The first bootable release of McOS Re 0.4.1 is available. McOS Re is a project dedicated to the re-creation of the Traditional Mac OS. The first goal is to be source- and binary-compatible with native PowerPC code for Mac OS.
McOS Re 0.4.1 Released
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I’d love to see a recreation of the traditional MacOS. Especially given that Apple is turning the legendary MacOS UI into a farce in OS X, I’d like to see the traditional one preserved.
Not that I in any way object to an OSS version of the classic MacOS, but the site seems a little light on details. Since I don’t have a mac to test this thing on, can someone maybe tell me, in some detail, the features of this thing?
You could try to run it on PearPC.
how is OS X a farce? this is not a flame, just someone who has only seens screenshots of OS X, never used one…
And people are having a go at the Haiku Project. At Least Haiku is a more original name than McOS.
Bring on the lawsuits, from both Apple Computer and McDonalds.
Reimplementing an existing OS for hardware which it already runs on in assembly language. It sounds like a non-starter. If you wanted to run the OS on newer hardware, it would be easier to patch the existing it.
@neoTheCat
Many Mac OS users are bitter about the transition to Mac OS X, and can find many flaws with it which any normal person wouldn’t notice. I would be one of those people, except I figure that Mac OS went so far downhill by the time Mac OS 9 was released, it didn’t matter. That, and Mac OS X gave me the Unix shell.
@Anonymous: re: Patching 9 to stay current
With a closed System and Finder, that would be a pain. If this were an open re-implementation of the System, thinking System=kernel, having access to Classic MacOS programs with a system that employs some modern features (memory management, multi-threading, etc) could be cool. As Linux fits in with GNU, so could this GPL’ed McOS
However, without being able to read a bunch about it, or the IDE it’s developed with, it’s hard to say if that’s a goal.
NOW! If this were something portable, say with a JIT recompiler a’la BasiliskII, running a *real* MacOS natively on speedy-as-all x86/64 hardware would be boss for sure.
| T () // /. T ()
The only reason I switched to a Mac is because of it’s unix based operating system. If they still had that OS9 crap many people who switched to Mac wouldn’t even consider them.
Sometimes a user just wants a hobbyist OS with which to do hobbyist type tasks. So more power to this guy.
But having never used MacOS Classic, let me ask this: can you still get software? Is there any freeware or shareware available? Are old, unsupported titles available on eBay for purchase, or have any been released into the public domain?
Just wondering.
I’d like to second your comment. MacOS 9 and below were jokes in my opinion.
@It misses the point …
OS 9 wont boot a current Mac hardware. A little know issue is that the Mac’s used to have 4MB of ROM with all the hardware drivers needed to run the system.
@itomato
Speedy-as-all x86/64 hardare would be a dog running any Mac OS program. A JIT compiler would have to create the register space on the stack (larger memory overhead) or keep all register data in global memory.
PPC: mul r1,r2,r3
x86: mov eax,[r2]
mov ebx,[r3]
mul ebx
mov [r1],eax
The above is a simple multiply operation. The x86 doesn’t have enough registers space to store all 64 general registers of the PPC. You would also have a hard time optimizing the code for the x86 since most math functions can only be done against the EAX register.
Subject says it all.
While I’m pretty sure the original poster meant porting McOS to the x86 platform, I would like to point out that MacOS 8.1 runs faster emulated on a modest current PC then it did on its original hardware. Assuming that the host system is set up properly .
MacOS X, however is very slow in emulation. At least it was at vers .2 of PearPC.
Can you cut out the “crap” and “jokes” bit about Mac OS 9. While you may hate it, other people may like it. And there are good reasons for that. This isn’t to degrade your own choice, but it is to say that different people measure OSes in different ways.
On the availability of Mac OS 9 software: it still exists and people are still producing it. The last time I checked, over half of Mac users were using Mac OS 9 or earlier. This is nothing to sneeze at given that OS X was release over three years ago. In all likelyhood, a lot of these people will never upgrade to Mac OS X because Mac OS X is “just another Unix” or “just another Windows” in many ways. That is to say that Mac OS X is developed for the developer rather than the end user, that Mac OS X doesn’t offer the simplicity and easy maintainance of Mac OS 9 (and earlier), and that Mac OS X is as buggy as heck (fortunately most of those bugs are trivial, but they are still annoying). If you look at Mac OS X good and hard and decide that you are not interested in the Unix bits, then you are probably going to be happier with Windows.
Going back to McOS: if you want to develop a hobby OS, then you should develop something unique. If you want to clone something, then you should do it properly. (For example: write it in a more portable language.)
On LIDE: there is plenty of information on the company website. It is developed by somebody else, and is offered up for free these days. Alas, it is to frail to operate under OS X and it sounds like it won’t run on a 68k so I cannot try it to report on it.
“Speedy-as-all x86/64 hardare would be a dog running any Mac OS program. A JIT compiler would have to create the register space on the stack (larger memory overhead) or keep all register data in global memory.”
The memory overhead of emulating all integer and floating-point register on the stack would be 4x2x32 = 256bytes. It is nothing. But a JIT implemented in that way would be slow anyway…
Your comment about global memory I really can’t understand, is it something to do with structured programming and evils about global variables? In a JIT compiler global variables are no problem.
“PPC: mul r1,r2,r3
x86: mov eax,[r2]
mov ebx,[r3]
mul ebx
mov [r1],eax”
More like:
(mapping: r2=ebx, r3=ecx)
(if both r2 and r3 must be preserved)
mov eax, ebx
imul eax,ecx ; JIT ‘renames’ r1 to eax
(if the JIT decides r2 will not be used anymore)
imul ebx, ecx ; JIT ‘renames’ r1 to ebx
(if the JIT decides r3 will not be used anymore)
imul ecx, ebx ; JIT ‘renames’ r1 to ecx
“The above is a simple multiply operation. The x86 doesn’t have enough registers space to store all 64 general registers of the PPC.”
That is true but isn’t as limiting as it can appear, it makes writing JIT compilers a PITA and loses some performance but thats all.
“You would also have a hard time optimizing the code for the x86 since most math functions can only be done against the EAX register.”
What? No. The only real limitations are that some mul/div instructions uses eax/edx for a specified purpose but that is it.
The real problems in PPC emulation are emulation of the virtual memory subsystem and the Altivec instructions. And those are both _very_ difficult to get fast enough :/
I would think that the Toolbox ROM is fairly well known to Mac users, particularly Mac developers. Saying that it had the drivers you need is a bit of an overstatement anyway. The so-called New World Macs didn’t have a toolbox ROM. That was loaded from disk. (About Mac OS 8.6 to Mac OS 9.x.) Also, the ROM wasn’t sufficient. The OS had to be aware of and tailed to your machine. This was true before 8.6, and it is true after 8.6.
OS 9 was ok, but heck I wouldn’t of stayed with the mac if Apple continued to use it. OS X is Applized, but it was first a PC OS(next) if I’m thinking right and a good one for the time!
MacOS X was a step with regards to the underlying technology, but I previous versions of MacOS to be much more elegant and consistent. The stuff Apple is doing with OS X (eg: changing around UI elements, mixing themes, generally overusing eye-candy) didn’t happen with MacOS classic. The guts of the OS were trash, of course, but the UI was nifty.
IIRC, NeXT computers were initially using Motorola MC68k chips, just like Macs of the time.
It was only when hardware sales were getting tough that they ported the code to other platforms, and sold the OS by itself for x86, etc.
if a bit dated – though it might well have taken the shock of losing on so many markets simultaneously, to wake Apple up to the fact its previous policies on IT were counterproductive. Someone cloned the MacOS back in the early 1990s, but I never heard much about it and it seems to have folded and faded.
It sounds a lot like AROS and other proprietary-OS cloning efforts – So go for it! The more that is out there with freely-available source code, the less we have to worry about being corralled like wild horses and branded by the robber barons.
…still has got its place IMHO. Sure, it’s an oldfashioned, outdated OS, but it runs flawless here. The big plus for many is the installed base of tools (software (design-tools), hardware (scanners, etc)). I can work with OS9 all the time, can get all my jobs done (design).
If i would switch 100% to OS-X i had to spend mucho $$$ for x-ready software & hardware to get the same results in OS9, nothing more. Oh, the speed of workflow would be slower, bc. X is more demanding on the same hardware. I can’t stand the laggy finder, btw.
The real downside of 9 is that it’s dead. so hardly any new drivers, software support.
If these developers want to recreate and improve the classic Mac OS, wouldn’t it make more sense to use Copland as a base to start from?
That was supposed to be a refined version of the classic Mac OS, but with modern features like pre-emptive multitasking etc
In my opinion, that would really be furthering the classic Mac OS, because it was what Apple originally planned to do.
I was a big copland fan, watched and read all the articles about it (nice UI!) – too bad that apple set unrealistic goals and dealines and killed it. I would like to see someone take on copland, now THAT would be a beauty! – Not just a re-encarnated OS, an OS that was sort of dead and now is reborn
I honestly don’t get some of you old Apple users. If it wasn’t for Mac OS X, I would *never* have even looked at a Mac. OS 9 and earlier are jokes. I loaded classic to have a look at it on here, and it was as ugly as win95. And what’s with criticizing the Mac OS X UI? It’s _easily_ the most easiest to use operating system I’ve ever tried (note I use Linux on my desktop, and used to be a Windows user).
You old Apple users are a fussy bunch. Long live OS X !
I like the idea of recreating the old Mac UI (never having used one), but why in Assembly, and why not just layered on the Mac OS X WindowServer, or at least on the Linux or Darwin kernel?
The project will face serious troubles if it wants to recreate a whole system without a solid multitasking base…
Anyway, I’m happy that my iBook is Mach/Unix based.
If you have to ask why you are not the target audience
Anyways, OS9 wasn’t that bad really. Heck 8.1 wasnt that bad. Ran great on my old low spec’d 8100/100. I’m also one of the people that really doesn’t like the OSX gui. it has to much eye candy. To bubbly for me. OS9/8 can look pretty good with theming. I had mine looking like beos
Dont get me wrong, I run panther, I love it and I would never switch back to OS 9…well…if they gave me a million dollars invested in a 2% interest baring account I would, but in any other case no.
i am still interested in the “what could have been” aspect of this OS because I am an OS junky (I have quite an extensive selection), and this OS, as far as I understand, wants to be a “what if OS 9 had continued, which better memory management, more feature, etc”.
Personally I would love to see a “what if…” of Copland + OpenDOC + PowerTalk + other dead apple technologies.
Not to mention that I would love to have someone re-invent the NewtonOS with binary/source compatibility with the original one but overcome the original OS’s shortocomings…oh oh! and TalOS as well! and the star-trek program by apple….man I want too much
Where?
I think this project is a “dead man walking”.
I expect an incoming legal action from Apple.
Apple is NOT like Amiga.
“Not to mention that I would love to have someone re-invent the NewtonOS with binary/source compatibility with the original one”
Uhm, yes. I would love this, too.
PS about OS8/9: I used Mac OS 8/9 since december ’98 to last april, than I used Mac OS 10.3.x in these months and I want to say I’m very happy about Panther. I like it more than OS9. And, belive me, I was a great fan of OS9.
Well yes, this was the reason I have finally switched to Mac, the unix side of the O/S turned me into a great Mac fan.
I agree with others posters too….getting software for previous O/S, will get harder and harder, whereas with the new OSX, one can have best of breeds, OS 10 which is a great GUI and OS X which brings the wealth of Unix/Linux software.
I agree entirely. I acquired a Powermac and the first thing I did was wipe the hard-drive. After Windows 98 and Slackware Linux, OS 9 looked and felt not right. I went out and bought OS X, never been happier. Not only have I now got a O/S that I can easily install, but my Linux knowledge learned over the past 1 year is not wasted. For me it is now a pleasure to work in either the GUI or Terminal.
I am happy with Panther.
Being interested in Photography I got some great downloaded GUI software from Yellow Mug Software (EasyCrop and EasyFrame and EasyBatchPhoto) make short work of digital imaging tasks I perform and GIMP 2.0 handles all other graphical jobs.
mySQL works perfectly out of the box and is in my opinion is a brilliant piece of software.