Snow Leopard on Unsupported PowerPC Macs

[B S Magnet]

Nifty update re: QuickTime!

While moving QuickTime 7.6.4 components into place manually has more or less worked for builds 10A96 and 10A190, one feature which hasn’t worked is the ability to glance at rich media previews from Finder using QuickLook, such as previewing an mp3 or mp4 with the convenience of the space bar.

I have good news: manually moving QuickTime 7.7.0 for Leopard’s components into place, in lieu of QuickTime 7.6.4 (or the time-expired 7.6 version of QuickTime bundled with the development build of Snow Leopard), not only appears to work as QuickTime 7.6.4 did, but it also restores Finder’s QuickLook rich media functionality (including under “File > Info…”) for the first time on the SL-PPC builds!

In addition, one of the problems with QuickTime 7.6.4 was the inability to select and change the pull-down options under the “Export…” menu. With 7.7.0, these pull-downs are now accessible, and I ran a full export (A/V-passthrough) from a clip I pulled into QuickTime from PPCMC 7.2.3’s “Stream YouTube video in QuickTime”).

I have updated the WikiPost to reflect 7.7.0’s improvements over 7.6.4.

[EDIT: I’ve added two proof-of-performance pics.]

 

[armdn]

Hello guys! Just read out "wiki". I think in the end of all this journey, we have to write a big shell script (with archive of components), to patch whole system...

 

[B S Magnet]

Nifty update re: QuickTime!

While moving QuickTime 7.6.4 components into place manually has more or less worked for builds 10A96 and 10A190, one feature which hasn’t worked is the ability to glance at rich media previews from Finder using QuickLook, such as previewing an mp3 or mp4 with the convenience of the space bar.

I have good news: manually moving QuickTime 7.7.0 for Leopard’s components into place, in lieu of QuickTime 7.6.4 (or the time-expired 7.6 version of QuickTime bundled with the development build of Snow Leopard), not only appears to work as QuickTime 7.6.4 did, but it also restores Finder’s QuickLook rich media functionality (including under “File > Info…”) for the first time on the SL-PPC builds!

In addition, one of the problems with QuickTime 7.6.4 was the inability to select and change the pull-down options under the “Export…” menu. With 7.7.0, these pull-downs are now accessible, and I ran a full export (A/V-passthrough) from a clip I pulled into QuickTime from PPCMC 7.2.3’s “Stream YouTube video in QuickTime”).

I have updated the WikiPost to reflect 7.7.0’s improvements over 7.6.4.

[EDIT: I’ve added two proof-of-performance pics.]

Hot on the heels of finding that QuickTime 7.7.0 restores functionality to QuickLook and exporting QT media, I decided to try something with the iTunes problem — namely, that of iTunes 10.6.3 on 10A96 not launching.

OK. So, iTunes 10.6.3, whose minimum system requirements are OS X 10.5.8, was released in June 2012 — not only well after Snow Leopard, but also well after Lion. Priority for PowerPC functionality was probably lowest on the development team’s bucket list.

The first release of iTunes 10, however, dates to September 2010. Security fixes, bug fixes, and minor additions notwithstanding, there were also updates to Frameworks/PrivateFrameworks between 10.0.0 and 10.6.3.

The important milestones here, I think, happened with iTunes 10.3.1 (June 2011) and 10.4.0 (July 2011) — 10.3.1 being when 32-bit Lion support quietly began and 10.4.0 being when 64-bit Lion support started (released simultaneously with Lion’s launch).

However symbolic or unrelated to the actual supported operation of iTunes 10 on PowerPC this milestone was, one can guess this is when the legacy support ball for PowerPC functionality (and, for that matter, Rosetta) began to be dropped deliberately — though not necessarily in ways which an end-user of a G4 or G5 running OS X 10.5.x would have been aware of.

So… instead of testing iTunes 10.0.0, I chose to test a later version which folded in some initial bug fixes and minor updates to improve stability, but not so new that the version fell into Lion support. In other words, Something before 10.3.1.

This prompted me to try iTunes 10.2.2 (April 2011).

I am excited to share that basic iTunes 10 functionality works with 10A96! It still has the same graphics-related glitching issues on the play display window (same as iTunes 9.2.1), but iTunes 10.2.2 does work!

From here, I plan to test 10.3.0, and work my way up the minor version updates — 10.4, 10.5, etc. — until I can pinpoint when iTunes 10 finally broke with 10A96. Once that’s figured out, I can update the WikiPost.

As for the glitching window carrying over from iTunes 9.2.1, I don’t know how to troubleshoot fixing that, or whether it can be fixed. I think this may be related to some alpha-level code (like a Framework or PrivateFramework) in 10A96 and not the iTunes code base.

Onward!

[UPDATE: So far, all these 10.x versions of iTunes I’ve tested (10.2.2 through 10.4) will not quit cleanly. All hang into “not responding” following Cmd-Q and need to be force-quit. This could be related to problems with either saving the library XML file or the iTunes prefs file, but that isn’t yet clear. (Re-launching does show the playlist items from last session, so the hang’s cause remains unsolved.) BUT… if you’re following along, it appears that following minor changes to iTunes’ UI from 10.4 — notably, the traffic-light buttons are moved from vertical to horizontal — the play display window no longer shows glitches. The off-white window area, however, is missing. See third attachment.]

[SECOND UPDATE: I can confirm that iTunes will not launch on 10A96 beginning with iTunes 10.5.0. The final version to launch on Build 10A96 is iTunes 10.4.1. What changes with the 10.5.0 installation package are two additional components within the install pkg: CoreTypes.bundle (located in /S/L/CoreServices/) and CoreFP1.framework (located in /S/L/PrivateFrameworks/). Both of these replace the versions which installed with 10A96. It also appears that iTunes 10.5.0 — the first version of iTunes to support iOS 5.0/iPhone 4S — will not run without these components and it will not run with these components if the system isn’t, I’m deducing, at least OS X 10.5.8; OS X 10.6.0 Build 10A190; or higher (much like with iTunes 10.6.3). I could be wrong, of course. ¶ Further, when reviewing crash logs for iTunes 10.4.1 and below, three of the four threads linked with crashes are related to the dylib called libSystem.B.dylib. This dynamic library, in /usr/lib/, is installed with Build 10A96. I do not know whether this and/or other dylibs are newer revisions in 10A190 or even 10.5.8. But I can say the first thread, that of running iTunes itself, is what invokes the crash. See fourth attachment.]

Conclusion: iTunes 10.4.1 is as high as it goes for Build 10A96, but the dirty force-quitting of iTunes has yet to be solved.

LINKS

• All legacy iTunes versions between 7.3.0.54 and 12.8.2
• Release dates for iTunes 10.x revisions

 

[ChrisCharman]

Actually though, have any of you tried rebuilding Leopard Webkit? It obviously builds on 10.5 PPC, and Tobias said the code could be built for 10.6 Intel relatively easily, so it seems like 10.6 PPC ought to be doable as well.

You won't get the latest version of Webkit, obviously, but it would be a heck of a lot more up-to-date than what comes with Snow Leopard!

(I don't own any PPC Macs, I'm just watching from afar! )

Yes @vddrnnr has attempted this and the creator of LeopardWebkit responded earlier in the thread regarding the endeavour

 

[ChrisCharman]

Hello guys! Just read out "wiki". I think in the end of all this journey, we have to write a big shell script (with archive of components), to patch whole system...

I think if we can eventually patch the installer, much as @dosdude1 has done with more recent releases, that would be an ideal end goal. Probably we’ll end up with two separate viable builds. At least in my experience as time moves forward my build of 10A190 diverges increasingly away from 10A096 and 10.5 and depends more and more upon 10A432 sources which breaks compatibility with the earlier versions.

 

[ChrisCharman]

Hot on the heels of finding that QuickTime 7.7.0 restores functionality to QuickLook and exporting QT media, I decided to try something with the iTunes problem — namely, that of iTunes 10.6.3 on 10A96 not launching.

OK. So, iTunes 10.6.3, whose minimum system requirements are OS X 10.5.8, was released in June 2012 — not only well after Snow Leopard, but also well after Lion. Priority for PowerPC functionality was probably lowest on the development team’s bucket list.

The first release of iTunes 10, however, dates to September 2010. Security fixes, bug fixes, and minor additions notwithstanding, there were also updates to Frameworks/PrivateFrameworks between 10.0.0 and 10.6.3.

The important milestones here, I think, happened with iTunes 10.3.1 (June 2011) and 10.4.0 (July 2011) — 10.3.1 being when 32-bit Lion support quietly began and 10.4.0 being when 64-bit Lion support started (released simultaneously with Lion’s launch).

However symbolic or unrelated to the actual supported operation of iTunes 10 on PowerPC this milestone was, one can guess this is when the legacy support ball for PowerPC functionality (and, for that matter, Rosetta) began to be dropped deliberately — though not necessarily in ways which an end-user of a G4 or G5 running OS X 10.5.x would have been aware of.

So… instead of testing iTunes 10.0.0, I chose to test a later version which folded in some initial bug fixes and minor updates to improve stability, but not so new that the version fell into Lion support. In other words, Something before 10.3.1.

This prompted me to try iTunes 10.2.2 (April 2011).

I am excited to share that basic iTunes 10 functionality works with 10A96! It still has the same graphics-related glitching issues on the play display window (same as iTunes 9.2.1), but iTunes 10.2.2 does work!

From here, I plan to test 10.3.0, and work my way up the minor version updates — 10.4, 10.5, etc. — until I can pinpoint when iTunes 10 finally broke with 10A96. Once that’s figured out, I can update the WikiPost.

As for the glitching window carrying over from iTunes 9.2.1, I don’t know how to troubleshoot fixing that, or whether it can be fixed. I think this may be related to some alpha-level code (like a Framework or PrivateFramework) in 10A96 and not the iTunes code base.

Onward!

[UPDATE: So far, all these 10.x versions of iTunes I’ve tested (10.2.2 through 10.4) will not quit cleanly. All hang into “not responding” following Cmd-Q and need to be force-quit. This could be related to problems with either saving the library XML file or the iTunes prefs file, but that isn’t yet clear. (Re-launching does show the playlist items from last session, so the hang’s cause remains unsolved.) BUT… if you’re following along, it appears that following minor changes to iTunes’ UI from 10.4 — notably, the traffic-light buttons are moved from vertical to horizontal — the play display window no longer shows glitches. The off-white window area, however, is missing. See third attachment.]

[SECOND UPDATE: I can confirm that iTunes will not launch on 10A96 beginning with iTunes 10.5.0. The final version to launch on Build 10A96 is iTunes 10.4.1. What changes with the 10.5.0 installation package are two additional components within the install pkg: CoreTypes.bundle (located in /S/L/CoreServices/) and CoreFP1.framework (located in /S/L/PrivateFrameworks/). Both of these replace the versions which installed with 10A96. It also appears that iTunes 10.5.0 — the first version of iTunes to support iOS 5.0/iPhone 4S — will not run without these components and it will not run with these components if the system isn’t, I’m deducing, at least OS X 10.5.8; OS X 10.6.0 Build 10A190; or higher (much like with iTunes 10.6.3). I could be wrong, of course. ¶ Further, when reviewing crash logs for iTunes 10.4.1 and below, three of the four threads linked with crashes are related to the dylib called libSystem.B.dylib. This dynamic library, in /usr/lib/, is installed with Build 10A96. I do not know whether this and/or other dylibs are newer revisions in 10A190 or even 10.5.8. But I can say the first thread, that of running iTunes itself, is what invokes the crash. See fourth attachment.]

Conclusion: iTunes 10.4.1 is as high as it goes for Build 10A96, but the dirty force-quitting of iTunes has yet to be solved.

LINKS

• All legacy iTunes versions between 7.3.0.54 and 12.8.2
• Release dates for iTunes 10.x revisions

My understanding is that libSystem.B.dylib contains the C Libraries as well as other important libraries that the system depends on and references for a number of essential functions and it does seem to have been continually updated from 10.5 through to 10A432 and beyond. The version in 10A190 is different from 10A096 but is still not the same as the version shipped in 10A432. It’s next on my list of shared libraries to compile once Libxml has finally finished compiling on the old iBook. Too many other system components and frameworks refuse to function without the correct version of Libsystem.

CoreFoundation is an example of a fundamental framework that is not available as opensource (only CFLite) so must be copied over from 10A432 and this won’t work without the correct libraries on the system - the same seems to be true for the other key frameworks i.e IOKit, Foundation etc though they can be compiled they require the more recent library headers. Further some parts of some Frameworks, particularly certain PrivateFrameworks are included during compilation of the XNU kernel meaning the more recent release kernels will not work properly without them.

I’ve also noticed that the security frameworks and related binaries and libraries play a much larger role in builds after 10A190 meaning they also will need to be updated.

 

[Macbookprodude]

I really applaud all those trying to get Snow Leopard to work on PowerPC and to make it truely the last OS to run on G4,G5 machines, but without 10.6.8 and other updates + an updated browser, though I think Firefox legacy should work, I don't see it going anywhere. I hope I am wrong. I will try it out on my G4 DLSD.

 

[B S Magnet]

I really applaud all those trying to get Snow Leopard to work on PowerPC and to make it truely the last OS to run on G4,G5 machines, but without 10.6.8 and other updates + an updated browser, though I think Firefox legacy should work, I don't see it going anywhere. I hope I am wrong. I will try it out on my G4 DLSD.

Yah, you’re wrong.

—sent from an A1138 PowerBook G4 running InterwebPPC on Snow Leopard 10A96

 

[Macbookprodude]

Yah, you’re wrong.

—sent from an A1138 PowerBook G4 running InterwebPPC on Snow Leopard 10A96

Thats why I stated that.. but with no QE/CI for my Titanium 1ghz - still a capable machine, it would be slow on that machine.. but you like myself have a DLSD and so I will try it out.. I forgot interwebPPC was also developed by WickNix for 10.6, but does everything work in 10.6 ??? Other than interwebPPC which will solve our internet browsing issues, what else does work under 10.6 by itself ? Photobooth ? For a G5 what about loading Snow in 64-bit ?

 

[B S Magnet]

Thats why I stated that.. but with no QE/CI for my Titanium 1ghz - still a capable machine, it would be slow on that machine.. but you like myself have a DLSD and so I will try it out.. I forgot interwebPPC was also developed by WickNix for 10.6, but does everything work in 10.6 ??? Other than interwebPPC which will solve our internet browsing issues, what else does work under 10.6 by itself ? Photobooth ? For a G5 what about loading Snow in 64-bit ?

You really don’t know that until you actually roll up your sleeves and test it yourself on the 1GHz.

While I wouldn’t expect the performance of an earlier 1GHz G4 (even with an L3 cache), on PC-133 RAM capped at 1GB, to be dramatically different between Leopard 10.5.8 and Snow Leopard 10A96, some aspects may move more quickly or efficiently with the Snow Leopard 10A96 build over 10.5.8 — once optimizations outlined in the WikiPost are done to the testing system.

As it is, it appears the builds appear to be able run best on late G4s (7447s, 7457s, and possibly 7448s) and G5s before development was halted on all PowerPC systems. These models featured faster buses, newer-generation RAM, and CPUs which ran more efficiently than the 7400/7410/7455 era of chips.

 

[B S Magnet]

My understanding is that libSystem.B.dylib contains the C Libraries as well as other important libraries that the system depends on and references for a number of essential functions and it does seem to have been continually updated from 10.5 through to 10A432 and beyond. The version in 10A190 is different from 10A096 but is still not the same as the version shipped in 10A432. It’s next on my list of shared libraries to compile once Libxml has finally finished compiling on the old iBook. Too many other system components and frameworks refuse to function without the correct version of Libsystem.

Thanks for clarifying that question I had. It would appear that bringing in another version of that dylib (or other dylibs) from, say, the 10.5.8 build or one of the subsequent 10.6 dev builds would not be a good use of testing time and effort unless, as you’re doing, home-compiling for specific builds. In other words, manually swapping dylibs, unlike kexts or even frameworks, would be ill-advised.

CoreFoundation is an example of a fundamental framework that is not available as opensource (only CFLite) so must be copied over from 10A432 and this won’t work without the correct libraries on the system - the same seems to be true for the other key frameworks i.e IOKit, Foundation etc though they can be compiled they require the more recent library headers. Further some parts of some Frameworks, particularly certain PrivateFrameworks are included during compilation of the XNU kernel meaning the more recent release kernels will not work properly without them.

If you’re feeling adventurous, or if you want a quick diversion from compiling dylibs and/or frameworks, might you be up to seeing whether this short utility could be amended to compile for big-endian PPC use?

I’ve also noticed that the security frameworks and related binaries and libraries play a much larger role in builds after 10A190 meaning they also will need to be updated.

Overall, how applicable do you foresee these updated/recompiled frameworks being of utility to the 10A96 build? Alternately, does the build environment you’re using (in Xcode, I’m gathering) permit for relatively straightforward building if built/compiled on a 10A96 box?

 

[ChrisCharman]

Thanks for clarifying that question I had. It would appear that bringing in another version of that dylib (or other dylibs) from, say, the 10.5.8 build or one of the subsequent 10.6 dev builds would not be a good use of testing time and effort unless, as you’re doing, home-compiling for specific builds. In other words, manually swapping dylibs, unlike kexts or even frameworks, would be ill-advised.



If you’re feeling adventurous, or if you want a quick diversion from compiling dylibs and/or frameworks, might you be up to seeing whether this short utility could be amended to compile for big-endian PPC use?



Overall, how applicable do you foresee these updated/recompiled frameworks being of utility to the 10A96 build? Alternately, does the build environment you’re using (in Xcode, I’m gathering) permit for relatively straightforward building if built/compiled on a 10A96 box?

I believe as 10A096 is closer to Leopard than Snow Leopard you may have more success copying over some libraries and frameworks from 10.5, as @vddrnnr has done - i think I remember that he is using LibSystem from Leopard along with some other components. Obviously this will continue to move the base system closer to 10.5 but does seem to increase stability and usability. On 10A190 backporting from 10A432 fixed some errors but then creates further errors that need addressing so it’s an ongoing process.

I’m still learning as i go but will attempt to compile that utility when i get a chance and report back. I may try to cross compile on the Aluminium MacBook that arrived a few days ago as its now running Retail Snow Leopard with Xcode, 4GB Ram and an SSD so is much faster at building projects.

I’m using Xcode and the Command Line Tools from 10A096 Server on my 10A190 build on the iBook as well as on 10A096. The development environment is the same, one thing i will note however is that i’ve had to compile the versions of autoconf, automake, glibtool and more from the 10.6 AOSP in addition to copying TargetConfig and Xcrun over from 10A432 in order to compile more recent Open Source Projects. I’ve chosen to update the system binaries as I’m slowly trying to bring 10A190 closer the GM but the tools could also be built using MacPorts, which may be a better option on 10A096 to minimise the impact on system stability.

 

[ediflorianus]

This Thread is epic, I still can't believe it. (and also mad I didn't find it till just now)
So "Snow Leopard to work on PowerPC" this is NEWS , this IS BIG , this is EPIC...
So you-wee are really doing this? making a testable build for PowerPC's? so great , probably my excitement will last me a few nights.... I have a few PowerMac's ; I am ready to test any build or anything if you wish it.
Hope this news will get World Wide Coverage.(I still can't believe my eyes - a new beta-os for my old macs).
Was accustomed to idea of being stuck - 4ever on Leopard 10.5.8.

I have (for testing purposes ) : A few -tower's - PowerMac G5 versions (including water cooled one) , a few G4's (mdd + 2x older spec) , A few Ibooks G4/G3's and A couple of PowerBooks (about 2? maybe 3 - one has linux on it.). all at your disposal. (just send me msg if anything is required).

And many.... many.... THNX
For all your efforts and great work . It's so appreciated you have no idea.(at least we all have hope for a better testing future-at least). I also have a few digital-books regarding the OS X Leopard/Snow Leopard/Tiger but not sure if it helps anyway.

 

[B S Magnet]

This Thread is epic, I still can't believe it. (and also mad I didn't find it till just now)
So "Snow Leopard to work on PowerPC" this is NEWS , this IS BIG , this is EPIC...
So you-wee are really doing this? making a testable build for PowerPC's? so great , probably my excitement will last me a few nights.... I have a few PowerMac's ; I am ready to test any build or anything if you wish it.
Hope this news will get World Wide Coverage.(I still can't believe my eyes - a new beta-os for my old macs).
Was accustomed to idea of being stuck - 4ever on Leopard 10.5.8.

I have (for testing purposes ) : A few -tower's - PowerMac G5 versions (including water cooled one) , a few G4's (mdd + 2x older spec) , A few Ibooks G4/G3's and A couple of PowerBooks (about 2? maybe 3 - one has linux on it.). all at your disposal. (just send me msg if anything is required).

And many.... many.... THNX
For all your efforts and great work . It's so appreciated you have no idea.(at least we all have hope for a better testing future-at least). I also have a few digital-books regarding the OS X Leopard/Snow Leopard/Tiger but not sure if it helps anyway.

Glad you found this project! Be sure to pass this along to folks you know who might want to do some nerding out.

If you want to dip your toes and do some testing, I’d offer that you pick one Mac you already enjoy using (for me, I’ve been using an A1138 PowerBook because it’s a late-PPC model and I also enjoy using its keyboard). Stick with that until you get more acquainted with all the quirks of the version you choose to try out and test — either 10A96 and 10A190.

If I were to hazard a guess, folks who have been participating over the past year have skewed to a 2:3 ratio between 10A96 and 10A190, respectively. (My own testing has been, so far, confined to the 10A96 camp.)

Also, be sure to have a good look through Table 2 to see whether your model(s) has been tested successfully by someone else already, and also Table 3 to get a sense of how your video card(s) might or might not be supported. Generally, the later the model, the more likely it should run reasonably well. G3s, however, are out (just as they were with Leopard).

The first post in the thread, the WikiPost, is packed with a lot of information and pointers on stuff. Bookmark it and keep it handy at all times!

Welcome!

 

[ediflorianus]

Glad you found this project! Be sure to pass this along to folks you know who might want to do some nerding out.

If you want to dip your toes and do some testing, I’d offer that you pick one Mac you already enjoy using (for me, I’ve been using an A1138 PowerBook because it’s a late-PPC model and I also enjoy using its keyboard). Stick with that until you get more acquainted with all the quirks of the version you choose to try out and test — either 10A96 and 10A190.

If I were to hazard a guess, folks who have been participating over the past year have skewed to a 2:3 ratio between 10A96 and 10A190, respectively. (My own testing has been, so far, confined to the 10A96 camp.)

Also, be sure to have a good look through Table 2 to see whether your model(s) has been tested successfully by someone else already, and also Table 3 to get a sense of how your video card(s) might or might not be supported. Generally, the later the model, the more likely it should run reasonably well. G3s, however, are out (just as they were with Leopard).

The first post in the thread, the WikiPost, is packed with a lot of information and pointers on stuff. Bookmark it and keep it handy at all times!

Welcome!

OK. I will. Tnx for notice.
P.S. Can't we edit the Reporting system from OS X (old ver 10.5-10.6serv) so any errors are reported into one place? (as the apple reporting servers don't work on older os anymore anyway).

 

[B S Magnet]

OK. I will. Tnx for notice.
P.S. Can't we edit the Reporting system from OS X (old ver 10.5-10.6serv) so any errors are reported into one place? (as the apple reporting servers don't work on older os anymore anyway).

I’m not sure.

I’m also unsure how one would go about setting up a server for receiving error reports from everyone who’s testing SL-PPC on their systems.

 

[B S Magnet]

Hey @Larsvonhier —

On the WikiPost, I noticed on Table 2, where you tested SL-PPC on your PowerBook G4/500, mention of a build called 10A343 which didn’t show up on Table 1. Was 10A343 another build we haven’t known about, or was this a typo? Thanks!

 

[Macbookprodude]

Yah, you’re wrong.

—sent from an A1138 PowerBook G4 running InterwebPPC on Snow Leopard 10A96

View attachment 1778173

How did you get dark mode on the G4 PB ?

 

[B S Magnet]

That's what it says.
View attachment 911795
I'll have to reboot into Leopard to see what it reads it as. My installation has all of the Leopard Geforce kexts installed.

Also I've manage to install MacPorts 2.6.2. I had to compile it from the source as the SL version was erroring out in the installer, and the Leopard version complained that it can't be installed on 10.6.
View attachment 911796

When you installed Macports, do you remember how you navigated the problem of installing ports which abort because of “bad CPU type”? I successfully compiled Macports manually, much as you did, but when pulling repositories for dependencies, it is pulling “darwin_10.i386” tarballs only and aborting, unsurprisingly, with an “error 2” (due to that bad CPU type).

 

[Project Alice]

When you installed Macports, do you remember how you navigated the problem of installing ports which abort because of “bad CPU type”? I successfully compiled Macports manually, much as you did, but when pulling repositories for dependencies, it is pulling “darwin_10.i386” tarballs only and aborting, unsurprisingly, with an “error 2” (due to that bad CPU type).

I don't remember what I installed I haven't even booted into it in forever. I know I installed neofetch without any errors though.

 

[B S Magnet]

I don't remember what I installed I haven't even booted into it in forever. I know I installed neofetch without any errors though.

Interesting. I’ve kind of run into a wall with that. I even tried to set aside last day’s compile of mp and brought over my mp build from Leopard, but that ends up being a non-starter all its own (with mp wanting a darwin9 system and finding 10 instead).

I guess I’ll keep at it. Thanks.

 

[Larsvonhier]

Hey @Larsvonhier —

On the WikiPost, I noticed on Table 2, where you tested SL-PPC on your PowerBook G4/500, mention of a build called 10A343 which didn’t show up on Table 1. Was 10A343 another build we haven’t known about, or was this a typo? Thanks!

You are right, it was a typo: I meant build 9A343 (Tiger). Corrected this in the table on page #1. Here's the link to what I found back then, credit to @LightBulbFun !

 

[MacPro2006VBox]

You are right, it was a typo: I meant build 9A343 (Tiger). Corrected this in the table on page #1. Here's the link to what I found back then, credit to @LightBulbFun !

9A343 is a Leopard beta

 

[Larsvonhier]

Correct.

 

[B S Magnet]

In my latest bout of, “Why are you doing this, B S M,” I present a slow and fairly janky Virtual PC 7, running a fresh build of XP SP3, on Build 10A96.

Because why not.

As you can see, there’s a VPC plug-in module which goes into hanging right after I open a .vp7 file, but I managed to (barely) reach DuckDuckGo after downloading the last 32-bit XP version Firefox (45) I could get installed onto it.

The last time I used VPC for anything was back in 2005, for a job I had then. Whatever optimization/troubleshooting skills I had from that time are long, long gone. I use a Windows box less than once every couple of years, at best.

Pleaes calp solwly.