AmigaOS 3.2 released!

[mahen]

Ah... Oh... Indeed, maybe the WB filename length doesn't correspond to the filesystem one... Gonna check ! Thanks for the suggestion. I really thought that was FS-bound !

[boemann]

Originally Posted by torturedutopian

Hi everyone ! I hope this was not reported before :-)
I use 3.2.1 on my A1230 ATM (because it fails to boot with the 3.2.2 probably due to my custom 3.2.1 ROMs).

Did you people notice that when your rename a file while being in "list" mode, the whole files listing flickers for a few seconds ? (more or less probably depending on the machine speed & number of files)

Does not occur in icons view mode.

cheers

Yes we are aware and even tried to see if we could fix it for 3.2.2.1 but it turns out to be a somewhat more difficult thing to solve. So I'm not even going to promise it can be fixed in 3.2.3 (whenever that lands).

It flickers as many times as there are files in the drawer, so you many be in for a long wait.

[mfilos]

I use the arexx script "S:AutoArrangeIons.rexx" VERY frequently (via an MCP hotkey combination) but there are rare times that I get this sort of error when I use it (and ofc it doesn't apply)...



If I reboot, then the problem goes away.
Any ideas?

[Olaf Barthel]

Originally Posted by torturedutopian

Hi everyone ! I hope this was not reported before :-)
I use 3.2.1 on my A1230 ATM (because it fails to boot with the 3.2.2 probably due to my custom 3.2.1 ROMs).

Did you people notice that when your rename a file while being in "list" mode, the whole files listing flickers for a few seconds ? (more or less probably depending on the machine speed & number of files)

This is a side-effect of how AsyncWB does what it does when it performs the renaming of the respective files.

Workbench "knows" that if the name of a directory or file changes, it will just have to "rub out" the old name and render it under its new name.

AsyncWB can only tell Workbench that the item under its old name no longer exists ("remove that, please") and that a new item exists now ("add this one, please").

So Workbench first removes the item under its old name, which then results in re-rendering the entire visible list of items, and then adds the item again under its new name, which requires another rendering pass.

Workbench certainly does much more than ought to be necessary (and the special effects firework you witnessed is much more pronounced on some Amiga systems than others: it just lacks appropriate sound effects), but we have not yet updated it to allow AsyncWB to do what it needs, without the overhead.

[Firestone]

I have noticed in this thread that there are a lot of people complaining about MagicMenu, and that's fine. I think most of us had this tool in our systems back in the 90s.
I still use it on my 3.2 installations, and it works "fine". The only issues I got is that it might lock up the system for a few seconds with the menu bar showing.

Anyway, the question I was curious about asking the devs... is it this popup-functionality from MagicMenu that is a challenge to implement, or is it the whole thing? I mean, would it be i theory possible to make the menu look like MagicMenu with drop shadows and that MagicWb-look?

And there is on more tool I guess everyone had: CycleToMenu.
Is this a feature built in somewhere now that I have missed maybe...?

[Yulquen74]

Are anyone able to snapshot icon drawers with more than 30 chars in icon filename? I find that i have to rename the drawers so that the icon filename is 30 chars or less first, snapshot, and then change back to the longer name. Using 3.2.2, and pfs3aio with fnsize set to 107.

[No.3]

I do not know what fnsize is, but most likely you did not change the Workbench preferences.

[Yulquen74]

You are right, did not notice that particular setting until now.

[Thomas Richter]

Originally Posted by Firestone

Anyway, the question I was curious about asking the devs... is it this popup-functionality from MagicMenu that is a challenge to implement, or is it the whole thing?

I already explained that elsewhere. The issue is the intutuition state machine. You cannot, from the outside of intuition, play against the state machine of intuition itself. When intuition shows a menu, it is aware that certain operations are at that point not possible, and does not attempt to proceed with them. If another program is grabbing the locks of the layers or layer infos, then intuition cannot know, and thus can run into a situation where nothing goes anymore because intuition waits on a situation that can no longer happen because MagicWB cannot progress either.

[kamelito]

Regarding the issue with recoverable alerts using Apollo V4, I tried to look at what changed in Exec 47.10 compared to 47.8 and apart form a shift in addresses to save $10 bytes due to optimizations and new version and copyright, plus the new way Gayle is detected, I have seen nothing obvious that could trigger the alerts. Does P96 rely on some Exec private structures to work that may have changed?
I will have to compare a bare rom vs one loaded in Apollo.

[mfilos]

After contacting some months back the author of VisualPrefs about a bug I experienced (concerning window title alignment if using specific window title height or font size) he found and fixed the bug. I tried a beta version yesterday and all work just fine.

The only reason I'm writing this here is that this version also fixed the issue I have stated in January concerning some screen font garbage while scrolling in some AmigaGuide files (like speciffic areas on HELP system of 3.2.x)
http://eab.abime.net/showpost.php?p=...postcount=1258

Apparently he will release an update on Aminet in the following days...
I'm very happy that so many bugs get hunted and fixed so quickly in 2023 - It's alive!

[Firestone]

Originally Posted by Thomas Richter

I already explained that elsewhere. The issue is the intutuition state machine. You cannot, from the outside of intuition, play against the state machine of intuition itself. When intuition shows a menu, it is aware that certain operations are at that point not possible, and does not attempt to proceed with them. If another program is grabbing the locks of the layers or layer infos, then intuition cannot know, and thus can run into a situation where nothing goes anymore because intuition waits on a situation that can no longer happen because MagicWB cannot progress either.

Got it, thanks for explaining again, Thomas.

[bubbob42]

Originally Posted by kamelito

Regarding the issue with recoverable alerts using Apollo V4, I tried to look at what changed in Exec 47.10 compared to 47.8 and apart form a shift in addresses to save $10 bytes due to optimizations and new version and copyright, plus the new way Gayle is detected, I have seen nothing obvious that could trigger the alerts. Does P96 rely on some Exec private structures to work that may have changed?
I will have to compare a bare rom vs one loaded in Apollo.

Don’t bother. ApolloTeam has received the source code changes made to exec from us, checked it and approved. The cause lies elsewhere.

They‘re still searching, so let’s hope the riddle will be solved, soon.

[kamelito]

Originally Posted by bubbob42

Don’t bother. ApolloTeam has received the source code changes made to exec from us, checked it and approved. The cause lies elsewhere.

They‘re still searching, so let’s hope the riddle will be solved, soon.

That’s Amiga Spirit, nice to see good cooperation!
Thanks

[Etze]

Originally Posted by bdb

The AmigaOS 3.2.2.1 Hotfix is available;

Yes...:

Originally Posted by boemann

The 3.2.2.1 hot fix is out ! check the hyperion website

[boemann]

The AmigaOS 3.2 developers will have a table at Amiga38 in October. So come by and talk to us

[Don_Adan]

I dont have AmigaOS 3.2, but seems to be bug in handling "echo" command in s-s. Executing s-s is stopped if "echo" command is executed. I dont know if this is bug or feature (in Bill Gates style). Anyway this bug/feature is not very good idea for me.
More infos here:

https://www.ppa.pl/forum/amiga/45769...i-komenda-echo

[boemann]

Originally Posted by Don_Adan

I dont have AmigaOS 3.2, but seems to be bug in handling "echo" command in s-s. Executing s-s is stopped if "echo" command is executed. I dont know if this is bug or feature (in Bill Gates style). Anyway this bug/feature is not very good idea for me.
More infos here:

https://www.ppa.pl/forum/amiga/45769...i-komenda-echo

I just checked and don't see that problem. Can other readers please confirm or deny this bug

[StevenJGore]

Originally Posted by boemann

I just checked and don't see that problem. Can other readers please confirm or deny this bug

I'm on 3.2.2.1 and I have "Echo" in my startup-sequence, and I don't see that problem either.

[Don_Adan]

More tests was done, seems that "echo" works if NIL: is used, without NIL: doesnt works correctly.