Commit by sir_richard :: r46904 reactos/ntoskrnl/mm/ARM3/miarm.h

[breakoutbox]

Hello,

"[NTOS]: 1MB is not 1000 * 1KB... "


http://en.wikipedia.org/wiki/Megabyte

1 KB = 1000 byte (decimal)
1 KiB = 1024 byte (binary)

1 MB = 1000 KB
1 MiB = 1024 KiB


Would be nice if ReactOS goes across the border
and also uses the modern SI unit definitions  ;-)
If in code definitions or shown as text to the user ..

Cheers,
Peter

[Daniel Reimer]

  WIndows uses internally the classic Definitions of MB and KB and we 
should do so, too. If you want it called MiB and KiB then its a GUI 
thing, nothing internal.

Am 18.04.2010 12:10, schrieb breakoutbox:
>   Hello,
>
>  "[NTOS]: 1MB is not 1000 * 1KB... "
>
>
>  http://en.wikipedia.org/wiki/Megabyte
>
>  1 KB = 1000 byte (decimal)
>  1 KiB = 1024 byte (binary)
>
>  1 MB = 1000 KB
>  1 MiB = 1024 KiB
>
>
>  Would be nice if ReactOS goes across the border
>  and also uses the modern SI unit definitions  ;-)
>  If in code definitions or shown as text to the user ..
>
>  Cheers,
>  Peter
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>

-----Ursprüngliche Nachricht-----
Von: Daniel Reimer <daniel.reimer(a)stud-mail.uni-wuerzburg.de>
Gesendet: 18.04.2010 12:46:51
An: ReactOS Development List <ros-dev(a)reactos.org>
Betreff: Re: [ros-dev] Commit by sir_richard :: r46904	reactos/ntoskrnl/mm/ARM3/miarm.h

>    WIndows uses internally the classic Definitions of MB
> and KB and we 
> should do so, too. If you want it called MiB and KiB then its a GUI 
> thing, nothing internal. 

Well, that's a valid argument.

WINDOWS still does it the wrong way.
Then why should ReactOS do it right ..  ;-)



Cheers,
Peter
___________________________________________________________
NEU: WEB.DE DSL für 19,99 EUR/mtl. und ohne Mindest-Laufzeit!
http://produkte.web.de/go/02/

[Olaf Siejka]

Mainly for compatibility reasons. If Windows does it that way, so would do
Windows apps and drivers, depending on it. If we change this default
behaviour, even with introducing some kind of aliases for the previous one,
we risk breakage, that would could be hard to determine/locate. When ReactOS
gets stable enough, i see no obstacle to change this, but right now, let us
not put too many mushrooms into single basket.

2010/4/18 <breakoutbox(a)web.de>

>   -----Ursprüngliche Nachricht-----
>  Von: Daniel Reimer <daniel.reimer(a)stud-mail.uni-wuerzburg.de>
>  Gesendet: 18.04.2010 12:46:51
>  An: ReactOS Development List <ros-dev(a)reactos.org>
>  Betreff: Re: [ros-dev] Commit by sir_richard :: r46904
>   reactos/ntoskrnl/mm/ARM3/miarm.h
>
> >    WIndows uses internally the classic Definitions
> > of MB and KB and we
> > should do so, too. If you want it called MiB and KiB then its a GUI
> > thing, nothing internal. 
>
>  Well, that's a valid argument.
>
>  WINDOWS still does it the wrong way.
>  Then why should ReactOS do it right ..  ;-)
>
>
>
>  Cheers,
>  Peter
>  ___________________________________________________________
>  NEU: WEB.DE DSL für 19,99 EUR/mtl. und ohne Mindest-Laufzeit!
>  http://produkte.web.de/go/02/
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>

[breakoutbox]

Olaf Siejka schrieb:
>   Mainly for compatibility reasons. If Windows does it
> that way, so 
>  would do Windows apps and drivers, depending on it. If we change this 
>  default behaviour, even with introducing some kind of aliases for the 
>  previous one, we risk breakage, that would could be hard to 
>  determine/locate. When ReactOS gets stable enough, i see no obstacle 
>  to change this, but right now, let us not put too many mushrooms into 
>  single basket.
>  
Hm,

how does a change to these #define names probably risk breakage ?


#define _1MB (1024 * _1KB)

in /trunk/reactos/ntoskrnl/mm/ARM3/miarm.h

[Olaf Siejka]

I am talking in general, about attempts to "correct Microsoft".

2010/4/18 breakoutbox <breakoutbox(a)web.de>

>   Olaf Siejka schrieb:
>
>   Mainly for compatibility reasons. If Windows does it that way, so would do
> >   Windows apps and drivers, depending on it. If we
> > change this default
> >  behaviour, even with introducing some kind of aliases for the previous one,
> >  we risk breakage, that would could be hard to determine/locate. When ReactOS
> >  gets stable enough, i see no obstacle to change this, but right now, let us
> >  not put too many mushrooms into single basket.
> >
> >  
>  Hm,
>
>  how does a change to these #define names probably risk breakage ?
>
>
>  #define _1MB (1024 * _1KB)
>
>  in /trunk/reactos/ntoskrnl/mm/ARM3/miarm.h
>
>
>
>
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>

[Andrew Faulds]

I don't think Microsoft was wrong, I think Hard Disk manufacturers have
tried to make them wrong in this case.

On 18 April 2010 20:58, Olaf Siejka <caemyr(a)gmail.com> wrote:

>   I am talking in general, about attempts to
> "correct Microsoft".
>
>  2010/4/18 breakoutbox <breakoutbox(a)web.de>
>
>  Olaf Siejka schrieb:
> >  
> >   Mainly for compatibility reasons. If Windows does it that way, so would
> > >   do Windows apps and drivers, depending on it. If
> > > we change this default
> > >  behaviour, even with introducing some kind of aliases for the previous one,
> > >  we risk breakage, that would could be hard to determine/locate. When ReactOS
> > >  gets stable enough, i see no obstacle to change this, but right now, let us
> > >  not put too many mushrooms into single basket.
> > >
> > >  
> >  Hm,
> >
> >  how does a change to these #define names probably risk breakage ?
> >
> >
> >  #define _1MB (1024 * _1KB)
> >
> >  in /trunk/reactos/ntoskrnl/mm/ARM3/miarm.h
> >
> >
> >
> >
> >
> >  _______________________________________________
> >  Ros-dev mailing list
> >  Ros-dev(a)reactos.org
> >  http://www.reactos.org/mailman/listinfo/ros-dev
> >  
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>  


-- 
Andrew Faulds (andrewros)
http://ajf.me/

[ShadowFlare]

Both sides have their own historical reasons for keeping things the way they
are.

On Sun, Apr 18, 2010 at 2:03 PM, Andrew Faulds <ajfweb(a)googlemail.com>wrote;wrote:

>   I don't think Microsoft was wrong, I think Hard
> Disk manufacturers have
>  tried to make them wrong in this case.
>
>
>  On 18 April 2010 20:58, Olaf Siejka <caemyr(a)gmail.com> wrote:
>
> >   I am talking in general, about attempts to
> > "correct Microsoft".
> >
> >  2010/4/18 breakoutbox <breakoutbox(a)web.de>
> >
> >  Olaf Siejka schrieb:
> > >  
> > >   Mainly for compatibility reasons. If Windows does it that way, so would
> > > >   do Windows apps and drivers, depending on it. If
> > > > we change this default
> > > >  behaviour, even with introducing some kind of aliases for the previous one,
> > > >  we risk breakage, that would could be hard to determine/locate. When ReactOS
> > > >  gets stable enough, i see no obstacle to change this, but right now, let us
> > > >  not put too many mushrooms into single basket.
> > > >
> > > >  
> > >  Hm,
> > >
> > >  how does a change to these #define names probably risk breakage ?
> > >
> > >
> > >  #define _1MB (1024 * _1KB)
> > >
> > >  in /trunk/reactos/ntoskrnl/mm/ARM3/miarm.h
> > >
> > >
> > >
> > >
> > >
> > >  _______________________________________________
> > >  Ros-dev mailing list
> > >  Ros-dev(a)reactos.org
> > >  http://www.reactos.org/mailman/listinfo/ros-dev
> > >  
> >
> >  _______________________________________________
> >  Ros-dev mailing list
> >  Ros-dev(a)reactos.org
> >  http://www.reactos.org/mailman/listinfo/ros-dev
> >  
>
>
>  --
>  Andrew Faulds (andrewros)
>  http://ajf.me/
>
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>

[breakoutbox]

ShadowFlare wrote:
>   Both sides have their own historical reasons for
> keeping things the 
>  way they are.
>
>  On Sun, Apr 18, 2010 at 2:03 PM, Andrew Faulds <ajfweb(a)googlemail.com 
>  <mailto:[e-mail]>> wrote:
>
>      I don't think Microsoft was wrong, I think Hard
>      Disk manufacturers have tried to make them wrong in this case.
>  
I didn't talk about history.

But if You like :: let's look into history. One day in ...

".. December 1998 the International Electrotechnical Commission (IEC), 
the leading international organization for worldwide standardization in 
electrotechnology, approved as an IEC International Standard names and 
symbols for prefixes for binary multiples for use in the fields of data 
processing and data transmission. The prefixes are as follows .."  
http://physics.nist.gov/cuu/Units/binary.html

... and since 1998 things changed.

Cheers,
Peter

[Timo Kreuzer]

It's not about wrong or right or compatibility.

It's a name. For a constant of value 1024*1024. We could rename it to
_1MiB.
But does that really help anyone? I'm sure it would rather confuse
people, as a lot of people sadly don't know the difference.
Also from a scientifical perspective it's still wrong. _1MiB would mean
1024*1024*bytes, but the constant doesn't contain bytes, it's a number
without a unit. So should we rename it to _1Mi ? No, because Mi is a
prefix and useless without any actual unit.

This discussion is completely useless. And as usual in reactos, it gets
the most attention.

Timo



breakoutbox schrieb:
>   ShadowFlare wrote:
> >   Both sides have their own historical reasons for
> > keeping things the
> >  way they are.
> >
> >  On Sun, Apr 18, 2010 at 2:03 PM, Andrew Faulds <ajfweb(a)googlemail.com
> >  <mailto:[e-mail]>> wrote:
> >
> >      I don't think Microsoft was wrong, I think Hard
> >      Disk manufacturers have tried to make them wrong in this case.
> >  
>  I didn't talk about history.
>
>  But if You like :: let's look into history. One day in ...
>
>  ".. December 1998 the International Electrotechnical Commission (IEC),
>  the leading international organization for worldwide standardization
>  in electrotechnology, approved as an IEC International Standard names
>  and symbols for prefixes for binary multiples for use in the fields of
>  data processing and data transmission. The prefixes are as follows
>  .."  http://physics.nist.gov/cuu/Units/binary.html
>
>  ... and since 1998 things changed.
>
>  Cheers,
>  Peter
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>

[ShadowFlare]

On Sun, Apr 18, 2010 at 2:48 PM, breakoutbox <breakoutbox(a)web.de> wrote:

>   ShadowFlare wrote:
>
> >   Both sides have their own historical reasons for
> > keeping things the way
> >  they are.
> >
> >  On Sun, Apr 18, 2010 at 2:03 PM, Andrew Faulds <ajfweb(a)googlemail.com<mailtomailto:
> >  ajfweb(a)googlemail.com>> wrote:
> >
> >     I don't think Microsoft was wrong, I think Hard
> >     Disk manufacturers have tried to make them wrong in this case.
> >
> >  
>  I didn't talk about history.
>
>  But if You like :: let's look into history. One day in ...
>
>  ".. December 1998 the International Electrotechnical Commission (IEC), the
>  leading international organization for worldwide standardization in
>  electrotechnology, approved as an IEC International Standard names and
>  symbols for prefixes for binary multiples for use in the fields of data
>  processing and data transmission. The prefixes are as follows .."
>  http://physics.nist.gov/cuu/Units/binary.html
>
>  ... and since 1998 things changed.
>  
Sorry, thought you were talking about the size and not the name.  Thanks for
clarifying.

[ShadowFlare]

On Mon, Apr 19, 2010 at 12:01 AM, ShadowFlare <blakflare(a)gmail.com> wrote:

>   On Sun, Apr 18, 2010 at 2:48 PM, breakoutbox
> <breakoutbox(a)web.de> wrote:
>
> >   ShadowFlare wrote:
> >
> > >   Both sides have their own historical reasons for
> > > keeping things the way
> > >  they are.
> > >
> > >  On Sun, Apr 18, 2010 at 2:03 PM, Andrew Faulds <ajfweb(a)googlemail.com<mailtomailto:
> > >  ajfweb(a)googlemail.com>> wrote:
> > >
> > >     I don't think Microsoft was wrong, I think Hard
> > >     Disk manufacturers have tried to make them wrong in this case.
> > >
> > >  
> >  I didn't talk about history.
> >
> >  But if You like :: let's look into history. One day in ...
> >
> >  ".. December 1998 the International Electrotechnical Commission (IEC), the
> >  leading international organization for worldwide standardization in
> >  electrotechnology, approved as an IEC International Standard names and
> >  symbols for prefixes for binary multiples for use in the fields of data
> >  processing and data transmission. The prefixes are as follows .."
> >  http://physics.nist.gov/cuu/Units/binary.html
> >
> >  ... and since 1998 things changed.
> >  
>  Sorry, thought you were talking about the size and not the name.  Thanks
>  for clarifying.
>  
Oops, I didn't realize you were responding to my response to someone else's
message.  My original message was not intended for you, but for Andrew
Faulds, since he did seem to be talking about the sizes that were picked.