Apple’s claims that the iPad Air is around twice as fast as the iPad 4 have been validated by Geekbench scores posted by Primate Labs. The iPad’s multi-core performance score increased by 88 percent, while the single-core performance jumped by 90 percent …
Geekbench 3 is a set of processor speed tests based on real-world tasks, and so reflecting the performance a user can expect to experience in everyday use.
With these kind of results, it’s perhaps unsurprising that Intel is joining the ARM party. CNET reports that the chipmaker has just announced it will be making its own 64-bit ARM processors, though it plans to push even further ahead: its chip will be quad-core, rather than the dual-core of the A7. Intel last made ARM chips back in 2007, when they were used in the Blackberry and Palm smartphones.
The benchmarks also reveal the dramatic difference between the iPad 2 – which Apple has retained as an entry-level full-sized iPad – and the Air: the Air is more than five times faster!
Primate Labs wonders why Apple has chosen to retain the older model given a relatively small price difference. We suspect it’s a decision heavily influenced by the education market, a key focus for Apple, where a difference of $100 per student adds up to substantial sums.
Primate Labs show that Apple has pushed the A7 chip in the iPad Air to 1.4GHz, 100MHz faster than the iPhone 5s, and speculates that this is due to the larger battery, larger chassis (allowing for better cooling) or a mix of the two. As mentioned in our review roundup, AnandTech’s tests suggest the latter may be a significant factor, performance in heavy-duty tasks dropping off only slightly over time compared to the iPhone 5s.
With this rate of progress, talk of future Macs running ARM chips instead of Intel ones does start to look a little less far-fetched. If the annual performance gains could be maintained, we’d have an ARM chip beating the performance of this year’s MacBook Air in less than three years.
Of course, by then Intel performance would also have advanced significantly, but the idea of ARM chips closing the gap doesn’t look to be an entirely crazy one.
FTC: We use income earning auto affiliate links. More.
The A7 has been a very impressive processor in the 5s thus far, so I’m looking forward to see what the future holds for ARM based chips.
In fact, last year, I said as much on my site http://www.tame-geek.co.uk/tech/apple-chips-the-move-from-intel-would-make-sense/
My iPad 3rd gen is getting a 766 in geek bench so now sure where these numbers are coming from.
I enjoy using my iPad 3, don’t have an issue with it performance wise. It’s kind of sad to essentially have a device that’s as powerful as something that was released in early 2011. Guess that’s the price you pay for being an early adopter of retina.
Apple screwed over iPad 3 owners in two major ways — one by discontinuing it, and two by lying about the performance and the “advanced graphics processor”
I will never again be an early adopter of any iDevice because of what Apple did with the iPad 3. It was / is / will always be inexcusable.
Did your iPad 3 suddenly stop functioning when they announced the iPad 4? Did it suddenly start becoming slower or less useful? I don’t understand your complaining. Technology is advancing faster than ever. I wonder, if Apple never released the iPad 4 and went straight to the Air after the year in between, would you be saying the same thing?
You don’t understand my complaining? The iPad 3 was taken off shelves faster than any other iPad in the history of the product line. It overheated, it’s slow, and it was touted as advanced, and it was expected that it would follow the same product life cycle as previous iPads. Yes, technology advances every day, so why not release a new iPhone every 7 months? Oh, that would piss a lot of people off wouldn’t it. But that’s exactly what they did with the iPad 3 — and nearly all the problems it had were remedied by the iPad 4; not to mention the iPad 4 is THREE TIMES faster than the iPad 3, while the iPad 3 aligns with the iPad 2 performance wise. And yes, when I upgraded my iPad 3 to iOS 7, it became very slow and stopped functioning what I would consider to be normally.
Well I have an iPad 3 and I’ve never had any issues with overheating or sluggishness. Neither have the 7 other members of my family with the same thing. I also manage iPads for an entire school district (600+ iPad 3’s) and have literally zero complaints about anything you talked about.
Maybe it’s just you.
Yeah, maybe it’s just me and the other thousands of people on Apple forums…not to mention that the iPad 4 was specifically re-designed to address the overheating issues. If it wasn’t a problem, why change the design to remidy a problem that never existed? :D
The iPad 3 was a overheating brick of a device and a big disappointment to me.
People will always defend Apple no matter what, but out of the iPads they have released it was definitely the “one to avoid.” Sadly, this year’s iPad mini is looking to fill the same kind of role. Thicker, heavier, almost certainly clocked lower than the full-sized one just to stay within the thermal envelope, and so very likely …. running hot as well.
If the current iPad mini could run iOS 7 with any skill or efficiency (it cannot), I would stick with it, but as a mini user, I’m pretty much forced to buy the new warm brick mini instead unless I switch back to the big one which isn’t really an option.
IMO if Apple cared about their customers, they should have put out an updated non-retina mini alongside the new one but I guess that wouldn’t make them as much money would it? The current mini, with just a slight increase in system RAM, no retina, and even a two year old processor (instead of the three year old one it currently has), would be a far lighter, cooler, and more capable machine than this year’s retina mini.
Retina adds nothing but bragging rights. It doesn’t make the mini a better computer and the baggage needed to enable it actually makes things worse.
The Mini 2 will almost certainly be clocked at the same 1,3GHz as the iPhone 5s, where it seems to run just fine.
Agreed. When I went to buy my iPad 3 originally, I was actually planning on getting a refurb’d iPad 2. But the refurb’d iPad 2 apple sent me had the light leaking problem. So I said to hell with it, I’m not going to both with that, and I just bought the iPad 3 instead. I guess I would have been better off going for a replacement refurb iPad 2 looking back, knowing what we know now. The iPad 2 gets better battery life than the iPad 3, it’s faster when running iOS 7, and it’s lighter, and doesn’t overheat. iPad 3 = terrible product. And that’s why they pulled it only 7 months after it hit store shelves. For the first 2 months, demand was higher than supply, so really, the life cycle of the iPad 3 for average consumers was FIVE months! This is ridiculous and totally outrageous…I understand why they did it though; they had to have something to compete with Android tablets — and that was the retina display…nevermind that at the time, they barely had a GPU advanced enough to power it, and it caused terrible battery life and overheating issues that were never resolved — it actually made more sense for Apple to financially release the crap product iPad 3; BUT; they had to have known about all the problems it had, and chose to release it anyway — knowing the iPad 4 would be out in a few months and replace it — hence the anger and lack of respect towards early adopters and loyalists. Of course none of US knew that at the time; if we did, I doubt many people would have bought the iPad 3 at all.
To put the device in some perspective. It would be as powerful as a 13″ 2010 model Macbook Air. (Intel Core 2 Duo L9400 1.86 GHz (2 cores))
Geekbench 2 (MacBook Air) and Geekbench 3 (iPad Air) are not comparable. Get a glance at primate labs blog.
Not directly, agreed, but they do give a steer to the rate of development.
Me and my iPad 3 are really sad.
So glad I bought the iPad 3 when it came out…oh wait…what a joke.
Interesting article, but benchmarks rarely predict real world performance. The dirty little secret about mobile benchmarks: http://mostly-tech.com/2012/09/29/the-dirty-little-secret-about-mobile-benchmarks/
– Rick
Geekbench scores were designed to reflect real-life usage, so get closer than most. Of course, the processor isn’t the roadblock in many tasks, so they do give an exaggerated view of the overall user experience, but definitely more indicative than many.
Thanks for responding and approving my link. Since Geekbench says the iPad Air is almost TWICE as fast as the iPad 4, I’d love for you (or anyone else) to show me one REAL WORLD example where it’s really twice as fast. I’ve tried to do this before and it’s never the case.
– Rick
Once I get my iPad Air, I’ll be able to do a few comparisons on things like opening large apps. Won’t be scientific, but should give a feel as to whether it’s in the right ballpark.
Sounds good. Thanks. I would be interested in seeing startup and shutdown times as well. It should be faster, but I suspect the difference will be minimal.
– Rick
Why did you delete my analysis of the Mashable video. I removed the link in the first version and toned down my second attempt at posting it — which also was not approved. Do you delete the posts of everyone who disagrees with your claims? I was just trying to quantify the results of their tests and the fact there tests do not back up the fact that Geekbench is any more real-world than the other mobile benchmarks.
– Rick
Posts tend to take longer to get approved at the weekend, as we’re not online all the time. I’ll be posting a review based on real-life experience after a week or so.
Wouldn´t it make sense to put an ARM Chip aside the Intel Chip?
Like they did with the M7 and the A7? It could do a lot of the “basic” tasks, for example Mail, web browsing and notifications. These Apps are already running on the A7, so it´s all about connecting the dots again.
Although it probably would be more radical and Apple style to replace the Intel Chip at all.
That would be a software developer’s nightmare. Besides, ARM pretty much owns Intel when it comes to mobile and embedded devices…faster speeds, less power useage, less heat emission, better graphics…list goes on. This is why with Haswell Intel specifically focused on two things — graphics and power.
Why would that be a nightmare?
ARM Apps would run on the ARM Chip and the other on the INTEL Part. So if you´re just check your mail and browse the web, the system only uses the power efficient ARM Chip, while other apps like Adobe CS still rely on the Intel part.
IMHO i would be the much worse nightmare to work with the first two versions of an ARM-ported Adobe CS ;)
because you’d have to write your app for the intel part and the arm part and make them work together seamlessly, unless apple only allows some things to work on the arm processor, like on the iphone 5s, for example checking notifications in he background, or some other basic services, but I don’t think it would work for bigger things like a full fledged app :D
They really need to add an option for editing the comments
Some people are not looking at the big picture here…. What we need to understand is that smartphones and tablets will be our future PC’s and laptops in a near future so it makes totally sense that you’ll see PC like performance in smartphones. Maybe right now it might seem a gimmick but everything is taking us to a time where your laptop will be your smartphone and you’ll just have to dock it and start working with your external monitor and keyboard as you do it right now with your laptop at work..
We’ve certainly already reached the stage where many people are buying a tablet instead of a PC. Laptops will continue to have a role for at least a minority of us for quite a few years yet, though.