Slashdot Mirror
Apple iPad 2 As Fast As the Cray-2 Supercomputer
An anonymous reader writes "Presenting at the IEEE High Performance Extreme Computing conference, a researcher from the University of Tennessee presented evidence that the iPad 2 is as fast as the original Cray-2 supercomputer. Performance improvements were made to the iPad 2 LINPACK software by writing Python for generating and testing various Assembly routines. The researcher also found that the ARM Cortex-A9 easily beats the NVIDIA/AMD GPUs and latest Intel/AMD workstation CPUs in performance-per-Watt efficiency."
Is more powerful than the Atanasoff machine!
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
9.80665 m/s^2
Seriously.
What fanbois won't say about Apple!
Now, were's the "Imagine a Beowulf cluster of iPads!" jokes?
Oh my god. If I have to read one more BS Apple story like this on the internet, I'm going to go nuts.
Apple lovers must be stopped. They're driving ad revenue and hits to all these *retarded* articles. They keep writing them because people keep clicking on them. STOP IT people!
Maybe I should just follow "if u can't beat em, join em." I should just post "Using an iPhone gives you crabs" or "iPhone as valuable as cream of wheat" and watch the money roll in.
I just laugh. Remember that new screw hoax? They said "they just make it too easy."
Need to keep your genetically engineered amusement park attractions under control? There is an app for that!
Newer smaller computer is faster than older larger computer! Some didn't think it possible, one of those people submitted the article under the false impression that anyone gave a fuck.
There is no memory shortage. yes I have heard of XFCE. Go away.
Does it run Linux?
Give me Classic Slashdot or give me death!
> Performance improvements were made to the iPad 2 LINPACK software by writing Python
You lost me there.
> ...for generating and testing various Assembly routines.
That makes more sense.
I was privileged to program on the Cray-2 back in the day. It was an awesome machine if you had the right kinds of problems for it to solve. My hat is off to the company who let me use the fastest computer in the world for my vi sessions :). That said it;s hardly surprising that the march of Moore's law has resulted in an iPad today beating a computer 13 or so years its senior.
This just proves that Cray copied off of Apple. Hopefully Apple will sue them into oblivion.
But the iPad 2 doesn't have reciprocal approximation in place of the outdated mathermatical operation called division. Wake up Apple, it's the 21st century.
. . . the article taketh away.
From the Phoronix article: "When benchmarking the Apple iPad 2, the University of Tennessee employee achieved 4 GFLOPS per Watt on the ARM SoC (measured at the chip level)."
The linked graphs don't have units on them, so I have to assume until proven otherwise that the article is correct. But performance per watt, while a valid comparison, doesn't equate to "faster than a Cray-2" in the sense I read the headline, since I assume the Cray-2 pulled quite a bit more power than the iPad. To be "faster than a Cray-2", you really would need a Beowulf cluster of iPad processors.
I am not a crackpot.
who has been publishing the Top 500 Supercomputer list for many, many years. I would bet that he ran Linpack himself on the Cray-2.
Decades of Moore's law and clever mniaturization lead us to freaking hipster filtering JPEG crap on Instagram, faster. Slow clap.
That article is a poorly-written piece of shit.
1. Is this operations per second per watt or operations per second?
2. Is this LINPACK metric something that exercises the Crey's massive pipeline architecture, where huge arrays of numbers (the vectors) were operated on at lightning speed through pipeline (assembly line-style) chip design? Or is it just a looping test?
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
Will the cray2 blend as well as the ipad2?
ARM Cortex-A9 easily beats the NVIDIA/AMD GPUs and latest Intel/AMD workstation CPUs in performance-per-Watt efficiency
So, it is saying that a car with an engine that can get 400mpg is more economical than one with 30mpg, but they leave out the important part that it will take you 10x longer to get to your destination. I hate the trite "typical marketing", but that is what this is
"When life gives you lemons, don't make lemonade. Make life take the lemons back!" -- Cave Johnson
The Cray 2 was a child's toy?
I guess that kid grew up to be a programmer or something.
I would hope we had advanced since then. The point of this 'revelation' was what? Click ad revenue generation?
---- Booth was a patriot ----
It has been demonstrated that the ipad 2 is lighter than an Apple II.
The ipad 2 user interface has been tested and proven much better than the Zilog Z80's.
On a blind test, the ipad's screen resolution has been voted subjectively better than the MSX's!
And an independent research confirmed that it has more available apps than the HP41C!
In a random test with a control group, 3 out of 5 teenagers prefer the ipad when offered the option of an ipad or a Newton, and 2 out of 4 girls prefer the ipad over Justin "Beaver".
Oh my God, the ipad is really the best thing in the whole universe! No, it has been demonstrated that it is better than 5 universes put together with whipped cream and strawberries on top!!
This far in, and still no comments about a Beowulf Cluster of iPads.
What has Slashdot become?
#DeleteChrome
All these great things that have been done on a Cray now equal the numbing stupidity of things like Facebook on an iPad?
"The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
The Cray connection made me thing of The Last Starfighter.
With all of the artist wannabe advertising approaches, you would think that Apple would be all over the idea of creating 3D models and rendering them on your portable touch screen Cray.
A Pirate and a Puritan look the same on a balance sheet.
... if dropped on your foot.
Wansu, th' chinese sailor
I find that to be simply amazing that hand-held consumer devices are only now matching and exceeding computing hardware that was invented over 27 years ago.
Apparently, Jack Dongara is smarter than you.
One of the slides is titled "Let's Do Assembly ... in Python"
Decades later, technological advancements have been made! Everyone (especially those brainwashed Apple cultists) was astounded when science decided it wasn't going to sit there and do nothing!
GP said: "A Cray 2 can pump out 1.9GFLOPS, an ipad2 can put out about 300-400Mflops. About as much as a early PentiumPro"
I am not sure a cray 2 can pump 1.9GFlop/s but an ipad 2 can do more than 300MFlop/s. The processor is clocked a 1 Ghz. It is dual core and it supports NEON vectorial instruction that can do multiple single floating point instruction at a time. (I am not sure exactly how many, I'd need to go to the actual spec, there might be fused multipy-add.) So I am pretty sure the theoretical peak performance is over 2GFlop/s. Probably around 4GFlop/s.
After that, I am not sure you can run linpackwith that speed. But 300MFlop/s seems quite low.
Moreover, I know Jack Dongarra: I trust his experiments.
http://bits.blogs.nytimes.com/2011/05/09/the-ipad-in-your-hand-as-fast-as-a-supercomputer-of-yore/
you need to remember, however, that the software for these consumer devices is nowhere close to that on the Crays. no optimization is done any more... for you script kiddies, "optimization" means you manually with the assembly language, or automatically in the compiler, try several things and pick the one that uses the least memory/processor cycles/OSPF if multithreaded/whatever based on what you want to gain by optimizing code. all this "include.kitchensink" stuff just packs in extra code crap in case any of it is needed.
and Clippy or Bob never ran on a Cray, either.
if this is supposed to be a new economy, how come they still want my old fashioned money?
imagine a Beowulf cluster of nuclear submarine fire-control computers...
imagine a Beowulf cluster of Beowulf clusters...
imagine a Beowulf cluster of fscks....
if this is supposed to be a new economy, how come they still want my old fashioned money?
how many furlongs per fortnight did each perform?
if this is supposed to be a new economy, how come they still want my old fashioned money?
Guess what? Computing power has increased exponentially in the last three decades. If this surprises you you're an idiot. I mean, ten years ago I paid $2,000 for a Toshiba laptop with a whopping 1 GB of RAM and a 20G hard disk; now I can get a Tracfone with better specs than that.
Slashdot sure seems to have its collective mouth wrapped tightly around the iSchlong lately.
According to ARM's documentation of the Cortex-A9 FPU it takes two cycles to perform one double-precision FMAC operation. 1 GHz * 2 cores / 2 cycles per operation = 1 GFLOP, assuming ideal conditions and zero overhead. In practical scenarios, 300-400 MFLOP is probably about right. And now that I think about it, the figure they list may already be for a dual-core, since Cortex-A9 was never intended to ship in a single-core configuration.
I remember during the 32 & 64 bit era a lot of people would mention console performance vs X-rays. Big deal, it's old hardware. Of course new stuff is more lots of power in a much smaller bit of kit.
And in other news, the Asus Transformer Prime is 4x as fast as the Cray. Android (NVIDIA Tegra 3 T30 1300 MHz (4 cores) ) vs Apple (Apple A5 (32nm) 1000 MHz (2 cores) )
I hate how everything must be compared against Apple iProducts. I don't recall every comparisons of yesteryear being brand specific. I don't care if the iPhoneX is 2x as fast as iPhoneX-1, or the iProductY is 2x as fast as the Cray. Give me damn benchmarks or clock speed of current day standards, and not a commercial.
Jimmy Kimmel recently went out on the street with an iPhone 4S and passed it off as the new iPhone 5 and asked people what they thought of it. Not one of them realized it was the old iPhone 4S
Why would they? No-one has seen an iPhone 5, so they would be inclined to believe someone who said it was regardless of what you gave them. I didn't get the point of this at all, even as a joke it was just absurd.
I mean, a guy with a camera crew shows up and hands you an object and says it is "X" and asks for thoughts. Are you going to disbelieve it's what they say it is?
"There is more worth loving than we have strength to love." - Brian Jay Stanley
They think the phone is faster because they're told it's faster, not because it is.
The new phone is actually faster, in many measurable ways. I assure you that when I get the newer phone I will be able to tell.
My point was that people who buy Apple products buy them because of brand identity
People buy things repeatedly because they liked how the older one worked and they'd like to repeat the same positive ownership experience with an updated model that has some improvement they like. It's about utility of products produced by a company and the customers trust level with a given company.
So your post seems to think of "Brand identity" as something to sneer at. But why is it bad for a person who had a good experience to think they can repeat it from the same source? That seems like a very reasonable assumption.
they believe it's because it's better, faster, whatever
But when it is in fact better and faster then why is it a problem that they believe it to be so? Shouldn't believing true things be desirable?
it's the result of marketing in general.
In general, I agree with some of the points you are making about marketing. They are trying to tell you something has some qualities or attributes you crave; sometimes they mislead.
But then you are trying to paint Apple users as misled and unable to discern truth from reality. In fact Apple as a company is better than most at giving you real information in marketing; showing actual apps being used, giving you realistic battery life figures.
Apple users on the other had are no different than anyone else, they are bombarded by marketing messages from all over, not just Apple. Again going back to the general case, people everywhere, not just Apple users but including them , are FAR more wary of marketing messages now than they have been in the past. And that is why people are happy to buy from the same company again when they have a good experience - not because of any marketing but because they don't have to RELY on marketing to help discover if a product is really good or not. That's not just true of Apple, it's just that Apple has had a better track record than a lot of companies in repeatedly making products that work well for most people who buy them.
It's also why people generally go to the same grocery stores and restaurants all the time. When a place does something you like, it's a lot less work to simply rely on them to provide a good experience going forward.
In the end misleading marketing has exactly the opposite effect intended; you draw a few more people in to start with but then when word of mouth gets around they are misleading, sales drop like a stone.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
New tech catches up to 20 year old tech! More at 11....
I think the Cray 2 could print something and it anticipated the future by not having USB or an SD card slot.
Sorry, but gray text on gray background is making my eyes bleed.
Processor clock speed doesn't equal sustained MFLOPage. As soon as the program counter jumps somewhere outside the instruction cache, you get hit with an eternity (in processor terms) delay waiting for the cache to get filled again by some algorithm that tries to predict where the Program execution will go next. Modern CPUs, both x86 and ARM types, rely on cache size to a huge degree. At the end of the day, the bigger cache size wins for real life problems, and that's a crucial spec that's hard to find for these vendor specific devices.
Otherwise referred to as "Link Bait"!!
You are right caches are important. I quoted theoretical peak. But here we are talking about dense linear algebra kernels. Matrix vector multiply for instance is a O(n^3) algorithm on O(n^2) memory. So overall you do more computation than memory transfer. With good caching techniques, tiling, software prefetching, you might be able to correct for it.
That's what these benchmarking techniques are designed to do. find the optimal blocking at every cache level and register blocking and software prefetching policy.
There are traces of single core cortex A9 on arm's website. But I don't know if there is one FPU or 2 FPUs. But I think there are two.
The scenario here is not a practical one, it is LinPack which is a quite dense computation. Many architecture reaches 90% of peak performance on linpack.
Also you are talking about double precision computation, I am not sure linpack is double precision, it might be single precision. (actually wikipedia confirms it is double precision http://en.wikipedia.org/wiki/LINPACK_benchmarks)
Some of these architectures (not sure about cortex a9) have a separate pipeline for the FPU and the ALU, allowing to perform integer operations and jumping logic in parallel to the floating point operations.
Finally, as you said, cortex A9 can perform one 64 bit FMAC in a cycle. FMAC is fused add multiply, that is to say out = in1 + in2 * in3, so there are 2 Floating point operations performed. ( http://infocenter.arm.com/help/topic/com.arm.doc.dui0068b/Bcfbbjgf.html )
In brief, if there is a single FPU, thanks to fused add multiply, peak performance is 2GFlop/s. If there are 2 FPUs, it is 4GFlop/s.
Maybe, in order to make news relevant to readers, they chose to compare to something most readers are familiar with? That's pretty much the point of analogies and comparisons.
Why do you think we ever talked about storage in terms of "Libraries of Congress" in the first place?
> Give me damn benchmarks or clock speed of
> current day standards, and not a commercial.
RTMFA! It has numbers. OF COURSE the summary has the appealing bits. Welcome to journalism. Welcome to the Internet. Welcome to the human species.
More info from a year ago: http://bits.blogs.nytimes.com/2011/05/09/the-ipad-in-your-hand-as-fast-as-a-supercomputer-of-yore/
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Two things come to mind:
1. Cleanliness of the phone. Phones get dirty in pockets. Screens get smudged with grease. Pocket lint gets inside screen protectors and cases. Ever clean your phone and then love how "new" it feels? Ever take off a beat-up old case?
2. Age of the OS. I don't have an iPhone, but it stands to reason that a fresh install will be more responsive than the one in your pocket that's been collecting pictures, emails, etc for a year. I know my old Android phone (HTC Magic with 2.2) behaves like this.
To "Motorola droid 10x faster than new iPad AND Cray-2"
Owners of a Cray can run any software they like without obtaining permission from it's manufacturer.
Advantage: Cray
-Lod
When you are talking about a battery operated device, performance per watt isnt a bad choice of metric.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
Why didn't those idiots just buy an iPad 2 then? Just goes to show how much money people will waste to be anti apple
it doesnt have nearly as much wirewrap as a Cray 2, thus making it lame
call us when you have an army of workers individually wiring up each bank of gates with a screw driver and a bit of wire ... IN PRODUCTION
And in other news, the Asus Transformer Prime is 4x as fast as the Cray.
It's much more impressive to compare a common toy of today to a serious workhorse of yesterday.
Turned out my reply made exactly the same point you did at the end of your post. I totally agree with you.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Other things that are as fast as Cray 2 supercomputer - about a million ancient PCs... but putting Apple in the title suddenly makes this news.
Suddenly, the iPad 2 (or a high end iPhone 5) doesn't seem all that expensive
Of course! It was even demonstrated in "Iron Sky" movie: http://www.imdb.com/title/tt1034314/
*hiding*
... of how far we've come is that a £250 graphic card (eg. OpenCL on a Radeon 5870) which can compute at 2.72 TFLOP/s would have been the world's fastest super computer in 1999.
BS Apple fluff is Phoronix's new gig, now that Steam is actually coming to Linux?
I still think it's pretty awesome that you can carry a device with the equivalent processing power of a device in 1979 that was so massive it had its own bench.
I wouldn't mind seeing more benchmarks of computational power/storage, etc. For instance, how much magnetic storage was there forty years ago? I have 13.6 tb on my home media server, and it sure would be nice to say that I have more storage than the entire world did in 1970...
Well, for a decade, everything was compared to the original IBM PC XT.
Non-Linux Penguins ?
Apple is the catalyst that will begin (or has already begun) the Stupocalypse where society becomes increasingly impressed and influenced by trivial nonsense to its ultimate demise.
I haven't thought of anything clever to put here, but then again most of you haven't either.
And in other news, the Asus Transformer Prime is 4x as fast as the Cray. Android (NVIDIA Tegra 3 T30 1300 MHz (4 cores) ) vs Apple (Apple A5 (32nm) 1000 MHz (2 cores) )
I hate how everything must be compared against Apple iProducts. I don't recall every comparisons of yesteryear being brand specific.
Well, then you have a memory like a sieve. Comparisons were always to something well known to the audience addressed - and judging by sales, even most Fandroids assume "Asus Transformer Prime" is an Autobot.
Of course news about a fake are Fake News.
Maybe, in order to make news relevant to readers, they chose to compare to something most readers are familiar with? That's pretty much the point of analogies and comparisons.
Or maybe it's what we call "Marketing".
Yeah, I'm sure Apple paid for that research. Which wasn't about comparing an iPad to a Cray BTW.
Of course news about a fake are Fake News.