Mac OS X 'Snow Leopard' Features and Parallel Computing

[Stridder44]

Why is this being touted as 10.6?? It sounds like they are just patching and making 10.5 more reliable and effective?? Sounds like service Packs to me..


Without any new added features and just background features this should be a free upgrade or just 10.5.8 or something not 10.6

I don't think you understand what a service pack is because this is WAY more than a service pack would ever do. They fix holes and patches. This is going to reduce file sizes, update programs, etc. But yeah, I hope it's not $129. $20 for Leopard users would be nice.

 

[morespce54]

...

Back to SNow Leopard, though it won't make the same media splash, and won't bring as much love to APple, I see this as a very important thing for a number of reasons:

1. It shows Apple cares about bloat, and about improving things as time goes on at a level the average person doesn't even understand.

2. Apple can implement under-the-cover technology that does not give Apple the wow factor, but allows developers to give the wow factor.

3. It's all work that could be used directly towards other devices, say, devices that may not have as much processor or memory as modern day personal computers.

4. It will make 10.7 all that much better. Imagine with all this solid foundational work done, how much they can focus on the UI/ground breaking interaction that we know Apple for (at least know them for more than most other companies).
...

~Tyler

Agree. Good things are always welcome, as much as improvements.

But my question is: Wouldn't that be strange, marketing wise (and by a company that is really good in that matter), to put out a new OS without any flashy things to show up?

I can see that tagline:
The new OS X: better, faster, more stable... and the same old desktop backgrounds!

Seriously, I can see IT teams being excited about a more stable version but how is that gonna sell to John Doe?

 

[flopticalcube]

Agree. Good things are always welcome, as much as improvements.

But my question is: Wouldn't that be strange, marketing wise (and by a company that is really good in that matter), to put out a new OS without any flashy things to show up?

I can see that tagline:
The new OS X: better, faster, more stable... and the same old desktop backgrounds!

Seriously, I can see IT teams being excited about a more stable version but how is that gonna sell to John Doe?

Hence the name. I would question it if they didn't offer it as an upgrade to Leopard users. I would probably wait until 10.7 then.

 

[haravikk]

I'm still hoping they add at least one new feature; a new bloody Finder! I've wanted one since OS X.0, the thing doesn't do a damned thing that I want it to, is asking it to remember window and side-bar sizes, or view-options REALLY so much to ask? Not to mention all the extra things they could do like add tabs, or visual clip-boards and things, tabs would be great for me as I can then open one finder window on my second monitor and do loads with that instead of NEEDING exposé all the time. Sure it's a great feature, but being able to use it and needing to use it are two different things.

Rant on the Finder aside, I'm excited to see what these improvements are like. Personally I don't have much issue taking advantage of multiple cores since I typically work with thread-pools and queues of smaller tasks, but better capabilities are intriguing, being able to leverage the GPU in addition to that could be very interesting.

In-built exchange support could rock too, since I might be moving to a company that uses it heavily.

 

[irun5k]

But to categorize what they're doing is just "patching" then it clearly shows you do not understand what Snow Leopard is. Fair enough, since it's new, but the improvements here are far beyond fixing a few bugs and optimizing some algorithms.

Right, but the question is, how much value does this have to the average consumer? For example, if I write a shareware app that sells for $20 and then I decide to re-write the whole thing in a new language, and sell the new version for $20, it is doubtful I'll get many buyers. Even if the app takes up slightly less disk space and runs a wee bit faster..."big deal" most users will say.

Now, presuming I had a good reason for re-writing it... i.e. it makes it easier for me to build upon and add new capabilities in the future... well- guess when I'll reap most of my financial benefit... in the future, when I actually release something that takes advantage of my re-write.

It is up to Apple, but honestly, I do not thing many people will pay $129 for this upgrade. Personally, I'd pay $20, $30 bucks for it. Most importantly though, I will probably pay $129 for 10.7, since presumably it will kick ass due to all the work being done for 10.6.

 

[guzhogi]

I think computers today are a lot more powerful than we know, it's just that very few OSes & apps are completely optimized for multiple cores and all the technology we have today.

 

[Hattig]

That person that posted the .app size comparison between Snow Leopard and Leopard should take a look inside the files to see what is there and what has been removed - some of the bundles are positively miniscule in comparison to what was there before - yet they're still Universal (binaries actually aren't the big part however, it's other media - images, etc, that take up space) and presumably have the same functionality.

Aside from application clean-ups to remove unwanted crud, what is there?

1) One binary that is run on a virtual machine at execution time (LLVM) or similar JIT compilation type scenario - would save some space, but not a lot because binaries aren't that massive in comparison to supporting files.

2) Removal of duplicate code - over time a lot of duplicate code must have been created for these applications, especially custom GUI widgets and the like. If these are put into system libraries, that's a lot of space saved (GUI widgets require supporting graphics files as well).

3) Vector UI instead of bitmap - could save space by no longer having bitmap images for many aspects of the system. However in memory they'd be bitmaps again, so not so much a runtime advantage.

4) Compressed .apps - maybe, but a lot of media content cannot be compressed much further.

5) Only available in English currently, no translations yet configured (installed upon demand maybe, rather than at application install time?)

What else can people think up? How in the hell could iTunes be a >120MB application anyway? The executable is >32MB alone, so moving to a single LLVM-style binary could save 16MB or more alone. Lots of 150kb icons in there too ...

 

[Eric S.]

Source aside, the points made are compelling against Apple actually licensing their operating system.

That was a joke (hence the smiley). I would say these points are just obvious. I don't see any serious arguments for Apple getting into the can of worms that would result from trying to support OS X on generic PCs. I once did software development for a company that produced Unix for PCs. The amount of third-party HW that needed to be supported, and the variety of bugs encountered, was extreme. The situation is a little better now (at least in theory), with standard interfaces like USB. But other HW like different graphics, audio, NICs, etc., would be a major headache.

 

[zioxide]

3) Vector UI instead of bitmap - could save space by no longer having bitmap images for many aspects of the system. However in memory they'd be bitmaps again, so not so much a runtime advantage.

Making the entire UI vector based would allow for scalable interfaces (resolution independence) to be implemented much easier too.

 

[haravikk]

2) Removal of duplicate code - over time a lot of duplicate code must have been created for these applications, especially custom GUI widgets and the like. If these are put into system libraries, that's a lot of space saved (GUI widgets require supporting graphics files as well).

I'd like to note on this point that a ton of code is likely now covered by various in-built libraries like Core Audio, Core Image etc. which many of these applications predate and thus probably didn't use.

 

[jettredmont]

So, I just scanned the article on Wikipedia about ZFS and not being too computer illiterate, I'm not gonna lie... it came out as a LOT of greek to me. It just seems like a terrific file system when it comes to handling TB's of information... lots and lots of TB's.

So how exactly does it benefit the average, normal end user? I can start to understand the need on a server side of things, so building it into SL would be advantageous, but why would the avg iMac consumer benefit from it any better than HFS+ is doing?

There's a MacBreak Tech show specifically on ZFS, to start: Zulu Foxtrot Sierra.

The main end-user advantage is that ZFS can enable a Drobo-style hard disk array where you can add another disk and it all just appears as a single volume. For example, if you have a 250GB ZFS volume, and pop a new 500GB drive into your box, you can have a single 750GB ZFS volume.

The "real" benefits, though, are sexy only to developers and geeks. It's a nice file system and offers many features which developers could take advantage of. Like much of Snow Leopard, it's not necessarily a selling point, but it enables developers to make their own selling points on top of it.

 

[Stridder44]

Basically, no. There are 2 CPU cores, but they use the same RAM, the same hard disk, the same I/O bus. A single core CPU is hardly ever utilized for 100% of its speed (despite what your tools tell you), a multi-core CPU wastes even more cycles while waiting for the data to be available. Of course, you will see some improvement if most of the system services utilize both cores, but not 70-90%. Leopard is already somewhat multi core optimized, so the difference between Leopard and Snow Leopard won't be very big.

This might seem a bit off topic, but I wonder how good Vista is at utilizing multi-core stuff (honest, fact-based answers, not "NONE cuz micro$oft is teh suxzor$!!1").

Can't wait for Snow Leopard.

 

[milo]

"RAM -- up to a theoretical 16TB, or 500 times what is possible today."

So by next year can we expect to see economical terabyte RAM chips?

Just *theoretical* economical terabyte ram chips...but since Leopard already supports 4TB, those must be shipping already...

Do you work for free? No? Then why do you expect Apple's development team to work for free?

When you work, do you do an incomplete job, then require your boss to pay you again to fix what you said you were going to do in the first place?

If people buy something that doesn't work as promised, they expect a fix for free, not to have to buy it again.

 

[winterspan]

True. the problem with wikipedia is that almost anyone can update it, and thus there has been articles about false or inaccurate information in it. I read anything on the web with a grain of salt.
but in the case of parallel computing - the whole point of multi CPU/processors is for the hardware to break down a command and offload each part to a separate processor than assemble it all back together again. faster than one person (per say) doing all the work.

simple enough terms?

A lot of the hype about wikipedia inaccuracy is bullocks. The vast majority of sections are properly managed and any modifications made have to be backed up by references to legitimate sources. An average person can't just anonymously rewrite a whole topic and expect for no one to notice. The advantages to a massive, open encyclopedia FAR OUTWEIGH the problems with maintaining accuracy, consistency, etc. For most routine topics, wikipedia is a *GREAT* source of information. Controversial topics, on the other hand, usually ignite editing propaganda wars and the articles get locked down. lol.

 

[diamond.g]

MS Exchange Support

Does anyone else chuckle as much as I do when they read about how excited Apple fans are to get support for MS proprietary technology?

 

[WilWheaton]

I'm still hoping they add at least one new feature; a new bloody Finder! I've wanted one since OS X.0, the thing doesn't do a damned thing that I want it to, is asking it to remember window and side-bar sizes, or view-options REALLY so much to ask? Not to mention all the extra things they could do like add tabs, or visual clip-boards and things, tabs would be great for me as I can then open one finder window on my second monitor and do loads with that instead of NEEDING exposé all the time. Sure it's a great feature, but being able to use it and needing to use it are two different things.

I would pay $129 for a konqueror file browser for OS X, or hopefully, with the upcoming native OS X support for QT4 we will see it for free . I'm not sure there's any way to obviously override finder as the default file browser though.

 

[DanB91]

A lot of the hype about wikipedia inaccuracy is bullocks. The vast majority of sections are properly managed and any modifications made have to be backed up by references to legitimate sources. An average person can't just anonymously rewrite a whole topic and expect for no one to notice. The advantages to a massive, open encyclopedia FAR OUTWEIGH the problems with maintaining accuracy, consistency, etc. For most routine topics, wikipedia is a *GREAT* source of information. Controversial topics, on the other hand, usually ignite editing propaganda wars and the articles get locked down. lol.

i say this alot but some people don't believe me

 

[gcdc]

Exchange 2007 only?

I noticed in the image and text on the apple website, they only indicate "out-of-the-box support for Microsoft Exchange 2007." So what if your company still runs Exchange 2003 or an even earlier version? Will it work at all with 2003? It seems odd for Apple to implement something that could conceivably cause at least a few companies (i.e. Exchange-using IT company with users on Windows and Macs), to be more likely to upgrade their Windows products. I would be upset if I upgraded to Snow Leopard expecting Exchange support only to find it wasn't compatible with what my company runs. They make it pretty clear it is Exchange 2007...but that seems odd to me. And some home users hoping for better access to work email/calendars/etc. could easily miss that detail.

On another note, I assume the Exchange Active Sync for iPhone will work with whatever relatively recent Exchange version your company uses?

 

[Trip.Tucker]

I would like to point out that most applications (actually all save for Bootcamp Assistant) are still Universal.

????

All applications in the world?
All applications on your computer?
What? What are you talking about?

 

[flopticalcube]

????

All applications in the world?
All applications on your computer?
What? What are you talking about?

In one of the pictures of 10.6 of the original article.

 

[i.mac]

Can you not see new products coming along that will only run on SL?

Those new products will makeup for what we may see as lacking in the new os. Besides, there will be new features, just not flashy ones.

 

[Trip.Tucker]

So those of us with 8 Core Mac Pros are in for a real treat when this all start to take effect!

Only if the application is multicore aware and able to take advantage of it.

In the one of the pictures of 10.6 of the original article.

I think I know what you're saying. Multicore applications may still appear as universal, I haven't seen any indication they may arrive at a new name.

i say this alot but some people don't believe me

I don't believe you

 

[gcdc]

Does anyone else chuckle as much as I do when they read about how excited Apple fans are to get support for MS proprietary technology?

Mac user @ home + forced PC/Exchange use @ work = very happy when apple improves Exchange/PC support.

Easier telecommuting, checking things on-the-go with a laptop or iPhone, etc. MobileMe (even though I hate that name) will make my life easier b/c I will be able to keep all my calendars in sync, assuming it works with an Exchange-based Outlook calendar. If it doesn't, SL should fix that. (I've tried google calendar, but really like iCal, and being able to use it if I'm not online.)

Also, the more Apple plays nicely with Enterprise standards, the more they will be able to increase their market share and be a real threat to Windows stranglehold on corporations as the Enterprise customer realizes that Apple makes pretty good stuff.

 

[zioxide]

Only if the application is multicore aware and able to take advantage of it.

That's the whole point of 10.6.

 

[diamond.g]

Mac user @ home + forced PC/Exchange use @ work = very happy when apple improves Exchange/PC support.

Easier telecommuting, checking things on-the-go with a laptop or iPhone, etc. MobileMe will make my life easier b/c I will be able to keep all my calendars in sync, assuming it works with an Exchange-based Outlook calendar. If it doesn't, SL should fix that. (I've tried google calendar, but really like iCal, and being able to use it if I'm not online.)

Also, the more Apple plays nicely with Enterprise standards, the more they will be able to increase their market share and be a real threat to Windows stranglehold on corporations as the Enterprise customer realizes that Apple makes pretty good stuff.

Overall I agree, I still think it is funny that some of us are excited about Exchange support. More than anything I would hope that Snow Leopards push support would help Apple to actually displace Exchange.