Hacker News Reader

Comment by ctdonath | Hacker News Reader

Yawn. (Admittedly at a glance...) They still don't get it.

Apple doesn't sell spec sheets. They sell end-to-end integrated systems. A 7 second boot time doesn't mean as much when it gets me to a relentless stream of "the webpage you're at wants to open a webpage you trust - are you sure?" and other user-unfriendly operations. Total experience sells. The iPad is beating the pants of tablets with twice the specs - but half the user experience.

Intel is aiming at where Apple is now. When Intel arrives there, Apple will have moved on. Don't build what I want, build what I'm going to want a year after I buy it a year from now.

And for that $300M plan, Apple has $30,000M cash to counter it. Get in the right order of magnitude first before picking a fight.

View on HN

Having spent about 1 hour installing a clean copy of Snow Leopard and then Lion, followed by a seven hour SNAFU trying to get Vista > Win 7 upgrade to work on a Boot Camp partition, the misery of Windows is fresh in my mind.

In the Apple paradigm, I click on the Lion upgrade in the App Store (it knows I already purchased it). It installs, and we're done – no reboot. Just a video tutorial on new features.

In the Microsoft paradigm, I'm digging up DVD cases to manually type product codes that require me to needlessy recreate my upgrade path and make decisions about things like 32bit vs. 64bit. It's still installing Vista I type this, and the progress bar has been stuck at 75% for about an hour. The user experience: a scaled-up 640x480 installer that has been on the "Completing installation..." step for two hours.

It baffles me why anyone would affirmatively seek out this experience. As for the Intel "ultrabooks" – more vaporware from the 1990's-era PC industry. The "rest of us" have moved onward and upward.

EDIT: Windows installer just spit the DVD out and froze. Can't move the mouse. Time to start over! (I can almost hear them: Have you tried turning it on and off again?)

EDIT 2: If anyone is following this, attempt #2 failed the same way. This is hour 10 of trying to install Vista on a clean MacBook Pro 2010.

EDIT 3: Gave up and installed a VM through Fusion. Ironically, this is easier than a native Windows install.

"This is hour 10 of trying to install Vista on a clean MacBook Pro 2010."

Have you tried installing Lion on a non-Apple notebook? You seem to be making a very unfair comparison.

Why? Are other Intel-based laptops better at installing Windows?

FYI – I've had Vista/Win7 installed on this machine before. It's a question of how horrible the process is. Typing in product codes, inserting DVDs, activation codes, warnings about anti-virus, etc.

Yes, all of them are. Apple notebooks are the worst at running Windows.

You don't have to type product codes -- you can install without one and copy/paste from a text file works fine once you're all set up. You don't have to insert DVDs, Windows installs fine from a USB stick (except on Apple notebooks, so there). Installing Microsoft Security Essentials will shut up any AV warnings.

Besides, you don't even need to install Vista to get an upgrade copy of Windows 7 working. A Google search would have told you that.

That's not true! When Vista came out the 15" MacBook Pro was the machine which had the best performance in Vista. Think a bit about that.

Installing Lion on a non Apple Mac is illegal and that's why it is so hard to do. Windows is not and should run smoothly like on every other machine.

That's not true! When Vista came out the 15" MacBook Pro was the machine which had the best performance in Vista. Think a bit about that.

Having had all this trouble, I don't believe that for a second. Turning off AHCI and not supporting USB boot in 2011 is inexcusable.

Installing Lion on a non Apple Mac is illegal

Depends on the jurisdiction. In many non-retarded ones it isn't.

Windows is not and should run smoothly like on every other machine.

It should, yes, but it doesn't.

I don't get it. If my boot drive is wiped clean, what are the hardware barriers that explain the differences that you point to?

EDIT: Not sure why I'm downvoted for asking this. Can anyone advise as to what hardware issues would make a 2010 Macbook Pro more difficult to install Windows on?

The EFI firmware with "legacy" BIOS support that Apple uses. It doesn't help that Windows actually supports booting from UEFI but not the EFI 1.1 hack that Apple uses.

Thanks for pointing me to that. I wasn't able to determine the real-world impact of the limited BIOS support.

Anyway, I use Windows for exactly one thing: accessing (horrible) web apps that require IE. My original point was that the installation process is a mess from the perspective of a licensed user – Apple hardware or not.

>> Besides, you don't even need to install Vista to get an upgrade copy of Windows 7 working. A Google search would have told you that.

Win 7 won't accept my product code because it is an upgrade code. I googled it.

http://www.winsupersite.com/article/windows-7/clean-install-...

Good to know. Looks like a pain, though.

Editing a registry entry and typing a command in couldn't possibly be a pain to anyone on Hacker News...

Again, it's not that I can't grok it – it's that I shouldn't have to. I get no sense of pride from hacking a Windows upgrade into a clean installer. Empty victory, etc.

Also, my benchmark is the Lion upgrade. That's how it's supposed to work.

Are you sure Apple even supports running Vista on a 2010 MacBook Pro? I've had a huge number of problems with Windows on an Apple computer, and I now think Apple computers are simply terrible at running Windows.

I have Win7 running on Lion, the only problem is the graphics driver that crashes every now and then. I think Vista is not compatible with Lion's bootcamp though.

I blogged about some of the problems I faced: http://monogatari.doukut.su/2011/07/my-experiences-with-appl...

I have had a similar experience. MBP is a mostly ok Windows laptop, but it's clear they want you to run Mac OS.

'As for the Intel "ultrabooks" – more vaporware from the 1990's-era PC industry.'

These don't look like vaporware.

http://www.engadget.com/2011/09/01/lenovo-announces-u300s-ul... http://www.engadget.com/2011/09/02/acer-aspire-ultrabook-s3-... http://www.engadget.com/2011/09/01/toshibas-portege-z830-is-...

Can you buy any of those today? No. Will they last more than six months on the market? Nobody knows. How many times have PC makers threatened some Apple Killer product that went no where. Too many to count.

Some specs are more important than others - eg battery life, weight, form factor and price. Nailing a placeholder product while you figure out the reinvention prevents you from losing by default. Also I would be interested to hear what strategy you think Intel should follow to stop a future arm+apple duopoly.

Honestly? It's time to dump the "win" part of wintel. Intel simply needs to face facts: The chasm between hardcore gamers and the general computer-buying public is so wide and yawning it may as well be on another planet. Yet that is still where Intel aims with its wintel strategy: More games than on Mac. I mean, seriously, WHO CARES at this point that I can run Crysis on my netbook?

Here's a bold plan that will kill two birds with one stone: Buy Canonical, throw the bulk of your $300 Million at their development, poach/cajole/trade for every UX guru not employed by Apple, and create a killer Linux desktop OS to go along with that ultrabook. It will scream, it will run a battery all night long and beg for more, it will have no "hells" (.dll, dependency, driver, symbol, whatever "hell" is inherent in Windows and Linux), it will run a plethora of online apps and services, and it will host a complement of installed applications that take care of 95% of a mainstream consumers needs.

No need to install flash, it's already there. Need to play an .mkv file? No problem when VLC is installed by default. Netflix on demand? At your fingertip. Amazon mp3 store embedded in banshee? Sure, why not? Writing a term paper or presentation? LibreOffice to the rescue.

The thinking goes that, over time, specs of gaming PCs have been a leading indicator of where productivity PCs would be years later. More so than looking at the specs of PC servers.

The probable 40-50 million on Steam alone is not really a number you can ignore.

It is for these Ultrabooks ...

No, it really isn't. There's a reason Intel's integrated graphics have improved so significantly with Sandy Bridge.

Intel is already developing a linux distribution in-house.

http://en.wikipedia.org/wiki/MeeGo

Intel should make a single board that does it all. And offer it in 2-3 sizes/varieties. Dell can then just use the Intel Ultraboard and not have to worry about the hardware (CPU, GPU, RAM, Wifi, Ethernet, Audio, SSD, PCI/USB etc.) Let Dell worry about offering Pink and Magenta cases. Let Intel work with Microsoft closely so Windows 8 rocks on Ultraboard. There should be only one driver per Ultraboard and it should include the Chipset, Video, Audio, Network drivers. I never want to install Dell/HP/Asus drivers. Ever. They are all horrible across the board.

Reinstalling Windows should not take more than 15 minutes, it should require no user-interaction, and it should leave the data partition alone. The data partition is where users images/photos/videos/music go. Intel should work with MS to ensure that executables do not run from this partition. If my mom downloads a virus, she shouldn't have to call me to fix it. Hardware DEP hasn't done much to stop viruses. Intel needs to come up with hardware features to remedy root-kits without my assistance.

Intel thinks of itself as one of the hardware makers with competitors. It makes CPUs, like AMD or Samsung. It makes boards like MSI or ASUS. This is red ocean. It needs to make its own blue ocean and of all the companies in the PC industry, it is actually capable of doing that. Instead of offering $300m to others, it should put $3B into creating the single Ultraboard that any PC seller can build a case around and sell tens of millions of these. Dell can't do it. Sony can't do it. Intel can. Just like Apple did.

When Apple sells a Mac, Air or otherwise, Intel wins a bit. They're inside. Because of that, they are definitely shooting for ahead of now - to the part where Apple decides they may not need their chips.

I agree that the user experience needs to improve across the Windows 8 platform for this to really matter, but that is mainly in Microsoft's court.

For me, I like having both a Mac and PC on hand. An Ultrabook would be nice if only for the size.

Except that Apple's and 3rd-party Mac code is much much more portable. They can switch from x86 down the road in a way that Windows can not.

Please explain. Most recent Windows code is written in C# which makes portability easier but even for C/C++ code as long as the api level compatibility is maintained ( and no one frankly is better than Microsoft at that game) Windows code should be no less portable.

It might be an issue of scale but not portability. Much less software exists for Mac than does for Windows.

Apple's done two successful architecture changes on their desktop OS (68k->PPC->x86). Microsoft's done none.

Traditionally, MS has been very good at maintaining compatibility through different versions. Maybe they'd make an architecture change easily, but they haven't had to do so yet.

They have always supported multiple architectures - Win NT ran on Alpha, CE had ARM port, they had IA64 version of Windows server. Technically x86 to x86-64 also is a architecture change and they sailed pretty well thru it given most shipping Windows version today is 64 bit.

It's true!

I even know someone who's seen a non-x86/64 Windows system!

I am sure it doesnt exist as you sir marshray havent seen it. I must have just woken up from a dream where Windows machines running on Itanium also ran Oracle database ported to Windows Itanium.

You know him too?!

Sorry. I concede that there is a major application or three that runs on Itanium. For now. Microsoft is dropping the platform.

As for ARM all I'm saying is we'll wait and see.

Well for one, Apple transitioned from ppc->intel REALLY well with Mach-o binaries and Rosetta.

Using a vm/jit isn't the only way to handle architecture changes. In addition it doesn't remove the need to handle kernel changes/etc... that the vm itself runs on. Nor does it mean that the vm and all of its underlying assumptions are valid on a new instruction set. It took java a fair amount of time to run fast on multiple architectures, I doubt .NET will fare any better.

When Windows actually manages to (again, it ran on alphas/mips/ppc and something else back in the 4.0 days) run on more than one platform I might partially buy into .NET being "portable" across windows releases and architectures. But I suspect a fair amount of things will have to change or move to new apis. Similar to Microsofts go it alone 64bit model in Windows, another architecture in the fray will force things into spaces they didn't originally expect.

Basically, I don't trust anything coming out of Microsoft until they deliver the goods. Apple has shown they can handle multiple platforms transparently, and without the need of a vm/jit. Which can help out things like battery life with caches/etc...

http://en.wikipedia.org/wiki/Mach-O http://en.wikipedia.org/wiki/Fat_binary#NeXT.27s.2FApple.27s... http://en.wikipedia.org/wiki/Rosetta_(software)

From CES2011

The technology demonstration included Windows client support across a range of scenarios, such as hardware-accelerated graphics and media playback, hardware-accelerated Web browsing with the latest Microsoft Internet Explorer, USB device support, printing and other features customers have come to expect from their computing experience. Microsoft Office running natively on ARM also was shown as a demonstration of the full depth and breadth of Windows platform capabilities on ARM architecture.

I dont think there is reason to doubt Microsoft on API compatibility which i s what this ultimately boils down to. Windows recently ran on IA64 BTW. .NET helps but Win32 in itself is proven enough to need little more than just a recompile for porting.

I was thinking more along the lines of C/C++ WIN32 api and things like 64bit calling conventions etc... that might not remain quite the same on ARM.

There is no such issue. You dont seem to understand what you are talking about. Look up ABI. They couldnt possibly port software as big as the OS itself and MS Office / Oracle DB to Itanium without having stable C/C++ and Win32 ABI/API.

Ok fine I'm a retard.

Yeah everybody knows there's no way a C API from Microsoft could ever be a leaky abstraction. Lol.

Or that there's no way the win64 calling conventions weren't influenced by win32.

But in fact their 64bit abi on x86_64 was in fact derived from their 32bit calling conventions.

http://stackoverflow.com/questions/4429398/why-does-windows6...

Correct me if I'm wrong, but there is an exact case of where 32bit windows on x86 directly influenced 64bit. This is what I'm talking about when I talk about other 64bit platforms may not behave the way Microsoft expects. And this directly influences c/c++ register allocation.

Though I will state, what Microsoft did for its fastcall on 32bit windows and 64bit is rather ingenious in that the same call can work on both platforms.

Right, being mainly an extension the x64 is hugely influenced by x86. The 64-bit Win32 ABI is obviously influenced too, to the point that on x64, 32-bit process is largely just a native 64-bit process with 32-bit modules loaded.

Most of the time, the differences are hidden by abstractions in the Win32 C API. The calling conventions do a pretty good job of this (especially the 64-bit ones). Still, there is a nontrivial amount of software that does need to use a few (or many) assembly instructions or interact with the CPU in low-level ways. Windows is a relatively friendly OS for applications that rely on non-standard behavior.

You can't "just port" an antivirus package. Certainly many malware samples won't run on ARM either, but many will (like .vbs). It will take time to sort all that out. In the meantime, many companies simply won't deploy Windows without AV.

You keep saying "without using a vm/jit" like it's something Apple did, but in reality, Rosetta is a JIT and not every Mac OS X binary for PPC was a Universal one (which, incidentally, caused a few problems when Rosetta's translation wasn't up to par, either).

Also, .NET is "portable" not just across Windows releases, but through the work of Mono large portions of it also work on Linux, and on other architectures via Linux as well -- so presumably, anywhere Mono or Microsoft's official .NET are compiled, .NET will run so long as no weird x86- or binary-specific things (like P/Invoke) are done. There are other problems with Mono, however, but those are for another discussion.

This is true that Rosetta is a JIT, but my point was more that the Mach-O binary format allows for multiple architecture binaries, therefore somewhat obviating the need for a jit.

The only use Rosetta served was running cross architecture binaries.

Mono also doesn't run the full .NET stack, e.g. think the Netflix drm components, so I can't quite consider it cross platform. Cross platform windows wise sure, but that isn't saying much.

There really isn't that much C# in commercial software, even from Microsoft. You might find an accounting package here or there, but first-tier apps are almost entirely C and C++ under the hood.

For example, MS Office and IE are still used primarily in 32-bit mode on 64-bit systems.

It doesn't matter how much software exists or is in use. We are talking about portability here. For. Net it is is not an issue at all and for Win32 C/c++ only a recompile is necessary. Oracle DB exists for Windows IA64 and office does for Arm. Those are proofs that like any other sane platform portability is not a big issue for Windows.

I think you don't have any idea how much it costs to port a major application like that.

Even if everything goes smoothly with the codebase, just setting up another build environment for something like ARM means installing a new compiler version and toolset. Many (or most) dev shops are stuck on 6-10 year old compilers.

You also multiply the QA regression cycle costs by some incremental factor.

There are lots of applications small and large, that to this day, don't even ship a 64-bit build.

Yes, there are some applications that have been ported. Even if you were to come up with a not-short list of them I would submit everything else in the world that has not.

I think you are confused as to what exactly we are debating here. Porting costs are real - somehow Mac software isn't magically a free port to ARM. The porting costs depend on code base, complexity, various other things. That's however not what we were debating - the debate was Microsoft has never done it and so it might be hard for them. The counter point I was making was Microsoft has been there and done that and there is no reason to believe Windows software is any harder to port to another platform than Mac Software except for the scale part - Windows just has a ton of more complicated software than Mac has had and porting "toy" Mac apps will be less of an effort than say porting a big Enterprise Windows app. However it isn't fundamentally easier or harder.

Microsoft already has Windows 8 and Office running on ARM. The fact is, in both cases, supporting a different processor is a recompile away (with edge cases failing in both cases). Windows NT and NeXTStep were both supporting different processor architectures at the same time.

The big different is not technology, it's culture. Apple will happily make all your old software obsolete as long as it doesn't seriously interfere with them selling new hardware. Microsoft has the much harder job: they need businesses to buy OS licenses (that is what they sell) and those businesses have a lot of old crappy expensive software that must continue to run to make the sale.

Yes, I know Microsoft has Windows and Office running on ARM.

Wake me up when 50% of the shrinkwrapped Windows-logo compliant and hardware software sitting on store shelves runs on it at native speed.

Fun fact: Apple had OS X running on x86 since the first release (iirc Jobs dicslosed this announcing the transition to Intel processors).

That's less interesting when you know that NeXTStep (the direct predecessor to OS X) was actually released for x86.

Fun fact: Windows NT was released for PowerPC.

I'm surprised you didn't mention the great gretzky quote: “I skate to where the puck is going to be, not where it has been.”

Reply on news.ycombinator.com