If you are typsetting in Word, then I truly feel sorry for you, there are so many good DTP applications that handle text flow and text formating much much better.
When we don't even have OSes that support Font features to correct font errors like ligatures, how can we expect Glorified word processors to do so.
I actually hope the updated font system in Longhorn takes us a step into the future of onscreen typography. Even the Mac font rendering capabilities are pathetic and they pretend to cater to the graphic design market.
What are ligatures, and why would I care about them in typing up a letter?
Ligatures are actually something the poster would NOT want, so I assume they mean more advanced recognition of ligatures and kerning adjustment.
Ligatures are when non-propotional fonts spacing have two letters that interfere with each other. A common example would be 'fi' notice that the dot on the i may run into the f, and it shouldn't.
Basically Word only uses the standard Font system that is built into the OS, and OS font systems don't always correct items that typsetters or typography people need corrected in publication.
As for the ligature problem, it is more a problem with the FONTS themselves than the fact that Word doesn't compensate for the problems in the font.
A real DTP program, like Quark or Pagemaker does more advanced kerning using an internal system of textual placement.
This allows them to notice and fix things like ligatures and also gives you more features like when doing justified margins they can do 'real' justification where the letter spacing is adjusted and not just the spaces between words is adjusted, as Microsoft Word does now.
I'll eat someone's tinfoil hat if this works plugin-free with any browser other than IE7 on Windows.
Better buy the tinfoil and get your web cam ready. The forms are SERVER objects that are represented to a browser using standard XML,HTML and very generic JavaScripting which means the data services of the form Server side.
Go read the notes on Web Forms development with Visual Studio 2005 Beta 2. Same thing, just instead of creating them in a development environment, it lets managers and business people create data collection systems as easily as creating the form.
Just tested an application of Web forms from Visual Studio 2005, using the same concepts and technology. It works in IE on Windows, IE on my Mac, Safari on my Mac, Firefox on my Linux installation, and even IE 4.0 on Mac 8.1 Emulator.
Enjoy eating the hat, and maybe comment on something you actually are informed on next time.
Too bad Microsoft hasn't copied Apples concept of bundled files (i'm not positive of the exact term) that act as a single file. It really is nice. (and it is used by their 'pages' software in the same way microsoft is zipping files together).
Microsoft has had 'paired' folders and files since like Win2k. Windows also can treat a Zip Binary as a single archive folder.
Do you just assume every feature on a Mac is unique to a Mac or don't get out in the real world much?
The "System Controller" is an exclusive Apple part, built in IBM's fabs, but designed by Apple. It's responsible for connecting and arbitrating the G5's FSBs to memory and the HyperTransport bus.
Plus, cooling and power management are unique to Apple, and heavily integrated with the system software. This includes the passive liquid cooling system in the top-end G5s. Details can be seen in the Developer note [apple.com].
Ya, they helped designed the Memory Transport, but out of the wonderful Mac, that is about the thing anyone comes up with, other than the obvious Firewire.
Liquid Cooling System... Ooh, that is really 'cool'. This has also been available in the PC world for years and years. Do you really thing Apple was the first to come up with this 'innovation'? Come on, get out of the Apple 'elitist' marketing...
The sleep mode on a Mac just works. A mac is always available in the time it takes a monitor to warm up... about two seconds for an LCD.
A mac notebook is well balanced. The user's hands don't cover the speakers. It doesn't bristle with fiddly rubber bump knobs that inevitably break off.
A mac notebook can be mounted as a firewire target disk on any other firewire based computer. This means instant filetransfers of any size with no configuration.
There are hundreds of things like this, and no, they'll never show up on a spec sheet. If you're the kind of person who doesn't care about these things, it doesn't matter.
Still, for many of us, these features are critical. They make the difference between a 1990s style machine we're constantly fiddling with and a modern machine that just fades into the background any time it isn't in use, and most of the time that it is.
The kind of engineering you're talking about is chip design. No, apple doesn't design chips. They used to! They used to design all sorts of hardware standards, like NuBus and ADB. Once USB and AGP came around, though, there wasn't any point. The PC world had finally caught up and started making products that could support the Macintosh experience, so Apple went with the PC standards, cut prices, and all was good.
Maybe it has been a while since you have used a modern NON-APPLE notebook.
Yes, my notebook can outbenchmark a G5. It has a 64bit P4 3.8ghz Processor, and a GeforceFX 6800 Ultra video, and a 7200rpm Raid 0 Array in it. And I can run a full and real 64bit Operating System on it.
Ok, lets go through your list?
Sound? My notebook is virtually silent unless running high end Video, and the fans will kick on for a moment or two. No louder than a freaking powerbook.
Sleep Mode? Ever hear of Hibernate on WindowsXP or Standby mode. I can even choose which I want. Standby (comparable to Apple Sleep mode) I can turn on my notebook in under 1 sec. Hibernate, which it returns from a FULLY POWERED off state, I can be on and online in less than 5secs. Obviously you read the Apple marketing when they tell you they have the fastest and best sleep method, sorry, they don't.
Well Balanced? I don't cover the speakers with my hands, didn't on my 3yr old notebook either. And it, like this one has built in subwoofer with sound only a Apple notebook could beg to offer. Current laptop has 4+1 Speaker system even.
As for 'fiddly rubber bump knobs'? My case is aluminum, and all my buttons are also aluminum, no plastic crap here either.
Firewire target disk? And you think WindowsXP doesn't do this? Do you live in a barn?
The thing you DON'T realize, is that I use Mac Notebooks, work with people that use them every day as their primary computer. Trust me, there is no comparison.
As for Apple design... That ended years ago. It wasn't just the PC world caught up, it was the PC world had been running ahead of Apple for too long, Apple had no choice without dumping tons in R&D, which they had failed to do in the bad timeline of the 90s, pre-OSX, when the OS was so aged, it was even an embarrassment to hard core Apple fans.
Ya, Firewire is a Apple design, how many years old is it now?
Actually it has a GeforceFX Go 6800 Ultra, running in benchmarks only second to the current Geforce Ultra 6800, 7800, and ATI 850XL DESKTOP video cards.
Not too bad for a notebook.
Go to tomshardware.com and look it up if you are really have no idea.
And yes it is faster than what Apple is offering for their G5s.
I can do Doom on a freaking Notebook faster at a higher resolution than any Dual G5 Apple even with the lastest NVidia cards offered by Apple.
In fact the only way to get the resolution my NOTEBOOK has is to buy the 30" Apple display, sad, since they were once revered for their Graphical advantages.
I hope you know what a 'northbridge' is, and why it's called a 'chipset'.
Mod up the Grandparent if you want, but the post totally MISSED my point. I don't care what chipsets Apple engineered or assisted with.
My point was this, Apple is too use to using off the shelf technologies, even if they do some internal engineering, most parts inside a G5 have NO apple innovation. There are things like the memory bus adaptation to improve RAM speed in the G5, and there is Firewire, etc...
However, my argument is that Apple is getting to use to taking off the shelf technology and using it, instead of innovating, even the OS.
Apple should of took the G5 processor technology, fully extended it to the 3ghz range, and even adapted it to tri-core instead of waiting on IBM to do it for them. Microsoft easily did this, and this was JUST for the XBOX 360 gaming platform, let alone something as important to Apple as the central CPU in its entire product line.
So thank you for the post, but I kind of understand the difference between a CPU and Chipset, probably a bit more than I should even let on.
That's not correct -- Apple designed the G5's chipset, including the superfast memory bus that almost equals the Opterons'.
Apple engineers may change the name to G5, and slap their logo on the technology, but you need to go look up the POWER achitecture.
The G5 is nothing more than a IBM designed CPU, from the original IBM POWER designs, and is identical in operation to the IBM POWER designs which Apple has NO hand in...
This is why Apple is fed up with IBM's slow migration to the 3ghz version, and is moving to the Intel architecture.
If Apple was 'designing' the G5 or PowerPC CPUs, they would move them to the 3ghz range, instead of waiting for IBM to do it for them, and getting upset because it isn't happening.
after all, Apple puts a lot more engineering into their boxes than the typical PC vendor does
That's so sad that you actually believe what you say... As I sit here typing this on a Notebook made by a PC Vendor that can out benchmark a Dual G5 Mac, especially in graphics, let alone a Apple Notebook.
So tell me again, how Apple puts more 'engineering' into their boxes again? Their only engineering over the past few years is sucking parts from vendors, not building original hardware.
ATI and now NVidia has been throwing their PC based designed video cards (with features guided by DirectX and Microsoft) to Apple as scraps so Apple's video didn't totally suck.
And take even the G5, Apple did NOTHING to it beyond what IBM provided. Even freaking Microsoft was smart enough to license the PowerPC G5 technology, extend it, merge it into a tri-core CPU for their next XBOX (Yes with some help from NEC, etc) - but at least they did it.
Apple has done NOTHING with engineering 'their boxes', but taking what vendors are giving them. Apples does have some innovations under their belt, but their overall 'internal system' design is NOT one of them.
We know they used G5 computers. Do we know that they used OSX? This, to me, would be a MUCH bigger deal. It's one thing to go from DirectX on x86 to DirectX on PPC. The NT HAL takes care of that anyways. It seems a much bigger thing deal to go from DirectX on W2K kernel to DirectX on the Darwin kernel.
No OSX, No Darwin, they don't have the speed or compatibility of DirectX/XNA on the Windows NT Core. (Which is an argument I have had with both Apple and Open Source for a long time. Neither of which can EVER compete in gaming with Microsoft until a comperable technology exists on these platforms. Every one thinks DirectX is just video. IT IS NOT, it is a full set of tools and APIs for developers that manage sound, input, and all other aspects of gaming, as well as provide a rich 3D feature set so developers can just write games and NOT have to worry about the specific hardware it runs on.)
Back to the XBox 360....
I believe the OS Core on the development XBOX 360 platforms would be a about Windows 2003 timeline Windows NT Core running on the the Dual G5s.
Also something the Mac fans here DIDN'T seem to read is that the part where TWO Dual G5 Macs, with NVidia 6800 Video Cards that were running the XBOX360 demos were estimated to be more than HALF as fast as the XBOX 360 Tri-Core CPU and ATI-GPU - and that is also without providing any anti-aliasing as the XBOX 360 provides with no performance penalty.
So why would someone buy a console? Maybe because even two dual G5 Macs, running an optimized OS for it still can't produce the graphics of a 300-400 console platform.
People also like to quote the Original XBOX as just a 733 PIII, la la, but it was released like 5 years ago. At the time, there was NO PC that could do the comperable graphics the XBOX could.
NVidia's technology they put in the original XBOX didn't even hit PCs for a year or so later. (pixel shader's, etc)
The Orignal XBOX when it was released was seen as a powerhouse at the TIME. (Remember the quotes, if one bit was a gallon of water it could fill the Pacific ocean every second, etc).
The API used by Microsoft on the Alpha version of NT, and the only API anyone outside Microsoft has even seen for Alpha NT, is 32-bit. Whatever happens under the covers is irrelevant... it doesn't matter if the kernel is internally 8-bit, 16-bit, 32-bit, 64-bit, 48-bit, 60-bit, 80-bit, or 128-bit... only the exposed programming model matters. And that programming model under Windows NT on the Alpha is 32-bit. 32-bit address space, 32-bit words in every API... the fact that the Alpha has 64-bit registers stops at the OS interface, stops at the GUI libraries, stops at the memory manager, stops at the kernel interface.
This is such crap...
WindowNT on the Alpha was 64bit, used 64bit integer, 64bit floating point, and 64bit address space (only curtailed by the CPU Address Space limitation).
All applications that compiled to RUN on the Alpha were ALSO 64ibt applications, even though they used an API that was similar to the Win32 API, they just were able to take advantage of VLM and 64bit integers and floating point operations natively.
Do you even get out in the real world? You act like all this information is secret? If it was so secret HOW WAS ORACLE able to release an WindowsNT Alpha version of their database that was a 64bit application using all the 64bit abilities of the Alpha CPU and its VLM which greatly benefited the database.
You act like none of the Applications written for WindowsNT on the Alpha CPU existed...
Wake up Neo...
Not "I can't write GUI code", but "I can't write ANY code". Microsoft has never even published the kernel APIs except under NDA to companies like Softway Systems that were developing their own subsystems.
This isn't true... You can pick up books off Amazon even that fully explain the NT kernel and the API set. It may be a bit geeky, but not something Microsoft has EVER NOT disclosed to the public.
In fact If you want to write a subsystem for NT, you need this informaiton, even on MSDN you can find this informtion if you want to create somethin like this.
Even if Darwin was implemented with hamsters and tiny steam engines, it would still be more 64-bit than that.
But it would still be less of a 64bit OS than WindowsNT on the 64bit Alpha 10 years ago, and alot less of 64bit OS than WindowsXP 64 for Itanium released in 2001. Or even the new X86 version of WindowsXP 64bit versions available TODAY.
Microosft for gods sake is running a 64bit version of Windows on G5 PPCs for XBox 360 development even.
That was one of my original points even. Microsoft actually has a 64bit OS (Windows) running on the G5 before Apple does, and it is for gaminng development. How freaking ironic...
Windows users have SEVERAL options when it comes to 64bit architectures and 64bit versions of Windows.
Windows users are not waiting on a 64bit version, one has been available for years. Apple users are STILL Waiting, and will be for quite some time until Apple finally tells everyone that a 64bit OS is a good thing.
Apple just not have the technology development advtantage they did in the 80s. Windows has been running on various processors in various confirguartions for years. Microsoft got a solid NT core with cutler team, a NT client/server kernel that still is far above modern kernel incarnations and applauded by even *nix architects that are honest.
Apple instead is piggybacking on BSD technologies, and even their GUI rendering system is a Postscript/PDF/Adobe hack. (Not that BSD or the Adobe's technolgy isn't good, they just wern't something Apple created themselves.)
See the thing people don't GET from my posts, is I'm not a BIG FAN of Microsoft. I just think it is STUPID for any of us in the Open Source world to not know the strengths of what is on the market and we are competing with.
I also used to love Apple, I still wish OSX would be what it should have been back in 1993 when they were going to create an OS like it. However it isn't. and
I've been working on 64-bit UNIX for over ten years now... that's longer than Windows NT, 32-bit or 64-bit, has been a viable product. Why would I be doing that if there was no reason to use 64-bit applications?
I am growing tired of this discussion, and you are right, we are not communicating, as we probably agree more than we don't if we did converse better in these posts.
However in your statement above, you do realize that a 64bit version of WindowsNT has existed for over 10 years? It is what the Alpha CPU was premired with and was a 64bit OS, with the only 64bit restrictions being due to the capabilities of the first generation Alpha CPU. (i.e. 36bit memory address space).
Don't pretend like the only 64bit applications or OSes that have been around for a while are only *nix based. And yes there was a considerable market for the Alpha systems running WindowsNT - enough that until Compaq bought DEC and halted WindowsNT support there was a strong market for Windows 2000 at RC1 for Alpha that Microsoft worked with DEC and also invested a considerable amount of development into this OS.
Don't mislead people, even if you do know what I say is true already.
Um, Darwin is the UNIX underpinning, that's the only place the 64-bit stuff lives yet. It's Quartz and Aqua that's 32-bit only.
Oh one more thing. Darwin utilizes the cross 32-64 abilties of the PPC processor, but this is more of an accolade for the PPC not Darwin.
Let me clearly repeat this for everyone...
Darwin is not even Remotely a 64bit OS, not even close.
Calling a few 64bit tricks to grab more than 4GB on the PPC, because the PPC allows it does not make Darwin a 64bit OS.
It would be like saying that WindowsNT32 that runs on multiple CPU 32bit systems that can access 64GB of ram is a 64bit OS. Sure it can page past the 32bit address space limitations, but it is STILL a 32bit OS.
That makes a bigger difference on Windows than on UNIX, because UNIX doesn't run multiple apps in the same address space or map the kernel into the application's address space, so even in 32-bit mode every app already has its own 4G of address space to play with.
So on the Mac this is really only an issue if you have more than 4G of physical RAM. But even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage, you just need an OS that doesn't waste half the available address space (or 1/4 of it, even on NTAS) on shared memory.
Once again you have no understanding or are simply making up crap...
1) Unix environments DO MAP the Kernel into the Application Address Space...
2) Unix applications on a 32bit platform do not get a full 4gb of Address space.
Lets take a look at Linux for Example...
The Kernel not only takes 1GB from the application address space, but the kernel is LIMITED to 1GB. This creates many problems for the Linux platform.
Here maybe you need to actually read about memory management and get out of your Apple fan-boy/girl club.
But even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage, you just need an OS that doesn't waste half the available address space (or 1/4 of it, even on NTAS) on shared memory.
This is because it was running on a 64PPC CPU that allows for this, it has nothing to do with it being a UNIX base. Geesus...
WindowsNT-32bit can and has been able to even use PAGES to get access to more than 4GB of RAM of physical RAM. For example Multi-CPU x86-32bit systems running Windows 32bit versions would use this paging trick to access up to 64GB of RAM. (Something even far above the GREAT OSX and G5)...
Go read up on memory management and what really happens and try to understand not only the architectures of the CPUs in discussions but the OSes running on them.
BTW when you say that 'even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage', you do realize that 10.4 is ALSO not a 64bit OS.
I should just bow out because logic and facts will not win your mind over from the Apple marketing machine. You are a lost lemming that will believe everything and only what Apple marketing tells you.
Unfortunately, as nice as OSX and the Mac platform is, it is NOT always what the marketing would lead everyone to believe.
Um, going from x86 emulation to native x86-64 code does require OS support, yes, but that doesn't mean the performance increase is due to a change in the OS. In fact, it's due to a change in the chip. Switching from Windows-32 to Windows-64 on an Opteron means you're switching from the Opteron's legacy x86-support to its native instruction set. That's as big a change as swapping out the CPU.
Ok, you have sufficiently proven you know very little about how the x86 64bit CPUs from AMD work.
Secondly, my whole point was why a 64bit OS is IMPORTANT. You just confirmed that IT IS IMPORTANT... Maybe you should join with some of the others in asking Apple to really make a 64bit verion of OSX. (And yes the PPC architecture would ALSO greatly improve if ran in the native 64bit mode.
Geesh.
Here also go read up on the performance differences and modes of the x86-64bit architecture.
You are so basically knowledgeable, yet there are so many erroneous facts in your statements.
Let take this one for example: No, RDP is not part of the Windows GUI, it runs on top of it and does "screen scraping" to transfer bitmaps back and forth. I'm talking about the design of the GUI itself.
The Windows GUI is not a client-server protocol, it's a library based design that involves very tight coupling between GDI and the application,
RDP is a Client-Server Protocol, and it DOES NOT sit on top of the GUI, but actually just UNDER the GUI, as the RDP information is intercepting the GDI calls, and transferring them across the network - not images only. Sure bitmaps are transferred but only when GDI information cannot create the bitmaps, just LIKE XWindows in effect.
I never said the Windows GUI was EVER designed to be a remote PROTOCOL. EVER. In fact it was my argument that it was NOT and therefore did not have the inherent SPEED problems of the XWindows Model on a desktop. As you admit yourself.
However, RDP (Remote Desktop Protocol) WAS Designed to be a REMOTE PROTOCOL, and as it does sit between the GUI and DOES NOT do 'SCRAPPING'.
The older applications like PCAnywhere would be an example of "SCRAPPING", the RDP in Windows is NOT... Maybe read up on some of this before you make statements of assumptions.
Also if you will note the RDP progression for Longhorn ALSO grabs Calls from the GDI, and the AVALON system. Hence the full 3D Vector Model used In Longhorn will actually be FASTER than either XWindows or Windows RDP today, as all the majority of the protocol information for newer applications that have abandoned BITMAPS will only have to send Vector and coordinate information, which is a lot faster than rendered Bitmaps.
64 bit platforms are not the "next generation" in anything. They've been "this generation" since 1994. I had a 64-bit desktop before even Mac OS 9 and Windows 98 were "this generation".
They will Be the next generation whether you like it or not. I had this same freaking argument with a co-worked at the University back in 1988 about 286s and 386s platforms. The also argued that 32bit computing was a waste of CPU performance and there would NEVER be a reason we would need to write applications in 32bit mode. 17 years later, and geeks are now telling me the same thing about 32bit applications and 64bit applications. Geesh.
Do you really believe that all desktop CPUs will not move to 64bits in a few years? Truly do you think technology will stop and say 32bits are enough? Dumb...
BTW I was at the Alpha launch(Comdex 1992 - Demoed using a native version of Windows NT 3.1 at the time), and also had one of the first Alpha systems around. It was doing 'supercomputer' class performance at the time not 'in spite of' but was helped by the fact it was a 64bit CPU.
And unless address space is an issue, "64 bit platforms" are no faster than "32 bit platforms", all other things being equal.
So computing two 64bit integers once as opposed to two 32bit integers (TWICE) is really going to be equal? You are kidding right?
It is NOT all about the transferring of bits, but the computation that does not have to break the data into small chunks.
Because they're doing 64-bit integer operations, but Windows 64-bit model uses 32-bit integer operations, so there's no performance advantage to 64-bit code just from the 64-bit instruction set...
Where are you making this up from... Give me the links?
Windows 64bit, boots into Long Mode (64bit mode) (if we are talking about the AMD 64bit x386 design, if you want to talk about how it boots on the Itanium or Alpha, the process differs).
And after booting the OS and NT Kernel run in a NATIVE 64bit mode using and computing with 64bit integers.
For gods sake, even 32bit applications running on Windows64 run faster because in long mode of the OS they get the use of the 64bit Registers, as well as
1. You redesigned all your services to use 64-bit large-address space code where you had previously been using read/write to access a large disk-based object through a 2G "windows".
Um, actually Windows 32bit (NT in fact since 1993) has supported a 17 terabyte partion address space.
2. Newer computers are generally faster than old ones, and Opteron's instruction set is faster than IA32 because of the slightly cleaner ABI and the larger register file.
Only the OS changed for the performance increase.
3. You're running Oracle, and Oracle through utterly heroic measures has managed to take advantage of the "32-bit-segmented" mode of the Xeon to get option 1 working, so your speedup is a special case that most people can't take advantage of.
Sorry, I don't let my friends do Oracle, nor do I.
4. You don't know, and you don't care.
Maybe because I knew you like making cute lists on topics you don't have any informative input on, and I thought I would give you something to ponder for your next top 5 list in your exciting life as it must be.
And this is a problem... why? I mean, this is a completely normal way to build GUI applications
You're right this is not a problem; however it is a kludge. It would be nice for the developers to be able to have an application that didn't have to perform a bit of 'magic' to run the 64bit code.
It would also be nice to have real 64bit libraries and a 64bit GUI to build the application in and not have to have separate builds for the interface and the code that does any 64bit calculations.
Everyone in this post keeps saying that 64bits is slower and not needed; however, this is borderline compulsive in an attempt to save face for Apple, when the Apple users should be directing their angst towards Apple for leaving them short.
64bit platforms are the NEXT generation in both SPEED and ADDRESS SPACE that will be needed for the next generation of applications.
If 64bits is in fact 'often slower' as one person quoted, then why are the GPUs on EVERY high speed video card now running with 64,128 and higher bits to 'get the performance' that was NOT possible on 32bit GPUs.
Why? Because Microsoft had to come up with a scheme that would let them link 64-bit applications with 32-bit libraries without having to port all the existing libraries to 32-bit
This isn't even factually correct. Are you sure you have developed for Windows or did you find this on some great Google search to prove your point.
Most windows applications port to the 64bit version with little more than a simple recompile, and with integer declarations that do need to be expanded, does the concept of search and replace ring a bell?
Additionally, you assert this is because of the need for 64bit applications to reference 32bit DLLs/libraries. Again this is NOT true. A 64bit application on Windows has NO direct mechanism for calling 32bit code/DLLs/Libraries. PERIOD.
Go read the Free Developers SDK at least.
and unlike UNIX Windows doesn't use a clean client-server display interface.
'clean'? are you splitting hairs of comparing RDP to the X-Windows Protocol? You seem to think this is a 'feature' of UNIX's GUI subsystem, but if you really know GUI developers that are trying to get performance out of XWindows they will sit you down and explain why a network graphical 'protocol' is truly not that best Graphical subsystem on a desktop system.
It is great if 10 users are accessing the computer via the network, but on a single desktop with a single user, it is a big bottleneck in not only RAM but in performance on *nix environments.
Now we could argue if Xwindows or RDP was the better remote GUI protocol, and I could honestly argue either way, as both have benefits over the other, but if you are going to tell everyone that this makes any *nix better or faster on the desktop, you will probably have a few GUI application developers stepping in to dismiss your comments.
There IS a reason why OSX's Quartz does NOT sit on top of XWindows and does not itself have a 'clean' client-server display interface. They wanted performance for their users.
I'm going to skip your rant about the PPC/x86 debate. I find a lot of good things about the PPC architecture, however it is a bit sad that Microsoft will be popping out a tri-core 3.2ghz gaming console running Windows64 on it before Apple even offers a CPU with that speed or configuration, or has a 64bit version of OSX running on it. Sad that on Apple's own turf, Microsoft could pull off what so far Apple does not even have planned to do, especially with a full 64bit version of OSX. (And again this is a testament to NT architecture that not only has it in the past, but still can run very easy on PPC systems and even do so in the newer 64bit modes and having a fully ported DirectX subsystem for this architecture as well for gaming performance.
For people who actually need it, Tiger is a BETTER 64-bit environment than Windows 2003/XP, because their existing code that they've bee
So you're comparing two completely different architectures, one of which was register starved until the move to 64bit, and the other which had no such problems, and claiming that moving to a fully 64bit architecture made a difference for one, so it must for the other?
Ok, I will give you ONE chance to re-read my post.
As you will note I did not say that by switching to the 64bit CPU servers we gained the performance, we were already running the 64bit hardware, but with the 32bit version of Windows.
It was when we moved the server OSes to the 64bit versions of Windows we got the dramatic performance increase and we were able to reduce the servers in use to 1/5.
So if your argument that the only benefit in the 64bit x86 platform is from the extra registers, then we would of had the dramatic performance increase when we put in the 64bit hardware even with the 32bit version of Windows. Instead the performance increase came when we changed the OS to the 64bit version.
I think you should do a little research on the various 64bit architectures and OSes and the benefits before totally looking like a fool.
Or for god's sake, at least fully read and comprehend the post you are responding to next time.
The point is, making OSX fully 64-bit would actually slow it down because things that work just fine with 32 bits would suddenly have to move around twice as much data and take twice as much memory, most of which would be unused. The speedup whem moving to a 64bit version of Windows is simply because the ia32 architecture is so bloody awful that AMD had nowhere to go but up with x86-64.
Thsi is just not true. The only way an 64bit of OSX would be a slower is if the Apple truly kludged it up, and I don't think they would.
x86 32bit architecture was OLD, but the current implementations of the 32bit x86 design are quite advanced and are technically can complete neck and neck the teh x86-64 version of themselves. It is just when dealing with LARGE amounts of data that is either being computed or pushed in and out, the extra lanes in a 64bits really help out for performance.
There used to be an old argument that an optimized 16bit application would run faster than a 32bit application as well. This is only partially true, and only holds true then you are not trying to move large amounts of information or do high end calcualtion where the utilization of he extra bits pay off.
The same is true of 32bit vs 64bits. There are exmaples where a 32bit applicaiton could be technically faster, but there are more expamples where the utilization of the 64bit nature of the CPU pays off not only in performance but in the long term with the increasing demand for RAM we are starting to face.
Apple will someday and will be smart to do so someday to move OSX to a pure 64bit OS. It may be later than Microsoft and other OS vendors, but they will eventually have to do it to keep up.
But I understand, it's either to flame Apple and claim they don't have a 64-bit OS. Because we all know that 64-bit is 1) twice as much as 32-bit (2^32*2=2^64, welcome to maths!) 2) always faster than 32-bit (yeah! exactly for, like, word processing! dude, i can type in 64-bit!) 3) a long-needed change especially for desktop computers (i've always wanted to calculate the newest genom information at home... finally I can do it)
Cute...
In my world though, moving our servers to the 64bit version of Windows took our server requirments to less than 1/5 the equivalent systems when running the 32bit version of Windows 2003 Server.
So maybe that fuzzy math stuff does work in the real world.
(I hope to God your next post isn't a boring rant on explaining 40bit/48bit memory addressing of 64bit CPUs, the history of the Alpha 64bit CPU, or any other nerd item that 90% of everyone here should know.)
Yes it is, and I am sick and tired of this debate. If you don't understand the 32/64 bit nature of PowerPC, then please, for the love of $DEITY, shut up.
Ok prove it.. Show the entire world where there are two sets of binaries of OSX Tiger. One set that runs on the G5 and truly is a 64bit OS and show us the second set that is a 32bit OS and runs on the G4 and G3.
You won't... Period. If Tiger was 64bit, #1 all the interface elements of OSX would run in 64bit mode, #2 All the drivers and Darwin layer of OSX would run in 64bit mode - and NEITHER DO.
And as I demonstrated before the easiest way to check for this, can Tiger run on a G3 processor? If it can USING THE SAME COMPILED BINARIES, then it is NOT NOR COULD BE a 64bit OS.
(Tiger does allow some applications to run in 64bit mode, but by doing so the applications are severed from working with many of the GUI and other libraries available to developers - mean virtually the Application would have to broken up to have a 32bit interface and only use the 64bit code for intensive calculations.
On the other hand you have *nix like Linux and even WindowsXP/2003 which have been doing a FULL and TRUE 64bit OS for many years...
Ok, first off I said "Shove calculation intensive operations"
Secondly I a pretty good understanding of how a CPU and its relation to its bit capability are relavent. 64bit CPUS are not limited to just pushing data in lanes of 64bits, but ALSO computing data in 64bit segments. It is not all about databus - trust me.
As for the reason Apple tries to convince its customers that moving to 64bits is NOT necessary is only because their OS is not 64bits. It is just that plain and simple.
If 64bits were NOT important, then why were we all blasted with Apple's Marketing of the 'first 64bit desktop computer' - which wasn't even true. But yet by the G5 having 64bits (even though not used) seemed to be pretty important to Apple if they went out of their way to market it and tell everyone it had 64bits.
So maybe you can tell us truly where Apple stand on the benefits of 64bit computing... 1) Good cause the G5 has 64bits... 2) Bad cause OSX doesn't support it?
Get past the marketing con artists for 5 seconds and maybe you can see outside your cave.
It took apple many years to move from the 68000 line to the PPC.
This is true, it took them 8 years longer than anticipated, which is really quite sad, as System 9 software didn't even fully utilize the PPC capabilities, wasting a lot of the power of the late 90s PPC based Macs.
I guess this follows Apple style, as they are wasting performance and capabilities in the G5 PPC today as well.
However I still assume that this "3 core" cpu can't be 3 full fledged G5 cpu's running at 3ghz. If it were, I am sure Apple would be, or will be, using it.
Heck they dont even have a single G5 ppc running at 3Ghz yet, do they?
Wrong again... In fact it will be at least the G5 technology as the game testing and development software for the XBox 360 was designed for G5 Macs.
It you would read the posts here, you would see why it IS 3GHz and Tri-Core, when Apple is still using older PPC G5 technologies.
Microsoft licensed the technologies and were able to further engineer the CPU themselves, that is why they own the rights to the technology and produce the CPUs at any fabrication plant, and not only produce them directly through the IBM route as Apple has to do.
Slashdot Mirror
If you are typsetting in Word, then I truly feel sorry for you, there are so many good DTP applications that handle text flow and text formating much much better.
When we don't even have OSes that support Font features to correct font errors like ligatures, how can we expect Glorified word processors to do so.
I actually hope the updated font system in Longhorn takes us a step into the future of onscreen typography. Even the Mac font rendering capabilities are pathetic and they pretend to cater to the graphic design market.
What are ligatures, and why would I care about them in typing up a letter?
Ligatures are actually something the poster would NOT want, so I assume they mean more advanced recognition of ligatures and kerning adjustment.
Ligatures are when non-propotional fonts spacing have two letters that interfere with each other. A common example would be 'fi' notice that the dot on the i may run into the f, and it shouldn't.
Basically Word only uses the standard Font system that is built into the OS, and OS font systems don't always correct items that typsetters or typography people need corrected in publication.
As for the ligature problem, it is more a problem with the FONTS themselves than the fact that Word doesn't compensate for the problems in the font.
A real DTP program, like Quark or Pagemaker does more advanced kerning using an internal system of textual placement.
This allows them to notice and fix things like ligatures and also gives you more features like when doing justified margins they can do 'real' justification where the letter spacing is adjusted and not just the spaces between words is adjusted, as Microsoft Word does now.
I'll eat someone's tinfoil hat if this works plugin-free with any browser other than IE7 on Windows.
Better buy the tinfoil and get your web cam ready. The forms are SERVER objects that are represented to a browser using standard XML,HTML and very generic JavaScripting which means the data services of the form Server side.
Go read the notes on Web Forms development with Visual Studio 2005 Beta 2. Same thing, just instead of creating them in a development environment, it lets managers and business people create data collection systems as easily as creating the form.
Just tested an application of Web forms from Visual Studio 2005, using the same concepts and technology. It works in IE on Windows, IE on my Mac, Safari on my Mac, Firefox on my Linux installation, and even IE 4.0 on Mac 8.1 Emulator.
Enjoy eating the hat, and maybe comment on something you actually are informed on next time.
Too bad Microsoft hasn't copied Apples concept of bundled files (i'm not positive of the exact term) that act as a single file. It really is nice. (and it is used by their 'pages' software in the same way microsoft is zipping files together).
Microsoft has had 'paired' folders and files since like Win2k. Windows also can treat a Zip Binary as a single archive folder.
Do you just assume every feature on a Mac is unique to a Mac or don't get out in the real world much?
No offense to other Mac fans.
The "System Controller" is an exclusive Apple part, built in IBM's fabs, but designed by Apple. It's responsible for connecting and arbitrating the G5's FSBs to memory and the HyperTransport bus.
Plus, cooling and power management are unique to Apple, and heavily integrated with the system software. This includes the passive liquid cooling system in the top-end G5s. Details can be seen in the Developer note [apple.com].
Ya, they helped designed the Memory Transport, but out of the wonderful Mac, that is about the thing anyone comes up with, other than the obvious Firewire.
Liquid Cooling System... Ooh, that is really 'cool'. This has also been available in the PC world for years and years. Do you really thing Apple was the first to come up with this 'innovation'? Come on, get out of the Apple 'elitist' marketing...
The sleep mode on a Mac just works. A mac is always available in the time it takes a monitor to warm up... about two seconds for an LCD.
A mac notebook is well balanced. The user's hands don't cover the speakers. It doesn't bristle with fiddly rubber bump knobs that inevitably break off.
A mac notebook can be mounted as a firewire target disk on any other firewire based computer. This means instant filetransfers of any size with no configuration.
There are hundreds of things like this, and no, they'll never show up on a spec sheet. If you're the kind of person who doesn't care about these things, it doesn't matter.
Still, for many of us, these features are critical. They make the difference between a 1990s style machine we're constantly fiddling with and a modern machine that just fades into the background any time it isn't in use, and most of the time that it is.
The kind of engineering you're talking about is chip design. No, apple doesn't design chips. They used to! They used to design all sorts of hardware standards, like NuBus and ADB. Once USB and AGP came around, though, there wasn't any point. The PC world had finally caught up and started making products that could support the Macintosh experience, so Apple went with the PC standards, cut prices, and all was good.
Maybe it has been a while since you have used a modern NON-APPLE notebook.
Yes, my notebook can outbenchmark a G5. It has a 64bit P4 3.8ghz Processor, and a GeforceFX 6800 Ultra video, and a 7200rpm Raid 0 Array in it. And I can run a full and real 64bit Operating System on it.
Ok, lets go through your list?
Sound? My notebook is virtually silent unless running high end Video, and the fans will kick on for a moment or two. No louder than a freaking powerbook.
Sleep Mode? Ever hear of Hibernate on WindowsXP or Standby mode. I can even choose which I want. Standby (comparable to Apple Sleep mode) I can turn on my notebook in under 1 sec. Hibernate, which it returns from a FULLY POWERED off state, I can be on and online in less than 5secs. Obviously you read the Apple marketing when they tell you they have the fastest and best sleep method, sorry, they don't.
Well Balanced? I don't cover the speakers with my hands, didn't on my 3yr old notebook either. And it, like this one has built in subwoofer with sound only a Apple notebook could beg to offer. Current laptop has 4+1 Speaker system even.
As for 'fiddly rubber bump knobs'? My case is aluminum, and all my buttons are also aluminum, no plastic crap here either.
Firewire target disk? And you think WindowsXP doesn't do this? Do you live in a barn?
The thing you DON'T realize, is that I use Mac Notebooks, work with people that use them every day as their primary computer. Trust me, there is no comparison.
As for Apple design... That ended years ago. It wasn't just the PC world caught up, it was the PC world had been running ahead of Apple for too long, Apple had no choice without dumping tons in R&D, which they had failed to do in the bad timeline of the 90s, pre-OSX, when the OS was so aged, it was even an embarrassment to hard core Apple fans.
Ya, Firewire is a Apple design, how many years old is it now?
shitbox notebook has a 6800GT
Actually it has a GeforceFX Go 6800 Ultra, running in benchmarks only second to the current Geforce Ultra 6800, 7800, and ATI 850XL DESKTOP video cards.
Not too bad for a notebook.
Go to tomshardware.com and look it up if you are really have no idea.
And yes it is faster than what Apple is offering for their G5s.
I can do Doom on a freaking Notebook faster at a higher resolution than any Dual G5 Apple even with the lastest NVidia cards offered by Apple.
In fact the only way to get the resolution my NOTEBOOK has is to buy the 30" Apple display, sad, since they were once revered for their Graphical advantages.
I hope you know what a 'northbridge' is, and why it's called a 'chipset'.
Mod up the Grandparent if you want, but the post totally MISSED my point. I don't care what chipsets Apple engineered or assisted with.
My point was this, Apple is too use to using off the shelf technologies, even if they do some internal engineering, most parts inside a G5 have NO apple innovation. There are things like the memory bus adaptation to improve RAM speed in the G5, and there is Firewire, etc...
However, my argument is that Apple is getting to use to taking off the shelf technology and using it, instead of innovating, even the OS.
Apple should of took the G5 processor technology, fully extended it to the 3ghz range, and even adapted it to tri-core instead of waiting on IBM to do it for them. Microsoft easily did this, and this was JUST for the XBOX 360 gaming platform, let alone something as important to Apple as the central CPU in its entire product line.
So thank you for the post, but I kind of understand the difference between a CPU and Chipset, probably a bit more than I should even let on.
That's not correct -- Apple designed the G5's chipset, including the superfast memory bus that almost equals the Opterons'.
Apple engineers may change the name to G5, and slap their logo on the technology, but you need to go look up the POWER achitecture.
The G5 is nothing more than a IBM designed CPU, from the original IBM POWER designs, and is identical in operation to the IBM POWER designs which Apple has NO hand in...
This is why Apple is fed up with IBM's slow migration to the 3ghz version, and is moving to the Intel architecture.
If Apple was 'designing' the G5 or PowerPC CPUs, they would move them to the 3ghz range, instead of waiting for IBM to do it for them, and getting upset because it isn't happening.
Source http://en.wikipedia.org/wiki/Powerpc
after all, Apple puts a lot more engineering into their boxes than the typical PC vendor does
That's so sad that you actually believe what you say... As I sit here typing this on a Notebook made by a PC Vendor that can out benchmark a Dual G5 Mac, especially in graphics, let alone a Apple Notebook.
So tell me again, how Apple puts more 'engineering' into their boxes again? Their only engineering over the past few years is sucking parts from vendors, not building original hardware.
ATI and now NVidia has been throwing their PC based designed video cards (with features guided by DirectX and Microsoft) to Apple as scraps so Apple's video didn't totally suck.
And take even the G5, Apple did NOTHING to it beyond what IBM provided. Even freaking Microsoft was smart enough to license the PowerPC G5 technology, extend it, merge it into a tri-core CPU for their next XBOX (Yes with some help from NEC, etc) - but at least they did it.
Apple has done NOTHING with engineering 'their boxes', but taking what vendors are giving them. Apples does have some innovations under their belt, but their overall 'internal system' design is NOT one of them.
You need to get out more...
We know they used G5 computers. Do we know that they used OSX? This, to me, would be a MUCH bigger deal. It's one thing to go from DirectX on x86 to DirectX on PPC. The NT HAL takes care of that anyways. It seems a much bigger thing deal to go from DirectX on W2K kernel to DirectX on the Darwin kernel.
No OSX, No Darwin, they don't have the speed or compatibility of DirectX/XNA on the Windows NT Core. (Which is an argument I have had with both Apple and Open Source for a long time. Neither of which can EVER compete in gaming with Microsoft until a comperable technology exists on these platforms. Every one thinks DirectX is just video. IT IS NOT, it is a full set of tools and APIs for developers that manage sound, input, and all other aspects of gaming, as well as provide a rich 3D feature set so developers can just write games and NOT have to worry about the specific hardware it runs on.)
Back to the XBox 360....
I believe the OS Core on the development XBOX 360 platforms would be a about Windows 2003 timeline Windows NT Core running on the the Dual G5s.
Also something the Mac fans here DIDN'T seem to read is that the part where TWO Dual G5 Macs, with NVidia 6800 Video Cards that were running the XBOX360 demos were estimated to be more than HALF as fast as the XBOX 360 Tri-Core CPU and ATI-GPU - and that is also without providing any anti-aliasing as the XBOX 360 provides with no performance penalty.
So why would someone buy a console? Maybe because even two dual G5 Macs, running an optimized OS for it still can't produce the graphics of a 300-400 console platform.
People also like to quote the Original XBOX as just a 733 PIII, la la, but it was released like 5 years ago. At the time, there was NO PC that could do the comperable graphics the XBOX could.
NVidia's technology they put in the original XBOX didn't even hit PCs for a year or so later. (pixel shader's, etc)
The Orignal XBOX when it was released was seen as a powerhouse at the TIME. (Remember the quotes, if one bit was a gallon of water it could fill the Pacific ocean every second, etc).
The API used by Microsoft on the Alpha version of NT, and the only API anyone outside Microsoft has even seen for Alpha NT, is 32-bit. Whatever happens under the covers is irrelevant... it doesn't matter if the kernel is internally 8-bit, 16-bit, 32-bit, 64-bit, 48-bit, 60-bit, 80-bit, or 128-bit... only the exposed programming model matters. And that programming model under Windows NT on the Alpha is 32-bit. 32-bit address space, 32-bit words in every API... the fact that the Alpha has 64-bit registers stops at the OS interface, stops at the GUI libraries, stops at the memory manager, stops at the kernel interface.
This is such crap...
WindowNT on the Alpha was 64bit, used 64bit integer, 64bit floating point, and 64bit address space (only curtailed by the CPU Address Space limitation).
All applications that compiled to RUN on the Alpha were ALSO 64ibt applications, even though they used an API that was similar to the Win32 API, they just were able to take advantage of VLM and 64bit integers and floating point operations natively.
Do you even get out in the real world? You act like all this information is secret? If it was so secret HOW WAS ORACLE able to release an WindowsNT Alpha version of their database that was a 64bit application using all the 64bit abilities of the Alpha CPU and its VLM which greatly benefited the database.
You act like none of the Applications written for WindowsNT on the Alpha CPU existed...
Wake up Neo...
Not "I can't write GUI code", but "I can't write ANY code". Microsoft has never even published the kernel APIs except under NDA to companies like Softway Systems that were developing their own subsystems.
This isn't true... You can pick up books off Amazon even that fully explain the NT kernel and the API set. It may be a bit geeky, but not something Microsoft has EVER NOT disclosed to the public.
In fact If you want to write a subsystem for NT, you need this informaiton, even on MSDN you can find this informtion if you want to create somethin like this.
Even if Darwin was implemented with hamsters and tiny steam engines, it would still be more 64-bit than that.
But it would still be less of a 64bit OS than WindowsNT on the 64bit Alpha 10 years ago, and alot less of 64bit OS than WindowsXP 64 for Itanium released in 2001. Or even the new X86 version of WindowsXP 64bit versions available TODAY.
Microosft for gods sake is running a 64bit version of Windows on G5 PPCs for XBox 360 development even.
That was one of my original points even. Microsoft actually has a 64bit OS (Windows) running on the G5 before Apple does, and it is for gaminng development. How freaking ironic...
Windows users have SEVERAL options when it comes to 64bit architectures and 64bit versions of Windows.
Windows users are not waiting on a 64bit version, one has been available for years. Apple users are STILL Waiting, and will be for quite some time until Apple finally tells everyone that a 64bit OS is a good thing.
Apple just not have the technology development advtantage they did in the 80s. Windows has been running on various processors in various confirguartions for years. Microsoft got a solid NT core with cutler team, a NT client/server kernel that still is far above modern kernel incarnations and applauded by even *nix architects that are honest.
Apple instead is piggybacking on BSD technologies, and even their GUI rendering system is a Postscript/PDF/Adobe hack. (Not that BSD or the Adobe's technolgy isn't good, they just wern't something Apple created themselves.)
See the thing people don't GET from my posts, is I'm not a BIG FAN of Microsoft. I just think it is STUPID for any of us in the Open Source world to not know the strengths of what is on the market and we are competing with.
I also used to love Apple, I still wish OSX would be what it should have been back in 1993 when they were going to create an OS like it. However it isn't. and
I've been working on 64-bit UNIX for over ten years now... that's longer than Windows NT, 32-bit or 64-bit, has been a viable product. Why would I be doing that if there was no reason to use 64-bit applications?
I am growing tired of this discussion, and you are right, we are not communicating, as we probably agree more than we don't if we did converse better in these posts.
However in your statement above, you do realize that a 64bit version of WindowsNT has existed for over 10 years? It is what the Alpha CPU was premired with and was a 64bit OS, with the only 64bit restrictions being due to the capabilities of the first generation Alpha CPU. (i.e. 36bit memory address space).
Don't pretend like the only 64bit applications or OSes that have been around for a while are only *nix based. And yes there was a considerable market for the Alpha systems running WindowsNT - enough that until Compaq bought DEC and halted WindowsNT support there was a strong market for Windows 2000 at RC1 for Alpha that Microsoft worked with DEC and also invested a considerable amount of development into this OS.
Don't mislead people, even if you do know what I say is true already.
Um, Darwin is the UNIX underpinning, that's the only place the 64-bit stuff lives yet. It's Quartz and Aqua that's 32-bit only.
Oh one more thing. Darwin utilizes the cross 32-64 abilties of the PPC processor, but this is more of an accolade for the PPC not Darwin.
Let me clearly repeat this for everyone...
Darwin is not even Remotely a 64bit OS, not even close.
Calling a few 64bit tricks to grab more than 4GB on the PPC, because the PPC allows it does not make Darwin a 64bit OS.
It would be like saying that WindowsNT32 that runs on multiple CPU 32bit systems that can access 64GB of ram is a 64bit OS. Sure it can page past the 32bit address space limitations, but it is STILL a 32bit OS.
That makes a bigger difference on Windows than on UNIX, because UNIX doesn't run multiple apps in the same address space or map the kernel into the application's address space, so even in 32-bit mode every app already has its own 4G of address space to play with.
So on the Mac this is really only an issue if you have more than 4G of physical RAM. But even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage, you just need an OS that doesn't waste half the available address space (or 1/4 of it, even on NTAS) on shared memory.
Once again you have no understanding or are simply making up crap...
1) Unix environments DO MAP the Kernel into the Application Address Space...
2) Unix applications on a 32bit platform do not get a full 4gb of Address space.
Lets take a look at Linux for Example...
The Kernel not only takes 1GB from the application address space, but the kernel is LIMITED to 1GB. This creates many problems for the Linux platform.
Here maybe you need to actually read about memory management and get out of your Apple fan-boy/girl club.
http://kerneltrap.org/node/2450
But even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage, you just need an OS that doesn't waste half the available address space (or 1/4 of it, even on NTAS) on shared memory.
This is because it was running on a 64PPC CPU that allows for this, it has nothing to do with it being a UNIX base. Geesus...
WindowsNT-32bit can and has been able to even use PAGES to get access to more than 4GB of RAM of physical RAM. For example Multi-CPU x86-32bit systems running Windows 32bit versions would use this paging trick to access up to 64GB of RAM. (Something even far above the GREAT OSX and G5)...
Go read up on memory management and what really happens and try to understand not only the architectures of the CPUs in discussions but the OSes running on them.
BTW when you say that 'even under 10.3 the Powermac G5 supported 8G of physical RAM. You don't need a 64-bit OS to get this advantage', you do realize that 10.4 is ALSO not a 64bit OS.
I should just bow out because logic and facts will not win your mind over from the Apple marketing machine. You are a lost lemming that will believe everything and only what Apple marketing tells you.
Unfortunately, as nice as OSX and the Mac platform is, it is NOT always what the marketing would lead everyone to believe.
Um, going from x86 emulation to native x86-64 code does require OS support, yes, but that doesn't mean the performance increase is due to a change in the OS. In fact, it's due to a change in the chip. Switching from Windows-32 to Windows-64 on an Opteron means you're switching from the Opteron's legacy x86-support to its native instruction set. That's as big a change as swapping out the CPU.
Ok, you have sufficiently proven you know very little about how the x86 64bit CPUs from AMD work.
Secondly, my whole point was why a 64bit OS is IMPORTANT. You just confirmed that IT IS IMPORTANT... Maybe you should join with some of the others in asking Apple to really make a 64bit verion of OSX. (And yes the PPC architecture would ALSO greatly improve if ran in the native 64bit mode.
Geesh.
Here also go read up on the performance differences and modes of the x86-64bit architecture.
http://www.devx.com/amd/Article/16101
You are so basically knowledgeable, yet there are so many erroneous facts in your statements.
Let take this one for example:
No, RDP is not part of the Windows GUI, it runs on top of it and does "screen scraping" to transfer bitmaps back and forth. I'm talking about the design of the GUI itself.
The Windows GUI is not a client-server protocol, it's a library based design that involves very tight coupling between GDI and the application,
RDP is a Client-Server Protocol, and it DOES NOT sit on top of the GUI, but actually just UNDER the GUI, as the RDP information is intercepting the GDI calls, and transferring them across the network - not images only. Sure bitmaps are transferred but only when GDI information cannot create the bitmaps, just LIKE XWindows in effect.
I never said the Windows GUI was EVER designed to be a remote PROTOCOL. EVER. In fact it was my argument that it was NOT and therefore did not have the inherent SPEED problems of the XWindows Model on a desktop. As you admit yourself.
However, RDP (Remote Desktop Protocol) WAS Designed to be a REMOTE PROTOCOL, and as it does sit between the GUI and DOES NOT do 'SCRAPPING'.
The older applications like PCAnywhere would be an example of "SCRAPPING", the RDP in Windows is NOT... Maybe read up on some of this before you make statements of assumptions.
Also if you will note the RDP progression for Longhorn ALSO grabs Calls from the GDI, and the AVALON system. Hence the full 3D Vector Model used In Longhorn will actually be FASTER than either XWindows or Windows RDP today, as all the majority of the protocol information for newer applications that have abandoned BITMAPS will only have to send Vector and coordinate information, which is a lot faster than rendered Bitmaps.
64 bit platforms are not the "next generation" in anything. They've been "this generation" since 1994. I had a 64-bit desktop before even Mac OS 9 and Windows 98 were "this generation".
They will Be the next generation whether you like it or not. I had this same freaking argument with a co-worked at the University back in 1988 about 286s and 386s platforms. The also argued that 32bit computing was a waste of CPU performance and there would NEVER be a reason we would need to write applications in 32bit mode. 17 years later, and geeks are now telling me the same thing about 32bit applications and 64bit applications. Geesh.
Do you really believe that all desktop CPUs will not move to 64bits in a few years? Truly do you think technology will stop and say 32bits are enough? Dumb...
BTW I was at the Alpha launch(Comdex 1992 - Demoed using a native version of Windows NT 3.1 at the time), and also had one of the first Alpha systems around. It was doing 'supercomputer' class performance at the time not 'in spite of' but was helped by the fact it was a 64bit CPU.
And unless address space is an issue, "64 bit platforms" are no faster than "32 bit platforms", all other things being equal.
So computing two 64bit integers once as opposed to two 32bit integers (TWICE) is really going to be equal? You are kidding right?
It is NOT all about the transferring of bits, but the computation that does not have to break the data into small chunks.
Because they're doing 64-bit integer operations, but Windows 64-bit model uses 32-bit integer operations, so there's no performance advantage to 64-bit code just from the 64-bit instruction set...
Where are you making this up from... Give me the links?
Windows 64bit, boots into Long Mode (64bit mode) (if we are talking about the AMD 64bit x386 design, if you want to talk about how it boots on the Itanium or Alpha, the process differs).
And after booting the OS and NT Kernel run in a NATIVE 64bit mode using and computing with 64bit integers.
For gods sake, even 32bit applications running on Windows64 run faster because in long mode of the OS they get the use of the 64bit Registers, as well as
1. You redesigned all your services to use 64-bit large-address space code where you had previously been using read/write to access a large disk-based object through a 2G "windows".
Um, actually Windows 32bit (NT in fact since 1993) has supported a 17 terabyte partion address space.
2. Newer computers are generally faster than old ones, and Opteron's instruction set is faster than IA32 because of the slightly cleaner ABI and the larger register file.
Only the OS changed for the performance increase.
3. You're running Oracle, and Oracle through utterly heroic measures has managed to take advantage of the "32-bit-segmented" mode of the Xeon to get option 1 working, so your speedup is a special case that most people can't take advantage of.
Sorry, I don't let my friends do Oracle, nor do I.
4. You don't know, and you don't care.
Maybe because I knew you like making cute lists on topics you don't have any informative input on, and I thought I would give you something to ponder for your next top 5 list in your exciting life as it must be.
And this is a problem... why? I mean, this is a completely normal way to build GUI applications
You're right this is not a problem; however it is a kludge. It would be nice for the developers to be able to have an application that didn't have to perform a bit of 'magic' to run the 64bit code.
It would also be nice to have real 64bit libraries and a 64bit GUI to build the application in and not have to have separate builds for the interface and the code that does any 64bit calculations.
Everyone in this post keeps saying that 64bits is slower and not needed; however, this is borderline compulsive in an attempt to save face for Apple, when the Apple users should be directing their angst towards Apple for leaving them short.
64bit platforms are the NEXT generation in both SPEED and ADDRESS SPACE that will be needed for the next generation of applications.
If 64bits is in fact 'often slower' as one person quoted, then why are the GPUs on EVERY high speed video card now running with 64,128 and higher bits to 'get the performance' that was NOT possible on 32bit GPUs.
Why? Because Microsoft had to come up with a scheme that would let them link 64-bit applications with 32-bit libraries without having to port all the existing libraries to 32-bit
This isn't even factually correct. Are you sure you have developed for Windows or did you find this on some great Google search to prove your point.
Most windows applications port to the 64bit version with little more than a simple recompile, and with integer declarations that do need to be expanded, does the concept of search and replace ring a bell?
Additionally, you assert this is because of the need for 64bit applications to reference 32bit DLLs/libraries. Again this is NOT true. A 64bit application on Windows has NO direct mechanism for calling 32bit code/DLLs/Libraries. PERIOD.
Go read the Free Developers SDK at least.
and unlike UNIX Windows doesn't use a clean client-server display interface.
'clean'? are you splitting hairs of comparing RDP to the X-Windows Protocol? You seem to think this is a 'feature' of UNIX's GUI subsystem, but if you really know GUI developers that are trying to get performance out of XWindows they will sit you down and explain why a network graphical 'protocol' is truly not that best Graphical subsystem on a desktop system.
It is great if 10 users are accessing the computer via the network, but on a single desktop with a single user, it is a big bottleneck in not only RAM but in performance on *nix environments.
Now we could argue if Xwindows or RDP was the better remote GUI protocol, and I could honestly argue either way, as both have benefits over the other, but if you are going to tell everyone that this makes any *nix better or faster on the desktop, you will probably have a few GUI application developers stepping in to dismiss your comments.
There IS a reason why OSX's Quartz does NOT sit on top of XWindows and does not itself have a 'clean' client-server display interface. They wanted performance for their users.
I'm going to skip your rant about the PPC/x86 debate. I find a lot of good things about the PPC architecture, however it is a bit sad that Microsoft will be popping out a tri-core 3.2ghz gaming console running Windows64 on it before Apple even offers a CPU with that speed or configuration, or has a 64bit version of OSX running on it. Sad that on Apple's own turf, Microsoft could pull off what so far Apple does not even have planned to do, especially with a full 64bit version of OSX. (And again this is a testament to NT architecture that not only has it in the past, but still can run very easy on PPC systems and even do so in the newer 64bit modes and having a fully ported DirectX subsystem for this architecture as well for gaming performance.
For people who actually need it, Tiger is a BETTER 64-bit environment than Windows 2003/XP, because their existing code that they've bee
So you're comparing two completely different architectures, one of which was register starved until the move to 64bit, and the other which had no such problems, and claiming that moving to a fully 64bit architecture made a difference for one, so it must for the other?
Ok, I will give you ONE chance to re-read my post.
As you will note I did not say that by switching to the 64bit CPU servers we gained the performance, we were already running the 64bit hardware, but with the 32bit version of Windows.
It was when we moved the server OSes to the 64bit versions of Windows we got the dramatic performance increase and we were able to reduce the servers in use to 1/5.
So if your argument that the only benefit in the 64bit x86 platform is from the extra registers, then we would of had the dramatic performance increase when we put in the 64bit hardware even with the 32bit version of Windows. Instead the performance increase came when we changed the OS to the 64bit version.
I think you should do a little research on the various 64bit architectures and OSes and the benefits before totally looking like a fool.
Or for god's sake, at least fully read and comprehend the post you are responding to next time.
The point is, making OSX fully 64-bit would actually slow it down because things that work just fine with 32 bits would suddenly have to move around twice as much data and take twice as much memory, most of which would be unused. The speedup whem moving to a 64bit version of Windows is simply because the ia32 architecture is so bloody awful that AMD had nowhere to go but up with x86-64.
Thsi is just not true. The only way an 64bit of OSX would be a slower is if the Apple truly kludged it up, and I don't think they would.
x86 32bit architecture was OLD, but the current implementations of the 32bit x86 design are quite advanced and are technically can complete neck and neck the teh x86-64 version of themselves. It is just when dealing with LARGE amounts of data that is either being computed or pushed in and out, the extra lanes in a 64bits really help out for performance.
There used to be an old argument that an optimized 16bit application would run faster than a 32bit application as well. This is only partially true, and only holds true then you are not trying to move large amounts of information or do high end calcualtion where the utilization of he extra bits pay off.
The same is true of 32bit vs 64bits. There are exmaples where a 32bit applicaiton could be technically faster, but there are more expamples where the utilization of the 64bit nature of the CPU pays off not only in performance but in the long term with the increasing demand for RAM we are starting to face.
Apple will someday and will be smart to do so someday to move OSX to a pure 64bit OS. It may be later than Microsoft and other OS vendors, but they will eventually have to do it to keep up.
But I understand, it's either to flame Apple and claim they don't have a 64-bit OS. Because we all know that 64-bit is
1) twice as much as 32-bit (2^32*2=2^64, welcome to maths!)
2) always faster than 32-bit (yeah! exactly for, like, word processing! dude, i can type in 64-bit!)
3) a long-needed change especially for desktop computers (i've always wanted to calculate the newest genom information at home... finally I can do it)
Cute...
In my world though, moving our servers to the 64bit version of Windows took our server requirments to less than 1/5 the equivalent systems when running the 32bit version of Windows 2003 Server.
So maybe that fuzzy math stuff does work in the real world.
(I hope to God your next post isn't a boring rant on explaining 40bit/48bit memory addressing of 64bit CPUs, the history of the Alpha 64bit CPU, or any other nerd item that 90% of everyone here should know.)
Tiger isn't' even a real 64bit OS"
Yes it is, and I am sick and tired of this debate. If you don't understand the 32/64 bit nature of PowerPC, then please, for the love of $DEITY, shut up.
Ok prove it.. Show the entire world where there are two sets of binaries of OSX Tiger. One set that runs on the G5 and truly is a 64bit OS and show us the second set that is a 32bit OS and runs on the G4 and G3.
You won't... Period. If Tiger was 64bit, #1 all the interface elements of OSX would run in 64bit mode, #2 All the drivers and Darwin layer of OSX would run in 64bit mode - and NEITHER DO.
And as I demonstrated before the easiest way to check for this, can Tiger run on a G3 processor? If it can USING THE SAME COMPILED BINARIES, then it is NOT NOR COULD BE a 64bit OS.
(Tiger does allow some applications to run in 64bit mode, but by doing so the applications are severed from working with many of the GUI and other libraries available to developers - mean virtually the Application would have to broken up to have a 32bit interface and only use the 64bit code for intensive calculations.
On the other hand you have *nix like Linux and even WindowsXP/2003 which have been doing a FULL and TRUE 64bit OS for many years...
shove calculation intensive opporations
Ok, first off I said "Shove calculation intensive operations"
Secondly I a pretty good understanding of how a CPU and its relation to its bit capability are relavent. 64bit CPUS are not limited to just pushing data in lanes of 64bits, but ALSO computing data in 64bit segments. It is not all about databus - trust me.
As for the reason Apple tries to convince its customers that moving to 64bits is NOT necessary is only because their OS is not 64bits. It is just that plain and simple.
If 64bits were NOT important, then why were we all blasted with Apple's Marketing of the 'first 64bit desktop computer' - which wasn't even true. But yet by the G5 having 64bits (even though not used) seemed to be pretty important to Apple if they went out of their way to market it and tell everyone it had 64bits.
So maybe you can tell us truly where Apple stand on the benefits of 64bit computing... 1) Good cause the G5 has 64bits... 2) Bad cause OSX doesn't support it?
Get past the marketing con artists for 5 seconds and maybe you can see outside your cave.
It took apple many years to move from the 68000 line to the PPC.
This is true, it took them 8 years longer than anticipated, which is really quite sad, as System 9 software didn't even fully utilize the PPC capabilities, wasting a lot of the power of the late 90s PPC based Macs.
I guess this follows Apple style, as they are wasting performance and capabilities in the G5 PPC today as well.
However I still assume that this "3 core" cpu can't be 3 full fledged G5 cpu's running at 3ghz. If it were, I am sure Apple would be, or will be, using it.
Heck they dont even have a single G5 ppc running at 3Ghz yet, do they?
Wrong again... In fact it will be at least the G5 technology as the game testing and development software for the XBox 360 was designed for G5 Macs.
It you would read the posts here, you would see why it IS 3GHz and Tri-Core, when Apple is still using older PPC G5 technologies.
Microsoft licensed the technologies and were able to further engineer the CPU themselves, that is why they own the rights to the technology and produce the CPUs at any fabrication plant, and not only produce them directly through the IBM route as Apple has to do.
The Xbox 360 will not even have a hard drive this time.
Actually it does, and it is removeable so you can take your saved games, video, and music with you.