anand's jaguar article

[Lepton87]

Too far from jaguar discussion...no one else has some insightful analysis?

Yep, but sadly the topic derailed a long time ago Also, there's probably nothing to add about jaguar that wasn't already said.

 

[Abwx]

You also know that Intel bribed OEM's to *exclude* AMD

In Europe they also bribed the retailers.

One of the biggest campaign in history to keep the consumer
from buying competing product while ripping him off with
high CPU prices that would had been unsustainable
without theses practices.

All this was paid at the consumer expenses.

 

[galego]

If a benchmark is bad you don't use it, regardless of why it's bad. If you realize a benchmark is bad you retract what you said about it. You aren't doing this. You instead keep going on about things like benchmarks where FX-8350 is 42% faster than i7-3770K (I did a summary of all of Phoronix's benches in their article and never found anything that fast) and ignore all the ones where it's slower. You say other benhcmarks are cheating but you automatically assume that some totally anonymous score that you can't even cite a reference for is not just legitimate but also fair. You don't know this.

You never even clearly wrote anything about your anonymous unixbench scores not being final or applicable. This is all you ever said about it:

It was the stuff about theoretical GFLOPS that you said wasn't final. Can't even keep your own arguments straight...

When a benchmark is bad for you, then you don't use, but different people has different meanings for "bad".

I don't know why you mix now unixbench with phoronix, because them don't use that benchmark. In any case the 42% faster (more exactly 41.6%) was first introduced by other poster in AT, and has been quoted before by several posters in AT. You can search the 41.6% here on AT or you can go to the source

http://openbenchmarking.org/embed.php?i=1210227-RA-AMDFX835085&sha=293f200&p=2

Of course, there are benchmarks where the i7 wins. Nobody said the contrary. But overall both the FX and the i7 are on the same performance envelope.

I have explained you my point. I have explained you that I am waiting for more benchmarks before evaluating the performance of jaguar. You continue denying it. Why don't look to what I said on #64, #74, #92, #94, #165,... (bold added now):

Now, I prefer to evaluate the performance of the jaguar cores using theoretical values and benchmarks as the one that cited in a previous post whereas I wait for more benchmarks.

I am evaluating the Jaguar benchmarks that I know, whereas waiting for more.

I will wait to see some realistic benchmarks (lacking the Cripple_AMD function evidently) and will run my own benchmarks as well, but it looks as AMD has clear winner here.

I am anxiously waiting for openbenchmarks on jaguar based chips!

And some of those are in replies to you. Now stop.

Are there any benchmarks for VIA CPUs with the default and the modified string? I'd be curious how much performance difference we are talking about. If it's 5% it's a completely different story then 40%.

Arstechnica got a 47.4% magic increase in performance on a PCMark score between the VIA chip detected as "CentaurHauls" and the same chip detected as "GenuineIntel". The chip moved from slower than the Intel chip used in the comparison to outperforming the Intel chip. The benchmark was giving the fake sensation that Intel chip was better.

According to Fog research ICC-based benchmarks can give to Intel an artificial gap of between a 10% and a 15% approx. Curiously, this is about the difference between one or two successive generations of Intel chips.

That is the psychology. AMD is pure and noble, uncorrupted and incorruptible. Everyone else is the son of the devil.

I doubt anyone believe that about AMD. But the facts are that Intel has been caught lying, cheating benchmarks, cheating drivers, and involved in other dishonest and illegal tactics many times and in many countries. Some facts:

There is a Cripple_AMD function in Intel compiler, but there is no Cripple_Intel function in AMD compiler...

We know AMD, VIA, and Nvidia abandoned BABPco with accusations of bias. Next are the words of one member of the SPECpower subcommittee when he was working at the BAPCo consortium’s benchmarking committee:

constantly butted heads with Intel, and some of the correspondence generated during that time demonstrates the extreme and outrageous measures Intel takes to influence that consortium.

Now explain to me, if Intel chips are so great why do they spend these extreme pressures to cheat benchmarks such as Sysmark 2012 for favouring their chips over the competence? If the Intel chips were so fast as some believe, any fair benchmark would show it. No?

These are rethoric questions. The reason why Intel cheat and lies on benchmarks, drivers... is because their chips are not so good as their marketing says us.

Under openbenchmarks (aka open source, i.e. no cheating) a FX-8350 is within a 10% difference of the performance of an i7-3770k (HT enabled).

That is why Intel expend a considerable amount of time and money to cripple the competence in any imaginable way: compilers, benchmarks, drivers, games, OEMs, reviews...

 

[Abwx]

Now explain me, if Intel chips are so great why do they spend these extreme pressures to cheat benchmarks as Sysmark for favouring their chips over the competence?

True that Sysmark is the typical rigged bench in favour of Intel ,
this dates back from the Athlon XP era when all tests that favoured
AMD were removed and replaced by tests using useless repetitives tasks
to brute force the final outcome in favour of the pentium 4.

The fact that Anand still use it in his CPU comparison charts
is a testimonybthat this site is not neutral.

Among other , according to the FTC :

64. Several benchmarking organizations adopted benchmarks that measured performance of CPUs running software programs compiled using the Intel compiler or libraries. Intel’s deception affected among others, the Business Applications Performance Corporation (“BAPCo&#8221, Cinebench, and TPC benchmarks.

http://sharikou.blogspot.fr/2009/12/ftc-accuses-intel-of-rigging-benchmarks.html

About Sysmark dark history :

http://www.vanshardware.com/reviews/2002/08/020822_AthlonXP2600/020822_AthlonXP2600.htm

 

[krumme]

Its easy to understand Otellini paying Michael. It was an unsecure time for Intel, and for the first time, they had to really fight. In that situation Otellini does what he is trained to do, and what his experience tells him.

Comming from a sales role, as account manager, he just did what he learned there, to use all available tricks from the sales handbook to win. I dont think he is a bad person at all, he just come from a profession that in many ways have turned more and more rotten the last 50 years. Sale today is full of sexual language - "lets go fuck xxx", power language, and for that reason host many half psychopats. Otellini have no such attributes, but beeing in that culture is bad, and thats what led him do the choises 10 years ago.

Failure to see the above also mean failure to see what Otellini really did that was of major importance to Intel, and why its so profitable today. He introduced a business sence in the culture, and a more innovative thinking; centrino, atom. There is lots of examples. As many engineer heavy organizations Intel is difficult to turn, but Otellini did more than anyone else could have done in his position.

 

[Abwx]

Failure to see the above also mean failure to see what Otellini really did that was of major importance to Intel, and why its so profitable today.

It is profitable today because he managed to keep the concurrence
from growing while they had the momentum , he has the same merit
as a bank robber that wasnt caught and used the money to open
a few businesses.

 

[Vesku]

Again there is cultural context, even after celebrities and business people have been convicted of crimes they are often still admired in the US by a surprising amount of people. There is also an appreciation for not getting caught/convicted even if the evidence is damning.

Heck if the Enron folks had done a bit less visible damage to people's livelihoods they'd probably have a few public fans.

Check out the current Governor of Florida - http://en.wikipedia.org/wiki/Rick_Scott

"In December 2000, the U.S. Justice Department announced what it called the largest government fraud settlement in U.S. history when Columbia/HCA agreed to pay $840 million in criminal fines and civil damages and penalties. Among the revelations from the 2000 settlement, which all apply to when Scott was CEO, were that Columbia overbilled Medicare for unnecessary tests and false diagnosis codes."

 

[ShintaiDK]

Again there is cultural context, even after celebrities and business people have been convicted of crimes they are often still admired in the US by a surprising amount of people. There is also an appreciation for not getting caught/convicted even if the evidence is damning.

Heck if the Enron folks had done a bit less visible damage to people's livelihoods they'd probably have a few public fans.

Check out the current Governor of Florida - http://en.wikipedia.org/wiki/Rick_Scott

"In December 2000, the U.S. Justice Department announced what it called the largest government fraud settlement in U.S. history when Columbia/HCA agreed to pay $840 million in criminal fines and civil damages and penalties. Among the revelations from the 2000 settlement, which all apply to when Scott was CEO, were that Columbia overbilled Medicare for unnecessary tests and false diagnosis codes."

When the entire political foundation is build on corruption. It becomes the norm and widely accepted as a standard.

 

[ViRGE]

Yeah, we're done here.
-ViRGE