'b' character at the end of inter in Cray fortran

"tolasma" <esillica@gmail.com> wrote in message news:2a0a7cca-f857-488e-97ee-c79cea94dcfe@o21g2000prh.googlegroups.com... > Now I'm compiling program written in CRAY. > There is weird thing. > ndash is integer type. > What is following??????? > ndash = 1777b 1023 Translate to ndash = int(o'1777') -- write(*,*) transfer((/17.392111325966148d0,6.5794487871554595D-85, & 6.0134700243160014d-154/),(/'x'/)); end

Thank you very much. And There are more weird. DATA MSK /777777777777777777777B/ DATA SIGNBIT/1000000000000000000000B/ When I compile that as following ndash =3Dint(o'777777777777777777777') I've got following message. [gkrun@hydra2 temp]$ g77 -o octal octalt.f octalt.f: In program `MAIN__': octalt.f:2: ndash =3Dint(o'777777777777777777777') ^ Integer at (^) too large How can I solve this problem? Thank you~ TN On 3=EC=9B=9428=EC=9D=BC, =EC=98=A4=ED=9B=842=EC=8B=9C41=EB=B6=84, "James V= an Buskirk" <not_va...@comcast.net> wrote: > "tolasma" <esill...@gmail.com> wrote in message > > news:2a0a7cca-f857-488e-97ee-c79cea94dcfe@o21g2000prh.googlegroups.com... > > > Now I'm compiling program written in CRAY. > > There is weird thing. > > ndash is integer type. > > What is following??????? > > =C2=A0ndash =3D 1777b > > 1023 > > Translate to > > ndash =3D int(o'1777') > > -- > write(*,*) transfer((/17.392111325966148d0,6.5794487871554595D-85, & > 6.0134700243160014d-154/),(/'x'/)); end

On Mar 28, 7:50=A0am, tolasma <esill...@gmail.com> wrote: > Thank you very much. > And There are more weird. > > =A0 =A0 =A0 =A0DATA MSK =A0 =A0/777777777777777777777B/ > =A0 =A0 =A0 =A0DATA SIGNBIT/1000000000000000000000B/ > > When I compile that as following > > =A0 =A0 =A0 =A0 =A0 ndash =3Dint(o'777777777777777777777') > > I've got following message. > [gkrun@hydra2 temp]$ g77 -o octal octalt.f > octalt.f: In program `MAIN__': > octalt.f:2: > =A0 =A0 =A0 =A0 =A0 ndash =3Dint(o'777777777777777777777') > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0^ > Integer at (^) too large > > How can I solve this problem? > > - Show quoted text - These two statements expect MSK and SIGNBIT to have a length of 64 bits. Regards, Mike Metcalf

On 2011-03-28 02:50:25 -0300, tolasma said: > Thank you very much. > And There are more weird. > > DATA MSK /777777777777777777777B/ > DATA SIGNBIT/1000000000000000000000B/ > > When I compile that as following > > ndash =int(o'777777777777777777777') > > I've got following message. > [gkrun@hydra2 temp]$ g77 -o octal octalt.f > octalt.f: In program `MAIN__': > octalt.f:2: > ndash =int(o'777777777777777777777') > ^ > Integer at (^) too large > > How can I solve this problem? 20 "7"s at 3 bits each is 60 bits. Just like a Cray word. Easiest solution is to use a Cray. :-) Otherwise you are left with understanding what was intended. That means knowing enough about both how a Cray worked and what you problem needs. Both aspects are somewhere between awkward and very awkward. The trailing B means Binary which is really Octal in case your recipe did not make sense to you. Why B for Octal remains a mystery that only the original Cray folks can explain. Most likely a case of choosing a name before they realized the possibility of generalizations. > Thank you~ > > TN > > > On 3월28일, 오후2시41분, "James Van Buskirk" <not_va...@comcast.net> wrote: >> "tolasma" <esill...@gmail.com> wrote in message >> >> news:2a0a7cca-f857-488e-97ee-c79cea94dcfe@o21g2000prh.googlegroups.com... >> >>> Now I'm compiling program written in CRAY. >>> There is weird thing. >>> ndash is integer type. >>> What is following??????? >>> ndash = 1777b >> >> 1023 >> >> Translate to >> >> ndash = int(o'1777') >> >> -- >> write(*,*) transfer((/17.392111325966148d0,6.5794487871554595D-85, & >> 6.0134700243160014d-154/),(/'x'/)); end

On 3/28/11 7:36 AM, Gordon Sande wrote: > On 2011-03-28 02:50:25 -0300, tolasma said: > >> Thank you very much. >> And There are more weird. >> >> DATA MSK /777777777777777777777B/ >> DATA SIGNBIT/1000000000000000000000B/ >> >> When I compile that as following >> >> ndash =int(o'777777777777777777777') >> >> I've got following message. >> [gkrun@hydra2 temp]$ g77 -o octal octalt.f >> octalt.f: In program `MAIN__': >> octalt.f:2: >> ndash =int(o'777777777777777777777') >> ^ >> Integer at (^) too large >> >> How can I solve this problem? > > 20 "7"s at 3 bits each is 60 bits. Just like a Cray word. The Cray word was 64 bits long, it was CDC that had 60 bit words. > Easiest solution is to use a Cray. :-) Otherwise you > are left with understanding what was intended. That means > knowing enough about both how a Cray worked and what you > problem needs. Both aspects are somewhere between awkward and > very awkward. > > The trailing B means Binary which is really Octal in case > your recipe did not make sense to you. Why B for Octal remains > a mystery that only the original Cray folks can explain. Most > likely a case of choosing a name before they realized the possibility > of generalizations. > It's not a mystery! ;) The Cray people chose "B" because that's what the CDC people choose and they were Cray's biggest competitor. CDC probably choose "B" because it doesn't look like zero; "O" and "0" were tough to distinguish on old line printers ;). Dick Hendrickson

"Gordon Sande" <Gordon.Sande@gmail.com> wrote in message news:impvbn$5k5$1@dont-email.me... | On 2011-03-28 02:50:25 -0300, tolasma said: | | > Thank you very much. | > And There are more weird. | > | > DATA MSK /777777777777777777777B/ | > DATA SIGNBIT/1000000000000000000000B/ | > | > When I compile that as following | > | > ndash =int(o'777777777777777777777') | > | > I've got following message. | > [gkrun@hydra2 temp]$ g77 -o octal octalt.f | > octalt.f: In program `MAIN__': | > octalt.f:2: | > ndash =int(o'777777777777777777777') | > ^ | > Integer at (^) too large | > | > How can I solve this problem? | | 20 "7"s at 3 bits each is 60 bits. Just like a Cray word. There are 21 digits = 63 bits. The second octal constant [10000....b] has 22 digits = 64 bits. Looks like 64-bit words to me. | Easiest solution is to use a Cray. :-) Otherwise you | are left with understanding what was intended. That means | knowing enough about both how a Cray worked and what you | problem needs. Both aspects are somewhere between awkward and | very awkward. Might not need any of that it you have a compiler supporting 64-bit integers. | The trailing B means Binary which is really Octal in case | your recipe did not make sense to you. Why B for Octal remains | a mystery that only the original Cray folks can explain. Most | likely a case of choosing a name before they realized the possibility | of generalizations. The letter 'O' was probably avoided because it was easily confused with zero.

Similiar Articles:

Where did Fortran go? - comp.lang.fortran
Though, I think Cray - or at least the Cray I use ... about a system for building GUI applications in Fortran. C::B ... > So you end up with a code that works with some ...

How to check whether malloc has allocated memory properly in case ...
... was, then, common to put the test at the end ... I agree that the latter dated from Fortran 77, because of the character ... new hardware became available, such as Cray ...

Comparing pointers in Cray Fortran and Fortran 90
Comparing pointers in Cray Fortran and ... 36 END Figure 1 illustrates ... to point to objects of type CHARACTER. An implementer (that is, a writer of a Fortran 90 ...

Calling Fortran Functions and Subroutines from a C or C++ Function
... or underscore characters and consist of 31 or fewer characters. In Cray C, Fortran ... INTS FORTRAN_ADD_INTS, Arg1,Arg2 = 10, 20 Result of FORTRAN Add = 30 End ...

7/9/2012 8:02:00 PM