Menu
▾
▴
Re: [B2-devel] Thoughts on SheepShaver and Basilisk II
|
From: howard s. <how...@ho...> - 2012-06-16 22:03:55
|
Hi all, The sudden increase in activity on this ML makes for interesting reading. It looks like the development community is looking for an agenda? We, the maintainers of the http://www.emaculation.com website have ample experience in why and how SheepShaver is used by various groups of users. And have ample experience in what its shortcomings are. So, we would like to chime in from the user perspective. First of, something about importance: In the discussion these past days, someone suggested that SheepShaver is not all that important because much old software can be run in Mini vMac. This is certainly not true for those who use SheepShaver on an Intel Mac, by far the largest group of SheepShaver users. After the demise of the Classic environment and later Rosetta, SheepShaver is the only way left to run Mac OS applications on an emulated PowerPC processor. Many Mac users simply need SheepShaver to keep some old applications running or to have access to files in formats that are not supported by today's Mac software. SheepShaver became a necessary tool that, as we know from the forum discussions and support questions, in some cases is even used to run irreplaceable software that is vital to professionals, businesses, or scientists. Indeed. I was not saying that Mini vMac and DOSbox covered all or most of the SheepShaver bases. I was merely saying that SheepShaver is not a necessity for everybody's legacy application needs. A lot of software from the nineties was also available on DOS and Windows and is much easier to emulate that way. (We recently switched to the Windows version of a particular application on our Macintoshes in order to allow for more seamless integration via Wine.) On the Macintosh-only front, HyperCard has tended to be, in my experience, the classic Macintosh thing that people most want to emulate, and I have been able to recommend Mini vMac (and sometimes RunRev) for that. The classic Macintosh did not start with a large user base. When one filters out people who are just enough satisfied with the OSX or Windows equivalents of the classic applications, the people who can successfully migrate to newer applications (like RunRev), and the people who just want HyperCard, the pool of potential contributors, developers, and users is relatively small. Someone called SheepShaver a niche product. While that in itself is true, we feel that some tender love and care (including some expansion in the feature set) could make it a bit less “nichy”. Looking around the Internet, one finds many failed attempts at getting SheepShaver running, or when it does run not providing the desired functionality. Once again, I was not saying that SheepShaver was not important. It is very important to several work-flows that I maintain. I was merely trying to point out why emulation of a PowerPC Macintosh had not enjoyed the same sort of attention that other projects such as DOS emulation have drawn. The programming difficulties are much greater and the number of users much smaller. If there were millions of people very interested in doing this, there would be multiple projects and probably a commercial product or two. >Yes we understand that. Our point would be: SheepShaver could be more helpful if it were more complete. A point we would like to make is that we, as volunteer supporters/build providers, are not in favour of a diverging code base for the supported platforms. Towards a better feature set for all supported hosts: -Improved CD/DVD handling. Like an eject key-stroke for disk images? >And real CD's. While changing images would be nice (changing cd's is not well supported), we were also thinking about being able to read other formats besides data cd-roms. Audio is not read correctly on all platforms, while DVD's are not supported at all. -USB support. -Video acceleration on the host. Which applications (games?) that you use have seemed to need this? >Perhaps that needs to be rephrased: support for OpenGL etc. -Memory management allowing running Mac OS above 9.0.4. -More opcodes supported, allowing more applications to work. (Office comes to mind). I still use Word 5.1a on occasion without problems. Which version of Office causes problems? >Office 98 and 2001 (all applications that are included in both) run into trouble. Some seem to think this has to do with the lack of virtual memory, but there are also signs that it is related to some opcodes not being supported. Also, are you sure that you aren't compiling with the PPC_REENTRANT_JIT flag set to 1? That disables a number of operations that are supported. >We compile with the default settings, if that setting is 1, then Yes. We should check. -Improve networking, including Appletalk. -Improve communication to serial/parallel ports. -A possibility to launch a SheepShaver built-in preferences editor without the emulator starting (by holding a modifier key?). That would make a separate external preferences editor superfluous. Wouldn't you prefer to have these separate? We actually disabled the configuration editor in our environment. It is very rare that one wishes to tweak a working configuration, and adding another step to start-up seems unnecessary. >This is more an issue for the OSX build, as that is the only one with built-in editor, which can only be opened after SheepShaver started. We would like to get rid of all the separate prefs editors floating around and a unified integrated solution would be great, specially if it could be started to configure SheepShaver without actually starting SheepShaver. The current GTK prefs editor in Linux does that. (But it has no "Save" option, it will always start SheepShaver or not save changes.) -Improve SCSI support. As in allowing SheepShaver to interact with real SCSI devices? This would be nice, certainly, but might produce some messy conflicts with host operating system drivers if enabled by default. >Yes, SCSI support was available in the old BasiliskII (build 142, 2001) for Windows through the ASPI driver. Perhaps that old code could point a way? Specific for OSX: -Clipboard exchange for text in 64-bit mode SheepShaver (it currently doesn’t work at all in 64-bit mode.) -Possibility to compile/build SheepShaver in MacOSX 10.7.x (Lion) and future versions. That requires reworking lots of assembly that llvm dislikes. And I am not sure what llvm likes. >That's where your expertise must provide the answer, we wouldn't know ;-) -Solution to the weird color problems on Intel Mac after SheepShaver window has been minimized or hidden and is brought back in view. (Possibly a SDL issue.) Specific for Windows: -Replace the reliance on the cdenable.sys cd rom driver for CD support. It currently doesn’t work on 64-bit hosts. -Replace the reliance on the BasiliskII Ethernet driver with for example WinPcap, as it currently doesn’t run on 64-bit hosts (requiring driver signing). -Replace reliance on Cygwin to build for Windows. Would you prefer to get X11 from Windows Services for UNIX? > Never tried that in Windows, but I guess it would take too much from the average user to go down that path. Also, is this a future-proof concept? >Thanks! >Howard Best regards, Ronald P. Regensburg and Howard Spoelstra basilisk-devel mailing list bas...@li... https://lists.sourceforge.net/lists/listinfo/basilisk-devel ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ basilisk-devel mailing list bas...@li... https://lists.sourceforge.net/lists/listinfo/basilisk-devel |
