Edited for clarity..
Quote: " Also I guess it has escaped your notice which no doubt it actually has, but C-Caps the x86 Processor's Floating Point Unit to 4x 64bit Registers... the 5x86 itself actually has 4x 80bit Floating Point Registers, which most don't support using ASM math ( Tasm which I was using with PureBasic only supports 32bit Floats)
(3x86/4x86 both have 2)
"
More techno smoke screen gibberish. But, Just for clarity sake, while the 387, 486, Pentium have 32/64 and 80bit floating point capability. Were talking about these 40bit FPU operations that you apparently require to retain authenticity. You simply don't need them.
Quote: "8086 and 286 in real-mode would = extra work because they don't operate the same as floating point."
Neither have a floating point unit. It wasn't integrated into the x86 chips until the 486 line. Before that the system needed a math coprocessor. i.e 8087,187,287,387
Quote: "AcornASM has different registers so assigning them isn't as simply and straight forward as assigning the BASIC structure...
A - Accumulator (4bit).......}
X,Y - Index Registers (40bit)} these can be access per bit
M - Memory Registers (32bit) }
SP - Stack Pointer
PC - Program Counter (32bit) (split into 2 bit)
F - Flag Register (8bit) {N, V, B, D, I, Z, C can't remember what they mean right now}
"
This completely bogus specification is nothing like the 6502, secondly the Electron doesn't appear to have a second 6800 (i assume you mean a 6809 derivative here) processor ?..
6502.. is 8bit, with 16bit address lines. Registers A ,X & Y .. it does not have floating point calculations period.
6809 is a hybrid evolution of the 6502, 16bit registers (A,X,Y), 16 bit address lines. It doesn't have float point support either.
Like many other users around here, I've spent a little under a decade programming these chips (vic20,bbc model b, c64,apple 2, c128.. and 6809 on the Tandy colour), thus statements like this, fail to make any cohesive sense....
Quote: " Whilst although setting up the math for the 40bit Ops was necessary, it was also necessary for authentisity to make sure that it was within a 9-Digit Significant range.
This has nothing to do with adding more work for the sheer hell of it, and everything to do with making the language work EXACTLY how it should.
I could've set it up for std x86 asm, but then I would've had to hack it later when ppl sit there and bitch about the real numbers not working right; Sorry but to me it is far more important to have something working first time CORRECTLY rather than just working and available to ppl.
"
Previously you said you wanted this to run on 386 processor. However, the 386 doesn't have FPU operations in it, it is purely integer. To do so, would require a 386 system with 387 math co processor installed. Or you'd have to roll your own FPU emulation as stated previous. (see bellow)
Quote: "It is shoddy workmanship to release somethign you know damn well is buggy."
You'd have to release something, it order for that to be a concern I would think.
Quote: " What the hell do I look like to you, a technical engineer creating a language from backengineering the damn machines?
I'm going by exactly what my manual says...
It has 1x 80bit Real Number Register (Split Hi/Lo or X/Y)
It then has 8x 32bit Integer/Memory Registers
"
More gibberish. See above. Hence, to restate my point. Floating point operations should be mapped on the cpu's FPU instructions. The internal accuracy of calculations isn't going to be relevant in order to compile probably 99.99% of existing acorn basic programs.
i.e
a=100/3
if a=33.33333
print " it's a third"
else
print "ouch, something is wrong "
endif
Quote: "6502A, Motorola 6800 Base ASM; but all of the major player register I did list above. Which can all be stripped and placed directly onto the x86 386+"
Previously, you stated the 6502 as a direct descendant of the x86. It is not. Moreover, In order to support Acorn basics inline assembly directives. You not only have to emulate the 6502's behavior, but the system architecture behind it. In order for it successfully execute anything but the simplest preexisting code. Since your after accuracy here, this is conceptually flawed from the get go.
Quote: "Why does it matter, i'm not going to tell anyone my real email in forum; what the hell do you take me for a moron?
And I especially am not going to sit here and say what my email is because quite frankly I know that Empty is part of the dev team for PlayBasic and I don't want that jackass knowing my email.
"
I'm the only person who has access to the Mailing List. Why does it matter?, the only legal access you could have had was via the mailing edition list a few weeks back. But now your claiming that you've apparently you've been accessing alpha's illegally 6/8 months ago... This, I will NOT stand for.
Quote: "Quite frankly tought titties.
I've been messing around with PlayBasic since 1.4-something, currently I believe I have the 2.0x build... i can't remember what one exactly.
"
Typical, rather than say, "oh i must have been mistaken", Your now going to sit there and continue lie to everybody (including the author of the language) ?. In order for such idiocy to be taken seriously, you'd have to prove this claim. You course you won't..
Send me the read me from any version between 1.40 to 2.06 then, or any file from it's installer for that matter.
Quote: "Now if your all finished, either stay in this thread and help me plan for the compiler; Or get the hell out and leave me alone.
Quite frankly you guys obviously don't like me, and honestly I can't say I have a partifular affinity for any of your either; So either make yourselves useful or piss off and bother someone else.
"
Oh please.. and who said I don't like you ?. Being a little paranoid aren't we. However like most people, I don't suffer liars or fools very well.
Quote: "
I don't think it is unreasonable to accept some modicam of acceptable behaviour around here. Quite simply put, if you have nothing to actually add to this topic that is within reasonable conversation then don't post in it.
"
It's long past time the point for you started growing up. I would thought, that as a self proclaimed professional developer, you'd be used to remaining objective. Hence, If you don't want your ideas subjected to public scrutiny, be that corrections, or whatever, then posting them in a public forum is probably not for you.
Kevin Picone
[url]www.underwaredesign.com[/url]
Play Nice!
Play Basic (Out Now!)
