louie@sayshell.umd.edu (Louis A. Mamakos) (06/26/91)
I was cleaning up my disk today, and happened across the NX_VOID game that the Lighthouse folks wrote and released years ago. And guess what? It works just fine on a 68040, Release 2.1 NeXTstation. I guess that I'm not supposed to be surpised or anything, but I'm rather pleased that you can write a program that does what NX_VOID does, and play by all of the rules. When you play by the rules, things keep working! NX_VOID is a asteroids-like outer-space multi-level shoot-'em up. Louie sez "Check it out." sonata.cc.purdue.edu:/pub/next/1.0-release/NX_VOID.Z louie
jiro@shaman.com (Jiro Nakamura) (06/26/91)
In article <1991Jun26.023443.27660@ni.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: > I was cleaning up my disk today, and happened across the NX_VOID game > that the Lighthouse folks wrote and released years ago. And guess > what? It works just fine on a 68040, Release 2.1 NeXTstation. I > guess that I'm not supposed to be surpised or anything, but I'm rather > pleased that you can write a program that does what NX_VOID does, and > play by all of the rules. When you play by the rules, things keep > working! I see another person who has been brainwashed by the Mac world. ;-) Yes, I think the NeXT is *much* more stable between version releases than the Mac ever was. It also helps that the OS isn't going to change much since it already has much of what we need. Only the appkit is going to change and NeXT has made it pretty apparent that things following the API will be backwardly supported at least one revision level. Most of the things that broke from 1.0 to 2.0 were doing strange things at the system level, or depended on files being certain places. Although I'm sad that we don't have a lot of technical documentation about the guts of NeXTStep (such as say: Inside Macintosh), it also makes the lives of end-users simpler, I would say. What I think should be a high, high, high priority for NeXT is guarranteeing that *nothing* can crash the window server. As a developer, I can conjure up apps that can crash the server consistently (unfortunately, I do this non-deliberately). This is a Bad Thing and harks of Mac-dom. No application should be able to crash WM, end of story. - jiro -- Jiro Nakamura jiro@shaman.com Shaman Consulting +1 607 277-1440 Voice/Fax/Data "Bring your dead, dying shamans here!"
ddj@zardoz.club.cc.cmu.edu (Doug DeJulio) (06/27/91)
In article <1991Jun26.144145.11129@shaman.com> jiro@shaman.com (Jiro Nakamura) writes: > Yes, I think the NeXT is *much* more stable between version releases >than the Mac ever was. It also helps that the OS isn't going to change ^^^^^^^^^^^^^^^^^^^^^^^^ >much since it already has much of what we need. Only the appkit is going >to change and NeXT has made it pretty apparent that things following the >API will be backwardly supported at least one revision level. I hope that this is not true. I would like to see NeXT change from a Mach 2.5 based kernel to a Mach 3.0 based kernel. A Macintosh emulator I'm working on would be much easier with some of the features of Mach 3.0 available (mainly external pagers). Of course, there's no reason that this change would have to break any software that works under Mach 2.5. -- Doug DeJulio ddj@zardoz.club.cc.cmu.edu
jiro@shaman.com (Jiro Nakamura) (06/27/91)
In article <1991Jun26.200004.1042@zardoz.club.cc.cmu.edu> ddj@zardoz.club.cc.cmu.edu (Doug DeJulio) writes: > In article <1991Jun26.144145.11129@shaman.com> jiro@shaman.com (Jiro Nakamura) writes: > > Yes, I think the NeXT is *much* more stable between version releases > >than the Mac ever was. It also helps that the OS isn't going to change > ^^^^^^^^^^^^^^^^^^^^^^^^ > >much since it already has much of what we need. Only the appkit is going > >to change and NeXT has made it pretty apparent that things following the > >API will be backwardly supported at least one revision level. > > I hope that this is not true. I would like to see NeXT change from a > Mach 2.5 based kernel to a Mach 3.0 based kernel. A Macintosh > emulator I'm working on would be much easier with some of the features > of Mach 3.0 available (mainly external pagers). > > Of course, there's no reason that this change would have to break any > software that works under Mach 2.5. As you guessed, I meant OS API. There aren't going to be great innovations like Mac System 6-->7 or other things since Mach is a pretty mature product. I too would like to see Mach 3.0 on the NeXT. I guess NeXT is trailing Mach a couple of steps to be on the safe side. After all, they have Avie on their team, so it isn't because they are incompetent. - jiro -- Jiro Nakamura jiro@shaman.com Shaman Consulting +1 607 277-1440 Voice/Fax/Data "Bring your dead, dying shamans here!"
songer@ei.ecn.purdue.edu (Christopher M Songer) (06/28/91)
In article <1991Jun26.200004.1042@zardoz.club.cc.cmu.edu> ddj@zardoz.club.cc.cmu.edu (Doug DeJulio) writes: > >I hope that this is not true. I would like to see NeXT change from a >Mach 2.5 based kernel to a Mach 3.0 based kernel. A Macintosh >emulator I'm working on would be much easier with some of the features >of Mach 3.0 available (mainly external pagers). ^^^^^^^^^^^^^^^^^^^^^^ > >Of course, there's no reason that this change would have to break any >software that works under Mach 2.5. >-- >Doug DeJulio >ddj@zardoz.club.cc.cmu.edu YES!!!!!!! (Let's say it again for emphasis, this is my personal pet NeXT peeve.) YES!!!!!!! External Pagers are crucial. They were a big performance hit under 2.5 and this is probably why they were left out of the NeXT version, but regardless of whether NeXT goes to the 3.0 microkernel, PUT EXTERNAL PAGERS IN NEXT MACH. There is so much you can do with them and, more importantly, so much you can't do without them.... SYSV shared memory compatibility libs would be easy to do with external pagers but rewriting code that uses SYSV shared memory is NOT easy! As NeXT OS currently stands, one cannot do a set of SYSV shared memory libraries. There are some coolo programs that use SYSV shm* calls. I'd like to run some of them. So, I whole heartedly agree with Doug's comment. An upgrade to the 3.0 microkernel would be fine, or patching the external pagers into the current would be fine too -- either way, having external pager would be a real boon! -Chris