Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
On 16 Oct 2001, at 8:50, Marcel Kilgus wrote: Das QL? I'd say it's masculine. Not in my ususal mode of address: Das Sch... *+=8% QL Dings, da wolfgang (hey, after thhat much dutch, we're entitled to some german...) - www.wlenerz.com
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
On 16 Oct 2001, at 17:15, [EMAIL PROTECTED] wrote: Glad its not feminine, Die QL sounds a bit rash in english! It's already happened, anyway. Wolfgang (runs to shelter) - www.wlenerz.com
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Is the QL masculine or feminine in gender ? In Italian masculine: Il QL And in Welsh, as we all know, masculine: Y QL hwn -- Dilwyn Jones [EMAIL PROTECTED] http://www.soft.net.uk/dj/index.html
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message 3BCD4547.6743.1997C9@localhost, Wolfgang Lenerz [EMAIL PROTECTED] writes On 16 Oct 2001, at 22:23, Malcolm Cadman wrote: In English 'objects' or artefacts are seen as being neutral, or inanimate. However, most English people do attach a gender to objects ... like cars = the 'old girl'. Yet, I guess the QL is a masculine object to most users ? To most non-users, it'll be an object of derision? He .. he .. a neat turn around :-) I guess we have established now that the QL is masculine in most European languages, ... and still capable of being a delight to ql-users :-) -- Malcolm Cadman
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Wolfgang Lenerz wrote: Is the QL masculine or feminine in gender ? For us Germans, of course, it's a neutrum... Das QL? I'd say it's masculine. Marcel
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Jerome Grimbert wrote: } Das QL? I'd say it's masculine. Isn't Das for Neutral, and Der for masculine (Die being for feminine ) ? Exactly. And Das QL just doesn't sound good ;-) Marcel
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], Jerome Grimbert [EMAIL PROTECTED] writes Malcolm Cadman makes some magical things to make me read } He .. he ... I got my 'le' and 'la' mixed up then :-) } } Is the QL masculine or feminine in gender ? In french, it is masculine. I do not know for german and english (neutral ?). Ah ... thanks, le QL then :-) In English 'objects' or artefacts are seen as being neutral, or inanimate. However, most English people do attach a gender to objects ... like cars = the 'old girl'. Yet, I guess the QL is a masculine object to most users ? -- Malcolm Cadman
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Malcolm Cadman makes some magical things to make me read } Alas, it's survival is limited to about 2 or 3 times the longuest subscription. } (But who cares about that, it's easy money now!, and maybe in the meantime } we can sell the magazine team to some other publisher..., just like selling } the old cow... ) } } C'est le vie ... ? Interesting gender inversion or typo. C'est le vit is not heard often, as it is just a kind of It is a male organ, so it might be taken as offensive by some bigot. Especially if you say it to someone (in which case, the spelling of vi* is not heard!!) C'est la vie is the classical quote. ObTopic: This post is ON topic, because of the subject line (flaming... :-)
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], Jerome Grimbert [EMAIL PROTECTED] writes } Alas, it's survival is limited to about 2 or 3 times the longuest subscription. } (But who cares about that, it's easy money now!, and maybe in the meantime } we can sell the magazine team to some other publisher..., just like selling } the old cow... ) } } C'est le vie ... ? Interesting gender inversion or typo. C'est le vit is not heard often, as it is just a kind of It is a male organ, so it might be taken as offensive by some bigot. Especially if you say it to someone (in which case, the spelling of vi* is not heard!!) C'est la vie is the classical quote. ObTopic: This post is ON topic, because of the subject line (flaming... :-) He .. he ... I got my 'le' and 'la' mixed up then :-) Is the QL masculine or feminine in gender ? -- Malcolm Cadman
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
On 15 Oct 2001, at 17:46, Malcolm Cadman wrote: Is the QL masculine or feminine in gender ? -- Malcolm Cadman For us Germans, of course, it's a neutrum... Wolfgang - www.wlenerz.com
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
At 09:54 ìì 15/10/2001 +0200, you wrote: On 15 Oct 2001, at 17:46, Malcolm Cadman wrote: Is the QL masculine or feminine in gender ? -- Malcolm Cadman For us Germans, of course, it's a neutrum... Wolfgang - www.wlenerz.com Ha! For Greeks is either a male or a neuter... (Especially when it crashes...)
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Malcolm Cadman makes some magical things to make me read } He .. he ... I got my 'le' and 'la' mixed up then :-) } } Is the QL masculine or feminine in gender ? In french, it is masculine. I do not know for german and english (neutral ?).
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe Phoebus - that well known computer murphy's law does state that by the time the last bug is removed, the program is obsolete (says Dilwyn still finding bugs in programs he wrote in the mid 1980s) -- Dilwyn Jones [EMAIL PROTECTED] http://www.soft.net.uk/dj/index.html
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Back in the late '70s/early '80s I used to buy computing mags like Practical Computing and Computing Today because they were very 'DIY' based with articles on programming and building bits of hardware, but when they changed to nothing but reviews of commercial software and hardware I lost interest and never renewed my subscriptions. QL Today still keeps that DIY spirit alive and I hope it lasts. I agree entirely with this, there were a number of magazines in those days which were more 'DIY' or 'tinkerer' orientated, and I miss that. These days they all assume you are a strict 'user' not a 'tinkerer' like many QLers are. One of the reasons for keeping a QL and Aurora system here is to occasionally take it apart (and that can mean software or hardware) and do something which is not run of the mill with it. While this PC/QPC2 combination is great for everyday use, I doubt I'll ever have the familiarity and confidence I do with a QL on any other computer platform. Thanks for the comments about QL Today, we like to think we do our best, and a huge chunk of the spirit of QL Toady is down to our contributors who keep coming up with such material for publication. -- Dilwyn Jones [EMAIL PROTECTED] http://www.soft.net.uk/dj/index.html
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], Jerome Grimbert [EMAIL PROTECTED] writes Malcolm Cadman makes some magical things to make me read } Ah ! I think it was just a french problem where magazines looks more an more } catalogues with lot of advertising and propaganda (guru Bill do this, guru } Bill thinks that...) } } -Message d'origine- } De : Malcolm Cadman } (...) } OT - yet computer magazines today are largely just product reviews, and } new software version reviews ... } (...) } } Probably a 'world wide' depressing period for computer magazines then } Nope, a classical evolution. Magazines are first made by enthusiasts, for other enthusiasts and hobbyists. When success is there, money start flowing, and marketing people gets in. At that time, the enthusiastic writers are usually exausted or outnumbered by 'will write for food' writers. Hence the content drop in level of specificity (Let's widen our audience), the commercial ads have more influence on the editorial, and the latest integrity enthusiasts stop being published or even writing. The magazine is then the optimal money-making scheme: - no hard content (counter productive with advertisement) - lots of advertisements and 'product'-review - good reputations (from the old time) in the newcomers' circles - hopefully a lot of running subscription. I cannot dispute your logic. I subscribe to specialist magazines for QL and RISC OS, which avoid this problem to a large extent. The only PC magazine I subscribe to is 'PC Magazine' by Ziff-Net. THis preserves a depth of discussion, despite the pressures that you so clearly outline above. Alas, it's survival is limited to about 2 or 3 times the longuest subscription. (But who cares about that, it's easy money now!, and maybe in the meantime we can sell the magazine team to some other publisher..., just like selling the old cow... ) C'est le vie ... ? -- Malcolm Cadman
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Ah ! I think it was just a french problem where magazines looks more an more catalogues with lot of advertising and propaganda (guru Bill do this, guru Bill thinks that...) -Message d'origine- De : Malcolm Cadman (...) OT - yet computer magazines today are largely just product reviews, and new software version reviews ... (...) -- Malcolm Cadman
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Marcel Kilgus wrote: Claude Mourier 00 wrote: I wonder what is QPC overhead (it's not to offense Marcel : QPC is a great product) : does anybody know the slowdown factor eaten by the emulator, or in other word, what is the factor to apply to the speed of SMSQ under QPC to have an idea of SMSQ speed written directly for x86 ? Depends on too many factors. As QPC's currently only an interpreter, 20 or more could be feasible. I expect much less for a JIT compiler version, but I don't know whether that will ever happen. Anyway x86 assembler is something nobody wants to program really, especially when coming from the 68k corner, believe me. Or even if coming from mainframe assembler! -- Peter S Tillier [EMAIL PROTECTED] or [EMAIL PROTECTED] Opinions expressed are my own and not necessarily those of my employer.
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
On 10 Oct 2001, at 22:13, Marcel Kilgus wrote: Anyway x86 assembler is something nobody wants to program really, especially when coming from the 68k corner, believe me. It sure isn't! Segments - B Wolfgang - www.wlenerz.com
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Wolfgang Lenerz wrote: Anyway x86 assembler is something nobody wants to program really, especially when coming from the 68k corner, believe me. It sure isn't! Segments - B Actually segments aren't used anymore (thank God). However, the instruction set is still quite horrible. And although current processors have internally more than 100(!) registers the set visible to the application is still very limited (7 if counted generously). Marcel
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], Claude Mourier 00 [EMAIL PROTECTED] writes Ah ! I think it was just a french problem where magazines looks more an more catalogues with lot of advertising and propaganda (guru Bill do this, guru Bill thinks that...) -Message d'origine- De : Malcolm Cadman (...) OT - yet computer magazines today are largely just product reviews, and new software version reviews ... (...) Probably a 'world wide' depressing period for computer magazines then ... -- Malcolm Cadman
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Hmmm, it's that time of year again So I had a good idea for a flaming (I need it since it's cold in Pennsylvania now ;-) What about rewriting QDOS/SMS (or converting the assembly sources) in order to run NATIVELY on x86 processors. There are a number of nice tools available for that (like the PortAsm/68K by MicroAPL) that could completely translate the assembly sources to optimised native x86 assembly. Then all SBASIC programs would run directly and we could also have QPC (or some other emulator) to work as a 68K native binaries on-the-fly translator (kind of what Apple did with PowerPC Macs in order to run older 68K apps) and in the future we could develop our own x86 apps (and take advantage of all the x86 based peripherals etc...) Not to mention that the middle-man (Dos/Win 9x/NT) will be vanish Imagine the speed/power potential of such a solution... :-) Now start flaming :-) Phoebus
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Phoebus, You are really sad :o) Why leave a perfectly good Greece for a home in Pennsylvania ? I can see a couple of (minor) flaws with your suggestion, but as an exercise, it appeals to my sense of humor quite a bit. The flaws are : what happens if we use PortAsm/68K and there are bugs in the generated code - we'd need to be seriously good Intel assembly language programmers to fix that - I suspect only Marcus would be able to sort it out. I'd have to learn Intel assembly myself, then do a series of articles in QL Toady - and you'd all get very very bored very very quickly :o) Have you SEEN Intel assembly - it is awful ! All the system variables, Basic variables, interrupts, vectors, traps whouldn't be there any more so we'd have to either code in C - oops, forgot, we can't, no C compiler :o) It just wouldn't be QDOSMSQ any more toto !! Nice mental exercise though. Norman. - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administrator Phone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd. URL:http://www.Lynx-FS.com -
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
At 01:50 ìì 10/10/2001 +0100, you wrote: Phoebus, You are really sad :o) Oh yeah! :-) Why leave a perfectly good Greece for a home in Pennsylvania ? Hmmm I still wonder, but fear not I'll be back to Europe soon (3/4 years... once I am done with college)... It's plainly too dangerous at this time to be a foreigner in the US... (Not to mention stupid things I hear nearly every day... but that's another story) I can see a couple of (minor) flaws with your suggestion, but as an exercise, it appeals to my sense of humor quite a bit. The flaws are : what happens if we use PortAsm/68K and there are bugs in the generated code I've seen PortAsm sources and it does really work with minimal bugs Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe - we'd need to be seriously good Intel assembly language programmers to fix that - I suspect only Marcus would be able to sort it out. And other people that might become interested (especially because of QDOS' approaches in many OS issues) through a forum like SourceForge... More assembly programmers exist for x86 than all the other platforms combined I am afraid. I'd have to learn Intel assembly myself, then do a series of articles in QL Toady - and you'd all get very very bored very very quickly :o) So we are now :-) (Just kidding :-) hehe Have you SEEN Intel assembly - it is awful ! As a matter of fact I have... Not worse than M68K for someone that STARTS learning now tho ;-) The goal (again) would be to attract new users... Existing ones are diminishing anyways :-) All the system variables, Basic variables, interrupts, vectors, traps whouldn't be there any more so we'd have to either code in C - oops, forgot, we can't, no C compiler :o) True partially. A compiler could be (relatively) easily written. For example, C68 exists already for many platforms (among them MS-DOS) and I believe Keith and Dave are ALWAYS up for a challenge :-) It just wouldn't be QDOSMSQ any more toto !! I don't agree with you... the essence of an operating system is its framework... This would be preserved... Furthermore, as we have discussed in the past in this list, a turn towards a more-Unix like approach for the OS kernel would benefit us greatly. (Higher level language for drivers etc...) Nice mental exercise though. It is aint'it? :-) Phoebus
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Have you SEEN Intel assembly - it is awful ! :O) The NASM syntax is better - using brackets for all address expressions and not bothering with % signs for registers makes it look clearer. I find 68k assembler easier though. My only gripe is all those .Bs, .Ws, etc. can make it look a mess. I do like the idea of having a standalone x86 SBASIC, but the maintenance issue probably makes it less attractive for developers than the emulator. Ian BTW I missed most of the assembler articles in QL Toady. Are they available online, or as back issues? -Original Message- From: Norman.Dunbar Sent: 10 October 2001 13:50 To: ql-users Cc: Norman.Dunbar Subject: RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors Phoebus, You are really sad :o) Why leave a perfectly good Greece for a home in Pennsylvania ? I can see a couple of (minor) flaws with your suggestion, but as an exercise, it appeals to my sense of humor quite a bit. The flaws are : what happens if we use PortAsm/68K and there are bugs in the generated code - we'd need to be seriously good Intel assembly language programmers to fix that - I suspect only Marcus would be able to sort it out. I'd have to learn Intel assembly myself, then do a series of articles in QL Toady - and you'd all get very very bored very very quickly :o) Have you SEEN Intel assembly - it is awful ! All the system variables, Basic variables, interrupts, vectors, traps whouldn't be there any more so we'd have to either code in C - oops, forgot, we can't, no C compiler :o) It just wouldn't be QDOSMSQ any more toto !! Nice mental exercise though. Norman. -- -- - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administrator Phone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd. URL:http://www.Lynx-FS.com -- -- - Visit our website at http://www.ubswarburg.com This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. This message is provided for informational purposes and should not be construed as a solicitation or offer to buy or sell any securities or related financial instruments.
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
I wonder what is QPC overhead (it's not to offense Marcel : QPC is a great product) : does anybody know the slowdown factor eaten by the emulator, or in other word, what is the factor to apply to the speed of SMSQ under QPC to have an idea of SMSQ speed written directly for x86 ? Claude -Message d'origine- De : Phoebus Dokos [mailto:[EMAIL PROTECTED]] Envoyé : mercredi 10 octobre 2001 15:41 À : QL Users' mailing list Objet : [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors Hmmm, it's that time of year again So I had a good idea for a flaming (I need it since it's cold in Pennsylvania now ;-) What about rewriting QDOS/SMS (or converting the assembly sources) in order to run NATIVELY on x86 processors. There are a number of nice tools available for that (like the PortAsm/68K by MicroAPL) that could completely translate the assembly sources to optimised native x86 assembly. Then all SBASIC programs would run directly and we could also have QPC (or some other emulator) to work as a 68K native binaries on-the-fly translator (kind of what Apple did with PowerPC Macs in order to run older 68K apps) and in the future we could develop our own x86 apps (and take advantage of all the x86 based peripherals etc...) Not to mention that the middle-man (Dos/Win 9x/NT) will be vanish Imagine the speed/power potential of such a solution... :-) Now start flaming :-) Phoebus
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Ian, BTW I missed most of the assembler articles in QL Toady. Are they available online, or as back issues? Not yet I'm afraid, but I'm sure that there are back issues available. The main problem is, I write one article but Jochen decides that I cannot be allowed to use up 80% of the magazine each time, so he splits it (with my blessing) into two or three separate articles. Then he numbers them differently,to what I have so I get confused, and he gets confused. I'll tell you how confused I am, I just referred to Marcel in an email, but called him Marcus - Sorry Marcel, I have some German customers here today, I got confused again. Anyway, I have also got the problem of errors I have noticed in the articles - some I have 'back ported' and other I have not. I intend to have the articles on-line at some point in the future - how near or how far remains to be seen - esp as I'm just getting stuff sorted out thanks to NTL :o( But, after all that, I'm sure Dilwyn or Roy would (Wood !) be happy to sell you some back issues. Failing that, I could send you shed loads of txt files which may or may not make up the series - I think I'm allowed to do that ! Norman. - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administrator Phone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd. URL:http://www.Lynx-FS.com -
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Phoebus, Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe No I don't remember (honest) I remember some problems with QLiberator though ! And what bugs do you refer to :o) Regards, Norman. PS. Wrap up well for winter. - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administrator Phone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd. URL:http://www.Lynx-FS.com -
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Trolling Writing an OS for a processor family is nothing, compared to the needed writing for peripheral supports. So, as well as generic x86 support, what about GENERIC x86 compatible supports of hardware in the OS. Natively of course, NO PC BIOS. (there is no Bios on some embeded card, so you cannot rely on that). And able to support all the fancy graphics cards (dual headed and DVI included), as well as all the fancy sound card/extension/native... with full capabilities and optimal operations... Disk access is another story too. All I give you is hardware IRQ (possibly multiplexed and shared), and the full addressing range of the CPU. Oh, last point, the system CANNOT assign the IRQ, it is always a choice of value you have to support... good luck...
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
At 02:52 ìì 10/10/2001 +0100, you wrote: Phoebus, Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe No I don't remember (honest) I remember some problems with QLiberator though ! And what bugs do you refer to :o) Hehehe... lemme search my QLT back issues and I'll get back at you Regards, Norman. PS. Wrap up well for winter. You have NO idea (And to think my friends in Greece complain about the weather being hot How ungrateful on their part ;-)
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
At 04:10 ìì 10/10/2001 +0200, you wrote: Trolling Writing an OS for a processor family is nothing, compared to the needed writing for peripheral supports. So, as well as generic x86 support, what about GENERIC x86 compatible supports of hardware in the OS. Natively of course, NO PC BIOS. (there is no Bios on some embeded card, so you cannot rely on that). And able to support all the fancy graphics cards (dual headed and DVI included), as well as all the fancy sound card/extension/native... with full capabilities and optimal operations... Disk access is another story too. All I give you is hardware IRQ (possibly multiplexed and shared), and the full addressing range of the CPU. Oh, last point, the system CANNOT assign the IRQ, it is always a choice of value you have to support... That's where the experience of other x86 OS programmers comes in play (Linux for example) We cannot have that with M68K platforms anymore :-( unfortunately
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
At 10:34 ðì 10/10/2001 -0500, you wrote: At 02:52 ìì 10/10/2001 +0100, you wrote: Phoebus, Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe No I don't remember (honest) I remember some problems with QLiberator though ! And what bugs do you refer to :o) Hehehe... lemme search my QLT back issues and I'll get back at you That should read get back TO you :-) ooopppsss (My fingers tremble from the cold - as does my -remaining- brain) Regards, Norman. PS. Wrap up well for winter. You have NO idea (And to think my friends in Greece complain about the weather being hot How ungrateful on their part ;-)
RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Oh, those bugs ! I remember now - I must get around to fixing them !!! Norman. - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administrator Phone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd. URL:http://www.Lynx-FS.com - -Original Message- From: Phoebus Dokos [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 10, 2001 4:34 PM To: [EMAIL PROTECTED] Subject: RE: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors At 02:52 ìì 10/10/2001 +0100, you wrote: Phoebus, Then again QDOS DOES have bugs anyways...every program does :-) (Even yours ;-)) Remember some funny things with FP numbers about 1 year ago?) hehe No I don't remember (honest) I remember some problems with QLiberator though ! And what bugs do you refer to :o) Hehehe... lemme search my QLT back issues and I'll get back at you Regards, Norman. PS. Wrap up well for winter. You have NO idea (And to think my friends in Greece complain about the weather being hot How ungrateful on their part ;-)
RE: Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Hmmm, an easier approach might be to write a version of SMSQ as an XFree86-style user interface to run on Linux. The hardware support is then already done and QDOS filesystems could be implemented with QXL.WIN files. Produce an SMSQ/QDOS API and you could write programs using familiar OS calls in NASM. Ian. -Original Message- From: jerome.grimbert Sent: 10 October 2001 15:10 To: ql-users Cc: jerome.grimbert Subject: Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors Trolling Writing an OS for a processor family is nothing, compared to the needed writing for peripheral supports. So, as well as generic x86 support, what about GENERIC x86 compatible supports of hardware in the OS. Natively of course, NO PC BIOS. (there is no Bios on some embeded card, so you cannot rely on that). And able to support all the fancy graphics cards (dual headed and DVI included), as well as all the fancy sound card/extension/native... with full capabilities and optimal operations... Disk access is another story too. All I give you is hardware IRQ (possibly multiplexed and shared), and the full addressing range of the CPU. Oh, last point, the system CANNOT assign the IRQ, it is always a choice of value you have to support... good luck... Visit our website at http://www.ubswarburg.com This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. This message is provided for informational purposes and should not be construed as a solicitation or offer to buy or sell any securities or related financial instruments.
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message Hb5f1151beb4.1002729565.ln4p1327.ldn.swissbank.com@MHS, [EMAIL PROTECTED] writes Norman has just sent me the texts of his articles, but the back issues are still of interest generally. Back in the late '70s/early '80s I used to buy computing mags like Practical Computing and Computing Today because they were very 'DIY' based with articles on programming and building bits of hardware, but when they changed to nothing but reviews of commercial software and hardware I lost interest and never renewed my subscriptions. QL Today still keeps that DIY spirit alive and I hope it lasts. Yes, I remember those days too ... when magazines had a decent content to them. I still have quite a few 'project type' magazines stored away for the 'rainy day' when I get to make some things. OT - yet computer magazines today are largely just product reviews, and new software version reviews ... I guess it could get interesting again in the future. -- Malcolm Cadman
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], Phoebus Dokos [EMAIL PROTECTED] writes All the system variables, Basic variables, interrupts, vectors, traps whouldn't be there any more so we'd have to either code in C - oops, forgot, we can't, no C compiler :o) True partially. A compiler could be (relatively) easily written. For example, C68 exists already for many platforms (among them MS-DOS) and I believe Keith and Dave are ALWAYS up for a challenge :-) It just wouldn't be QDOSMSQ any more toto !! I don't agree with you... the essence of an operating system is its framework... This would be preserved... Furthermore, as we have discussed in the past in this list, a turn towards a more-Unix like approach for the OS kernel would benefit us greatly. (Higher level language for drivers etc...) Nice mental exercise though. From what I gather, Intel are going StrongARM with the next generation of processors. That is abandoning the MIPS processors in favour of the RISC processors. It just so happens that StrongARM was developed in Cambridge, England, and scorned for many years by Intel and M$ :-) ... and emerged out of Sinclairs one time great rival Acorn Computers. So, recompiling QDOS / SMSQ for a new life on new hardware platforms is not such a bad idea. If the idea is good enough, there will be someone who will do it. Perhaps they have even realised it is the end of the road in sight for Windows ... and they need a better OS for the new hardware :-) Optimistic, or what ? -- Malcolm Cadman
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED], [EMAIL PROTECTED] writes QL Toady back issues:- Clip Never bought any as I've been a subscriber from day one :-)) Are you sure that wasn't the 'day after', day one ... :-) -- Malcolm Cadman
[ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
In message [EMAIL PROTECTED] k, Norman Dunbar [EMAIL PROTECTED] writes - Norman Dunbar EMail: [EMAIL PROTECTED] Database/Unix administratorPhone: 0113 289 6265 Fax:0113 289 3146 Lynx Financial Systems Ltd.URL:http://www.Lynx-FS.com OT - but what if LFS ? Have you sorted out your ISP ? -- Malcolm Cadman
Re: [ql-users] Need some flaming :-) - What about re-writing QDOS for x86 processors
Claude Mourier 00 wrote: I wonder what is QPC overhead (it's not to offense Marcel : QPC is a great product) : does anybody know the slowdown factor eaten by the emulator, or in other word, what is the factor to apply to the speed of SMSQ under QPC to have an idea of SMSQ speed written directly for x86 ? Depends on too many factors. As QPC's currently only an interpreter, 20 or more could be feasible. I expect much less for a JIT compiler version, but I don't know whether that will ever happen. Anyway x86 assembler is something nobody wants to program really, especially when coming from the 68k corner, believe me. Marcel
