A recent article by Tony Smith from The Register titled “Mac OS X 10.3 Panther will not be a 64-bit OS” caused a good deal of confusion with many people, including me. It is also caused a heated argument here on OSNews. The basic point of the article is that Mac OS 10.2.7 and 10.3 are not “true” 64-bit OSes, but the article does not clearly explain what a “true” 64-bit OS is. This had led to a lot of claims that the article is false or misinformed, rather than just unclear, which is certainly is.
It turns out that Smith’s article on The Register is more or less correct. More and more evidence is out there that applications on Mac OS 10.2.7 and 10.3 cannot address more than 4GB of RAM, which is the limit imposed by 32-bit operating systems. The basic evidence is from a post on Apple’s darwin-development list by Apple employee (or email account holder at least) Godfrey van der Linden. Van der Linden writes:
“Yes the kernel’s virtual address space will remain 32bit for the medium term but self-evidently we can address >32bit of physical memory from the kernel.”
So the kernel in current (10.2.x) and future (10.2.7 and 10.3 Panther) versions of Mac OS X will only support a virtual address space of 2^32 bytes = 4GB of memory. No one really cares how much memory the OS internals use, the key question is can real life applications break the 4GB barrier? One non-Apple poster on the darwin-development list, Shawn Erickson, speculates:
“Outside of that I believe the user mode libraries/frameworks will remain 32 bit and hence most processes/applications must remain 32 bit (if you are careful of what you use I see no reason you could not use 64 bit addressing in certain tasks). I have not seen anything yet that implies that new 64 bit addressing versions of the systems libraries and framework will be provided in Panther (I bet they will be recompiled to use 64 bit native math, etc. just not addressing). ”
Whenever you compile even the simplest application on an OS, you must link to libraries of some sort. While there are technical details yet to be revealed by Apple, if there are no 64-bit libraries on Mac OS X, the OS does not support 64-bit applications, in all practicality.
An anonymous poster on a Slashdot thread (OK, not the most relibable of sources) confirms this:
“Your code could use 64-bit addresses after being recompiled. But there is only one set of mmap(), etc… traps on the system at the moment (both 10.2.7 and Panther), and they only supports 32bit addresses. So, there is no way to put anything above 4GB in your address space to back those 64-bit addresses.”
Finally, The Ego on the same Slashdot thread says:
“An OS could be called “64bits OS” because it offers a 64bits ABI to userland applications compiled in “64bit mode”. My understanding is that this will _not_ be the case for Panther, all code will be compiled with the traditional ILP32 ABI.”
What does this mean? First, an individual application (or process) on current (10.2.x) and future (10.2.7 and 10.3 Panther) versions of Mac OS X can only address 4GB (2^32 bytes) of memory. So Panther does not change the status quo. As someone who writes Fortran programs that use lots of memory (but is not a computer scientist), MY personal definition of a 64-bit OS is one in which an individual application can use up to 2^64 bytes (17 billion GB) of virtual memory. OSes such as Solaris and 64-bit versions of Linux and Windows meet my test. Mac OS X, in what van der Linden calls the “medium term” (at least the next major release cycle, Panther), does not meet my test.
The other part of van der Linden’s post says that the kernel can support more than 2^32 bytes=4GB of physical memory. So, yes, the 8GB of memory that Apple claims the high end G5 model supports will indeed work. It is just that, from what I gather, no one application can grab all of it at once. Instead, each application can use only 4GB (the virtual address space limit which has not changed under Panther).
If this sounds suspiciously like 32-bit Windows, you are correct. The Windows NT kernel (found in Windows NT 4, 2000, XP and 2003) supports more than 4GB of physical RAM, but each process has only a virtual address space of 4GB, although the practical limits are much lower, as I will explain in a minute. For commercial market segmentation reasons this >4GB of physical RAM feature appears only in the 32-bit versions of Windows 2003 Enterprise and Datacenter Editions. Also for market segmentation reasons, hardware support for this feature is found only in the Intel Xeon chips, and not the nearly identical Pentium 4s. The 32-bit x86 Linux kernel also has support for the Xeon’s memory addressing abilities. Somewhat humorously, Red Hat is playing a similar segmentation game by charging extra for the more than 4GB of physical RAM feature.
As a side note, if you care, an Apple developer’s guide for the G5s says that they support 2GB DIMMs, so the high end G5 can really use 16GB of physical RAM, if you have the money. As a further side note, Motorola’s documentation for its recent G4 processors suggest that they can support 64GB of physical RAM, so Apple did not technologically have to wait until IBM’s release of the PowerPC 970s to support more than 2 or 4GB of RAM in PowerMacs and XServes. It’s not clear that Apple will support any of the 64-bit defining features of the PowerPC 970/G5, unless you are a cryptographer who cares about integer arithmetic with larger numbers.
Let’s go into some more detail about what the actual memory capabilities of Mac OS 10.2 Jaguar are. From some experimentation with C and Fortran codes, I have found that individual applications on 10.2 can address close to 4GB of virtual memory, even if the machine has much less physical RAM, as all currently shipping Macs do. The catch here is that only 2.25GB of this RAM is contiguous. So a simply written C or Fortran program cannot address more than 2.25GB of memory. The remaining 1.75GB of memory is not contiguous either. The problem here is that various shared libraries are loaded into memory at fixed locations in the 4GB of virtual address space, which splits up the available memory for the user.
How do 32-bit Windows and Linux compare? Windows by default allocates 2GB of RAM to each process. The other 2GB of RAM in the 4GB of virtual address space is reserved for the OS. In Enterprise Edition, there is a flag you can set to change the allocation to a more fair 3GB application/1GB OS breakdown. On top of this 2GB memory limit, 32-bit Windows allocates shared libraries in a poorly planned random manner, so the 2GB of address space is split up into chunks, and actual programs may only be able to access 1.5-1.7GB of contiguous memory.
32-bit Linux splits its 4GB of virtual address space per application into 3GB for the application and 1GB for the OS. However, of the 3GB for the application, only 2GB are contiguous, so again, simple C and Fortran programs (and many real life applications) can only use 2GB.
Apparently Microsoft has some hack so software vendors can use more than 2GB of RAM in Windows, and The Register article that started this discussion article alludes to the fact that Apple has added a similar hack to Panther. But these hacks are only practical for big money software companies such as Oracle and Adobe, and the performance of the hacks is often less than going to a true 64-bit application on a true 64-bit OS such as Solaris, AIX, HP-UX, Irix, zSeries, Tru64 and the 64-bit versions of Linux, Windows and the BSDs.
Why might someone want to use more than 4GB of virtual memory? First, some scientific problems require that much memory. Fair enough. Second, people editing video or other large multimedia files might want to put the entire file into memory, to work on all at once rather than only one piece at a time, which reduces loading from the disk. Third, databases are really big and getting bigger, and putting more of your data in memory speeds up access. Fourth, even more people with very little physical memory, 64-bit virtual memory can be a big help. Remember those annoying shared libraries that take memory away from the user? In a 17 billion GB address space, there is presumably plenty of room to stash those shared libraries so that they never cramp the space taken up by your actual data.
A good way to conclude this discussion is to summarize Apple’s market position with regards to memory handling. Mac OS X is not a 64-bit OS, by my definition emphasizing the ability of applications to use 64-bits of memory. Therefore, Apple is behind almost all commerical UNIXs, 64-bit versions of Linux for many platforms including Intel’s Itanium and AMD’s Opteron/Athlon64, and 64-bit versions of Windows for Itanium and Opteron (the latter is to be released in early 2004 to OEMs only, rumors say).
However, as far as 32-bit OSes go, Mac OS X is actually somewhat ahead in the memory game. The virtual address space available to an application is close to 4GB, which is more than the 2GB Windows provides and the 3GB Linux provides. And the Mac’s contiguous address space of 2.25GB exceeds the 1.5-1.7GB on Windows and the 2GB for Linux. Furthermore, Apple will not require you to shell out big money for some “Advanced Server” version of its OS if you want to install more
than 4GB of physical memory in your machine, which is what Microsoft and Red Hat want you to do.
The downside of Apple’s support of more than 4GB of physiscal RAM is that there is less need for it on a Mac. Most Macs are used by one (professional) user at a time, and this user is typically going only to use one big application at a time. Unless this user opens up two massive applications, he won’t need more than 4GB of RAM, and he cannot use more than 4GB of RAM in one application because he is not running a true 64-bit OS.
On the other hand, Windows servers often run Terminal Services, which, while expensive, lets multiple users remotely log onto a GUI session and run multiple applications. If each user opens a big application, that extra memory can become handy quickly. Apple does not offer a Terminal Services solution, instead letting you log on remotely via ssh or telnet to a command line or X-Windows environment, where you are pretty much limited to using ported open source programs.
If you want a 64-bit OS on the desktop, use the appropriate versions of Windows or Linux on an Itanium or AMD64 (Opteron, Athlon64) machine. I would suggest AMD64 because it’s fully backwards compatible with existing 32-bit software for the appropriate OS, althogh Intel is trying hard to improve x86 emulation on the Itanium. Near-future versions of the Mac OS can run on a 64-bit processor from IBM, but will not themselves support 64-bit applications.
Update: A similar article, at eWeek.
About the Author:
A sometimes Fortran programmer who uses Mac OS X on his desktop machine and Solaris to run his codes.
32-bit Linux splits its 4GB of virtual address space per application into 3GB for the application and 1GB for the OS.
By default, yes, but you can change this with patches on the Net. For example, I can choose the splitting I want (well, 1, 2, 3 or 3.5GB) with Gentoo’s custom kernel. However, I must admit that I never used that feature, so I don’t know if it’s safe or not, and I don’t know if you can change the contiguous address space.
The hacks the article refers to are the addressing window extensions. You can create a “window” of virtual memory and map in parts of a > 4GB chunk into that window. If you need to access memory outside the window, you move the window to where you need it. Its just like bank switching or segmentation for you old fogies that remember that. Its not pretty, but its hardly only suitable for large companies as the author suggests. Heck, anybody who wrote 16-bit code in DOS could handle it without problems.
Gee, in the 1970’s I was told a word length is the OS’s natural language size. A 64 bit word length OS is the only measure of what would run on a 64 bit computer is it not? In other words, can the OS understand a single 64 bit command? Not a hacked or doubled command. The instruction set of the processor must be 64 bit even it only the ability to use the total command is there.
“It’s 64-bit.” “No, it’s 32-bit!”
So Panther is the Windows 95 of OSX?
“Mac OS 10.2.7 and 10.3 are not “true” 64-bit OSes”…
whoopdie do. Apple has a fast chip now. I am happy. I don’t care if the OS is a “true” 64-bit OS or not.
Unfortunately I do care about whether Panther is 64-bit or not. I have a cube that is plenty fast enough for my daily operations, but I had planned on buying a new G5 to begin doing 64-bit scientific computing development. It would have been a very cost effective alternative to the Itanium platforms running Windows. I’m profoundly disappointed.
Right now, in HP/UX 11i, it is a jumble of 32/64-bit APIs, some are 32-bit data vs. 64-bit data, others are 32-bit addressing vs. 64-bit addressing, others use large files (>2G), etc. Every single combination of “bit-ness” requires a separate library and corresponding set of compiler switches. Once you roll in 3rd-party libraries, the mish-mash of ABI’s is a pain in the ass. Because of this, an HP/UX programmer using 3rd-party libraries is safer using 32/32-bit versions.
Does it really freaking matter if every single program used on the Mac is bound to a 4G barrier that no one approaches, as opposed to breaking every application, increasing the development and QA cycle, just for a single Fortran programmer?
Give me a break.
For those of you who are not aware OS X enforces that all applications run prebound.
During installation of an application, the final stage is “Optimizing System Performance” at which point the entire filesystem is scanned by a process called update_prebinding and all applications found on the system are pre-assigned a portion of the address space in which to run.
Unfortunately, this can be somewhat of a problem on a 32-bit architecture, considering that there’s only 4GB of address space available. I have no idea what OS X does if the virtual image size of all applications and libraries on the system exceeds 4GB, and this has always worried me.
So, my question would be what is the size of the virtual address space accessable by the kernel and the userland prebinding applications? Will all applications be prebound to a portion of a 64-bit address space?
With the number of people running Panther betas, I am surprised that more of this information isn’t available. Most of the information available as to which portions of Panther (if any) have gone 64-bit seems to be hearsay and speculation.
The most likely scenario I can see is a 64-bit kernel and (almost exclusively) 32-bit userland when operating on a PPC970 system.
One would hope that Apple would soon take the Solaris approach and start shipping 64-bit counterparts to all 32-bit userland libraries. From this article though, it seems dubious if this will happen in Panther as Apple doesn’t seem to have implemented a 64-bit ABI yet.
I Don’t see why they will not move XServers and OS X server over to True 64 bit server systems. there is little reason not to realy since a server can realy benifit from this.
so Hank….what do you think about buying a 64 bit Server for scientific programming?
so Hank….what do you think about buying a 64 bit Server for scientific programming?
As a back end processor that would be fine. Offline visualization with 32-bit program would be possible, but I would rather have a full 64 bit solution.
> whoopdie do. Apple has a fast chip now. I am happy. I
> don’t care if the OS is a “true” 64-bit OS or not.
Well, for that matter neither do I (not much anyway). It’ll mean that we won’t have to worry about application compatibility woes for some time, at least until apple switches to 64bit fully (which they WILL have to do at some time).
The main plus from all of this is that the processor wars look to be over. The MHz myth is now history and pc-buyers can buy computers based on software merit rather than superiority of hardware. Not to say PPC has ever been inferior technically, only speed wise.
But anyway, those wanting 64bit computational ability can use NetBSD, OpenBSD or Linux. I’m confident that these O.S.S OSses will soon have PPC970 ports available.
isn’t this all just speculation? won’t we have to wait and see what apple adds to gcc?
-hugh
RAM prices are going to deter all but the most determined from using the full capabilities of the G5.
A 2GB DDR RAM stick costs US$1300 or $10,000 for the full 16GB complement (8x2GB sticks) for a dual G5.
$US10,000 for RAM – That’s a lot of money.
“If you want a 64-bit OS on the desktop, use the appropriate versions of Windows or Linux on an Itanium or AMD64 (Opteron, Athlon64) machine.”
…in which case, you will still be running 32-bit applications under unless they’ve been recompiled, or you’re compiling it yourself.
From what I’m reading, G5 programs can be compiled to use PowerPC64 instructions but retain 32-bit addressing and object modes. Is it disappointing that they won’t do true 64-bit addressing in OS X 10.3? Sure. But what makes a CPU “X-bit” is not the amount of memory it can address without using segments. If address lines were what this was judged by, we would be talking about the PPC970 as a 42-bit processor–and even more absurdly, we would be talking about Apple IIs, TRS-80s and Commodore 64’s as 16-bit computers, not 8-bit, since they could all address 64K of contiguous memory (instead of the 256 bytes 8-bit addressing would allow).
>> If you want a 64-bit OS on the desktop, use the
>> appropriate versions of Windows or Linux on an
>> Itanium or AMD64 (Opteron, Athlon64) machine.
> …in which case, you will still be running 32-bit
> applications under unless they’ve been recompiled,
> or you’re compiling it yourself.
This is particularly true on WINDOWS. However, on O.S.S operating systems available for the powerpc such as Linux, OpenBSD and NetBSD have most if not all of the applications compiled from scratch *FOR* their appropriate archicture. This goes without saying for OpenBSD and NetBSD, particularly because of their ports system. Linux will be available on PowerPC 970 with Yellowdog (I know) and Debian (I believe). Both of these distros are more than likely to compile everything with 64bit instructions.
Admittedly, closed source applications such as Opera and Oracle will have to have their respective companies recompile them for us. Oracle will most likely be one of the first vendors to make such a move.
Most typical Mac users won’t even notice this. But, I acknoledge that some -such as you and I – will. Those that do are free to use one of the *BSDs, Linux or Fink to bring full 64bit applications to our desktops.
hehe, at those prices I definitely will never stress panther to its limits.
$US10,000 for RAM – That’s a lot of money.
Particularly since, even after currency conversion, RAM is about 2/3 – 1/2 the price in the US as it is everywhere else.
If one really wanted 64-bit applications on a G5 and wanted to address > 4GB of memory, could one use a 64-bit PPC Linux? Does such a thing exist? Is it stable? Do 64-bit Linux applications such as gcc exist for PPC? That could be an answer for some people in the short term if they really need 64-bit addressing. Then again, that wouldn’t work for people who depend on mac applications or proprietary applications that are compiled for specific archetectures (like Linux x86). However, if you are developing your own apps using gcc or whatnot Linux could be a solution for you. I am not suggesting this for the long term, but, if you need 64-bit NOW, Linux could be a viable option.
As an aside, if you really expected Panther to be fully 64-bit, I think that you were deluding yourself. It takes time to make a 64-bit kernel and userland, and I am glad that Apple is taking the time to get it right rather than releasing something half-baked just to get it out. You wouldn’t want to be stuck with some crappy 64-bit ABI that you had to live with because Apple took shortcuts early on. Windows 95 wasn’t fully 32-bit at first, OS/2 wasn’t fully 32-bit at first, Solaris wasn’t fully 64-bit until Solaris 8 and Solaris was running on 64-bit hardware for years. I could go on with examples, but, to sum it up, while this news is disapointing, it is not unexpected.
Oh, one final last note. If you really really need 64-bit now in MacOS X, and you have some kernel hacking knowledge, perhaps you could hack something together in Open Darwin, the source is available.
Skipp
>> If you want a 64-bit OS on the desktop, use the
>> appropriate versions of Windows or Linux on an
>> Itanium or AMD64 (Opteron, Athlon64) machine.
> …in which case, you will still be running 32-bit
> applications under unless they’ve been recompiled,
> or you’re compiling it yourself.
The reason for using a 64 bit OS is either to use 64bit applications or program 64bit programs yourself. I don’t see the problem here.
here’s a link that will answer your question
http://www.linuxppc64.org/
Linux at this moment can run natively on Power 4 systems from IBM, if yeah go looking through the workstation section of their site yeah notice that when you order a Power 4 system you can either have AIX or Linux installed on it.
http://www.terrasoftsolutions.com/news/2003-06-24.shtml
Yes, Linux has been ported to the Power4 for a while, the 970 is its little brother (less cache, added AltiVec – dunno what else is significantly different) so a port to the G5 should be no sweat. YellowDog apparently expects to have it done by the time Apple ships, spiffy keen. I’ll be buying my PowerMac G5 from YellowDog, gotta love ’em. Oh, it should be noted that much of the PPC64 Linux work is done by IBM engineers themselves, internally not some outside enthusiast (ala Ultra/Sparc Linux, which Sun doesn’t pitch in). I’m sure that makes a difference. After all, it is their hardware.
Not sure if I’m jumping into the PowerPC or 64-bit desktop computing anytime soon. Maybe in 5 years when all these nonsense is not experimental and their is complete support for application tools, drivers and hardware. It’s too risky right now and I don’t intend to be anyones guinea pig.
Regards,
Mystilleef
IBM is going to release PPC970 Servers running Linux, so yes, Linux will be available in a 64bit version for the PPC970. That’s a fact.
But I don’t know how long it will take to adapt that Linux to the PowerMac G5. For example its 9 fans are dynamically controlled by the OS. Without support for these special fans, Linux won’t run on the G5.
Look at the WWDC keynote on http://www.apple.com/quicktime
Watch the Wolfram Research guy talk about Mathematica on the G5.
The interesting part is from 1:43:20 to 1:43:40
He talks of 6GB of accessible memory if I understand that correctly and I’m quite sure that I do. He specifically mentions that the Dual Xeon does NOT support that.
Why would he say that if Panther did only support 2.25GB or 4GB of RAM?
The PowerMac G5 they used for their presentation had 8GB RAM. Let’s assume the OS still uses about 2GB of memory for shared libraries etc. What would remain if Panther DID allow applications to use 64bit pointers? About 6GB. And the Mathematica guy says 6GB.
I don’t believe that Apple implemented 64bit versions of all their libraries, but nevertheless there seems to be a way for applications to use 64bit pointers.
>Gee, in the 1970’s I was told a word length is the OS’s
>natural language size. A 64 bit word length OS is the only
>measure of what would run on a 64 bit computer is it not?
>In other words, can the OS understand a single 64 bit
>command? Not a hacked or doubled command. The instruction
>set of the processor must be 64 bit even it only the
>ability to use the total command is there.
Nope, the bit-ness of an OS is the size of the memory pointers it uses, your getting confused with the bit-ness of the CPU itself.
Example, AgigaOS was a 32-bit OS running (initially) on 16-bit hardware (which in fact also only had a 24-bit address bus, so some badly coded apps were actually only 24-bit..). The C64 had a 16-bit ‘OS’ running on 8-bit hardware.
It’s the original author here. My guess is that the Mathematica code is using the special hack discussed in The Register article. I have not denied the existance of this hack, although I admit my commentary at the end of the article (about the lack of need for more than 4GB of memory on the Mac and the lack of Terminal Services) downplays the importance of the hack. The experience on 32-bit Windows seems to be that very few programs that I am aware of use its version of the hack. Perhaps Apple has implemented its version better than Microsoft’s version.
My guess is that only companies with a very close relationship to Apple will implement its memory hack. Mathematica has been presented at previous Mac keynote presentations and spends a lot of effort on AltiVec optimizations. The Register mentions Photoshop as a candidate for the hack, while my best guess would be that Apple’s own Final Cut Pro gets the hack.
Companies will try to simiplify cross-platform support costs, such as MathWorks, the makers of Matlab, will be pretty unlikley to support this hack, and will presumably instead focus on true 64-bit OSes which require only recompiles with an additional compiler flag. Also, Quark, a big driver of professional system sales for the Mac, has had problems with a slow changeover to Mac OS X, and the CEO of that company made disparaging comments about Apple. I would expect little support there, although I am willing to be surprised.
Still, if Apple can get Photoshop, Final Cut Pro and Mathematica to support more than 4GB of RAM per application, they might just be able to generate a few very high end G5 sales.
The real question is when is IBM going to bring out PPC 970 Linux workstations and desktops so one doesn’t have to pay the Apple OSX tax to get a real 64 bit workstation on the PPC 970.
Sorry for the troll but I couldn’t resist it.
“The real question is when is IBM going to bring out PPC 970 Linux workstations and desktops so one doesn’t have to pay the Apple OSX tax to get a real 64 bit workstation on the PPC 970.”
It would be very bad for apple if they did. Just buy a retail copy of OSX and install it…against the EULA of course but so what!
Corps. have a lot fewer issues with Big Blue than Apple.
I suspect it does what Linux prebinding does, and calculate the most frequently linked to libraries using a topo sort and dropping any that can’t fit into the 4gig address space.
No it would be worse for MS and wintel than for Apple. The aim would be for different markets – the corporate market that isn’t buying Apple.
The people who buy Apple because it just works and because OS X has the smoothest desktop environment aren’t going to buy a copy of OS X and put it on an IBM box (even if it is cheaper which it may not necessarily be) because it won’t be guaranteed to just work – driver problems etc. A few Unix cross platform geeks may well try it for fun but thats not going to be significant for the market.
It might whowever hit Sun’s remaining workstation market quite hard though.
Mac OS X native object format is Mach-O. One of its features is that it supports storage of various architecture objects within a single file. So having a framework available for both G4 and G5 should be possible.
You can find more information on the Mach-O Runtime Architecture book, available at the Apple Developer Connection site has more information.
Today, most mac users don’t need to have a 64 bit processor.
They need a processor that goes faster. From their point a view, the PPC970 is a better (if not good) 32 bits proc ;-).
For the next months, it will be the real value of the PPC 970.
And within one or two years, the version of OS X at that time will resolve this 64 bits point.
Apple also says they use some code from FreeBSD 5, so the power users should be abble to do something soon.
Also it doesn’t resolve evrything, peoples using Cocoa can already use the Distributed Object systems to make some apps running and talking between each other to use as much RAM as needed.
I remember when I bought my Mac IIci that 16 MB ram cost me like $5000 – more than the computer – and when i bought a 7100/80, 32 MB cost me $2000. Nowadays, $100 for 512 MB DDR sticks seems like a sweet deal. Even $1300 for superfast 2GB sticks
I see Mac users on the forum, settling down, and encouraging each other by saying that thye “don’t need 64bit”. And that is fair. The question is why buy an expensive 64bit Computer to run 32bit MacOSX and apps? I personally think that it is not worth it, plus Apple is using the consumer to transition much later to a more powerful OS. This again was done with the 68K Macs and PowerPC migration. Its fine, as long as Apple is up front about it. But they are not [not surprising since many corp aren’t forward either]. Instead Apple flashes the hardware specs [high profile, keynote and all] but falls back to eye candy on the 32bit MacOSX.
I think Apple figure its users are not that sharp, they meainly want eye candy, an Apple will always trade power and efficiency and performance, for usability and eye candy. And for some people that’s ok.
Now, if you are a regular user, DON”T buy a G5, is hardware that you can’t fully utilize with 32bit MacOSX, in other words, overpriced over kill.
Now, if you are a power use, then do what I am planing to do, to buy a G5 asnd put either Linux or AIX on it, I would love to buy a 64bit PPC box backed by IBM! 🙂
I would think there would be a pretty long transition to all 64 bit apps. In fact, I understand the classic Environment is still going to be there in Panther. I’d hate to try and juggle all of that 🙂 Despite all of that, we will have really fast Macs, which still excites me to no end after living through the G4 era.
The one killer app for a 64 bit desktop is video edititing. Here the massive address space can be used to advantage. It is also the sort of creative area that Apple has targeted. If some one can produce a 64 bit video editing software program that runs on the the hybrid 64/32 bit version of OS X then both Apple and the program developers could make a killing.
If they don’t do it quickly then they might find this niche filled by Linux or Windows on IA64.
Anyone know when that is going to be released or will it co-incide with the first shipments of the PowerMac g5?
By Jay (IP: —.neo.rr.com) – Posted on 2003-07-11 14:52:39
>I would think there would be a pretty long transition to all >64 bit apps. In fact, I understand the classic Environment is >still going to be there in Panther. I’d hate to try and >juggle all of that 🙂
I agree. Also you bring up a good point. Makes me wander, if all this “juggling” has had an impact on Adobe, or the others. Is so much transition at Apple, hurting software maturity and developers in general? We hear from the users, some of them would buy a Mac no matter what. But are developers being affected by all this Apple transitions? One wonders.
>Despite all of that, we will have >really fast Macs, which >still excites me to no end after >living through the G4 >era.
That’s the good part. But I would still recomend to “regular” user to hold off and not buy the G5 until we get a MacOSX that is 64bit, with apps. If enough people would do that, it would send the signal to Apple, that they need to “OPTIMIZE” their OS as opposed to bringing Advance Hardware to compensated for their ineffiecient code. Funny that M$ does the same, apparently Apple has learned from them.
So instead of Mac users being exited about a new Zebra pattern buble theme, how about getting the OS to be more responsive and the apps to be faster? It won’t happen though with so many transitions. As someone who has use BeOS, MKLinux, AIX, and yellow dog on PPC and 68K machines [not to mention Amiga and TOS] I can tell you that Apple already has Great hardware, the OS is not written to take better advantage of it. Optimization, could bring Apple a winner on the desktop, that appeal to a wider audince, not just people who paint their faces with flowers.
I know many bussinesses that run Apple, and they certainly could use a boost from a *really* tweaked MacOSX. Even know someone that switched to YellowDog after trying it, he couldn’t beleive the same hardware was running so fast.
I wouldn’t mind having a PPC970 workstation with AIX on it, i’ve never used AIX, my *nix experiences are solely from Tru64, Irix, and Linux. What do people think of it as a system as oppose to other Unix’s?
I use, Irix, Linux, FreeBSD, AIX, some Solaris, and Tru64.
AIX is a very good Unix, you will find that some people hate it, but I happen to like it a lot. In my experience is more reliable than the others, I mean it needs less baby sitting. I usually set my AIX boxes and off they go, not much intervention is required after initial settings, with the others I am always adjusting stuff here and there, which is ok, as it provides employment. Just MHO.
Jeremy, nice article.
(You, doing your homework shows.)
Everyone else … don’t get your panties in a wad.
These things take time (to get right), quit being spoiled brats, (some of you) this is tough stuff.
I can’t tell you the times I’ve been floored by what these coders (from ALL the major OS houses) pull out of the hat and how quickly they truly can pull off the “merely” impossible.
Patience.
🙂 <— 64-bit smile.
Now, if you are a regular user, DON”T buy a G5, is hardware that you can’t fully utilize with 32bit MacOSX, in other words, overpriced over kill.
You’re missing one thing: if the G5 were a 32-bit machine the new G5 Macs would still be damn fast at a decent price.
And while you can’t utilize the 64-bit chip yet to access more memory I assume you will be able to write apps that do 64-bit math.
>You’re missing one thing: if the G5 were a 32-bit machine the >new G5 Macs would still be damn fast at a decent price.
Well, I don’t want to open that can of worms, but no is overpriced for a 32bitOS runing on it. You can do better with x86 Linux, or even with G4 linux.
It could be cheaper for the user if Apple stops the eye candy factory and Optimizes 32Bit MacOSX to run faster. Apple seems to shift the burden of their slow code onto higher end hardware. IMHO not worth it.
Now, if you buy a G5 with AIX, or to put Linux on it, and you have the computational requirements to do so, then go ahead. But Joe Eye-candy Panther, doesn’t need the machine, the price point and the inefficiency on top is brutal, IMO. Then again if someone wants to blow money on Apple, they can certainly do so.
So when do get a snappy, efficient MacOSX when Gn that is 128bit is introduced? No, Apple needs to face the music and fix MacOSX, instead of pumping flowers through people’s pupils. Don’t get me wrong, that is why I don use M$ either, they play the same game with Windoze, they always claims that you need more Ram and faster processor for their crappy OS to work better, although true for every OS, is still no excuse to shift the burden on hardware for their bad code. Look at BeOS and others, they run way better with less hardware.
Panther in beta has more eye candy than Jaguar and is dramatically faster than its predecessor. So they’re focussing on feature-set expansion and optimization – not one or the other.
>By Anonymous (IP: —.cpe.net.cable.rogers.com) – Posted on >2003-07-11 15:48:44
>Panther in beta has more eye candy than Jaguar and is >dramatically faster than its predecessor. So they’re >focussing on feature-set expansion and optimization – not one >or the other.
Well, historically eye candy affects performance, are you saying that Panther IS the exception? Do you use Panther? Do you run YellowDog on the same box?
Of course, Apple claims improvements, I am sure programmers are assigned to this, but clearly eye candy and not performance is Apple priority. So I could not say that Apple doesn’t work on performance, but I can say that they don’t do enough on that area, and do more on eye candy, as the beta Panther prooves.
Granted is a safe bet for Apple, becuase their user base, seem to be willing to sacrifice power for looks, but one could hope that Apple will break into other markets, they insinuated a Scientific market with the hardware [G5] but believe me it takes more than hardware, a powerful OS must go with it, and so far is not there.
“If they don’t do it quickly then they might find this niche filled by Linux or Windows on IA64.”
Note to self: Windows is not evan close to having a sub $1000, 64 bit Video Editing Application (period)
Longhorn is a 32-bit OS non-the-less (period)
64 bit Operating Systems are not truely defined yet (period)
What about Mac OS X 10.4 that will have an install base before Longhorn gets off the ground? Will it be 64 bit hands down?
>Longhorn is a 32-bit OS non-the-less (period)
I actually don’t care about M$ bloated ware.
>64 bit Operating Systems are not truely defined yet (period)
Nonsense! You are kidding right? Ever heard of Tru64 and the other 64bit gang, including Linux and AIX(when IBM ports it)?
64bit was not invented by Apple, and MacOSwhatever [if it ever gets there] is not going to be the first 64bit OS, live with it!
>What about Mac OS X 10.4 that will have an install base >before Longhorn gets off the ground? Will it be 64 bit >hands down?
Well on that one I don’t know, it seems that Apple concentrates or attracts a different market than M$. M$ tends to be more bussiness office oriented and MAcs more Artsyfartsy. Was hoping that Apple would take a different direction but for the most part it hasn’t happen yet, though there are exceptions to this on the Mac side.
“I see Mac users on the forum, settling down, and encouraging each other by saying that thye “don’t need 64bit”. And that is fair. The question is why buy an expensive 64bit Computer to run 32bit MacOSX and apps?”
Because this computer is significantly faster, and Apple’s power user base will definately make use of the speed
“I personally think that it is not worth it”
By that notion, you would suggest that they revert back to hardware which was slower. If this is the case you must be among those that believes that P4 users should revert to a slower pricessor….
“Apple is using the consumer to transition much later to a more powerful OS.”
Huh? Apple has encouraged all their user to update to new operating systems as released. I don’t follow your logic here.
“Apple flashes the hardware specs [high profile, keynote and all] but falls back to eye candy on the 32bit MacOSX.”
No, Apple is showing that they have powerful hardware, a powerful OS AND an aesthetically pleasing user interface which will will use a 64 bit OS in Mac OS X.
“I think Apple figure its users are not that sharp”
Nothing they’ve done would suggest that.
“they meainly want eye candy”
Not at all. We want powerful hardware, a powerful OS AND an aesthetically pleasing user interface which will will use a 64 bit OS in Mac OS X
“an Apple will always trade power and efficiency and performance for usability and eye candy.
Are you saying that because OS X incorporates a user interface that was initially slow that they swapped Aesthetics for performance? It’s important to keep in mind that the origional OS X release was very new and that the UI would later be optomized… as it was. Now, with Quartz extreme (which ofsets UI rendering to the GPU) Apple can have the same aesthetically pleasing UI, have efficiency AND performance. No tradeoff required.
“Now, if you are a regular user, DON”T buy a G5”
A better way to phrase that, “if you are a user that doesn’t need more speed, don;t buy any new computer… be it a pentium 4 or a G5.
, is hardware that you can’t fully utilize with 32bit MacOSX, in other words, overpriced over kill.
“if you are a power use, then do what I am planing to do, to buy a G5 asnd put either Linux or AIX on it, I would love to buy a 64bit PPC box backed by IBM!”
If you plan on using an alternate OS, that’s fine… but don;t give others reasons for not using OS X that aren’t legit… that’s irresponsible.
Regarding the eye candy vs. speed argument. What they (Mac proponents) are pointing out is the fact that there is no noticable slow down in responsiveness which is due to all the eye candy being shifted off to the GPU, and with the 8x AGP slot on the G5, it should be very interesting.
As for the over performance, the question I want to know is this. Apple has just updated/upgraded their PowerMac line, when will the unwashed masses see a speed increase in the consumer line of Apple desktops? is Apple quietly giving up and instead focusing on the professional line of workstations?
I’ve just had a look at IBM’s new (well, new to me) ThinkCentre series, and after having a look at the price, how can Apple justify charging $1899 for an eMac when I can pick up a ThinkCentre A Desktop for $1550 then add $400 or so for the screen. For an extra $50 one gets a 2.6Ghz P4 + 256MB + 17inch screen + 64MB DDR graphics card.
If Apple was selling the low end eMac with a 1.25Mhz PowerPC G4 + 64 DDR Graphics Card + 256MB memory, then the idea of purchasing one would be very tempting, however, I can’t see why one would purchase something of that lack luster performance?
Sure, their pro workstations are a bargin, but their desktop range is a bit of a ripoff in comparison to what is available.
“I personally think that it is not worth it”
>By that notion, you would suggest that they revert back to >hardware which was slower. If this is the case you must be >among those that believes that P4 users should revert to a >slower pricessor….
Nonsense! I never said that. Besides older hardware doesn’t have the price point of newer MAcs, which are expensive and bloated. Try running YDL on a MAc and compare it with Apple OS runing the same box: eye opening!
“Apple is using the consumer to transition much later to a more powerful OS.”
>Huh? Apple has encouraged all their user to update to new >operating systems as released. I don’t follow your logic >here.
No, you are lost as usual. The consumer is fooled to pay for expensive 64but hardware to run bloated 32bit software, so that they get a downpayment for future development at the users expense. Ok for hobbyist or Maczealot but not for responsible users. A responsible users should wait, until a G5 ships with a 64bit OS, whihc Panther is not, as any expert will tell you. I mean IBM build the chip and they stated that they are providing 32bit mode for Apple. I take IBM’s word over your propaganda any day.
“Apple flashes the hardware specs [high profile, keynote and all] but falls back to eye candy on the 32bit MacOSX.”
>No, Apple is showing that they have powerful hardware, a p>owerful OS AND an aesthetically pleasing user interface >which will will use a 64 bit OS in Mac OS X.
BS Panther is 32bit. The UI might please you, along with fancies of Steve speaches, but IT IS NOT powerful 64bit, it is bloated 32bit.
If you have not run YDL in the same box as MacOSX you’re pulling staments out of your rectum. Enough with propaganda already! Go Back to the eye candy and the flavors: grape, orange, etc.
Which goes RIGHT BACK to a post I made earlier that Apple should have licensed AIX core off IBM and worked their GUI magic on it. Too bad they had to go what they have now. Having run it several times (I was actually keen at one stage to buying one), I really see no advantage to Windows XP, 2000 or what ever mainstream operating system you can point out.
As for Windows 2000/XP stability, as a desktop I have been running it for a good 5 months with out a BSOD. The only reboot I did make was because of SP4 installation.
I can now actually see what Gil Bates was going on about previously, which I agresively flamed. Btw, sorry about that Gil. The fact remains, where is the justification for paying extra for Mac hardware and software? I’ve looked at it form both a Holistic point of view and individual components and I still can’t see an advantage.
Holistically, yes, there is greater chance of system stability because all the parts are chosen and the operating system is tweaked so that the hardware and software work well together. However, on closer inspection, how is that any different to the programmes that run between IBM and Microsoft? or infact, any vendor and Microsoft.
I then looked at the individual components. The operating system in itself is nice, however, it lacks a sufficient based of applications to select off for it to become more mainstream. Hardware wise, no matter how you slice it, they are way behind in terms of speed. The G5 has matched it, however, they’re focused on the high end. What is happening with the consumer end of the market.
If Apple is leaving the consumer market, why not just say so? IMHO, Apple would be alot better of focusing on the processional market where the premiums are better, the clients are willing to pay those premiums and that benchmarks are not always the deciding factor for purchasing something.
Is this in stone that Panther won’t be 64 bit? Is it from Apples own mouth?
“I would still recomend to “regular” user to hold off and not buy the G5 until we get a MacOSX that is 64bit, with apps.”
Not me. I would recomend to all users to go ahead and upgrade if they need the speed.
“If enough people would do that, it would send the signal to Apple, that they need to “OPTIMIZE” their OS as opposed to bringing Advance Hardware to compensated for their ineffiecient code.”
Ineffiecient code? Where do you get that? Jaguar is very efficient… especially with Quartz Extreme.
“So instead of Mac users being exited about a new Zebra pattern buble theme, how about getting the OS to be more responsive and the apps to be faster?”
I don;t know any Mac users getting excited about themes, and Jaguar is very responsive and the apps are definately fast will certinly be faster.
“It won’t happen though with so many transitions.”
Sure it will. The transition to OS X was the only major transition. Upgrading apps to take advantage of OS X’s 64 bit technology is somewhat trivial in the grand scheme of things.
“I can tell you that Apple already has Great hardware, the OS is not written to take better advantage of it.”
Sure it is. Why would you say that?
“Optimization, could bring Apple a winner on the desktop, that appeal to a wider audince,”
The OS is already optomized, and is getting even more optomized with ever release. Apple is already a winner on the desktop and is appealing to a wider audience.
“not just people who paint their faces with flowers.”
What?!
“I know many bussinesses that run Apple, and they certainly could use a boost from a *really* tweaked MacOSX.”
Well certinly, and business would benefit from increased productivity and performance. Thankfully, OS X already offeers a great deal of both.
“Even know someone that switched to YellowDog after trying it, he couldn’t beleive the same hardware was running so fast.”
he must have been using an earlier version of OS X, because it is already very fast.
Blah blah blah…. Linux rules! Blah blah blah… Apple’s bloated OS! Blah blah blah… So is Microsoft! Blah blah blah…
Just because the kernel only uses 32 bits for its address space does that preclude other programs running on the hardware from doing so?
A kernel is just a program running on bare metal. The kernel doesn’t need a 64 bit address space to do its work but does that preclude the kernel from assisting user space programs that might use 64 bit address space?
I know this probably sounds weird but if the kernel can set up stupid MMU tricks for user mode programs but use different tricks for itself then, in theory, one could have a 64 bit address space in use by a program but the kernel only using a smaller 32 bit address space subset.
This might be quicker/cheaper than having to convert the kernel to full 64bit from the get go.
If you don’t really *use* 64 bit addressing, extending all core OS functions to 64 bits only makes *everything* slower. If the upper 32bits of the registers can survive context switching under 10.3 and if you don’t need to address >2GB memory, you have best of both worlds.
BTW, all *BSDs are already 64 bit clean. I wonder why OS-X need more than a recompilation to be a 64bit OS – at least for kernel stuff.
“[the G5] overpriced for a 32bitOS runing on it.”
Not at all. As a matter of fact, its a sweet bargain.
“You can do better with x86 Linux, or even with G4 linux.”
The comparitive tests have shown otherwise.
“It could be cheaper for the user if Apple stops the eye candy factory and Optimizes 32Bit MacOSX to run faster.”
OS X has no eye candy, but it certinly has a lot of Aesthetically pleasing qualities all of which take advantage of Quartz extreme which means that there is no speed degridation. As a Matter of fact, Quartz extreme is overkill, so Apple could theoretically afford to add more Aesthetically pleasing qualities that might have otherwise slowed down the OS without it.
“Apple seems to shift the burden of their slow code onto higher end hardware. IMHO not worth it.”
This assumption assumes that Apple is not incorporating Quartz extreme which allows their aesthetically pleasing qualities to not happer processor speed in any way.
“Now, if you buy a G5 with AIX, or to put Linux on it, and you have the computational requirements to do so, then go ahead. But Joe Eye-candy Panther, doesn’t need the machine”
Who is joe eye candy and why would he buy OS X? It has no eye candy?
“the price point and the inefficiency on top is brutal, IMO.”
OS X is very efficient and the price point is EXCELLENT for those that could use the increased power.
“Then again if someone wants to blow money on Apple, they can certainly do so.”
The same is true for any computer. Thankfully, buying an Apple is a wise purchase.
“So when do get a snappy, efficient MacOSX when Gn that is 128bit is introduced?”
Nope. It’s sold right now. its called jaguar.
“Apple needs to face the music and fix MacOSX”
Face what music? OS X isn’t broken.
“[Apple needs to stop] pumping flowers through people’s pupils.”,/i>
What flowers are you talking about?
[i]”Don’t get me wrong, that is why I don use M$ either”
What flowers does microsoft use?
“[Microsoft] play[s] the same game with Windoze, they always claims that you need more Ram and faster processor for their crappy OS to work better”
Because Microsoft doesn;t not incorporate technology similar to Quartz Extreme, you may have a point there.
it is still no excuse to shift the burden on hardware for their bad code.”
What bad code are you talking about?
Overhyped + overpriced, that is Apple == exaggerated.
Let’s see next year. If the current situation doesn’t change much on Apple, well, most likely my money will be on AMD Athlon64.
“Apple does not offer a Terminal Services solution, instead letting you log on remotely via ssh or telnet to a command line or X-Windows environment, where you are pretty much limited to using ported open source programs.”
This is only partly true. Apple does have a product called Apple Remote Desktop which I’ve used very successfully to manage Macs without physical access. I think Apple is very aggressively developing remote user and multiple user environments. We’ll see a lot more of this technology as we move forward.
http://www.apple.com/remotedesktop/
“historically eye candy affects performance, are you saying that Panther IS the exception?”
Historically speaking you are correct, but previously, nobody ofset UI rendering to the GPU which allows for lots of UI improvement without a drag on the processor.
“clearly eye candy and not performance is Apple priority.”
Clearly? Where do you get that? Apple has performance AND aesthetics as a high priority (as it should be)
“[Apple doesn’t] do enough on that area [of performance], and do more on eye candy, as the beta Panther prooves.”
I don;t see where you get this. Pather is both aesthetically pleasing AND extremely efficient as was the case with Jaguar. Panther is simply more so… in both areas.
“Granted is a safe bet for Apple, becuase their user base, seem to be willing to sacrifice power for looks”
Not at all. Instead, we demand both.
“but one could hope that Apple will break into other markets, they insinuated a Scientific market with the hardware [G5]”
Apple has had a large presence in scientific fields for many years (even before the G5)
“but believe me it takes more than hardware”
Thankfully, Apple has the powerful hardware and much more.
“a powerful OS must go with it”
Thankfully, they have such an OS.
“and so far is not there.”
it certinly is
“Well on that one I don’t know, it seems that Apple concentrates or attracts a different market than M$. M$ tends to be more bussiness office oriented and MAcs more Artsyfartsy. Was hoping that Apple would take a different direction but for the most part it hasn’t happen yet, though there are exceptions to this on the Mac side.”
Apple’s current direction appears to appeal to Artsy Fartsy types and those in many other fields.
Which goes RIGHT BACK to a post I made earlier that Apple should have licensed AIX core off IBM and worked their GUI magic on it.
The reason Apple managed to produce OS X as quickly as they did is because they were starting from the NeXT codebase. Perhaps you haven’t examined Cocoa and CoreFoundation in-depth but the Cocoa event model is based around Mach message ports.
Why would Apple use a kernel which isn’t their own IP and doesn’t support Mach messaging? Does that make any sense from a business or technical perspective?
>Which goes RIGHT BACK to a post I made earlier that Apple >should have licensed AIX core off IBM and worked their GUI >magic on it. Too bad they had to go what they have now. >Having run it several times (I was actually keen at one stage >to buying one), I really see no advantage to Windows XP, 2000 >or what ever mainstream operating system you can point out.
I hear ya!
I wish that was the case too. You would be interested to know that I had an Apple AIX Server once, it was an RS6000 from IBM boxed in an Apple case, sold by Apple. This was before Apple made a deal with Steve Jobs, and before they rejected BeOS [big mistake].
The probl;em was that Apple knew nothing of AIX and they couldn’t not support the box. They tried, maybe if they had being more serious about it, history would be different.
The machine was a powerhorse. As it was a real server with 64bit JFS file system, with hot swapable drives. Solid compare to the flimsy G5 desktop design. [A desktop can’t compare with a real hardware server, this is the design, not the processor]. It could also take more than 2 processors, the G5 motherboard can’t do that today or a few months doan teh road.
Apple backed out of the AIX and went into the flower power thang.
“I’ve just had a look at IBM’s new (well, new to me) ThinkCentre series, and after having a look at the price, how can Apple justify charging $1899 for an eMac when I can pick up a ThinkCentre A Desktop for $1550 then add $400 or so for the screen.”
The eMac costs $799.
“If Apple was selling the low end eMac with a 1.25Mhz PowerPC G4 + 64 DDR Graphics Card + 256MB memory, then the idea of purchasing one would be very tempting, however, I can’t see why one would purchase something of that lack luster performance?”
The price to performance ration on the emac is very good.
“Sure, their pro workstations are a bargin, but their desktop range is a bit of a ripoff in comparison to what is available.”
Not at all.
Of you compare Apple’s consumer line of products in both hardware and software Apple’s products tend to be slightly more expensive, the same price or slightly less expensive.
My experience with older SPARC systems in the transition of Solaris is that even in 32bit address space mode, using the 64bit processor + “cheater” versions of the system librarys DID improve performance of applications.
Put another way, the 64 bit instructions and features didn’t need a 64bit address space to work so you could make use of performance gains in the CPU; cache, block instructions, 64bit math and stupid MMU/window tricks. I’d be real surprised if Apple didn’t have some form of “cheater” librarys in panther,
and maybe 10.2.7, to allow user mode gains from the new chip while the address space remains 32 bit.
As far as library pre-loading: Why preload when demand-paged VM should have already loaded in the most common shared librarys by the time the login: prompt appears on a UNIX(tm) box? Is this a MACH induced weirdness?
Historically speaking you are correct, but previously, nobody ofset UI rendering to the GPU which allows for lots of UI improvement without a drag on the processor.
Bullshit. If you have access to a windows or a linux machine,
– turn off video card accelaration under windows and see what happens.
– use vesa driver instead of your card’s driver under X and observe the performance
I remember co-processor enhanced GUI performance on a sub 2000$ computer back in 1984.
Its not really preloading, but prelinking. OS X relinks all programs at install time so they each load to a different chunk of virtual memory. This way, there are no address-conflicts at load time. This means no relocations have to be processed at load time (which saves a great deal of CPU for large C++ applications) and also means that relocations don’t cause pages of otherwise sharable memory to become unshared (which saves a lot of memory). Read up on the problems KDE has had with load-time relocations, and about the Linux prelinking mechanism.
Here is a picture and some information about this product, this was circa 1996 or so.
http://mirror.apple.com/server/serverview.html
Now at the time this was a different direction for Apple than the flimsy imacs. Different market as well, in all fairness.
Only a Maczealot would believe that Apples are cheaper than PCs and that MacOSX is an efficient use of PPC hardware.
Experience versus allucinations. Try YDL on a PPC, or even on a MKlinux on the older Macs, still show that through time Apple has produce slow OS, considering the excellent hardware that they use.
“newer MAcs, which are expensive and bloated.”
Not at all.
“Try running YDL on a MAc and compare it with Apple OS runing the same box: eye opening!”
I don’t understand what you’re getting at. Both are really great operating systems.
“The consumer is fooled to pay for expensive 64but hardware to run bloated 32bit software”
First of all, we don’t have any confirmation that OS X wont be 64bit, but even if it weren’t the Applications would be able to take advantage of 64 bits, but if for some unknwon reason that even THAT couldn;t happen (There’s no reason to think otherwise but lets play devil’s advocate) the G5 is INCREDIBLY faster than the G4 and P4, and is roughly on Par with XEON processors. So is the speed there? Absolutely. Does that speed cost less than an equally equipped PC… absolutely.
Also, What about OS X makes you think its bloated?
“Ok for hobbyist or Maczealot but not for responsible users.”
For ANYBODY that can use the increased power and performance that the G5 and Panther will give.
“A responsible users should wait, until a G5 ships with a 64bit OS”
Why would you say that? Even if the G5 isn’t 64 bit, the computer is INCREDIBLY faster. Anyone that can use that extra speed ought to consider a G5 purcahse.
“which Panther is not”
Its not even out yet!
“I mean IBM build the chip and they stated that they are providing 32bit mode for Apple. I take IBM’s word over your propaganda any day.”
No, they said that the 970 will retain full 32 bit compatibility without a hit in speed which can’t be said for alternative processors of the same calibar.
“BS Panther is 32bit.”
As far as we know, OS X will be able to take advantage of 32 bit and 64 bit technology.
“The UI might please you, along with fancies of Steve speaches, but IT IS NOT powerful 64bit, it is bloated 32bit.”
OS X is not bloated, and we don’t know if OS X will be one way or another it hasn;t shipped. We do know that it will be able to run 64 bit applications in 64 bit mode.
“Enough with propaganda already!”</I.
Me?
[i]”Go Back to the eye candy and the flavors: grape, orange, etc.”
That would mean buying a PC, because Apple doesn’t make such computers anymore. I prefer my Mac. I think I’ll stick with it.
The eMac costs $799.
…US. CooCooCaChoo seems to be from Australia. Remember when I said that Mac computers can be more expensive depending where you live?
” By goo.. (IP: —.adsl.ttnet.net.tr) – Posted on 2003-07-11 17:33:36
Historically speaking you are correct, but previously, nobody ofset UI rendering to the GPU which allows for lots of UI improvement without a drag on the processor.
Bullshit. If you have access to a windows or a linux machine,
– turn off video card accelaration under windows and see what happens.
– use vesa driver instead of your card’s driver under X and observe the performance
I remember co-processor enhanced GUI performance on a sub 2000$ computer back in 1984.”
* * *
Of course you are right. I do not have the time to correct all the propaganda and BS that that guy keeps putting in this forum. He will make you think that Apple invented evewrything, first 64bit OS, Specific hardware and what not.
If Apple was so good at creating OS, why did they went shopping for a diferent OS? AS they did with Next?
That guy only runs interferance, he will take every opposite position insinuating that macs are the best thing since slice bread. They are not, Apple concentrated in easy of use, to this day, hence a 1-botton mouse, never in performance or power usage. They always claim that PPC is superior, and I think PPC is great hardware, but I take exception with their slow OS. MacOSX is still slooooow, compared with other OS in the same box (period).
Now, a UI doesn’t have to be ugly to be fast as BeOS shows, Macos at the time never looked that good and could not compare in spped. I miss those 15 seconds boot up times. 🙂
That guy can go back to his kool-aid drinking fantasies.
“Solid compare to the flimsy G5 desktop design.”
Flimsy? That’s a new one.
“It could also take more than 2 processors, the G5 motherboard can’t do that today or a few months doan teh road.”
Assuming Apple retains the G5 naming convention what it uses the dirivitive of the Power 5, it will be able to utilize more than 2 processors.
“Apple backed out of the AIX and went into the flower power thang.”
Apple didn;t back out of an AIX deal to create the flower Power iMac. Where would you get that idea?
“a different direction for Apple than the flimsy imacs.”
That’s a new one. I’ve never heared anyone suggest that the iMacs were flimsy. As a matter of fact, they seem to be built rather solidly.
“Only a Maczealot would believe that Apples are cheaper than PCs and that MacOSX is an efficient use of PPC hardware.”
A Maczealot might as well as those that understand that when you compare both hardware and software the Mac tends to be either slightly more expensive, the same price, slightly less expensive, or significantly less expensive.
Try YDL on a PPC, or even on a MKlinux on the older Macs, still show that through time Apple has produce slow OS, considering the excellent hardware that they use.”
Apple has definately made some slow hardware in the past.
“CooCooCaChoo seems to be from Australia. Remember when I said that Mac computers can be more expensive depending where you live?”
As would be the case with other products… in this case a PC.
“Assuming Apple retains the G5 naming convention what it uses the dirivitive of the Power 5, it will be able to utilize more than 2 processors.”
BS, show were Apple has anounced a Quad G5? Current Mobo wont take more than 2. They can do one but they havent, all I hear is an idiot answering for them, after gulping kool-aid.
-“Apple backed out of the AIX and went into the flower power -thang.”
“Apple didn;t back out of an AIX deal to create the flower Power iMac. Where would you get that idea?”
Can you buy an Apple server running AIX? No, so they went the flower power route after that experiment.
Read again! And google for it. You just run BS, propaganda and interferance. Your are intellectually dishonest, your posts are prove of that.
Of course, Apple claims improvements, I am sure programmers are assigned to this, but clearly eye candy and not performance is Apple priority. So I could not say that Apple doesn’t work on performance, but I can say that they don’t do enough on that area, and do more on eye candy, as the beta Panther prooves.
Of course they are going to put a high priority on features. People just don’t pay for speed increases. But OS X has gotten significantly faster (on the same hardware) with each new release. It appears that Panther will be no exception.
It’s nice to see that the range of usable hardware is increasing with each new version.
“If Apple was so good at creating OS, why did they went shopping for a diferent OS? AS they did with Next?”
because at that time, they were trying to do too much with a legacy OS. When they finally realized that it would take too long, they were in to deep. They needed to buy their way out of the hole they dug. Thankfully, the end result turned out to be a nice choice.
“Apple always claim that PPC is superior”
Considering the advancements that are coming from IBM, i think they were right.. even if Motorola didn’t help PPC.
“MacOSX is still slooooow, compared with other OS in the same box (period).”
That’s not the experience that I’ve had at all. Rather, they are roughly the same speed 9depending on the task at hand) Early on, the UI was sluggish giving the illusion that the OS was slow, but that has since been fixed with Quartz extreme. (Thankfully)
“BS, show were Apple has anounced a Quad G5?”
What does that have to do with it. IBM is the primary driving force behind the chips that Apple uses and they have said that the Power5 can be scaled up to multiple processors. So, like i said, assuming that Apple retains the same naming convention (G5) when transitioning to the Power5 derivitive
“Can you buy an Apple server running AIX? No, so they went the flower power route after that experiment.”
Again, Apple didn’t choose to not adopt AIX so that they could create flower Power iMacs. That’s rediculious.
“You just run BS, propaganda. Your are intellectually dishonest, your posts are prove of that.”
Look in the mirrior when you make that statement next time.
“That’s not the experience that I’ve had at all. Rather, they are roughly the same speed 9depending on the task at hand) Early on, the UI was sluggish giving the illusion that the OS was slow, but that has since been fixed with Quartz extreme. (Thankfully)”
More BS, try running the lates MacOSX on an older Powerbook and see how bloated it is. Not even usable. Again showing that the code is bloated as compare to BeOS, Linux, MKLinux, the REAL BSD, even NExt etc.
What box are you running MacOSX on? What version? What are the Specs? Have you run YDL on the same box? What Version? What Apps have you use to test the speed on both OS? BS won’t do. Care to post benchmarks?
Of course, but the PC doesn’t necessarily follow the same raise. I’ll take an example. Intel and AMD CPUs are pretty much the same price in the US. However, in Canada, Intel CPUs tend to be more expensive. Now, Macs tend to follow the currency exchange here (so a 799$ eMac is 1199$ CDN, according to The Apple Store *website*). However, a 799$ PC isn’t necessarily 1199$ CDN here. They can be cheaper (or more expensive, depending on the brand). I don’t want to start a price debate again, but I hope you understand what I mean.
Eugenia,
I would like to request that you get a little more involved here in the moderation of these forums.
Mistik jogelour (IP: 209.11.79.—) is an OBVIOUS troll.
I respectfully request that you mod-down any further correspondence made by him that follow the current trend of posting that he has made.
15 second boot-up times? I don’t know about you, but I’m more than willing to wait a while longer for my box to boot. For my desktop I only do this once a day, my everyday fileserver has been up since I moved into this office a year and a half ago.
Boot-up time is hardly an indicator for OS speed. Frankly, I don’t care about raw speed of the OS. I have work that I need to get done and I can get it done faster on OSX than on Win2K. YDL isn’t an option for everyone. I need apps that are only available on OSX and Windows, and in my opinion OSX is clearly the better of the two.
What you refer to as eye candy is what regular desktop users prefer to call usability features. If I wanted blistering performance I’d install YDL+Blackbox on my system and I’ll be the last to stop you from doing this but some people actually ARE willing to sacrifice raw performance for usability. A desktop machine after all doesn’t sit in its closet untouched like a server, it gets used interactively, and for that purpose the OS better be as unobstrusive as possible.
Apple has done a great job in this respect. If you don’t like it, get a workstation from SGI but don’t forget to bring an extra wallet!
Anyone who uses MacOSX versus just READING about it will tell you that speed wise it is fine, not blistering but it hasn’t stopped millions of people from using it on a daily basis to do production level and professional work.
This garbage of MacOSX being slow is just that, garbage. Run MacOSX on a slow machine and it will be slow, run it on a decent machine and its fast. How hard is that?
Why should I get censored, when Anonymous (IP: 12.105.181.—) is the obvious troll?
Why don’t you put down the benchmark of MacOSX versus YDL and prove that they perform the same instead of crying to Eugenia?
Why don’t you put down the benchmark of MacOSX versus YDL and prove that they perform the same instead of crying to Eugenia?
Performance isn’t the be-all end-all.. I prefer the devices that I use for my everyday work to be usable too.
“try running the lates MacOSX on an older Powerbook and see how bloated it is.”
Considering the fact that that older powerboox probably doesn’t have one of the support Graphics processors you may see some degridation in performace.
Lest you misunderstand, that does not mean that OS X is bloated. Rather, it means that it is benefited by a GPU.
“Not even usable.”
That’s not the case at all. I have an older Mac that doesn;t have supported graphics card and yet installed Panther on it, and I was using it as a productivity machine for a few months.
“Again showing that the code is bloated as compare to BeOS, Linux, MKLinux, the REAL BSD, even NExt etc.”
All these OSes are very nice. I use some of them on ocation, but comparing them to OS X doesn’t suggest that OS X is bloated.
“What box are you running MacOSX on?”
I have a few. A G4 1GHz Tower, An 833 G4 iMac, a 12″ Powerbook and a second generation iMac.
“Have you run YDL on the same box?”
I’ve run YDL and LinuxPPC on both iMacs.
“What Apps have you use to test the speed on both OS?”
Pretty much standard fare stuff. Photoshop, Dreamweaver, Golive, MS Office, BBEdit etc…
“Care to post benchmarks?”
Not really.
Mistik pizza? What planet are you from? We run Jag, and 10.1 on 233 iMacs at work! The helpdesk PowerBook is a Pismo that is running Jag so that we can ditch TunnelBuilder and its funky VPN settings.
I’ve ran MkLinux on a 6100 and its slow, who are you trying to kid??? Drop MacOS7.5.5 to 8.6 on there and its fine.
Yellow Dog is hardly impressive compared to MacOSX. Why would you buy expensive Mac hardware to install an OS that was reverse engineered to run on it?
Your talking out of your azz!
” By Bas (IP: —.adsl.xs4all.nl) – Posted on 2003-07-11 18:19:57
Performance isn’t the be-all end-all.. I prefer the devices that I use for my everyday work to be usable too.”
Fair statement. Which proves that Apple is into a different market, not in the power computing, but of usability, which I have said all along.
I bet Apple is the right product for you. That’s ok. What I object to is outrageous claims of performance [not the HW] but MacOSX.
To paraphrase Obi-wan
“Who’s the more trollish – the troll, or the troll who responds to him?”
LOL
Being a stubborn fool doesn’t make on a troll (not saying who – just a general comment
Seriously, though. Comparing YDL to OS X isn’t exactly even. OS X has more “built in stuff” and has a better (and OO) development environment. That OO stuff typically doesn’t come free ya know.
“Why should I get censored”
Because every comment you’ve made thus far has been an outright troll, or has been a semi-cliverly disquised troll
“Anonymous (IP: 12.105.181.—) is the obvious troll?”
So says the troll.
“Why don’t you put down the benchmark of MacOSX versus YDL and prove that they perform the same instead of crying to Eugenia?”
Because I don;t need to prove to myself that OS X isn’t slow. Everybody here that has used OS X already knows that. I’m calling on Eugina because she’s the only one that can fix this problematic situation.
“Fair statement. Which proves that Apple is into a different market, not in the power computing, but of usability, which I have said all along.”
You are misconstruing what he said. he said that performace isn’t the the be-all end-all. It isn’t, yet nothing in that statement said that Apple’s computers were slower, or slow or bloated… which means you are wrong. it doesn;t prove anything.
“What I object to is outrageous claims of performance [not the HW] but MacOSX.”
Nothing autrageous about it. OS X is a good performer.
“You are misconstruing what he said. he said that performace isn’t the the be-all end-all. It isn’t, yet nothing in that statement said that Apple’s computers were slower, or slow or bloated… which means you are wrong. it doesn;t prove anything. ”
STOP TROLLING! Stop running interferance.
Apple’s market IS NOT power computing(period). That’s SGI. SUn, AIX, Tru64, HP-UX and to an extent Linux.
so the pro PPC guy who wrote an article the other day got reamed by readers despite a ton of research, and this guy quotes anon from /. and we accept his say? funny stuff indeed!
But I am sure a 128Bit machine will show that it wasn’t MacOSX fault, it was slow hardware, eh?
My argument about lack of power is as it applies to serious scientific computing. I don’t care how fast MacOSX can spin lilies and throw flowers, that alright but it is a different market.
This argument is similar to M$ argument for Windows 3.11, 95 and 98 performance. Although now with XP [which is based on NT] they have a better product.
Bet you MorphOS[if they ever get there] will be faster on a G5 than MacOSloowX.
How can 8GB ram cost over 10.000$? The 1.8 and 2.0 G5 mac does have 8 ram sockets, and it should be posible to buy 8x1GB ram for something like 2500$ (or less, but theese are the prices in Denmark).
Martin Tilsted
“My argument about lack of power is as it applies to serious scientific computing.”
So far, The scientific comparisons that have been completed thus far only confirm the reports made by each of the demonstrators at MWDC and by Veritest.
“I don’t care how fast MacOSX can spin lilies and throw flowers, that alright but it is a different market.”
That’s wierd. I’ve never seen OS X spin lilies or throw flowers. That would be mere “eye candy”.
“This argument is similar to M$ argument for Windows 3.11, 95 and 98 performance.”
Not at all. Microsoft incorporated bloat into their OS which demanded greater hardware albeit without much increased functionality.
Jaguar, doesn’t require much newer hardware and yet offers much more features and increased speed.
“Although now with XP [which is based on NT] they have a better product.”
Agreed.
“Bet you MorphOS[if they ever get there] will be faster on a G5 than MacOSloowX.”
We’ll see if its faster, but OS X cerinly isn’t slow…
“so the pro PPC guy who wrote an article the other day got reamed by readers despite a ton of research, and this guy quotes anon from /. and we accept his say? funny stuff indeed!”
Am I the only one that caught a slight sense of Anti Apple bias in the article… (unlike the overt FUD by Mistik jogelour) the article was a thinly veiled attempt at FUD.
“Bet you MorphOS[if they ever get there] will be faster on a G5 than MacOSloowX.”
We’ll see if its faster, but OS X cerinly isn’t slow…”
Agreed.
MacOSX is not simply slow, it is slower than other OS on a similar platform, plus bloated.
PPC is a great processor that Apple doesn’t fully exploits, maybe Apple should drop the eye candy budget, sin’t there enough already, and hire IBM engineers to step up their code. Everybody would win.
You don’t even own or use a Mac, how can you even know that its slow? Your a troll, aren’t you supposed to be under a bridge somewhere?
Almost a quarter of the 100 or so comments on this article are from you talking about some flour stuff and how MacOSX is slow because you read aboout it somewhere or saw some screenshots and the screenshots weren’t moving.