Exactly, and can we quit it with the uniformative and insulting posts, Sailor? As to who makes the G4, its still Motorola and there’s tons of info here:
Derelth sez:
and jti sez:
The older Mac, called “68k”, ran on Motorola 68030 and 68040 processors.
The PowerPC chip was made by Motorola but that is all I remember about the PowerPC chips.
The G3 was made initially by Motorola and later IBM joined the production team.
And the G4 were first made by Motorola but since they could not keep up with demand IBM joined in the production.
Some of you may remember the fiasco when Apple first introduced the G4 and said the processor speeds would be IIRC, 400, 450, and 500. Unfortunetly Motorola could not manufacture the G4 500 fast enough for Apple, so Apple had to lower the line to 350, 400, and 450. But what pissed everyone off was that the people who had ordered and paid for a 500 were given a 450 at no discount. Eventually, through customer complaints, Apple decided to honor the orders for the 500. Then Apple made the agreement with IBM to produce the G4 chips and that was when Apple introduced the G4 500 to the market.
-N
Hey, thanks. I didn’t know Motorola still made ICs. I viewed it as a two-way radio and cell-phone maker.
-
Apple once had a little research project going to evaluate the wisdom and lucrativity of porting MacOS to the Intel-standard hardware platform. This research project was code-named “Star Trek”. A few fragmentary good things came out of it (most notably QuickTime for Windows) but it never really went very far. In order for it to have succeeded, companies like Macromedia and Quark would have had to compile and release MacOS for Intel versions of their products as well as Windows and MacOS for Apple Mac versions, otherwise no one would have any incentive to buy and run MacOS for Intel. This becomes a chicken-versus-egg problem because Macromedia and Quark aren’t likely to dump development money into a MacOS for Intel version of their product if no one is using MacOS for Intel. Then, of course, there was the little matter of Apple having a manufacturing process for making traditional Macintoshes but no experience manufacturing Intel-standard PCs, and would they begin manufacturing BOTH platforms if they released and supported the other version of MacOS? So it was spiked.
-
MacOS X is a derivative of OpenStep, which ran on Intel processors. OpenStep, in turn, is a derivative of NextStep, which ran on Motorola 68K processors just as MacOS historically has done. It is far more portable than the classic MacOS; if Apple had sufficient reason to do so, they could release versions that would run on Intel x86 chips, Alpha chips, MIPS chips, or whatever. For quite some time during the initial development of MacOS X (back when it was code-named “Rhapsody”), Apple had actual compiled versions running on Intel hardware. However, the decision was made not to put much subsequent focus there, and there may never be a full operating system called MacOS X that runs on Intel hardware. However, there may instead be a set of API’s called “Yellow Box” that can be installed as part of a program’s installation routine that allow programmers to write once, compile twice (with minimal tweaking), and deploy on MacOS X and Windows / Windows2K platforms. This “yellow box” is a set of toolbox routines that support code written in Objective C. The hype on this is that it would be much easier to write a program that would run in yellow box for Wintel and yellow box for MacOS X than to write an application for MacOS Carbon (i.e., in C++) then port it to run as a Windows app, or vice versa.
-
FYI, the original Macintosh ran on the Motorola 68000 CPU at 8 MHz. The 68000 was a mixed-bag of a processor; imagine an 8086 (an XT processor) capable of some 32 bit and many 24 bit operations and running a GUI. A few subsequent Macs ran on the faster 68020 processor (like a 286 sort of), a longer string of Macs used the 68030 processor (which, like the 386 on the other side of the fence, supported the paging of memory and could support a fully 32 bit OS; unlike the Wintel world, of course, it was matched to a fully 32 bit GUI OS poised to make full use of it). Later Macs used the 68040, the last of the 68000 Motorola series processors to be used by the Mac. The Intel 386 and 486 processors tended to be faster–in any given year rather than cycle for cycle, since the Motorola processors were faster per cycle but you would see Intel processors running significantly faster CPU speeds in similar boxes. The Macintosh platform switched to the PowerPC platform with the PPC 601 chip being the first generation PPC chip. These chips competed with the first generation of Pentium chips. PowerPC chips were theoretically manufactured by both IBM and Motorola, but in practice most 601s seem to have been of IBM manufacture. The next generation of PPC chips were 603s and 604s and their improved variations the 603e and 604e. These competed against the Pentium Pro and Pentium II chips from Intel and Cyrix and AMD “686” chips. Motorola and IBM both made these and both make current PPC chips. By this time the PPC chips were pulling away from the Intel chips through most of their respective production cycles, but Apple was busy doing a business nose-dive so no one cared. The G3 chip is really the PPC 750 chip; it is called G3 because it is the 3rd generation PPC chip. It is a variation on the 603e chip but is blazing fast in comparison. The newer G4 chip, often described as simply a G3 with AltiVec hardware added on, is really a variation on the G4 chip; it is less streamlined and more robust for symmetric multiprocessing than the G3 chip.
AHunter3 sez:
Where do you get this information? I was at the Macworld SF when Steve announced the new OS and he said that it was a derived from Free BSD. And all the magazines I have read have said that it was derived from Free BSD. But most convincing of all is this quote from Apple’s web site:
**“Mac OS X supports POSIX file system semantics and NFS file sharing, as well as standard services like telnet and FTP, allowing easy operability with UNIX systems and applications.”
“The system’s kernel, which does the heavy lifting to support all those rich applications, is based on Mach 3.0 from Carnegie-Mellon University and FreeBSD 3.2 (derived from the University of California at Berkeley’s BSD 4.4-Lite), the most highly regarded core technologies from two of the most widely acclaimed OS projects of the modern era. We also took the famous Apache web server—which runs over half the websites on the Internet—and made it friendly enough to use on your desktop for personal file sharing.”**
Now the kernel, Mach, was taken from Caregie-Mellon Univ. several years ago and Apple introduced this kernel to an Apple supported version of Linux called MKLinux (Micro Kernel Linux), which I run on two older PPC Macs. Then Apple ditched the MKLinux venture and took the information it learned from that and incorporated Mach into FreeBSD. I have never heard anyone mention OpenStep.
I am also kind of curious about where you recieved your information about “Star Trek.” Like everyone else, I have heard the rumors but I have never talked to anyone who could verify this rumor. Apple is one of the most closed-mouth companies I know of, so I am curious because the way you talk about it, it sounds like you worked for Apple at the time, or were a personal friend of someone high up. If you have just heard it as a rumor, like everyone else has, then you should say so in your post so people are not decieved. Rumors should not be taken as fact until they are proven. I have heard a hundred and one other rumors about Apple, all of then plausible, but none of the verifiable, i.e a color Palm Pilot like device, a multi-processor G4 (which I do believe), bringing back G3’s at speeds of 1 Ghz, etc… I am really not trying to cut you down, say you are lying or anything like that but I think we need to make sure we know what are facts and what are rumors.
-N
Apple Computer Inc. bought NeXT Inc., acquiring both OpenStep and Steve Jobs. The result of the Steve Jobs thing, I guess you know. OpenStep is indeed based on BSD Unix. OpenStep ran on the Intel hardware platform. Before NeXT was releasing OpenStep for the Intel platform, they were manufacturing their own computers, little cubes that used the Motorola 68030 or 68040 processor and ran NeXTStep, a special modification of BSD Unix. Steve Jobs once said NeXT would either be the last successful launch of a proprietary hardware platform or the first wholesale failure of a good proprietary hardware platform. The latter happened and the little NeXT cube disappeared and the software OS was ported to the Intel platform, where it was known as OpenStep.
After its acquisition by Apple (and the reemergence of Steve Jobs as Apple’s honcho), OpenStep was ported to the PowerPC hardware platform. The GUI has elements of both OpenStep and the classic Macintosh OS in addition to the new GUI elements developed explicitly for Mac OS X called “Aqua”.
OK, now I get it. I knew about NeXT and Steve et al. But what I did not know was that OpenStep was based on BSD Linux. So OS X is based on OpenStep and OpenStep is based on Free BSD. It is kind of interesting that you do not hear anyone talk about the OpenStep part of the equation. And it does sound kind of…scandalous?..that Apple buys NeXT and then Steve becomes the CEO of Apple, and then Apple decides to use the OS that Steve developed while at OpenStep. So do you work for Apple or what? This seems like a lot more information than the normal layman would know. I swear that I know too much about Apple with all these ridiculous pieces of information in my head that won’t help me IRL.
-N
Well…the part about the code name “Star Trek” is strictly the repetition of rumor. The part about why Apple didn’t go forward with the project is mainly my own surmise (it strikes me as “common sense”). The part that is other than rumor or conjecture comes from a fateful ride on the Long Island Rail Road a few years back in which the fellow in the seat behind me was trying to explain to his companion why certain features of QuickDraw acceleration made it a really cool feature, and I realized he was a MacOS programmer from Apple; after his companion got off the train, I asked a shy question and found him to be enthusiastic and talkative. He was part of the Copland team (“You’ll like Copland, I promise you that. The underlying kernel is more stable than anything else that’s out there, even in alpha.”). He said they had played with the idea of porting the OS to Intel standard hardware but it had become a programmer’s hobby and no one thought it would ever reach release for reasons I detailed above. He also said the rest of the PC architecture (i.e., everything besides the Intel x86 CPU chip) was an old and hoary mess with its ISA bus and interrupts and other awkward chipsets with vague standards and a plethora of implementations among various PC manufacturers and MacOS for Intel would have to support that mess as well, of course.
Maxwell asks:
As has been said, there are many technical reasons why this would not be a reasonable thing to do, but the main reason is a financial one. Apple makes most of it’s money (today) from it’s hardware. If the Mac OS were available on the Intel platform, that would do nothing to forward Apple’s hardware oriented business model. In fact, it might be a detriment. Some users might opt for an Intel box instead of an Apple box, thinking that if the OS is the same, they’re getting a cheaper Mac.
Chronos wrote:
You probably heard that the GUI (the window manager) will run on any unix platform. The GUI is not the same as the OS - it’s just the look and feel - merely the ‘shell’.
I haven’t heard that Intel emulation was going to be built in, but the Mac has been able to run Windows binaries for several years using VirtualPC or SoftWindows.
jti wrote:
While that is true, it’s a bit misleading. The main reason that the MacOS is less susceptible is that there are design features that make it more difficult to develop viruses and make it much simpler to detect viruses.
Derleth wrote:
You’re not doing it right… I used a 3400c running OS 7 for nearly four months straight without a crash or freeze. It was a laptop and I used it nearly everyday. What was even more amazing was that I was constantly trying out new software and experimenting. I finally realized how long it had gone without a reboot and decided to do it manually to clear out all of the cobwebs. My G3 tower running OS8.5 has only crashed 3 times in the last year - all three times were a result of trying out some new shareware. My PCs no where near approach this level of stability.
As for underwhelming marketshare, you seem to be making the mistake of comparing Apple to the entire Wintel market. If you compare Apple to Dell, or HP, or IBM, or Compaq or Toshiba or Acer or Gateway or Hitachi or Sony or NEC or Packard Bell or Fujitsu or any of the other hundreds of Wintel manufacturers individually – then I think you’ll see that Apple’s marketshare is not so “underwhelming”.
Max wrote:
There are other reasons for keeping part of the OS fixed in ROM. It offers better protection for Apple’s IP (intellectual property), it adds to the OS stability and protection from virus attacks, and it keeps the user from having to waste a lot of RAM on software that’s not going to change (short list - I can name more).
Actually, this is not a function of having the mouse routines in ROM - it’s a function of having the mouse routines interrupt driven. Same goes for the keyboard. In the Wintel world the mouse and keyboard are serviced through polling.
TheNerd wrote:
A very good observation. Why do you think the Windows OS is so ‘fat’ and slow and buggy? It’s because it has to work on every bastard implementation of the hardware platform… Well, that and the fact that Microsoft doesn’t really care about quality…
It’s interesting, though, that the very reason you refuse to buy a Mac is precisely the reason I won’t spend my own money on anything else.
handy wrote:
These emulators will only run 68K emulation and they require the user to obtain a copy of a Mac ROM. Performance is horrible. A 300MHz Pentium II running 68K emulation is lucky to achieve the performance of a comparable 20MHz 68K Mac. Contrast this to the opposite situation A 300MHz Mac running VirtualPC can compete quite well in head-to-head benchmarks against a 300MHz Pentium II (a benchmark I’ve tested personally).
For the record, Motorola makes both the older 68K platform chips and the newer PowerPC platform chips. IBM co-developed the PowerPC technology (I think the architecture is entirely Motorola’s, but IBM helped develop the manufacturing process). IBM, from time-to-time has been called on to help with the manufacture of the PowerPC, but Motorola is the primary supplier. And for the record, Motorola still makes a lot of chips. You’d probably be amazed at the number of Motorola chips you use from day-to-day.
Maxwell wrote:
However, the point is that the Macintosh platform is both hardware and software, with most of the revenue coming from the hardware. Apple has no incentive to spread the software without the hardware. Now you can question Apple’s business model if you wish, but I suggest you take a look at their financial history first.
Strider wrote:
and 68000, and 68010, and 68020… just picking nits.
AHunter3 wrote:
I’m not sure what you mean by “mixed-bag”? The 68000 is a full 32 bit processor internally with a 16 bit external data bus. This means that if you need to do 32 bit operations, you need two fetches per datum. However, 32 bit instructions are not always required so the 68000 supports 16 bit and 8 bit instructions, as well. This is simply a compromise that is mostly driven by the packaging and system costs - fewer pins equals lower cost for the processor, the motherboard, the memory, etc… The 68020 supports the full 32 bit adress and data busses. The 68000 was not the first or only micro to make this sort of compromise, in fact, the 8088 did the same thing with an internal 16 bit architecture and an external 8 bit bus. The Intel architecture didn’t see a true 32 bit processor until the 80386.
I am a new Mac person, (still use my PC) and I have been reading up on All Things Mac lately. I remember hearing about how they had the Mac OS running off an Intel Chip a while back. The reason I read (or heard, can’t remember) that the idea was shelved was because all the other computer companies that provided PCs (Dell, Gateway, etc.) already had themselves tied up with Windows, had some sort of “contract” or agreement to use only Windows. So no one would want to take on the Mac OS. Sort of like it has been slow going to get desktop computers available, out of the box, running Linux. Now, I’m sure this rumor/whatever I heard was just that. I think I read it in some Mac magazine, or on a Mac mailing list. But there might be a little something to it.
On a side note - I have two Macs now, and a PC. An old 66 MHz PowerMac, a 266 MHz iMac (just got it! Woo Hoo!) and a 333 MHz AMD K6-2 PC. The PC crashes a lot. Blue Screens of Death, weirdness, crap. I get a lot of USB problems lately as well. My USB tablet and mouse are always getting lost on the PC, same with the USB CD-RW drive. Lots of hassle getting the USB to work again with it. On my Mac, (especially the iMac, before I upgraded to OS 8.6) I’ve gotten a few freezes, but usually when installing new software. However, the same USB CD-RW drive I use on the PC runs great on the iMac. Much more stable - never a USB problem. Also, the 266 MHz iMac is blazingly fast compared to the 333 MHz PC. Granted, the PC has a little less RAM (128 compared to 160 on the iMac) but the speed is dramatically faster on the iMac, even though it has a lower MHz speed (266 compared to the PC’s 333.)
Another side note (I have plenty of those) I ran a comparison of Photoshop 4 on the old Mac and the PC. I timed how long it took to open an identical 19 MB file on both computers. A 66 MHz Mac w/ 40 megs of RAM compared to a 333 MHz PC with 128 megs of RAM. And the little Mac took about 3 times the amount of time to open the file compared to the PC. Which, if you ask me, is odd. Using my pathetic math skills, I am guessing that the 66 MHz Mac should have taken at least 5+ times the amount of time to open the file, probably a lot more than 5 times, considering the RAM difference. I mean, it is a SIXTY-SIX MHz computer, compared to a 333 MHz computer! And only 40 megs compared to 128 megs! My point is - you get more power with a Mac than you do with a PC, even though they may have the same MHz speed. (Especially with Photoshop.)
The mhz of the chip is only one factor of the speed of your computer. Another factor, more significant as the chips are gettting faster, is the mhz of the motherboard. That is where AMD shines over Intel.
There is a limit of speed of which a computer can go depending motherboard it is on. A 333 mhz chip on a 66mhz-ATX motherboard will run approximately 33% slower than that same chip on the 100-mhz board.
I’m pretty sure my PC and iMac both are running with a 66 MHz bus speed, if that’s what you mean… And yet the iMac is definitely faster than the PC’s AMD chip. (But I am glad to know that AMD “shines” - I specifically wanted it instead of Intel.)
Yosemitebabe,
Some chips and chip architectures are quite a bit faster than other chips running at the same MHz. There are a lot of reasons this could be so–for example, there’s an article at
http://forum.arstechnica.com/cpu/1q00/g4vsk7/g4vsk7-1.html
explaining how and why the AMD Athlon chip has a lot of its architecture devoted to preprocessing of the x86 code; a competing CPU chip that did not have these preprocessing features could run at the same clock speed but perform like a turtle on tranquilizers.
Anyway, megahertz for megahertz, the PowerPC chip architecture gets a lot better mileage out of its advertised clock speed than x86 Intel chips do. A PPC chip of the 601 vintage (like the 66 you mentioned) gets, in general, the performance of a Pentium of 1.5 time the MHz. A PPC chip of current vintage (G4) gets close to 2 times the performance of a comparably clocked Pentium III.
Meanwhile, there are things that each operating system seems to do better (more efficiently, therefore faster). Even with hard drives of comparable speed, it has been my experience that Windows computers launch applications faster, whereas Macs are so much faster at copying, moving, or deleting files and folders that a Mac emulating a PC under VirtualPC can do it faster than a PC of the same speed. Adobe Photoshop, although not optimized for the Mac at the expense of the PC (Adobe isn’t stupid), tends to utilize a lot of calls and routines that seem to work better and faster on the Mac. So, as you said, “especially with Photoshop”. (You should see what a G4 does with Photoshop…WHEWWW!!!)