I'm no expert on Roman numerals, but I recently saw a movie released in MCMXLIX. Now, we know what the MCM means, but I submit that the intention of chiseling as little stone as possible, the romans avoided the use of 4 successive numerals by use of bi-quinary counting. However, the use of IX for 9 as opposed to VIV, as in the MCM above, is a bit ambigous. I'm not planning any trips to the former Roman Empire, so I doubt I'll do better than trusting the conclusion reached in this forum. Does anybody have an authoritative reference?
Dick
-----Original Message-----
From: Hans Franke <Hans.Franke_at_mch20.sbs.de>
To: Discussion re-collecting of classic computers <classiccmp_at_u.washington.edu>
Date: Tuesday, April 20, 1999 3:10 PM
Subject: Re: Program Challenge (was Re: z80 timing... 6502 timing)
>
>> > > Convert a binary value to Roman, using ASCII characters (or
>> > > the native character set if applicable) into a string
>> > > with a termination character (if ASCII, use the NUL (0)
>> > > character).
>> > Is a length terminated record valid ?
>
>> Hmmmm ... I might be inclinded to say yes, but is there any reason it
>> can't be NUL terminated (or terminated with the appropriate termination
>> character for the character set)?
>
>Just since it is the usual way within the OS I'm trying to code.
>NUL is like any other caracter valid within a string. Also, it
>gieve me an disadvantage - instead of just adding '\0' I have
>to calculate the string.
>
>> > > The valid Roman characters used are IVXLCDM
>> > Is Space valid ? Romans did include them at will.
>> I did not know that, but for this program no, spaces are not allowed.
>
>Ok.
>
>> > > > There are a few details which have been left out of the specification for
>> > > > this task.
>> > > > Does it require input validation?
>> > > I think I specified that. The valid range of Roman numerals is 1 through
>> > > 3,999 inclusive. The routine does have to check that and construct a
>> > > special string ( "*" ) if the input is not in that range.
>> > > > Is the binary input pure binary, or is it BCD?
>> > > Okay, that might be a valid point, but it's pure binary, not BCD.
>
>> > Fine, you stated max input 3999, but to be checked, and you
>> > stated binary, but in what format ? Half Word (16 Bit),
>> > Word (32) or Double Word (64) ?
>
>> Uh, I was unaware that halfwords had a specific size (on the DEC Alpha, a
>> half word would be 32 bits in size).
>
>Like Words, they are just machine dependant ... in a given 32 Bit CPU
>a halfword is just 16 - so whats the input ? (in fact, I'll assume 32 Bit)
>
>> > To be transfered at a given location, or via a pointer,
>> > or via register (attention, might again be processor specific).
>
>> Appropriate to the CPU or the situation. Obviously, a 6502 can't pass it
>> in a single register, but there are other options. That's why I didn't
>> specify how to pass the data in.
>
>:))) And why specifying a '\0' byte ?
>
>> > > I liked Sam's suggestion of ``printing to memory'' as a way to avoid the
>> > > complications of I/O in this, and if I didn't make this clear that the
>> > > conversion was to be stored in memory, I'm sorry.
>>
>> > Output, to a given location (pointer) or static buffer ?
>> > Check for buffer length or asumption of an buffer, always big
>> > enough ? (if you check the binary number you also have to check
>> > the buffer (if given).)
>
>> Again, it's up to the programmer. But be prepared to justify your answers
>> 8-)
>
>:))
>
>> -spc (Welcome to the world of programming 8-)
>
>All simple, all the same, isn't it ?
>
>And again, more questions:
>If I'm right at Megans description, she just include the next lower
>digit when it comes to these subtraction rules, and your Algo seams
>to be weak at the same point. Let me give an example:
>
>49 would be normaly coded as IL (always remember, it was kind of a
>system to reduce writing as much as possible - there are even examples
>where the number 248 is written CIIL) while Megan seams to code it as
>XXXXIX - basicly wrong - or did I miss something ? I'm not realy
>what one can call a DEC-Geek.
>
>So do we only have to supporte the one-less rule, or the rule
>of one subtraction numeral - or the full possibility with the
>goal to reduce writing to a max ?
>
>Gruss
>H.
>
>Oh, BTW: this contest is exacty into the direction where we didn't
>want to go - we are comparing algorythms and not CPUs - for a more
>CPU dependant contest we need to fix the algo to use, so the difference
>will reflect the differences in processor design and not into algo
>design (of course there's more sex appeal in the my-algo-is-bether-
>than-yours, than in the my-(what-ever-algo)-implementation-is-better).
>
>So I still go for the idea of implementing one algo as good as possible
>for the different CPUs - more on Friday (I'm not full time available
>until then).
>
>
>
>--
>Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
>Vote against SPAM: http://www.politik-digital.de/spam/en/
>Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
>Ich denke, also bin ich, also gut
>HRK
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://www.classiccmp.org/pipermail/cctalk/attachments/19990420/6b3b8ef4/attachment.html
Received on Tue Apr 20 1999 - 18:44:59 BST