion will be their saving grace.


See that bolded yellow part in your own post. Not professional or helpful criticism in the slightest. That set the tone from there. This is the very behavior that needs to be curbed in any thread, but seems quite prevalent in AMD threads.

[/FOR EVERYONEB] not anyone in particular for this thread. This article on its own WILL not sell a single extra AMD product other than the possible person that felt he was professional and decent, which likely wont happen. So going the Marketing ...blah,blah,blah has no impact nor direct reference to the article. It as has been establish could setup the scenario to do the aforementioned sales marketing pitch but that intent is quite devoid in this article.

PURPOSE: change the current benchmark landscape.
EFFECT: showcase what he believes is a more accurate depiction of his products.
POSSIBLE RESULTS OF EFFECT: Increased sales and general knowledge of said product.

He doesn't even mention Intel or even use AMD anywhere in the article itself. Of course we know it is AMD from the header and the description of the author at the end but also the mention of APUs and HSA, AMD sponsored acronyms. This leads to the conclusion he wasn't asserting his company to be dominate or even superior. He does allude to the current state of benchmarks don't adequately show the true performance of his companies product but doesn't state where they actually would stack in the hierarchy.

So sorry I don't see this as MARKETING PR. Had he mentioned the 7850k beating the Extreme series CPUs from Intel then yes MARKETING PR, as we know that is a specific set of circumstances for that to happen.

Still so many in here agree with his assertion yet mark the whole of the article off to PR. That makes no sense at all, in rational thought. So if he is correct about benchmarks and their inability to keep up with the current state of hardware and available metrics, for the most part, then how is it PR(not speaking in the broad sense but in regard to the usual terms of spin and damage control as some have referred to here)?

 

[xTrident]

Still so many in here agree with his assertion yet mark the whole of the article off to PR. That makes no sense at all, in rational thought. So if he is correct about benchmarks and their inability to keep up with the current state of hardware and available metrics, for the most part, then how is it PR(not speaking in the broad sense but in regard to the usual terms of spin and damage control as some have referred to here)?

It's easy - David Bennett is Corporate Vice President, Worldwide Commercial Sales at AMD.

If that's true I'd think it's rather obvious it's a marketing ploy.

I agree with David's thoughts, but I don't think that's his point... I really don't know why that doesn't make sense...?