• Sign in to Neowin Faster!

    Create an account on Neowin to contribute and support the site.

Sign in to follow this  

Is Apple Stable more than windows?

Recommended Posts

Digitalx    17

they both have their upsides and down sides in terms of security. stuff like yes because of the unix core, osx has better controller of UAC. but microsofts now adapting that into vista etc OSX is good XP is good too and so is vista. they are all good and have their own uniqueness and it really depends what the security or stability you are looking for is for... but one thing i do hate it mac's very high prices, i want to buy a 17" imac :(

Share this post


Link to post
Share on other sites
acies    0

^ yea me too. Macs prices are just too high for the average person^:(

Other wise i'd switch.

Share this post


Link to post
Share on other sites
Lasker    0

MAC OS X just rocks! I was a windows user, I tested also linux but when I tested os x is just amazing how stable and rock solid foundation is built in.

Share this post


Link to post
Share on other sites
_Pablo    0

conflating stability with security.

Agreed - I mistakenly typed "secure" instead of "stable". Mea culpa.

Where did the limitation that it had to be an x86 Mac come from?

Because that's what people will be buying now...unless they really have to have a multi-CPU Mac right now.

Except that the comment I quoted wasn't.

Purhaps the sun was casting glare on your screen. I'll quote it for you:

See all those places where you used the personal pronoun "I"? That's your sample. You. One.

Given this is a forum where individuals post, it seems redundant at best to point out that one persons post is the result of one persons experience - the sample size would equal the number respondents (minus the trolls and fanboys).

Get yourself a bucket of crap and put it on the steps. Come back 5 years later and put a spoonful of crap beside it. Wait 6 months and the flies will have consumed the small one, but still be working at the big bucket.

That assumes that all piles of crap are equally alluring to flies. But it's a confusing analogy - does the act of flies consuming more Apple Crap than Microsoft Crap equate to greater or lesser stability in Apple Mac OS X ?

But to keep your analogy going, I would suggest that using Vista and Leopard would be fairer as we would be putting out the same fresh crap at the same time, so we could see which the flies preferred.

Finally, I agree that you can polish a turd as much as you want, but it's still a shiny turd.

Share this post


Link to post
Share on other sites
the evn show    138

Agreed - I mistakenly typed "secure" instead of "stable". Mea culpa.

Because that's what people will be buying now...unless they really have to have a multi-CPU Mac right now.

Apple's Pro desktops and servers are still PowerPC systems. It's fair to restrict the argument when Apple stops selling them, not before. Does it really matter though? Is there any evidence that x86 Macs are more or less stable than PPC versions? Of what we can see of the core sub-system we know that the C code isn't significantly different.

Given this is a forum where individuals post, it seems redundant at best to point out that one persons post is the result of one persons experience

Your post was all about you. I said that one persons experience is not enough to draw any meaningful conclusions in this thread. Now you're talking about all the other posters in this thread or forum, something you neglected in the post I quoted.

That assumes that all piles of crap are equally alluring to flies.

Are they not?

The market share argument doesn't hold up on it's own (see apache/iis as the typical example)

Perhaps the big one is more attractive simply because it stinks so much worse than the small one.

But to keep your analogy going, I would suggest that using Vista and Leopard would be fairer as we would be putting out the same fresh crap at the same time, so we could see which the flies preferred.[

If it's not the huge numbers of Windows vulnerabilities that makes it attractive, then what is it? If it's some environmental factor like say install base, what's going to keep that from being a factor when Apple and Microsoft ship updated versions of their software?

Comparing the stability of operating systems a year from now isn't going to be any more fair than comparing the current ones. It'll just be different.

On the other hand, saying things like "XP is slightly more stable than Mac OS X due to it's maturity" is going to be valid when it's based on a reasonably large sample of users, and not on what's happened in your basement/office/bedroom since Christmas.

Attributing any advantages to Window's maturity without evidence is about as spurious as attributing it to all the blue and green in the default theme. It's not as if old code rots and becomes decrepit, nor is it true that OS X in whole is newer than XP. Major portions were first written before 32-bit Windows existed.

Share this post


Link to post
Share on other sites
_Pablo    0

Is there any evidence that x86 Macs are more or less stable than PPC versions? Of what we can see of the core sub-system we know that the C code isn't significantly different.

The PPC and x86 architectures are different beasts. The kernel and drivers will need at least a little assembler to talk to the hardware. The platform has changed significantly between PPC and Intel so many if not most of the drivers would have changed. Many optimisations in the PPC used Altivec which has now been replaced with either non-optimised code or SSE and of course the endian-ness is different. There are enough differences here to make any assumption that the PPC and Intel versions will behave the same invalid.

Your post was all about you. I said that one persons experience is not enough to draw any meaningful conclusions in this thread. Now you're talking about all the other posters in this thread or forum, something you neglected in the post I quoted.

Are they not?

At no point did I say that my experience was the rule and it's disingenuous of you to suggest that was the case. I gave my answer to the posted question over the past six months because that is where I have direct evidence.

Are they not?

I am not a fly so I can't answer the question but my guess would be not...there must be some gourmet s**t out there!

The market share argument doesn't hold up on it's own (see apache/iis as the typical example)

More users = more testers = more bugs found. Sure if you start with an idealised completely bug free system then an infinite number of users still won't find a bug...but every non-trivial system has bugs so users x time = more bugs found. Programmers can only target bugs they know about.

Comparing the stability of operating systems a year from now isn't going to be any more fair than comparing the current ones. It'll just be different.

So by your logic no operating system would ever become more stable, Windows XP is exactly as stable in it's current SP2 patch level as it was at RTM?

On the other hand, saying things like "XP is slightly more stable than Mac OS X due to it's maturity" is going to be valid when it's based on a reasonably large sample of users, and not on what's happened in your basement/office/bedroom since Christmas.

I said Windows XP SP2 today is more stable than Windows XP RTM was due to it's maturity - i.e. the work that has gone into stabilising it over five years. If Apple spend five years patching Mac OS X 10.4.7 then it will be more stable than it is today.

Attributing any advantages to Window's maturity without evidence is about as spurious as attributing it to all the blue and green in the default theme. It's not as if old code rots and becomes decrepit, nor is it true that OS X in whole is newer than XP. Major portions were first written before 32-bit Windows existed.

No, suggesting that an operating systems stability is a product of the patches made to it over five years has is a technically feasible argument, whereas suggesting it is a product of its GUI colours is not. Code doesn't rot, but the platform it runs on changes and the environment it lives in changes so code has to change to better survive in the environment (firewalls, protection against buffer overflows etc) and to take advantage of hardware advances (MMUs, DEP etc).

We are not talking about NextStep 1.0 though are we?

Share this post


Link to post
Share on other sites
oldaccount1    3

I think they both have their perks, which is why I'm all for having both installed on one machine. And the only way to do that legally is to get a Mac. Mmm.

Share this post


Link to post
Share on other sites
the evn show    138

The kernel and drivers will need at least a little assembler to talk to the hardware.

There aren't significant differences (in terms of lines of code or algorithms) between the x86 and PPC branches.

There's also very little assembler in the Mac OS X source code (that we can see).

Many optimisations in the PPC used Altivec which has now been replaced with either non-optimised code or SSE and of course the endian-ness is different.

Pure altivec code is almost not existent in the stuff we can see. Most is written against the fastmath, vimage, blas, etc. libraries. I can't remember the last time I saw an endian problem in OS X: it's been agnostic for ages - since before Apple was even using most of it.

Steve made mention that been conscious of making sure OS X was platform agnostic for ages, I could dig up some sources if you like

nt. There are enough differences here to make any assumption that the PPC and Intel versions will behave the same invalid.

And yet the behave almost exactly the same. The most significant changes were in the boot process where openfirmware gave way to bios and later EFI. There's no evidence so far to suggest that these changes have compromised or improved Mac OS X stability or security.

So by your logic no operating system would ever become more stable, Windows XP is exactly as stable in it's current SP2 patch level as it was at RTM?

I said that a comparison made between two products available today will not be any more fair if we were to compare to other products at some random point in the future. Whether Mac OS X or Windows++ becomes more or less stable in 1, 10, or 100 years won't matter because we'll still be comparing it against contemporary competition.

The fairness of the comparison won't change in any way, just the standing of one product against the other.

It sounds like you're trying to say that the measure of fairness is determined by who wins and who loses, or who was better prepared. To hell with that, if Apple isn't shipping a good operating system today then they deserve to be blasted for it. I don't care what the reasoning is, if your products suck then everybody should know about it. Likewise, if 15 years of desktop operating system dominance hasn't solidified Windows then Microsoft deserves to be called on their ineptitude and Apple does deserve some praise.

More users = more testers = more bugs found.

By one count, IE6/Win had 17 to Firefox's22. It doesn't follow that the number of users of a piece of software proportional to the number of vulnerabilities found in that software.

I propose another equation

More bugs in software = more bugs discovered in software.

rogrammers can only target bugs they know about.

Good practices and well designed processes combat bugs you don't know about. God invented unit, functional, integration, and interface testing for a reason. Good algorithms are resistant to new attack vectors.

Look at how little OpenBSD changes from release to release. Once the system was hardend there was very little need to update things like Apache or Bind. Core pieces of the kernel see updates to take advantage of new hardware or to improve performance, but in the end a good practices have lead to a pretty solid security track record.

Of course maybe everybody is just too afraid of Theo to actually try breaking it.

Share this post


Link to post
Share on other sites
_Pablo    0

There aren't significant differences (in terms of lines of code or algorithms) between the x86 and PPC branches.

A bug can be caused by a single character, let alone a line of code or a function.

There's also very little assembler in the Mac OS X source code (that we can see).

But rest assured it's there - at some point you have to talk to the hardware directly.

Pure altivec code is almost not existent in the stuff we can see. Most is written against the fastmath, vimage, blas, etc. libraries. I can't remember the last time I saw an endian problem in OS X: it's been agnostic for ages - since before Apple was even using most of it.

But again, it's there and it needed to be ported so there was room for a bug to slip in.

Steve made mention that been conscious of making sure OS X was platform agnostic for ages, I could dig up some sources if you like

And yet the behave almost exactly the same. The most significant changes were in the boot process where openfirmware gave way to bios and later EFI. There's no evidence so far to suggest that these changes have compromised or improved Mac OS X stability or security.

Indeed, on the whole the source code is the same and excluding any compiler issues, alignment, endian and miriad of other issues it should function exactly the same - but with something as complex as an OS there are no guarantees. Sure the whole Marklar project meant that OS X x86 was moving forward in step with the PPC build, but the PPC build was seeing millions of users whereas the x86 build was tested in a lab. At the end of the day I do agree that my experience of OS X x86 is solid.

I said that a comparison made between two products available today will not be any more fair if we were to compare to other products at some random point in the future. Whether Mac OS X or Windows++ becomes more or less stable in 1, 10, or 100 years won't matter because we'll still be comparing it against contemporary competition.

It sounds like you're trying to say that the measure of fairness is determined by who wins and who loses, or who was better prepared. To hell with that, if Apple isn't shipping a good operating system today then they deserve to be blasted for it. I don't care what the reasoning is, if your products suck then everybody should know about it. Likewise, if 15 years of desktop operating system dominance hasn't solidified Windows then Microsoft deserves to be called on their ineptitude and Apple does deserve some praise.

I agree that the OP is asking about the stability of the OS's as they stand today. But I say it's doing Apple a dis-service to compare a relatively new OS against a five year old pile of patches, so for me it'll be a fairer comparison to compare Vista against Leopard when both OS's are at roughly the same point in their life cycle.

By one count, IE6/Win had 17 to Firefox's22. It doesn't follow that the number of users of a piece of software proportional to the number of vulnerabilities found in that software.

Again, you are ommiting time. IE 6 is as old as XP and Firefox is a relative as it only hit 1.0 at the end of 2004. So in the four years preceeding the measurements you quote IE6 had many many more bugs fixed.

I propose another equation

More bugs in software = more bugs discovered in software.

Yes, but if there are a very limited number of users then the combinations necessary to expose some of the less trivial bugs are less likely to get hit. So I would propose:

time x number of users x bugginess of the code = number of bugs discovered

Good practices and well designed processes combat bugs you don't know about. God invented unit, functional, integration, and interface testing for a reason. Good algorithms are resistant to new attack vectors.

No, good practices and well designed processes stop bugs being created - they do not combat bugs you do not know about. God didn't invent any of these things, man did and for that reason alone no programmer worth their salt will guarantee their code as bug free (except contractors!).

Look at how little OpenBSD changes from release to release. Once the system was hardend there was very little need to update things like Apache or Bind. Core pieces of the kernel see updates to take advantage of new hardware or to improve performance, but in the end a good practices have lead to a pretty solid security track record.

Of course maybe everybody is just too afraid of Theo to actually try breaking it.

Conflating stability with security.

Share this post


Link to post
Share on other sites
the evn show    138

Still nothing but hand waving to support your claims.

But again, it's there and it needed to be ported so there was room for a bug to slip in.

They can also be audited out.

Again, you are ommiting time. IE 6 is as old as XP and Firefox is a relative as it only hit 1.0 at the end of 2004.

Version numbers mean very little, Firefox started (before a run of identity crisis) had it's first release in fall 2002, Windows XP was less than a year old at that time. Work on core components dates as far back as the mid 90s.

Also note that links provided only counted issues found in 2005 after both systems have had several years of 'real world' use.

So in the four years preceeding the measurements you quote IE6 had many many more bugs fixed.
As did phoenix/firebird/firefox.
time x number of users x bugginess of the code = number of bugs discovered

Changing the first two coefficents will do nothing if the code is infinitely buggy or infinitely not buggy. On the other hand infinite users or infinite time will not find bugs in flawless code, nor will it miss something in infinitely broken code.

The part that matters is the number of bugs in the code.

At least you're starting to accept that the quality of the software plays some roll, your original equation made no mention of that. Once you just need to stop fixating on the tagents you come to that nice clean one I posted earlier.

No, good practices and well designed processes stop bugs being created - they do not combat bugs you do not know about.

The process that happens before you open an IDE prevents you from creating bugs. The process that happens after you've made a bunch of code is what catches the stuff you didn't know you've written.

What's the point of testing if you know there aren't any bugs?

Conflating stability with security.

I was talking about the common (in the bsd community) notion that reboots were for installing new versions of the operating system or changing hardware.

quick recap:

it needed to be ported so there was room for a bug to slip in.

Fear? Check.

with something as complex as an OS there are no guarantees.

Uncertainty? Check.

[if there are]a very limited number of users then the combinations necessary to expose some of the less trivial bugs are less likely to get hit

Doubt? Check.

We have a winner!

Share this post


Link to post
Share on other sites
drshdw    0

OSX is just as bad as XP. We have macs at work, the whole building uses macs, from iMacs to G5s to iMac Core Duos, they all love to kernel panic, make funny noises, spin fans up and down, etc. I'm sure PCs do it too. Also programs in OSX tend to boggle down WindowServer (for those who know what it is), and it's a bitch to fix if that program hasn't been updated. :rofl:

Also a common misconception is that OSX's built in defrag is pretty decent, however over time, the disk will still become defragmented and OSX's speed will slow down, just like XP. TechTools and Diskwarrior are great time savers, however need be to boot off a seperate OSX partition, annoyingly.

In the end, OSX & XP are about the same, both have flaws, regarding which is better, it's more of what you're more comfortable with using.

Share this post


Link to post
Share on other sites
RD_SG    1

I think OS X runs a bit smoother than XP does. The problem is, there are much more third-party applications developed for Windows than there are for OS X, and that's the reason why XP seems to crash more. I think OS X is more user friendly than Windows, it's easier to install and remove applications and such, however there are still a bunch of programs (i.e: Adobe Photoshop) which are not universal binaries yet.

OS X is definitely more impressive visually and has features that XP doesn't have. (i.e: Dashboard, Spotlight)

I'm sure if more third-party applications were developed for OS X, we'd be seeing a lot more viruses on the machine.

They both have their pros and cons, and they excel in different areas.

Share this post


Link to post
Share on other sites
BajiRav    2,137

Both are same in terms of stability problems and overall system freezes,

only difference is that OS X needs more RAM and Windows XP demands more attention towards security.

Share this post


Link to post
Share on other sites
stadsport    0

Don't be fooled by the simplicity of the OS X UI. Once you dwelve into the command line you'll find the power of unix lurking just beneath the surface. Windows feels opaque and GUI-dependent by comparison.

I would agree with this, but I don't feel getting into Command line is even necessary. OS X's UI seems simple on the outside, but everything is there. It's not lacking any features, it's just well put together.

Take a bunch of legos and throw them on the floor and it looks like a huge messy pile of legos. Assemble those legos into something neat and compact and it's still all there. It's just not a messy pile of legos anymore.

Share this post


Link to post
Share on other sites
_Pablo    0

quick recap:

Fear? Check.

Uncertainty? Check.

Doubt? Check.

We have a winner!

Wow, i've been served!

Share this post


Link to post
Share on other sites
lawownsu    0

if you got a crappy windows pc and a good mac yea the mac is definatley more stable but if you got a good windows xp on the other side they are probably equal :D!

Share this post


Link to post
Share on other sites
SionicIon    76

I love the performance and stability of OSX!

Share this post


Link to post
Share on other sites
kogepan888    0

I think it is. Although i have to admit i spent more time on PC yet when im using my mac i feel must safer than when im using my PC. :D

Share this post


Link to post
Share on other sites
DigitalN.    2

I have definately froze my iBook more times than I have froze my Windows PC since december. although that could be due to me doing severe multitasking on 512MB of RAM and a 1.43GHz CPU compared to 2GB of RAM and a 3GHz A64 :D

Share this post


Link to post
Share on other sites
hagjohn    2,241

My 2 cents.... I think both OS X and XP are both pretty stable. Of course it depends on what software, hardware and drivers your using. If your using cheap hardware, bad software and/or badly written hardware drivers, then any system will not be stable.

My XP has only crashed once or twice in the 5 yrs I've had it, of course, I only use decent hardware on it and it was the 3rd party software I was trying to use that caused the crash.

On the Mac side, macbooks (and I believe some mbp's) are shutting down, probably due to heat issues, so I think it's pretty much a wash between the 2 systems. Go to any Mac releated forum and look around, you can see that Mac's suffer from similar issues as Windows systems do.

Share this post


Link to post
Share on other sites
Septimus    9

@op

No. Both about the same.

Share this post


Link to post
Share on other sites
+warwagon    13,409

one could argue, that Mac's are more stable because of the lack of spyware and viruses, but it also has to do with the lack of computer configurations all of which apple controls, unlike on the PC end.

Share this post


Link to post
Share on other sites
stadsport    0

one could argue, that Mac's are more stable because of the lack of spyware and viruses, but it also has to do with the lack of computer configurations all of which apple controls, unlike on the PC end.

This is debateable. Without going into too much detail, I've had OS X running extremely stable on non-Apple hardware.

Share this post


Link to post
Share on other sites
hotwire    0

hm linux rulez all. but it seems Mac runs smoother but of course it's only because most are not 3rd party software.

Share this post


Link to post
Share on other sites
primexx    372

in theory, not really, in reality, yes.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.