Slashdot Mirror
Better Business Bureau Targets Apple's G5 Ads
deathazre writes "The Council of Better Business Bureaus has suggested Apple Computer withdraw its claims of the world's fastest, and first 64-bit, PC after a complaint by Dell. However, even having one of their ads banned in the U.K. didn't stop them here in the States."
I have never understood.....when companies lie (or make questionable claims) about "fastest" or "first" people get all up-in-arms. But when companies lie about quality or other things people don't care.
Case in point: I didn't see Apple (or anyone else) complaining that Dell was lying in ads where they implied that the night call center workers were actually in the US.
And Dell directly implies in their ads that if you're running Unix there are all sorts of negative things that will happen to you that are pretty much not true.
Whatever. The BBB is just another piece of the marketing machine I guess.
I was wondering how Apple gets reprimanded for these claims, yet MS goes unheeded with all of its comparative TCO mumbo jumbo, and the fact that "exagerated advertising" is, well, normal. My only guess is that Apple was just too straightforward with its claims (gee, what a novel concept!), and didn't throw in enough technical buzzwords and marketing FUD to confuse everybody.
Apple's current stance: Our G5's are damn fast, faster than Wintel stuff, and we'll stand by that claim.
Average Consumer: Really? Wow.
Dell: Crap.
Apple's new stance: Our G5's have multi-threaded double buffered optimized 256 bit parallel pipelined 64 bit x-streaming architecture!
Average consumer:
Dell: much better.
ce n'est pas un Sig.
Hmmmmm....I recall Windows NT Workstation and Windows NT Server. Ive never seen Windows NT PC.
As the Apple Turns has an insightful take on the whole thing. I had never heard of the Dell cluster in Buffalo. Bummer for Buffalo.
Boom Shanka
Made out?
Hmm.
While Apple's iPod sales remain strong -- the company sold 730,000 iPods in its last quarter and had pre-release orders for 100,000 iPod mini music players -- Bachman said the company's G5 sales could turn out to be disappointing for the second quarter in a row.
For its fiscal first quarter, which ended Dec. 27, Apple shipped 206,000 G5 computers, short of many analysts' estimates. Bachman has estimated that Apple would ship 195,000 G5s in its current quarter.
Sounds like it didn't work to me. They hardly 'made out' on that one.
That is if you believe nVidia's claim that the Xbox runs at 80 gigaflops.
It may be fastest, and may have been the cheapest 64 bit desktop PC system on introduction. A problem with making the claim now is that that eMachines has been selling Athlon 64 systems at about $1000 for three months now.
Apple has long shelved the ad campaign though.
If you want very specialized fast floating point performance for certain scientific applications, Itanium is where it's at.
If you want business performance, a multi-processor Opteron trumps them all.
If you want super fast video and audio encoding, a dual Xeon with hyperthreading will probably kill all of these.
If you want the fastest mobile processor with good battery life, Centrino is the way to go.
If you want the fastest computing cluster, you'd probably have to go with UVA's Mac cluster.
All of the above are probably temporarily in those positions and subject to considerable debate. I hate these "fastest computer ever" statements. Computing is far too dynamic and varied to put all-encompassing labels on. No one platform is ideally suited for all tasks. You make compromises and go from there.
Yeah. I especially like the Intel Centrino laptop commercials where someone is in the middle of the fucking ocean or on the airport runway surfing the internet.
"Intel Centrino Lets You Unwire Your Life. Now you can enjoy a truly mobile lifestyle!"
Unless Intel advocates wardriving, I don't see how it's truly mobile. This is far more deceptive to me than G5 ads, which were just designed to let people know that Apple isn't behind times with its processor speed anymore, albeit the wording was amplified to drive the point home.
Tim
Trying to see whether an ad campaign has succeeded or failed based on the number of sales versus what was predicted in a given quarter is kind of like trying to infer causal effects for a graph showing correlation between two independently measured variables with no other data--a big mistake.
Whether an ad succeeds or fails often cannot be measured based on sales in a fluctuating economy with a variable product interest. The critical question is would they have sold 206k G5 computers if they had not run the ad campaign? There is also the corollary question of would they could have sold more with a different ad campaign?
The first of these questions is nontrivial to answer and requires good, well researched data on why your customers are buying the product. Apple may have that data--you most certainly do not. The latter is almost impossible to infer even with good data on people's purchasing.
Integrate Keynote and LaTeX
Just for fun I installed Panther on a beige G3 300 box and it still boots faster than my faster 700 Mhz XP system. So... What about the GUI is slow?
.deviatefromtheabsolute.
Apple had the claim to fame for a few months fair and square. They don't run those commercials anymore anyway...and again why does Dell care...they don't sell AMD chips either!!!
He's also mad becuase HP went with Apple and not them and now HP will have it's own branded iPod.
First, the process is NOT tedious, at least in North Carolina. You fill out a form online, and mail it in. Also, our BBB shows how many complaints the company has had total, how many resolved, and how many unresolved. That tends to show people what type of business is being run. For example, our local Compusa has over 1000 complaints. 'Bout 1/2 of those are actually resolved.
I know this because I had to file a complaint against a local company. They were pretty upset to have any mark on the BBB site.
I've filed several Better Business Bureau complaints, including one against a company in the North Carolina (against Ubi Soft, Inc). I've never seen the BBB change the complaint statistics as a result of my complaints. Of the three times I've gone through the complaint process I have never received any satisfaction from the companies involved and the BBB has not added my complaint to the total complaints against the companies.
The BBB is a paper tiger that does little more than to allow people to feel like they are complaining to someone who can make a difference. The fact is that the BBB is not for the consumer, it is an organization which is paid by companies in order for them to get a feel-good BBB logo to plaster around and to serve as a buffer between themselves and the public.
If you have a problem with a company go to the Federal Trade Commission or your State Attorney General. They are much more likely to be able to give you some serious assistance in getting satisfaction from an abusive company.
Sapere aude!
How do you measure the speed of a computer?
Is a 3 GHz P4 with 4 MB RAM faster than a 1.6 GHz G5 with 16 GB RAM? Not likely. So it's not [only] the clock frequency of the CPU that matters.
A fast computer is, IMHO, one which allows me to perform a certain task in a short time. Think about it - isn't this what _really_ matters? Can I write my essays in a shorter time on a 2 GHz G5 than an old iMac? Not really.
Comparing Macs and Windows machines, I'd say Macs have always been faster, and will likely be for quite some time. Since the G5 is the fastest Mac, I'd say it's the fastest computer.
My two cents.
Sig Nature
My point was that Intel uses a circuit that's exactly as efficient as the G5: a P4 does four floating point add/multiply operations per clock cycle, just like the G5. When it comes down to absolute performance, using a well optimized compiler and the same algorithm, there's no difference between a P4, Athlon, or G5. I have actually tested this, using some linear algebra benchmarks that really depend on CPU power. Apple and AMD are lying when they claim superior performance.
Real math is more than just adds and multiplies. Try doing vector operations, or multiple input instructions, or any other number of complex operations. THIS is where the CPUs different.
Also, notice that most DSPs (and the G4/5s) have multiply and add instructions, that take one clock cycle, because that operation occurs so frequently in DSP. That's certainly not in base x86, and I'm not sure if it is or is not in any of the SSE instructions.
Also, your mention of compilers is telling. It takes 5 or 10 years for compilers to really mature on any given architecture. So when we're talking about processors that are only a year or two old, compilers can make a HUGE difference.
Not to mention all the scheduling of operations that takes place on the CPU, and the differences in how many parallel things they can do.
Just because the most efficient multiplier circuit may be well known doesn't mean there aren't still tradeoffs in CPU design.