V5.01A Distribution of RT-11 - VMS

This is a discussion on V5.01A Distribution of RT-11 - VMS ; Does anyone have access to V5.01A of RT-11? I would prefer the RL02 image which has a Volume ID / Owner of: BC-P607C-BC RL2 1/1 However, any other media would be great if the RL02 image is not available! I ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 21

Thread: V5.01A Distribution of RT-11

  1. V5.01A Distribution of RT-11

    Does anyone have access to V5.01A of RT-11? I would prefer
    the RL02 image which has a Volume ID / Owner of:

    BC-P607C-BC
    RL2 1/1

    However, any other media would be great if the RL02 image is
    not available!

    I already have V5.00, V5.01, V5.01C (V5.01B does not exist)
    and V5.02 of RT-11. So it is ONLY V5.01A of RT-11 that
    I do not have. All of these distributions (any prior to V5.03) are
    allowed to be used under the Supnik emulator by hobby users of
    RT-11, so It would be "nice" to be able to have all of the versions
    that are allowed and make them available to other hobby users.

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  2. Re: V5.01A Distribution of RT-11

    Jerome H. Fine wrote:

    > It would be "nice" to be able to have all of the versions
    > that are allowed and make them available to other hobby users.


    Hopefully you are making device images of all these as
    ordinary disk files somewhere and are archiving them on
    CD-R.


  3. Re: V5.01A Distribution of RT-11

    >"Douglas A. Gwyn" wrote:

    > > Jerome H. Fine wrote:
    > > It would be "nice" to be able to have all of the versions
    > > that are allowed and make them available to other hobby users.

    > Hopefully you are making device images of all these as
    > ordinary disk files somewhere and are archiving them on
    > CD-R.


    Jerome Fine replies:

    That is the goal. And since the RL02 is the only standard
    distribution media that is random access and has sufficient
    capacity to hold ALL of the files for the RT-11 binary
    distribution on a single media, that is the form I have chosen
    rather than the RX02 or the RX50 floppy images.

    However, in order to allow as many RL02 images as
    possible to fit into an RT-11 partitions of only 32 MBytes,
    I normally will remove from the RL02 image all of the
    zero blocks which follow all of the files. There is absolutely
    no loss of any information as a result of doing this. In
    addition, the same RT-11 command is used:
    COPY/DEVICE/FILES RLimag.DSK DL0:
    in restoring the file to the RL02 media, although there
    will be a warning message and the remaining blocks will
    NOT be automatically zeroed for the user. But having
    each RT-11 partition hold up to a dozen RT-11 binary
    RL02 distributions rather than ONLY three is an
    excellent advantage.

    But since I also found it to be a challenge to be able to
    used the SAME files for the ISO9660 file structure
    AND for the RT-11 file structure. Since the CD-R is
    the media and it is READ ONLY, I found that it
    is possible to make a copy of the first cut of the
    CD-R to a file on the hard drive. One I had verified
    that the file on the hard drive was the same (using
    RT-11 no less), I added an RT-11 file structure
    for RL02 image of V5.03 of RT-11 in blocks zero
    through eleven. The only change needed to allow
    RT-11 to "see" the files was to add to offset from
    the start of the CD-R image to where the V5.03
    DSK file started - at the fifth word in each RT-11
    file directory segment. SIPP was very useful to
    do this operation. Following that, I also did:
    COPY/BOOT DU0:RT11XM.SYS DU0:
    which allowed the CD-R to be booted on a
    SCSI CDROM drive under RT-11 connected
    to a real PDP-11 or under an emulator that was
    able to look at the whole CDROM drive rather
    than just one of the files.

    After the RT-11 file structure was added to the
    hard disk copy of the CD-R, I just burned a second
    CD-R with the extra information. The rest of the
    blocks were still identical!

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  4. Re: V5.01A Distribution of RT-11

    Jerome H. Fine wrote:
    >>"Douglas A. Gwyn" wrote:

    >
    >
    >>>Jerome H. Fine wrote:
    >>>It would be "nice" to be able to have all of the versions
    >>>that are allowed and make them available to other hobby users.

    >>
    >>Hopefully you are making device images of all these as
    >>ordinary disk files somewhere and are archiving them on
    >>CD-R.

    >
    >
    > Jerome Fine replies:
    >
    > That is the goal. And since the RL02 is the only standard
    > distribution media that is random access and has sufficient
    > capacity to hold ALL of the files for the RT-11 binary
    > distribution on a single media, that is the form I have chosen
    > rather than the RX02 or the RX50 floppy images.
    >
    > However, in order to allow as many RL02 images as
    > possible to fit into an RT-11 partitions of only 32 MBytes,
    > I normally will remove from the RL02 image all of the
    > zero blocks which follow all of the files. There is absolutely
    > no loss of any information as a result of doing this.


    As long as they *really* are zero. I hope you are aware of
    the common practice of putting "unofficial" files on the
    distribution kit, then deleting them. The data is still
    there and the knowledgable user simply had to undelete the
    file (e.g. TECO). Thus, it wasn't part of the official
    distribution and not subject to support.

    Alan


  5. Re: V5.01A Distribution of RT-11

    >Alan Frisbie wrote:

    > > Jerome Fine replies:
    > > That is the goal. And since the RL02 is the only standard
    > > distribution media that is random access and has sufficient
    > > capacity to hold ALL of the files for the RT-11 binary
    > > distribution on a single media, that is the form I have chosen
    > > rather than the RX02 or the RX50 floppy images.
    > > However, in order to allow as many RL02 images as
    > > possible to fit into an RT-11 partitions of only 32 MBytes,
    > > I normally will remove from the RL02 image all of the
    > > zero blocks which follow all of the files. There is absolutely
    > > no loss of any information as a result of doing this.

    > As long as they *really* are zero. I hope you are aware of
    > the common practice of putting "unofficial" files on the
    > distribution kit, then deleting them. The data is still
    > there and the knowledgable user simply had to undelete the
    > file (e.g. TECO). Thus, it wasn't part of the official
    > distribution and not subject to support.


    Jerome Fine replies:

    While I was not aware that placing "unofficial" files was
    a standard policy in that manner (I never noticed it myself
    on an RL02 media), I have ALWAYS checked to make
    sure that this was the case. The simple test was was to:
    DIFF/BIN DL0:/START:zeros_block_number DL0:/START:zeros_block_number+1

    after I did a:
    DUMP/TERM DL0:/ONLY:zeros_block_number
    to check that all of the words were definitely zeros.
    I realize that the above use of BINCOM could NOT
    be done with the DEC distributed copy, so I modified
    my version to allow that ability.

    There are many other ways to do the same thing. But
    I definitely made sure that all of the blocks that were
    removed were all zeros and were after ALL of the files
    on the RL02 image - thus a DSK subset could also
    see all of every file if just one file needed to be copied,
    i.e. the last file.

    If RT-11 had a convenient Zip/UnZip utility, I would
    probably have used that as well since the use of the
    distributions is usually rare enough that the extra blocks
    of a hard disk drive would have been helpful.

    NOW however with such large SCSI drives available for
    even real DEC hardware (YES - I realize that the host
    adapters are still VERY expensive) and file space
    under SIMH being so easily available when a 100 GByte
    drive has become common place (I have used a 40 GByte
    drive myself for over a year and still have only 2 GBytes
    on the C: partition of 31 GBytes and nothing on the other
    partition on a permanent basis), that aspect is usually not
    a problem when all distributions of RT-11 can easily fit
    on a single CD of 650 MBytes. Probably all of the
    binary distributions of RT-11 that were ever released
    on an RK05 media / RL02 media can fit on a total of
    JUST 2 RT-11 partitions or just 64 MBytes (assuming
    that ONLY *really* zero blocks are removed!).

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  6. Re: V5.01A Distribution of RT-11

    On Thu, 02 Oct 2003 10:29:58 -0700, Alan Frisbie
    wrote:

    >Jerome H. Fine wrote:
    >>>"Douglas A. Gwyn" wrote:

    >>
    >>
    >>>>Jerome H. Fine wrote:
    >>>>It would be "nice" to be able to have all of the versions
    >>>>that are allowed and make them available to other hobby users.
    >>>
    >>>Hopefully you are making device images of all these as
    >>>ordinary disk files somewhere and are archiving them on
    >>>CD-R.

    >>
    >>
    >> Jerome Fine replies:
    >>
    >> That is the goal. And since the RL02 is the only standard
    >> distribution media that is random access and has sufficient
    >> capacity to hold ALL of the files for the RT-11 binary
    >> distribution on a single media, that is the form I have chosen
    >> rather than the RX02 or the RX50 floppy images.
    >>
    >> However, in order to allow as many RL02 images as
    >> possible to fit into an RT-11 partitions of only 32 MBytes,
    >> I normally will remove from the RL02 image all of the
    >> zero blocks which follow all of the files. There is absolutely
    >> no loss of any information as a result of doing this.

    >
    >As long as they *really* are zero. I hope you are aware of
    >the common practice of putting "unofficial" files on the
    >distribution kit, then deleting them. The data is still
    >there and the knowledgable user simply had to undelete the
    >file (e.g. TECO). Thus, it wasn't part of the official
    >distribution and not subject to support.


    One of the early magtape distributions we got had source past EOT...
    they must have been reusing tapes.


    --
    Ian
    Impressive If Haughty - Q Magazine

  7. Re: V5.01A Distribution of RT-11

    >paramucho wrote:

    > One of the early magtape distributions we got had source past EOT...
    > they must have been reusing tape.


    Jerome Fine replies:

    While reading files past the EOT file marks would take
    some non-standard utilities for the first file, once the
    first set of blocks had been read, the rest of the files
    could be copied with:
    COPY/SYS/QUERY TAP:*.*/POS:-1 DK:*.*

    The key point would be to NOT rewind the tape before
    starting to read the next file. If all of the files were to
    be copied, just leave out the "/QUERY" switch.

    NOTE that some duplicates may be found in this
    manner. If the originals should be retained, then
    use the "/NOREPLACE" switch. Otherwise, first
    "UNPROTECT" all the files on DK: and use the
    "/NOPROTECT" switch during the COPY operation
    so that any duplicate files can be replaced by the latest
    version.

    NOTE: PLEASE - Still looking for V5.01A of RT-11.

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  8. Re: V5.01A Distribution of RT-11

    In article <3F82C7FF.9001FE8F@b9rvnospamcompsys.to>,
    "Jerome H. Fine" wrote:
    >>paramucho wrote:

    >
    >> One of the early magtape distributions we got had source past EOT...
    >> they must have been reusing tape.

    >
    >Jerome Fine replies:
    >
    >While reading files past the EOT file marks would take
    >some non-standard utilities for the first file, once the
    >first set of blocks had been read, the rest of the files
    >could be copied with:
    >COPY/SYS/QUERY TAP:*.*/POS:-1 DK:*.*
    >
    >The key point would be to NOT rewind the tape before
    >starting to read the next file. If all of the files were to
    >be copied, just leave out the "/QUERY" switch.
    >
    >NOTE that some duplicates may be found in this
    >manner. If the originals should be retained, then
    >use the "/NOREPLACE" switch. Otherwise, first
    >"UNPROTECT" all the files on DK: and use the
    >"/NOPROTECT" switch during the COPY operation
    >so that any duplicate files can be replaced by the latest
    >version.


    Note that you may be restoring a file that was broken and fixed
    later. If you guys are taking erased bits off distribution tapes,
    you're asking for a ton of troubles. Sheesh! This is why
    TOPS-10 had the rule of making its Software Distribution Tapes
    on magtapes out of the box. It was against our packaging rules
    to reuse a tape that was going to go to the SDC.

    /BAH

    Subtract a hundred and four for e-mail.

  9. Re: V5.01A Distribution of RT-11

    >"Douglas A. Gwyn" wrote:

    > > It would be "nice" to be able to have all of the versions
    > > that are allowed and make them available to other hobby users.

    > Hopefully you are making device images of all these as
    > ordinary disk files somewhere and are archiving them on
    > CD-R.


    Jerome Fine replies:

    While that will be my final goal, right now I just want to
    be able to re-produce "pristine distributions". As Tim
    Shoppa pointed out, for the RL02 images, the correct
    Volume ID / Owner should also be used.

    Also, I want to thank MOSES for providing access to
    both V5.01B and V5.01C of RT-11. THANK YOU!!

    Well, I think I have solved what seemed to be a mystery
    (to me) for a number of years. The answer seems to be:


    Date RL02 Volume ID
    ----- -------------------
    V5.00 - 12-Mar-1983 - BC-P607A-BC
    V5.01 - 01-Feb-1984 - BC-P607B-BC
    V5.01A - missing - does not seem to exist
    V5.01B - 19-Mar-1984 - BC-P607C-BC
    V5.01C - 01-Sep-1984 - BC-P607D-BC
    V5.02 - 17-Jun-1985 - BC-P607E-BC
    V5.03 - 20-Dec-1985 - BC-P607F-BC
    V5.04 - 03-Sep-1986 - BC-P607G-BC

    As far as I can tell, all V5.01 were composed of 198 files.
    The V5.01B that I just acquired access to had 154 files
    which were the same and 44 files which were different.
    Further, since there were ONLY 47 days between the
    release of V5.01 and V5.01B, it seems VERY HIGHLY
    probable that there NEVER was a V5.01A that was
    released by DEC.

    SO!!!! I conclude that I was totally incorrect!!!!!!!!!!!!!!
    V5.01A is the missing distribution and V5.01B DOES EXIST!

    At this point, I would appreciate some help in regard to JUST
    one aspect. While I realize that the actual order of the files
    within an RL02 image is not important, I would VERY MUCH
    like to have the original order to the files. For the images that
    I now have, the order of the monitors differs from the order that
    the same files appear in V5.00 and V5.02 of RT-11. I suspect
    that it is likely that I have the wrong order on the RL02 images.
    If anyone might have a copy of the MAP KITS for any of the
    RT-11 distributions from the V5.0x distributions, especially for
    the distributions listed above, I would appreciate a copy so
    that the RL02 images that I now have access to can be verified
    and corrected if need be.

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  10. Re: V5.01A Distribution of RT-11

    In article <3F84C954.42DF3B12@b9rvnospamcompsys.to>,
    "Jerome H. Fine" wrote:
    >>"Douglas A. Gwyn" wrote:

    >
    >> > It would be "nice" to be able to have all of the versions
    >> > that are allowed and make them available to other hobby users.

    >> Hopefully you are making device images of all these as
    >> ordinary disk files somewhere and are archiving them on
    >> CD-R.

    >
    >Jerome Fine replies:
    >
    >While that will be my final goal, right now I just want to
    >be able to re-produce "pristine distributions". As Tim
    >Shoppa pointed out, for the RL02 images, the correct
    >Volume ID / Owner should also be used.
    >
    >Also, I want to thank MOSES for providing access to
    >both V5.01B and V5.01C of RT-11. THANK YOU!!
    >
    >Well, I think I have solved what seemed to be a mystery
    >(to me) for a number of years. The answer seems to be:
    >
    >
    > Date RL02 Volume ID
    > ----- -------------------
    >V5.00 - 12-Mar-1983 - BC-P607A-BC
    >V5.01 - 01-Feb-1984 - BC-P607B-BC
    >V5.01A - missing - does not seem to exist
    >V5.01B - 19-Mar-1984 - BC-P607C-BC


    Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    What this tells you is that the Software Distribution Center
    never got a submission for V5.01A. Also look at the dates
    of submission (I'll trust that you are picking up the correct
    dates). Since there is only six weeks' difference, my guess
    is that the B was a beginning of shipping what the -10 called
    Autopatch service, which was a quarterly update of the last
    major release including all fixes. It means that fixes only
    got a unit test and the package didn't go through the full-fledged
    field test. Just getting a sign-off for submission to SDC would
    take three months.

    /BAH




    /BAH

    Subtract a hundred and four for e-mail.

  11. Re: V5.01A Distribution of RT-11

    >jmfbahciv@aol.com wrote:

    > >Well, I think I have solved what seemed to be a mystery
    > >(to me) for a number of years. The answer seems to be:
    > > Date RL02 Volume ID
    > > ----- -------------------
    > >V5.00 - 12-Mar-1983 - BC-P607A-BC
    > >V5.01 - 01-Feb-1984 - BC-P607B-BC
    > >V5.01A - missing - does not seem to exist
    > >V5.01B - 19-Mar-1984 - BC-P607C-BC

    > Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    > What this tells you is that the Software Distribution Center
    > never got a submission for V5.01A. Also look at the dates
    > of submission (I'll trust that you are picking up the correct
    > dates). Since there is only six weeks' difference, my guess
    > is that the B was a beginning of shipping what the -10 called
    > Autopatch service, which was a quarterly update of the last
    > major release including all fixes. It means that fixes only
    > got a unit test and the package didn't go through the full-fledged
    > field test. Just getting a sign-off for submission to SDC would
    > take three months.


    Jerome Fine replies:

    I left out ONE extremely important point when I presented
    the above information. The "RL02 Volume ID" is NOT
    necessarily correct, although if an inspection is made of
    the file STARTA.COM, there is substantial confirmation
    that the "letter" after P607 is the "letter" that I have used.

    The problem is that the "Volume ID" symbols that are
    defined in STARTA.COM for the RL02 media are
    probably incorrect for the RL02 (as far as I am able
    to figure out) for V5.01 and V5.01B of RT-11. For
    these two distributions, the STARTA.COM file shows
    that the Volume ID is identical to the first RX02 volume.
    However, while the numbers are different, the "letter"
    that is used is the same in each case. Thus there is
    some confirmation that the correct "RL02 Volume ID"
    is what I have displayed for RL02 distributions V5.01
    and V5.01B of RT-11. The RL02 image that I now
    have for RL02 distributions of V5.01B and V5.01C
    were produced from RX02 images. It is also possible
    that the RL02 distribution of V5.01 of RT-11 was also
    produced from RX02 images since I can't ever remember
    not having access to V5.01 of RT-11, so I have
    had access to it for more than a decade.

    Thus the conclusions about the dates for the releases
    V5.01 and V5.01B made by jmfbahciv are valid, as
    far as I am concerned, since I have no reason to doubt
    the dates on the files.

    NOW!!!! There is still one aspect which is, in my opinion,
    VERY important - the order of the files in the RT-11
    directory may not be correct for those RL02 distributions
    which were produced by making copies of the files from
    RX02 images or in one case from a TK50 image. In all
    cases, the actual order of the files was made to match the
    order of the files from a known RL02 image that was usually
    the one immediately before or after. However, it is still
    very possible that the RL02 images produced from other
    media may have an order different from a DEC "pristine
    distribution" of that particular version. If ANYONE
    has a MAP KIT for any of the versions at:
    http://www.classiccmp.org/PDP-11/RT-11/dists/
    it would be appreciated if the order could be checked
    against the directory for an RL02 and corrected.
    Please send a private e-mail or post that you have
    an RL02 MAP KIT from DEC!

    If you have any other questions, please ask!!!!!!!!!!

    ONE final request - I am able to use PUTR to produce
    a file which is 2,147,450,880 bytes of all zeros. BUT
    I actually want a file which is 2,147,483,648 bytes or
    exactly 2 GBytes! When I used bzip2.exe to compress
    the 2,147,450,880 bytes, the result was only 1518 bytes.

    So if you can produce the 2,147,483,648 byte file of
    all zeros, the compressed file via bzip2.exe should not
    be very large. I called the file BIGZERO.DSK and the
    compressed file was BIGZERO.DSK.bz2

    Recently, I received a compressed image of a file that
    is all zeros with 2,147,483,647 bytes or ONE byte
    short. I want to be able to use these files with the
    Ersatz-11 emulator and have EXACTLY 64 RT-11
    partitions. The above file shows (with DUSTAT -
    thank you Tim Shoppa) that it is one block short
    because Ersatz-11 truncates files which are not
    a multiple of 512 bytes. I am assuming that if
    a FAT32 file system under Windows 98 SE can
    handle a file of 2,147,483,647 bytes, then a
    file of EXACTLY 2,147,483,648 bytes is also
    possible. Can anyone help????????????????

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  12. Re: V5.01A Distribution of RT-11

    jmfbahciv@aol.com wrote:
    > Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    > What this tells you is that the Software Distribution Center
    > never got a submission for V5.01A.


    I seem to recall that there was also an RT11V3 and RT1V3B
    but not RT11V3A. Apparently the second in the series got
    the "B" tag.


  13. Re: V5.01A Distribution of RT-11

    In article ,
    "Douglas A. Gwyn" wrote:
    >jmfbahciv@aol.com wrote:
    >> Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    >> What this tells you is that the Software Distribution Center
    >> never got a submission for V5.01A.

    >
    >I seem to recall that there was also an RT11V3 and RT1V3B
    >but not RT11V3A. Apparently the second in the series got
    >the "B" tag.


    Interesting. If true, there may have been a reason to skip
    the A name; it may have become tradition. The real way
    to find the info of how things were packaged and why is to
    talk to the gal or guy who did the work. I don't know if
    RT-11 group had a bona fide Release Engineering group like
    the PDP-10s had. But, the quarterly release of V5.01 points to
    the Autopatch method which was becoming the way of all things
    in the 80s at DEC. Do you recall if the positioning of files
    on a piece of medium important to the installation of the OS?
    The OP seems obsessed with order of files; with few exceptions,
    the order of the bits on a disk were not important with PDP-10 OSes.
    I don't know enough about the internals of RT-11 to judge if
    it is important.

    /BAH



    Subtract a hundred and four for e-mail.

  14. Re: V5.01A Distribution of RT-11

    jmfbahciv@aol.com wrote in message news:...
    > Do you recall if the positioning of files
    > on a piece of medium important to the installation of the OS?


    For RT-11 on a floppy or tape distribution, it was. The floppies were
    too small to hold the whole distribution, but were big enough to boot
    into something useful to install (or even SYSGEN - boy it's been a long
    time since I SYSGEN'ed from floppies) RT-11. For the tape distributions,
    it was important to put the most vital files at the start of the tape, and
    after those were copied to disk then the disk was booted from and the
    remainder installed.

    RT-11 had a data file and an intricate command procedure for preparing
    the command procedures that made the distribution kits. ("I'd rather
    write programs that write programs than write prograsm".) The data file
    also contained the volume labels for each different media.

    Tim.

  15. Re: V5.01A Distribution of RT-11

    In article ,
    shoppa@trailing-edge.com (Tim Shoppa) wrote:
    >jmfbahciv@aol.com wrote in message news:...
    >> Do you recall if the positioning of files
    >> on a piece of medium important to the installation of the OS?

    >
    >For RT-11 on a floppy or tape distribution, it was. The floppies were
    >too small to hold the whole distribution, but were big enough to boot
    >into something useful to install (or even SYSGEN - boy it's been a long
    >time since I SYSGEN'ed from floppies) RT-11. For the tape distributions,
    >it was important to put the most vital files at the start of the tape, and
    >after those were copied to disk then the disk was booted from and the
    >remainder installed.


    Puzzling. It would take just as much time to restore the rest of
    the files when booted from tape as booted from disk...ooohhhh! I see.
    Partition code. The boot device is hide-bound. No, that can't be
    it because it couldn't do segment retrievals from tape.

    >
    >RT-11 had a data file and an intricate command procedure for preparing
    >the command procedures that made the distribution kits.


    sure. Everybody's packaging did. But to foist it onto all
    customers is lunacy.

    > .. ("I'd rather
    >write programs that write programs than write prograsm".) The data file
    >also contained the volume labels for each different media.


    Bletch! Do you know if there was technical reason to tie a
    SDC cybercurd number to all installations? Tying a part
    number to an installation would open all kinds of CATCH-22s
    in a Pandora's box. We had enough troubles with the SDC
    naming system and BOMs without putting it into code!

    /BAH


    /BAH

    Subtract a hundred and four for e-mail.

  16. Re: V5.01A Distribution of RT-11

    >"Jerome H. Fine" wrote:

    > ONE final request - I am able to use PUTR to produce
    > a file which is 2,147,450,880 bytes of all zeros. BUT
    > I actually want a file which is 2,147,483,648 bytes or
    > exactly 2 GBytes! When I used bzip2.exe to compress
    > the 2,147,450,880 bytes, the result was only 1518 bytes.


    Jerome Fine replies:

    Problem solved - the compressed file is 1522 bytes.

    I had some help, but the solution was obvious after the fact!

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  17. Re: V5.01A Distribution of RT-11

    >jmfbahciv@aol.com wrote:

    > >>jmfbahciv@aol.com wrote:
    > >> Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    > >> What this tells you is that the Software Distribution Center
    > >> never got a submission for V5.01A.

    > >I seem to recall that there was also an RT11V3 and RT1V3B
    > >but not RT11V3A. Apparently the second in the series got
    > >the "B" tag.

    > Interesting. If true, there may have been a reason to skip
    > the A name; it may have become tradition. The real way
    > to find the info of how things were packaged and why is to
    > talk to the gal or guy who did the work. I don't know if
    > RT-11 group had a bona fide Release Engineering group like
    > the PDP-10s had. But, the quarterly release of V5.01 points to
    > the Autopatch method which was becoming the way of all things
    > in the 80s at DEC. Do you recall if the positioning of files
    > on a piece of medium important to the installation of the OS?
    > The OP seems obsessed with order of files; with few exceptions,
    > the order of the bits on a disk were not important with PDP-10 OSes.
    > I don't know enough about the internals of RT-11 to judge if
    > it is important.


    Jerome Fine replies:

    Sorry to disrupt your line of discussion, but after:
    V5.04 - 03-Sep-1986
    V5.04A - 05-Jan-1987
    V5.04B - 07-May-1987
    V5.04C - 02-Sep-1987
    V5.04D - 14-Nov-1987
    V5.04E - 01-May-1988
    V5.04F - 06-Sep-1988
    V5.04G - 19-Dec-1988
    V5.05 - 31-Oct-1989
    V5.06 - 31-Aug-1992

    As for the autopatch method, it was used to get from
    version to version if the system subscribed to the update
    service (as far as I am aware). But if a new system and
    licence and distribution were ordered, then the above
    distributions were explicitly produced for new users.

    By the time that V5.0x versions were produced, ALL
    monitor files in RT-11 contained an ASCII text string
    in block 4 of the file which was almost always displayed
    (there was a switch which could turn that off) during the
    start of the boot operation. In addition, the command:
    SHOW CONFIGURATION
    could also be used at any time to display the text string.

    So while it is entirely possible that the "A" versions
    were skipped in RT-11 releases prior to V5.04A,
    either someone forgot the tradition or it was finally
    abandoned. On the other hand, it also seems to have
    been the ONLY release that I have confirmed with an
    "A" letter.

    Again, with regard to the order of the files for both an
    RK05 and an RL01 / RL02 image, it makes absolutely
    no difference. However, for a "pristine distribution"
    (one that came from DEC originally) it would be "nice"
    to have the same order. Can anyone verify that the order
    of the files in the distributions that are available to run
    under the SIMH emulator that are available at:
    http://www.classiccmp.org/PDP-11/RT-11/dists/
    are correct, in particular V4.00, V5.00 and V5.01 of
    RT-11?

    In the next while, it is hoped to have V5.01B, V5.01C,
    V5.02 and V5.03 there as well. In particular, the order
    of the files for V5.01B and V5.01C will be the same as
    for V5.01 - so if that order is incorrect, V5.01B and
    V5.01C will also have the wrong order.

    Sincerely yours,

    Jerome Fine
    --
    To obtain the original e-mail address, please remove
    the ten characters which immediately follow the 'at'.
    If you attempted to send a reply and the original e-mail
    address has been discontinued due a high volume of junk
    e-mail, then the semi-permanent e-mail address can be
    obtained by replacing the four characters preceding the
    'at' with the four digits of the current year.


  18. Re: V5.01A Distribution of RT-11

    In article <3F934E50.1ED11082@b9rvnospamcompsys.to>,
    "Jerome H. Fine" wrote:
    >>jmfbahciv@aol.com wrote:

    >
    >> >>jmfbahciv@aol.com wrote:
    >> >> Look at the part numbers. V5.01 is 7B; V5.01B is 7C.
    >> >> What this tells you is that the Software Distribution Center
    >> >> never got a submission for V5.01A.
    >> >I seem to recall that there was also an RT11V3 and RT1V3B
    >> >but not RT11V3A. Apparently the second in the series got
    >> >the "B" tag.

    >> Interesting. If true, there may have been a reason to skip
    >> the A name; it may have become tradition. The real way
    >> to find the info of how things were packaged and why is to
    >> talk to the gal or guy who did the work. I don't know if
    >> RT-11 group had a bona fide Release Engineering group like
    >> the PDP-10s had. But, the quarterly release of V5.01 points to
    >> the Autopatch method which was becoming the way of all things
    >> in the 80s at DEC. Do you recall if the positioning of files
    >> on a piece of medium important to the installation of the OS?
    >> The OP seems obsessed with order of files; with few exceptions,
    >> the order of the bits on a disk were not important with PDP-10 OSes.
    >> I don't know enough about the internals of RT-11 to judge if
    >> it is important.

    >
    >Jerome Fine replies:
    >
    >Sorry to disrupt your line of discussion, but after:
    >V5.04 - 03-Sep-1986
    >V5.04A - 05-Jan-1987
    >V5.04B - 07-May-1987
    >V5.04C - 02-Sep-1987
    >V5.04D - 14-Nov-1987
    >V5.04E - 01-May-1988
    >V5.04F - 06-Sep-1988
    >V5.04G - 19-Dec-1988
    >V5.05 - 31-Oct-1989
    >V5.06 - 31-Aug-1992
    >
    >As for the autopatch method, it was used to get from
    >version to version if the system subscribed to the update
    >service (as far as I am aware).


    Autopatch was a software update service. This was so those
    who did not have access to sources could apply patches that
    solved show-stopping bugs to field image software. It was
    not used to get from version to version (Version has a
    significant meaning in DEC culture).


    > ... But if a new system and
    >licence and distribution were ordered, then the above
    >distributions were explicitly produced for new users.


    Huh? No they weren't. There existed no remastering for
    each new order. You're beginning to irritate me.

    >
    >By the time that V5.0x versions were produced, ALL
    >monitor files in RT-11 contained an ASCII text string
    >in block 4 of the file which was almost always displayed
    >(there was a switch which could turn that off) during the
    >start of the boot operation. In addition, the command:
    >SHOW CONFIGURATION
    >could also be used at any time to display the text string.


    Fine. Somebody added a tweak to the packaging. It was
    probably added so that customers could punt the distribution
    without installing it...I hope.
    >
    >So while it is entirely possible that the "A" versions
    >were skipped in RT-11 releases prior to V5.04A,
    >either someone forgot the tradition or it was finally
    >abandoned. On the other hand, it also seems to have
    >been the ONLY release that I have confirmed with an
    >"A" letter.


    So what? IT'S A ****ING PART NUMBER. Somebody thought
    it might be neat to tie the part number with the autopatch
    revision letter assignment.

    >
    >Again, with regard to the order of the files for both an
    >RK05 and an RL01 / RL02 image, it makes absolutely
    >no difference.


    Oh, good grief. If it makes no difference, why are you wasting
    so much time?

    > ..However, for a "pristine distribution"
    >(one that came from DEC originally) it would be "nice"
    >to have the same order.


    You are never ever going to reproduce a "pristine distribution"
    without the exact same tools, systems, and people who produced
    them.



    You people should be more interested in file integrity rather
    than prettiness.

    /BAH

    Subtract a hundred and four for e-mail.

  19. Re: V5.01A Distribution of RT-11

    jmfbahciv@aol.com wrote:
    >> I seem to recall that there was also an RT11V3 and RT1V3B
    >> but not RT11V3A. Apparently the second in the series got
    >> the "B" tag.

    > Interesting. If true, there may have been a reason to skip
    > the A name; it may have become tradition. The real way


    IIRC there was a V2A, I have the manual somewhere.

    Christian

  20. Re: V5.01A Distribution of RT-11

    In article <4vik61-p45.ln1@news.online.de>,
    Christian Corti wrote:
    >jmfbahciv@aol.com wrote:
    >>> I seem to recall that there was also an RT11V3 and RT1V3B
    >>> but not RT11V3A. Apparently the second in the series got
    >>> the "B" tag.

    >> Interesting. If true, there may have been a reason to skip
    >> the A name; it may have become tradition. The real way

    >
    >IIRC there was a V2A, I have the manual somewhere.


    It's certainly possible. My take on the skippage is that
    there was a set of sources set aside which were denoted as
    the "A" set. This could be the set that was used by a customer
    or outhouse field office to do their own hacks. Quite a number
    of field offices did their own development because they had
    a set of customers that needed code which (for whatever reason)
    couldn't be put into the general distribution. There are many
    reasons this was done: local modification service, proprietary
    mods, etc. Managing sources was a complicated business and
    keeping sets separated was easier if each had their own
    minor version number.

    My conjecture is further supported by the fact that A appeared in
    1987 which was when field offices were generally put out of
    business. There wasn't a need to have a separate minor version
    number for bookkeeping. Source management was not simple when
    you had multiple development efforts for the same product.

    /BAH

    Subtract a hundred and four for e-mail.

+ Reply to Thread
Page 1 of 2 1 2 LastLast