Report: Super-Lightweight 12-inch MacBook Powered By Apple Silicon To Launch This Year (macrumors.com) 103
Apple has designed a 12-inch MacBook powered by Apple Silicon that weighs less than one kilogram (2.2 pounds) and the company intends to launch it by the end of the year, according to a new report. MacRumors: Apple's first ARM-based Mac will use an A14X processor, which is codenamed "Tonga" and manufactured by TSMC, and the MacBook will have a battery life of between 15 and 20 hours, according to the Chinese-language newspaper The China Times. The report adds: According to Apple's supply chain, Apple is expected to launch a Macbook with a 12-inch Retina Display at the end of this year, using its self-developed and designed A14X processor, with the development code of Tonga, supporting a USB Type-C interface and weighing less than 1 kilogram, because of the low-power advantage of the Arm-based processor. The Macbook battery lasts 15 to 20 hours. The A14X processor will also be used in the new generation iPad Pro tablet."
Re:Pretty temping travel laptop (Score:5, Informative)
Emulation and translation are not mutually exclusive, as some people seem to think. Rosetta 2 exists to emulate an x86 Mac. It does so through code translation, but then again, how can it not? Why you push this narrative to support your tribe it is difficult to tell, but it's consistent with the generally stupidity of all your posts.
Also, small screens are terrible for photography, unless you think an iPhone is a serious photography tool, which you have also claimed in the past. It seems highly unlikely you've ever traveled as a photographer, or done anything productive, considering your constant advocacy for unsuitable tools.
At least with your acknowledgement of lack of x86virtualization you may be willing to admit that the Mac will no longer be the ultimate development platform, if it ever was. That must sting, eh SuperKendall?
Re: Pretty temping travel laptop (Score:2)
Re: (Score:2)
Not in response to SuperKendall on an Apple article. You need to be firm against super-fanboyism.
Re: Pretty temping travel laptop (Score:2)
Unnecessarily hostile.
Unnecessarily ignorant, more like.
Re: (Score:1)
Turds turn up in the toilet bowl. Apple zealots show up on apple.slashdot.org threads.
Re: Pretty temping travel laptop (Score:2)
Turds turn up in the toilet bowl. Apple Haters show up on apple.slashdot.org threads.
FTFY.
Re: (Score:2)
SuperKendall haters, not Apple haters, tribalists can't tell the difference
Re: (Score:1)
You don't know SuperKendall then
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
This definition is entirely without merit.
An emulator is something that emulates. How it emulates is immaterial.
In the Rosetta 2 case, the "original processor" is "mimicked", along with the entire platform and ISA, by Rosetta. Rosetta emulates and entire Mac using an alien processor.
Rosetta 2 "runs the original source code" as well. A virtual processor doesn't run anything, a physical processor does. A virtual environment, in this case a virtual CPU, IS AN EMULATION of a physical one.
Yes, there is a diff
Re: (Score:1)
I had an EVAKIT-75x back in the day. A rather expensive piece of NEC hardware. We used it to develop code for a uPD75328 processor. That was an NEC 4 bit microcontroller from the early 90s.
Back in the day of mask programmed controllers it was big bucks to use emulators to develop your code. Then you hoped the first run of parts (probably 50,000 of them, though you'd better commit to larger quantities if you wanted the salesmen to pay attention) would actually run.
Re: (Score:3)
Many modern emulators do no translation, e.g. MAME. If the goal is to perfectly emulate the original system then translation isn't ideal. One of the main issues is that it makes emulating the interactions between the CPU and various other bits of hardware at a cycle-exact level difficult, if not impossible.
Translation is good for speed but not for fidelity.
Re: (Score:2)
Many modern emulators do no translation, e.g. MAME.
Boggling.
That's a patently false assertion.
MAME has a very mature pluggable system for dynamic recompilation (i.e., translation from source instruction set to UML to target instruction set), with JIT front and back ends.
Re:Why do you seek to confuse what is simple? (Score:5, Informative)
since Rosetta 2 is different than the first Rosetta
Not really. It works the same way. Understand how Rosetta 2 works, you understand how Rosetta worked. You are just being obtusely stubborn, and have a very narrow definition of emulation that no one else uses. This really is a case of you saying tomahto when everyone else is uses with tomato. You are entirely correct about the translation, but entirely incorrect insisting it is not emulation. And semantics are actually pretty important, contrary to popular belief, otherwise, no one would understand each other. Your campaign against the Rosetta's translating code but not also being emulation is OCD. The Rosettas are emulators. They accomplish their emulation through code translation.
Re: (Score:2)
Re: Why do you seek to confuse what is simple? (Score:1)
First launch takes the compile time hit and adds the arm64 architecture it to the single architecture binary. Second launch sees the dual architecture binary and uses that.
Exactly correct; except, IIRC, Apple does the Cross-Compiling during Install Time.
There is also JIT Cross-Compiling when x86/64 code is encountered during Runtime.
BTW, Windows 10 for ARM does its Cross-Compiling during First Launch. Just FYI.
Since nobody expects an Install to happen quickly (so hiding a Cross-Compile and Rebuild there isn't really noticeable), IMHO, Apple's approach gives a better First Launch experience for the User; because there is no "First Run Beachballing", while Cross-Compilation an
Re: (Score:2)
""It is very reasonable to make some really good guesses from their documentation that they're not merely interpreting X86 code."
They are merely interpreting x86 code. It is not reasonable to guess otherwise.
"I think they're using X86 executables as source code."
Yes that what an emulator would do.
"...but source for a special compiler..."
or special emulator, or translator.
"...that reads both the relocation information and instructions for x86..."
information necessary to execute any application, yes
"...and e
Re: (Score:2)
An emulator is a simulator. It's an app that simulates an x86 chip. It has variables for all the registers in the CPU and walks the instructions one-by-one updating memory regions and register variables accordingly. This is NOT compiling.
It's the same difference between Python and C code. Python is interpreted and C is compiled.
Re: (Score:2)
as does any emulator, the question is only how it is "emitted", when and where.
That is simply wrong, sorry.
Re: (Score:2)
I can assure you if you insist that both is the same, when it clearly is not, in a german university test: you get no points for your answer.
Re: Why do you seek to confuse what is simple? (Score:1)
Why you fight against technical facts being clear, I will leave as an exercise to the reader.
Rosetta 2 exists to emulate an x86 Mac.
Technically incorrect, you are obviously one of the people who cannot grasp how code translation works.
Save your breath, SuperKendall (and everyone else).
dfghjk
Has vociferously been proving he doesnâ(TM)t know the fundamental difference between bare-metal emulation (like SoftPC) and Code-Translation (like Appleâ(TM)s 68k translation in PPC Macs, and now in Rosetta 2) on both AppleInsider (where he argued with those more informed on the subject for literally dozens of posts!) and Slashdot for months now.
Do not feed the troll. He's a third-class idiot. He has nothing better to do than prove his ignor
Re: Why do you seek to confuse what is simple? (Score:1)
Good point, in the future I'll just stick with a canned response summarizing why translation is not emulation, and just ignore whatever other trolling nonsense he spews.
Thumbs up!
But watch out; he's indefatiguable!
Re: (Score:2)
When assholes are beaten, they resort to patting each other on the back. Neither of you are even capable of engaging in the argument, you've run out of the ammunition that Apple gave you. All you have left is to tell each other how smart you are.
Re: (Score:2)
When assholes are beaten, they resort to patting each other on the back. Neither of you are even capable of engaging in the argument, you've run out of the ammunition that Apple gave you. All you have left is to tell each other how smart you are.
Speaking of patting ones self on the back...
Re: (Score:2)
Well, there's some additional terminology dropped here but nothing relevant nor is there any demonstration that these terms are understood, just more personal attacks from someone who would stand no chance in an objective discussion.
There is one thing that I have not done, I haven't watched the Apple presentation that you and SuperKendall clearly have, though I was aware of it when it was being broadcast. Thing is, I've worked in the industry for decades already, I don't need such simpleton concepts explai
Re: (Score:2)
"I push the "narrative" of technical fact so that people actually understands how Rosetta 2 works, since Rosetta 2 is different than the first Rosetta, so people might get confused."
No you didn't, you didn't even demonstrate that you understood "technical fact" or that you "understand how Rosetta 2 works".
"...apps will perform just as well being translated..."
You don't know this, you're just parroting Apple sunshine. This is unlikely to be the case.
"Technically incorrect, you are obviously one of the peopl
Re: (Score:2)
Transmeta's processor was supposed be one of those translation emulators. It had the same terrible performance for all platforms. Maybe it isn't too bad we never got to see it.
By the way, ad hominem much? I don't understand it.
Re: (Score:2)
Transmeta was a shitty company with a shitty product and they were a joke among the greater processor design community. They were somewhat know for employing Linus for a time. That says nothing about the underlying terminology, and to be clear, the only argument is terminology. Apple said Rosetta 2 is special, therefore Apple goons here cannot accept that Rosetta 2 is an emulator and they get twisted up in knots over what an emulator is. It can't be an emulator because Rosetta 2 is special.
As for "ad homi
Re: (Score:2)
An iphone may not be the best tool out there for photography, and a professional photographer is likely to have better (and much more expensive) tools at their disposal.
But professional photographers make up 0.0001% of the population, for the vast majority of people an iphone is adequate and professional tools are not worth the extra cost for the limited benefits they would provide.
Re: (Score:1)
Re: (Score:1)
Of course, but you're missing the entire point, which is understandable. SuperKendall claims to be a photo pro, and he makes this claim to promote the iPhone as a photographer's tool. Everything SuperKendall posts is in service to some tribal narrative and what I am talking about has nothing to do with whether an iPhone can satisfy photography needs of its users. It clearly does.
I was lumping his promotion of a 12" ultra-portable as a via photography notebook in with his promotion of an iPhone as a viabl
Re: (Score:2)
Re: Pretty temping travel laptop (Score:2)
Also, Rosetta 2 is only needed for FOSS and other 3rd party programs that haven't recompiled to support ARMx64. Anything you install via the Mac App Store in recent years will use the LLVM IL stream resident in the .app file that the developer uploaded to iTunes Connect to recompile for ARMx64 before you download it, so you'll already be running native ARMx64 code locally. Anything you install via Macports should be compiled for ARMx64 as well.
True about Mac App Store Delivering Arm code directly, and I'll be happy to take your most uplifting word that MacPorts operates in a similar manner. Both help keep .app bundles as small as possible...
However, there are plenty of other places where an unsuspecting ArmMac can encounter untranslated x86/64 code, not just FOSS s/w (although isn't that reason enough? ;-) ). And that is where Rosetta 2 comes in with some handy JIT Cross-Compiling.
So, I am not sure why you are downplaying dome damn-hard work by A
Re: (Score:2)
FOSS will be recompiled too, there's absolutely no reason not to. It's only going to be used for old closed source junk that hasn't been ported.
Re: (Score:2)
It is not necessary to include Bitcode in App Store apps. In fact, it can cause hassles that routinely lead developers to turn it off.
Re: (Score:3)
Since Rosetta 2 is not a emulation as some people seem to think, but instead a translation layer pretty much most apps should be good right out of the gate...
Rosetta 2 actually does both. It uses binary translation if it can, but there are situations where it can't and it falls back to emulation. (For ex. With any app that uses scripts which are compiled on the fly... web browsers, games, any Electron app, etc.)
Of course the important question is not how it works under the hood, but how fast is non-native code. The answer might not be as great as you think. Two big problems are: 1) Code-expansion because x86 instructions don't map 1-to-1 to ARM, because ARM inst
Re: (Score:2)
I agree with what you've said, but a couple things:
First, I object to your acceptance of "binary translation" being different from "emulation" as is claimed in what you responded to. The "binary" part is meaningless, its all binary, and translation has existed for a very long time. JIT is a form of translation, but JIT didn't magically converter emulators into something else when it arrived. What is different here is that Rosetta 2 is an Apple technology, and Apple fanboys are now making a distinction be
Re: (Score:2)
Also, remember that a critical aspect of this is the claim that the translated code will actually be FASTER than the native x86 code because of LLVM optimizations. That's an ignorant claim, of course, for reasons you've just given, but it is part of the "R2 is not an emulator" argument.
That claim is definitely laughable in the general unqualified context- however- I'd be remiss if I didn't note that it's absolutely possible in certain scenarios.
The Only Question is Thunderbolt Support (Score:3)
The only real question I have about these new machines is whether they will have Thunderbolt support so we can work around the terminally underpowered GPUs that Apple puts in so many of their machines (not to mention supporting all my FireWire-based audio and video gear).
I didn't realize just how underpowered the Intel GPUs were until I tried to run OBS on a Mac Mini, and it sputtered and stuttered and couldn't keep up, while a MacBook Pro with a significantly slower CPU had no trouble. It turns out that all the H.264 decoding for on-screen rendering was being offloaded to the GPU on the MacBook Pro, but was being emulated by the CPU on the Mini. One eGPU later, the Mac Mini was purring like a kitten.
I would expect the embedded GPUs that Apple ships in these machines to be similarly underpowered, hence my concern. Thunderbolt is the only reason that the Mac Mini didn't get returned to Amazon and replaced with a Linux box.
Indications are the GPU should be fairly powerful (Score:1)
I would expect the embedded GPUs that Apple ships in these machines to be similarly underpowered
At minimum, they would be as powerful as an iPad Pro, and that was already pretty decent even in a year-old model [notebookcheck.net]...
Way better than Intel integrated graphics, and for a desktop model they will probably have a GPU with even better performance. Apple has done a really good job with custom GPUs... I think in fact the mobile GPU scene is a big reason why they are switching the Mac line over to ARM as well.
That sai
Re: (Score:2)
I would expect the embedded GPUs that Apple ships in these machines to be similarly underpowered
At minimum, they would be as powerful as an iPad Pro, and that was already pretty decent even in a year-old model [notebookcheck.net]...
Way better than Intel integrated graphics, and for a desktop model they will probably have a GPU with even better performance.
I hope so. I mean, the iPad Pro's GPU is not nearly as bad as the integrated Intel abomination that comes in the current-generation Mac Mini (which pretty much just whimpers with its tail between its legs), but my 2017 MacBook Pro can still wipe the floor with it (at about 61% faster).
To put that in perspective, when you include the laptop's internal display in the calculation, a laptop with the iPad Pro's GPU would theoretically have almost as much GPU power per pixel when driving one external monitor as
Re: (Score:2)
Re: (Score:1)
Thunderbolt is the only reason that the Mac Mini didn't get returned to Amazon and replaced with a Linux box.
That's too bad. You could have saved several hundred dollars. I just can't rationalize the price of any Mac Mini.
Re: (Score:2)
The mini line wasn't always trash. I've purchased two over the years, but would never buy one today.
Re: (Score:2)
Re: (Score:2)
The Apple VT Hardware Encoder (by far the fastest encoder on MacOS) uses the Intel built-in GPU except on the iMac Pro.
Sadly, both of Apple's hardware encoders hang 100% of the time on every Mac I've used whenever I try to use them in OBS (in both Mojave and Catalina). The only hardware compression I'm aware of that actually works correctly in OBS is nvenc on Windows/Linux.
Either way, it wasn't the encoding that caused me problems; it seemed to be the mixing/layering. OBS doesn't let you designate a source as audio-only, so it is apparently doing heavy lifting for every source that could be onscreen, even if it is entirel
Re: (Score:1)
Re: The Only Question is Thunderbolt Support (Score:2)
Recent report just a few days ago said you can expect around Intel i9 performance.
Wow!
And just remember: This is only the beginning; whereas Intel has taken x86/64 about as far as it can go without resorting to room-temperature superconductors!
Re: (Score:2)
Thanks for the update. I haven't been paying nearly as much attention to tech lately as I probably should be. I've been spending too much time correcting people who are wrong on the Internet about various coronavirus hoaxes/disinformation. :-D
I think my Apple IIgs was faster than Intel's onboard offering. :-D
Okay, okay, so that's a slight exaggeration, but Apple should be embarra
Re: (Score:2)
It's unlikely to give the RTX 2080 pause for concern, but in terms of its purpose it's likely to be significantly ahead of the Intel graphics solutions.
Re: (Score:1)
It turns out that all the H.264 decoding for on-screen rendering was being offloaded to the GPU on the MacBook Pro, but was being emulated by the CPU on the Mini.
Either you were using a machine with a very old CPU [wikipedia.org], or Apple was dragging ass adding QuickSync decode support. Either wouldn't surprise me, as Apple has sold CPUs long past their prime in current models, and they literally took years before they enabled hardware HEVC decode support in iOS, even though the silicon already supported it.
apple arm Thin is in (just wait for the mac pro) (Score:2)
apple arm Thin is in (just wait for the mac pro)
Ash tray at half the size of the last one at X2 the cost.
So... (Score:2)
An iPad with an attached keyboard.
Earth shattering!
Re: (Score:1)
Rumor has it that it's also slightly jailbroken compared to the maxiPad. Or is Apple planning on 'securing' this thing in some sort of prison like their iOs gadgets?
Re: So... (Score:2)
Rumor has it that it's also slightly jailbroken compared to the maxiPad. Or is Apple planning on 'securing' this thing in some sort of prison like their iOs gadgets?
No.
I can't find the quote right now; but they have already publicly addressed that question.
Apple full well understands the difference between Mac and iPad markets.
Remember, they use these things, too!
Re: (Score:2)
Apple did say this, and prior to that I expected ARM Macs to not allow apps installed around the App Store. In hindsight, doing that was probably untenable since existing x86 apps would largely be ruled out unless installed thru migration.
Remember, though, that Apple also said absolutely NO 3rd PARTY APPs on the iPhone because Apple couldn't risk unvetted code corrupting the cellular network. That was all a lie of course, largely due to Jobs' ignorance, but it shows that Apple will change their mind on th
Re: (Score:2)
Apple traditionally releases locked down stuff and then quietly relents as time goes o
Re: (Score:2)
This was also replicated on the Mac where the first Macs did not have cursor keys. Intentionally - Apple wanted you to use the mouse to point and move the cursor around.
I have been using Macs since they were Lisas, and I can't remember one without cursor keys.
Can you actually find an example of this? Genuinely curious (and a bit skeptical).
Should be pretty sweet (Score:2)
Considering the reported performance already with the Developer Transition Kit's two-generation-old SoC running a Beta release of macOS Big Sur, this should be quite pleasant.
With that kind of battery life, so long as classes don't require Windows compatibility (most do not at this point), this should make a perfect back-to-homeschool laptop.
Re: (Score:2)
Worth it for college etc (Score:2)
If all the major apps (Adobe, Microsoft, browsers) support it the machine'll be a great performer. Because it's Apple it's unclear whether it'll be a good value, but hopefully it'll be cheaper than today's MacBook Air.
Keep in mind this is a Rumor (Score:2)
Posted on MacRumors, and already refuted within TFA itself!
Break the /. Rules and go read TFA before we all get wound up in speculation!
12 inch ??? (Score:2)
Re: (Score:2)
Re: 12 inch ??? (Score:3)
Protip: visit an optometrist.
My 11.6" netbook from 2008 is as functional as it ever was. But my vision began to deteriorate once I entered my 40s; a hereditary thing, my siblings got it too.
Otherwise, there's really no difference between a 12" and a 14" panel with the same number of pixels.
Large iPad+attached keyboard (Score:2)
Keyboard has bigger battery to increase run time.Runs a version of iOS that says "MacOS" when asked.
Re: (Score:1)
Its also probably somewhat jailbroken, at least for now.
Re: Large iPad+attached keyboard (Score:2)
Runs a version of iOS that says "MacOS" when asked.
Quit trolling, asshole.
Watch the WWDC 2020 Keynote starting at 1:26:03, and tell me that iPadOS can do any of that.
https://www.apple.com/apple-ev... [apple.com]
It's not a "self-developed and designed" processor (Score:1, Offtopic)
It's not a "self-developed and designed" processor when it's licensed ARM intellectual property.
This is like China claiming its "indigenous" 64-bit processor was "self-developed and designed" when it's licensed MIPS tech.
On the other hand, it's somewhat better than if China claimed Zhaoxin and Hygon chips were indigenous.
Re: It's not a "self-developed and designed" proce (Score:5, Informative)
It's not a "self-developed and designed" processor when it's licensed ARM intellectual property
Jesus Fucking Christ! Not this again!
For the Googolplexth time:
Apple has a Perpetual Arm Architecture Class License. That means they license the ISA.
Apple In-House designs their own Arm-compatible CPUs and Peripheral Subsystems (which is exactly why they kick other Arm SoCs).
Give us all a break, and do some research before you prove yourself to be any more ignorant, willya?
Sheesh!
Re: (Score:1)
Nice ad hominem attack, sir. I'm sure you feel great about pointing out my ignorance.
Re: (Score:2)
That was not an ad hominem, not even an insult. Perhaps check wikipedia ...
Re: (Score:2)
The architecture class licence grants Apple access to everything, not just the ISA. If they only wanted the ISA they would just licence that and save themselves a huge amount of money.
The architecture licence gets them access to things like the big.LITTLE IP (replaced by DynamIQ now) that lets them mix high performance and low power cores. They didn't design that part of the architecture, they licenced it from ARM.
In fact the architecture licence they have gives them access to internal ARM development infor
Well actually (Score:5, Informative)
Apple has been designing the A series chips. Apple isn't just taking ARM designs and burning them to silicon, which is why the Apple A-series chips are an order of magnitude faster than any other ARM chips out there.
Re: (Score:1)
They should call it an Altivec unit to make it seem more special.
Re: Well actually (Score:2)
Applevec here we come
Re: (Score:2)
They are competitive with the Snapdragon 865 in most synthetic tests, but lag behind by a long way in real world tests like AnTuTu.
https://nanoreview.net/en/soc-... [nanoreview.net]
Where is this "order of magnitude" coming from? What benchmark do Apple CPUs perform 10x better than the competition's in?
Re: (Score:3)
It's not a "self-developed and designed" processor when it's licensed ARM intellectual property.
Apple licensed the instruction set then designed their own CPU. Sort of like the relationship between AMD and Intel - only they are not competing.
It is unfair to say that ARM did not help Apple. The underlying architecture is all defined by ARM. But the implementation of said architecture on the Apple chip is designed by Apple. The ARM provided core designs are not used.
Pine book (Score:2)
Look better
Don't forget the additional silicon (Score:5, Interesting)
I'm probably buying one just because I write apps for a living and I really need to see what my iOS Apps look like and feel like to use on an aMac.
What I'm most intrigued to see isn't the weight, size, CPU benchmarks, MacOS, ports, etc, etc but what else Apple includes in its silicon other than just multiple ARM processors.
I suspect this, as well as Intel dropping the ball far too often of late, is one of the major factors in Apple's shift to their own silicon.
Will it support FaceID? It will no doubt have a neural engine like the A11 and later "Bionic" ARM's got. What will MacOS do with it?
If you followed anything in Apple's WWDC2020 you'll have seen that iOS14 is attempting to secure proprietary app data, specifically in-app AI models.
CoreML Model encryption was one notable announcement.
Apple has also updated their DeviceCheck API to include, in iOS14, an App Attest Service [apple.com] which lets developers check if an installation of an App is compromised before downloading data (e.g. AI model) which is at risk of being stolen.
My concern, with iOS Apps running on MacOS, is that they will be wide open to file browsing and stripping.
This isn't an issue on iOS (unless device is jail broken which Attest is supposed to address).
I think we'd already have seen more on device AI if it wasn't for the corporate fear of IP theft.
So expect more locking down and a higher garden wall coming out of Cupertino and, I suspect for these reasons I've given, a new wall around MacOS machines.
Isn't it time we had a shake up of the x86 platform dominance and ubiquity, that some one tries something new to force progress in new directions?
I think so. It has to be good for the industry as a whole. Remains to be seen if Apple pulls it off though.
Re: (Score:2)
If a device is in someone else's physical possession then that person has access to the data stored on it. Don't try to rely on software to restrict it, sooner or later it will be broken and the more valuable the data the sooner that will happen.
NEVER rely on client side security.
Re: (Score:2)
Right, there's no perfectly secure system. Cloud based systems are preferable for this reason, even though they themselves aren't 100% secure.
It's how close you can get to 100% that is the goal.
I think we'd already have seen more on device AI if it wasn't for the corporate fear of IP theft.
It's a shame this is the case because we'd all benefit from AI functionality on device instead of having to maintain a network connection to a cloud service.
Voice recognition and language translation is one such example that I've wanted to see on device and offline.
Software is too easily stolen.
Hardware is harder to
Re: (Score:2)
More Disposable Crap (Score:2)
What I expected (Score:2)
It's interesting how people here are focussed on the speed of executing Intel code. The reality is: Most of the code that runs will be compiled for iOS. Because most is in the OS, or in the Swift standard library, compiled for ARM. And building an app that runs native on A
Re: (Score:2)
Plus, this is the 12" MacBook. A lot of the usage will be the standard Mac apps plus Pages/Numbers...maybe Keynote. A lot will be native to start with.