0 INFO-VAX	Fri, 06 Feb 2004	Volume 2004 : Issue 72      Contents: carly's new compensation plan   Re: DECSet documentation online?) Re: HOW MANY BPI DOES A DLT7000 WRITE AT? P Re: Intel and Microsoft provide higher 32-bit applications performance         o? Re: Is Encompass a member of Interex (was : Hobbyist questions) ) Re: LK401-AA keyboard - no PS2 connector. ) Re: LK401-AA keyboard - no PS2 connector. - OpenGL Drawing, Motif on PseudoColor, ZLXp-E1 1 Re: OpenGL Drawing, Motif on PseudoColor, ZLXp-E1 G OpenVMS Pearl - OK for external use - another mixed architecture cluste C Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ????? C Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ????? C Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ????? # Re: Problems with File Access Dates C Re: SYS$STARTUP:MMOV$SHUTDOWN.COM -- Insufficiently discriminating? C Re: SYS$STARTUP:MMOV$SHUTDOWN.COM -- Insufficiently discriminating?   Re: VMS ODS-5 and Macintosh OS X  Re: VMS ODS-5 and Macintosh OS X Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ?  F ----------------------------------------------------------------------  # Date: Thu, 05 Feb 2004 20:45:53 GMT # From: "John Smith" <a@nonymous.com> & Subject: carly's new compensation planI Message-ID: <5YxUb.4950$UMB.3341@twister01.bloor.is.net.cable.rogers.com>   L http://quote.bloomberg.com/apps/news?pid=10000039&refer=columnist_crystal&si d=aXzdcu8XeDxQ    ? It's actually not too bad from a shareholder cash perspective..   I However from a product perspective (VMS in particular) it sets no special D goals... cash flow from operations can be boosted by NOT spending onD increasing market share for specific product lines. Since HP alreadyI considers VMS to be a marginal product, there is no incentive to increase  its visibility or sales.  H Now one might argue that Gorham, et. al. have that sort of incentive forI their fiefdoms, but when the head cheese has only an indirect interest in I the outcome of individual product lines it becomes easy to favor one over I another in the executive suite and at the BOD, to the detriment of Gorham  et. al's bottom lines.   ------------------------------   Date: 5 Feb 2004 13:53:24 -0600 4 From: kuhrt@nospammy.encompasserve.org (Marty Kuhrt)) Subject: Re: DECSet documentation online? 3 Message-ID: <VOOB3IMSGhz8@eisner.encompasserve.org>   a In article <K-ydnQ42CJTFPrzdRVn-jw@adelphia.com>, "John E. Malmberg" <wb8tyw@qsl.network> writes:  > Marty Kuhrt wrote:B >> Does anyone know if the DECset tools have online documentation?C >> I've searched high and low and found nothing.  The documentation < >> with regards to DECset on the hp site is terse at best... > + > http://h71000.www7.hp.com/doc/decset.html  > % > HTML and PDF formats are available.  >    Perfect, thanks!    8 Odd that there isn't a link to it on the DECset pages...  = http://h71000.www7.hp.com/commercial/decset/decset_index.html    ------------------------------  % Date: Thu, 05 Feb 2004 15:20:32 -0700 4 From: Norman Lastovica <norman.lastovica@oracle.com>2 Subject: Re: HOW MANY BPI DOES A DLT7000 WRITE AT?* Message-ID: <4022C1B0.EA53CA44@oracle.com>  7 get in touch with oracle rdb support directly.  current 8 releases have been enhanced to do a much better job with tape density handling.  	 MB wrote:  >  > "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com> wrote in message news:<11721EF39C7D7F47A55447158274CAF79A39C9@cossmgmbx01.email.corp.tld>... 
 > > MB wrote:  > > 4 > > > I am using Rdb(7.0) and trying to do a backup. > >  ..snip...7 > > > I have init the tape "Sony DLTtape iv tape " with  > > > /media=compaction but rdb  > > > seems to ignore it.  > > G > > Don't use any density.  As you say, INIT it at the VMS level.  Once I > > inited, mounts will default to the initialized values.  You shouldn't 4 > > have to tell Rdb explicitly what the density is. > > -- > > Brian Tillman  > > Smiths Aerospace" > > 3290 Patterson Ave. SE, MS 1B3 > > Grand Rapids, MI 49512-1991 B > > Brian.Tillman is the name, smiths-aerospace.com is the domain. > > > > > I don't speak for Smiths, and Smiths doesn't speak for me. > > . > > ******************************************J > > The information contained in, or attached to, this e-mail, may containG > >  confidential information and is intended solely for the use of the J > >  individual or entity to whom they are addressed and may be subject toK > >  legal privilege.  If you have received this e-mail in error you should K > >  notify the sender immediately by reply e-mail, delete the message from O > >  your system and notify your system manager.  Please do not copy it for any I > >  purpose, or disclose its contents to any other person.  The views or L > >  opinions presented in this e-mail are solely those of the author and doJ > >  not necessarily represent those of the company.  The recipient shouldL > >  check this e-mail and any attachments for the presence of viruses.  TheD > >  company accepts no liability for any damage caused, directly or8 > >  indirectly, by any virus transmitted in this email.. > > ****************************************** > " > Don't agree with that statement. > F > I have a database backup on disk which is 62,000,000 Blocks which is > about 31GBE > When  I try to use RMU/BACKUP to tape it ignores the density of the  > tape and wants" > three 70 GB tapes (yes three)... > G > I do a rmu/backup to disk first and copy it the tape I can squeeze on E > the same tape, two 31GB databases. So at this point the tape should C > have some 8GB of free space on it as I am using a Sony "DLTapeIV"  > cartridge in a DLT7000 drive.  >  > This is what I have to do. > * > $ init VMS1$MKE200: /MED=COMPACTION 0001  > $ mount/over=id/med=compaction" > $ copy database.bck vms1$mke200: > D > For me it looks like Rdb is not writing at the correct compression > rate or denisty. > ) > This is the spec on the sony cartridge.  > 20/40 GB ON DLT4000  > 35/40 GB ON DLT7000  > 40/80 GB ON DLT8000  > assuming 2:1 compression > Length 1,828 feet    --  	 - - - - - 0  opinions expressed here are mine and mine alone.  and certainly are not intended in any way to 0  express or represent any opinions or commitment  of oracle corporation.   *  norman lastovica / oracle rdb engineering   ------------------------------  # Date: Fri, 06 Feb 2004 05:35:16 GMT " From: GreyCloud <mist@Cumulus.com>Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance         o 7 Message-ID: <oIFUb.625$ss.21855@bcandid.telisphere.com>    Tom Linden wrote:   G > No offense inteded, Keith, but I fail to see what this has to do with H > VMS.  Moreover, If I want to run W2K, why spend the extra $s for IA64? >   I Good point.  Besides, M$ NT on Alpha systems never could compete against  H   VMS or Tru64 UNIX anyway.  If he wants 64-bit PC its a lot cheaper to * get the new HP a450e AMD64 for around $1k.   ------------------------------  % Date: Fri, 06 Feb 2004 00:19:58 +0100 " From: Didier Morandi <no@spam.com>H Subject: Re: Is Encompass a member of Interex (was : Hobbyist questions)2 Message-ID: <4022cfa3$0$274$636a15ce@news.free.fr>   ok.    Drop it.   D.   Clay M. Denton wrote:   R > I don't want to have a discussion over any organizations use of marketing terms. > N > The 3 organizations you have listed are separate corporations, with separateO > membership bases, representing different portions of the "HP User Community".  >  > Clay > I > On Wed, 04 Feb 2004 21:44:56 +0100, Didier Morandi <no@spam.com> wrote:  >  >  >>http://www.encompassus.org/   >>"Encompass, an HP User Group". >>" >>http://www.interex.org/home.htmlL >>"Interex, the International HP Customer Community, is an independent, not Q >>for-profit association providing information, education, and advocacy services  ! >>to members all over the world".  >> >>http://www.hp-interex.org/L >>"HP-Interex EMEA, the newly-formed federation of HP user groups in Europe R >>resulting from the merger of the Compaq Users Organisation EMEA (formerly DECUS L >>Europe) and interex Europe (the HP Enterprise User Group), represents the R >>largest body of Enterprise System Users in the EMEA region, uniting over 70 000  >>IT professionals". >>! >>EMEA today, Worldwide tomorrow. M >>Tell us how an HP User Group could not be a member of the International HP   >>Customer Community, Clay.  >> >>D. >> >> >>Clay M. Denton wrote:  >>9 >>>Encompass and Interex are VERY separate organizations.  >>>  >>>Clay Denton >>>Director  >>>Encompass US, Inc.  >>>  >  >    --  2 VAXUS - Your new helpful friend in the DEC Family!2 EHQ: 19 chemin de la Butte, 31400 Toulouse, France/       Phone: +336 7983 6418 Fax: +335 6154 1928 $                 http://www.vaxus.org   ------------------------------  % Date: Thu, 05 Feb 2004 13:28:34 -0800 , From: Ken Fairfield <My.Full.Name@intel.com>2 Subject: Re: LK401-AA keyboard - no PS2 connector.+ Message-ID: <bvuci3$jmi$1@news01.intel.com>    Paul Sture wrote: F > My LK461-A2, with PS2 plug is dead. I have a spare LK401-AA keyboardH > which is like new, but unfortunately has an MMJ? style plug which fits  > my trusty old VAXstation 3100. [...]   8     Seems to me just about every AlphaServer was shipped< with a keyboard, whether it was needed or not (usually not).< There must be tons of unused LK461 and LK462 keyboards to be8 had for the taking (they're only about US$50 _list_ from< HP/Compaq/Digital).  I saved two brand new LK462's, still in< bubble pack, from going to the dumpster for use at home (one: as a spare for 10 years hence when the first one dies :-).   	-Ken  --  6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield ! D1C Automation VMS System Support " who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------  * Date: Thu, 5 Feb 2004 23:14:54 +0000 (UTC)6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)2 Subject: Re: LK401-AA keyboard - no PS2 connector.1 Message-ID: <newscache$3jwmsh$5w22$1@news.sil.at>   ] In article <402235D7.612B7424@sture.homeip.net>, Paul Sture <nospam@sture.homeip.net> writes: E >My LK461-A2, with PS2 plug is dead. I have a spare LK401-AA keyboard G >which is like new, but unfortunately has an MMJ? style plug which fits  >my trusty old VAXstation 3100.   9 And in VT4xx and VT5xx and VAXstation 4000 and DEC3000...   D >Before I try taking the keyboards apart to see if I can simply swap& >cables, I though I'd try asking here. > H >Has anyone managed to do this, or I do I need to hunt down a converter?A >And yes, I've had a look at the VMS FAQ, but didn't see anything 	 >obvious.   I I don't see any chance for swapping cables (or for a converter). They are F totally different inside. An USB-to-PS2 converter is only a mechanicalG device, the electronic inside the keyboard supports both interfaces and I switches the mode. You currently find only USB-to-PS2 converters, not the H other way round, ie the most effective way to avoid people trying to use* older non-usb keyboards on USB interfaces.  M As for the other keyboard interface type (MMJ-to-PS2 converter), I don't know  but I doubt, too.    -Peter  J PS: In case anyone has spare (108key US style) LK463, LK462, LK461, LK411.L I do need 3 of them and I want to avoid buying new ones for over $100/piece.( Best would be 2 PS2 and 1 USB (or 3 PS2) --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   Date: 5 Feb 2004 18:23:24 -0800 / From: stuie_norris@yahoo.com.au (Stuart Norris) 6 Subject: OpenGL Drawing, Motif on PseudoColor, ZLXp-E1< Message-ID: <51262235.0402051823.c612ab7@posting.google.com>  
 Dear Readers,   A I am attempting to get an OpenGL / X / Motif program to work on a  hobbist A DEC Alpha running Open VMS.  I have a Renderer = ZLXp-E1, which I 
 understand> has just a single X colormap.  (from run sys$system:xglinfo)    @ =========== 10 Visuals for Screen 0 (default = 0x22) ===========; PseudoColor visual: ID = 0x22 (hex) 34 (decimal) screen = 0 :   DOUBLE buffered MONO COLOR INDEX visual with (Z Stencil))   GL Sizes: ColorIndex=8, Z=24, Stencil=4 #   Core X: depth=8, colormapSize=256   A If I let OpenGL choose the colormap I end up with the technicolor  screenC and I wish to get around this.  I have tried to install the default  colormapA for the screen into the creation of the drawing area widget.  See  attached code.   A I have tried to modify my program based on gl_ivp.c, but I cannot D figure out how to get it to work with Motif OpenGL drawing areas.  I! am at a loss as to what is wrong.      : I am using Open VMS 6.2. (The only media I have access to)  F What am I doing wrong here that is causing the program to not draw the/ strange colored window I am trying to get??????    Thanks   Stuart  7 ------------------------opengl.c-----------------------  #include <stdio.h> #include <stdlib.h>  #include <Xm/MainW.h>  #include <GL/gl.h> #include <GL/glu.h>  #include <GL/glx.h>  #include <GL/GLwMDrawA.h>    XVisualInfo *vi =NULL; Bool doubleBuffer = True;   A void resize (Widget w, XtPointer clientdata, XtPointer calldata); A void expose (Widget w, XtPointer clientdata, XtPointer calldata); A void initCB (Widget w, XtPointer clientdata, XtPointer calldata); A void inptCB (Widget w, XtPointer clientdata, XtPointer calldata);    int BufferAttributes [] =  {     GLX_DOUBLEBUFFER,      None  };  3 int *singleBufferAttributes = &BufferAttributes[0]; 3 int *doubleBufferAttributes = &BufferAttributes[1];     int main (int argc, char **argv) {     XtAppContext appContext;     Display *display;    XEvent event;    Widget shell;    Widget draw;     Colormap cmap;       Arg args[10];	    int n;       XtToolkitInitialize();   -    appContext = XtCreateApplicationContext();          display = XtOpenDisplay2 (appContext,NULL,"Test","Test",NULL,0,&argc,argv);  E    cmap = DefaultColormapOfScreen (DefaultScreenOfDisplay (display));       if (vi == NULL)    {;       vi = glXChooseVisual(display, DefaultScreen(display),  doubleBufferAttributes);       if (vi == NULL)        { >          vi = glXChooseVisual(display, DefaultScreen(display), singleBufferAttributes);B          if (vi == NULL) XtAppError(appContext,"No RGB visual with depth buffer"); '          doubleBuffer = False;                 }     }      n=0; 5    XtSetArg(args[n], XmNallowShellResize, True); n++; 5    XtSetArg(args[n], XmNtitle, "Main Window ");  n++; '    XtSetArg(args[n], XmNx,    10); n++; '    XtSetArg(args[n], XmNy,    10); n++; -    XtSetArg(args[n], XtNcolormap, cmap); n++; -    shell = XtAppCreateShell("shell", "shell", / applicationShellWidgetClass, display, args, n);   	    n = 0; 1    XtSetArg(args[n], GLwNvisualInfo, vi); n++;    0    XtSetArg(args[n], XmNcolormap, cmap); n++;   C    draw = XtCreateWidget("Draw", glwMDrawingAreaWidgetClass, shell,  args,n);C    XtAddCallback(draw, GLwNginitCallback,  (XtCallbackProc) initCB,  (XtPointer) NULL);C    XtAddCallback(draw, GLwNexposeCallback, (XtCallbackProc) expose,  (XtPointer) NULL);C    XtAddCallback(draw, GLwNresizeCallback, (XtCallbackProc) resize,  (XtPointer) NULL);    XtManageChild (draw);      XtRealizeWidget (shell);     XtAppMainLoop (appContext);   }    GLXContext glxcontext;    @ void initCB (Widget w, XtPointer clientdata, XtPointer calldata) { 0     GLwDrawingAreaCallbackStruct *callbackData =* (GLwDrawingAreaCallbackStruct *) calldata;       XVisualInfo *visinfo; 6     XtVaGetValues (w, GLwNvisualInfo, &visinfo, NULL);F     glxcontext = glXCreateContext (XtDisplay(w), visinfo, None, True);  .     GLwDrawingAreaMakeCurrent (w, glxcontext);  '     glClearColor(1.0, 1.0, 0.0, 1.0);   7     glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);  }   @ void expose (Widget w, XtPointer clientdata, XtPointer calldata) { /    GLwDrawingAreaCallbackStruct *callbackData = * (GLwDrawingAreaCallbackStruct *) calldata;-    GLwDrawingAreaMakeCurrent (w, glxcontext); !    GLwDrawingAreaSwapBuffers (w);  }   @ void resize (Widget w, XtPointer clientdata, XtPointer calldata) { /    GLwDrawingAreaCallbackStruct *callbackData = * (GLwDrawingAreaCallbackStruct *) calldata;-    GLwDrawingAreaMakeCurrent (w, glxcontext);     glXWaitX();>    glViewport (0,0,callbackData->width, callbackData->height); }     7 ------------------------opengl.c-----------------------  $cc opengl.c $link opengl,opt.opt  ( -----------opt.opt----------------------( sys$share:decw$openglwidgetshr.exe/share# sys$share:decw$openglushr.exe/share " sys$share:decw$openglshr.exe/share sys$share:decw$xmlibshr12/share  sys$share:decw$xtlibshrr5/share  sys$share:decw$xlibshr/share( -----------opt.opt----------------------   ------------------------------  # Date: Fri, 06 Feb 2004 03:30:57 GMT ! From: Andy V <nobody@nowhere.net> : Subject: Re: OpenGL Drawing, Motif on PseudoColor, ZLXp-E16 Message-ID: <RTDUb.158$FF5.41941@petpeeve.ziplink.net>   Stuart Norris wrote: > Dear Readers,  > C > I am attempting to get an OpenGL / X / Motif program to work on a 	 > hobbist C > DEC Alpha running Open VMS.  I have a Renderer = ZLXp-E1, which I  > understand@ > has just a single X colormap.  (from run sys$system:xglinfo)   > B > =========== 10 Visuals for Screen 0 (default = 0x22) ============ > PseudoColor visual: ID = 0x22 (hex) 34 (decimal) screen = 0 < >   DOUBLE buffered MONO COLOR INDEX visual with (Z Stencil)+ >   GL Sizes: ColorIndex=8, Z=24, Stencil=4 % >   Core X: depth=8, colormapSize=256  > C > If I let OpenGL choose the colormap I end up with the technicolor  > screenE > and I wish to get around this.  I have tried to install the default 
 > colormapC > for the screen into the creation of the drawing area widget.  See 
 > attached > code.  > C > I have tried to modify my program based on gl_ivp.c, but I cannot F > figure out how to get it to work with Motif OpenGL drawing areas.  I# > am at a loss as to what is wrong.   H You are using a color-index renderer rather than an RGBA renderer. This I works very differently in a number of ways -- first, no texture mapping.  C Second, lighting formulae select an index from a range of indices.  J Third, you need to use glColorIndex, as glColor is not used for rendering.  F I believe that you need to do three things with a C-I renderer to get  visible output.   I First, load the colormap you have created with some colors. This is done   with X11 calls.   C Second, set the clear *index* rather than the clear *color* -- see   glClearIndex().   C Third, ensure that the colormap attached to the window is actually  I loaded onto the board. This is done by telling your window manager to do  I this. I don't believe MWM, at least, does this by default. Unfortunately  I I don't recall the incantation for this, and I don't see anything on the   web that suggests it.   I Spoke too soon -- I found http://www.chemie.fu-berlin.de/cgi-bin/man?mwm   has this to say:  ;            colormapFocusPolicy (class  ColormapFocusPolicy) F                      This resource indicates the colormap focus policyB                      that is to be used.  If the resource value isG                      explicit, a colormap selection action is done on a D                      client window to set the colormap focus to thatG                      window. If the value is pointer, the client window G                      containing the pointer has the colormap focus.  If F                      the value is keyboard, the client window that hasE                      the keyboard input focus has the colormap focus. E                      The default value for this resource is keyboard.     F Hmm. Perhaps it was another window manager that had to be adjusted -- 
 possibly TWM.    ... B > void initCB (Widget w, XtPointer clientdata, XtPointer calldata) > { 2 >     GLwDrawingAreaCallbackStruct *callbackData =, > (GLwDrawingAreaCallbackStruct *) calldata; >  >     XVisualInfo *visinfo; 8 >     XtVaGetValues (w, GLwNvisualInfo, &visinfo, NULL);H >     glxcontext = glXCreateContext (XtDisplay(w), visinfo, None, True); > 0 >     GLwDrawingAreaMakeCurrent (w, glxcontext); > ) >     glClearColor(1.0, 1.0, 0.0, 1.0);      	glClearIndex (42);   9 >     glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);  > }  > B > void expose (Widget w, XtPointer clientdata, XtPointer calldata) > { 1 >    GLwDrawingAreaCallbackStruct *callbackData = , > (GLwDrawingAreaCallbackStruct *) calldata;/ >    GLwDrawingAreaMakeCurrent (w, glxcontext); # >    GLwDrawingAreaSwapBuffers (w);  > }   > I would put the glClear() call into expose rather than initCB.   -- Andy V, OpenGL Alpha Geek    ------------------------------   Date: 5 Feb 2004 12:33:25 -0800 1 From: susan_skonetski@hotmail.com (Sue Skonetski) P Subject: OpenVMS Pearl - OK for external use - another mixed architecture cluste< Message-ID: <857e9e41.0402051233.4712009@posting.google.com>   Dear Distribution lists,  E This is part of a de-brief from an event that took place on Tuesday.  ; Internal folks I promise this is the last Pearl of the day.   " The day just keeps getting better!   Sue    ____________________________  B The recent HP User Group event at HP Labs, Bristol, UK on 3rd Feb,E 2004 was given by Colin Butcher (XDelta Limited), Ray Turner (OpenVMS B Ambassador in HP UK pre-sales) and Martin Riley (HP AlphaServer UKD Product Manager). Judging by the feedback forms the day proved to beA extremely successful. HP Labs in Bristol provided a very pleasant  environment.  A The day started with a running cluster consisting of an Integrity C Server RX2600 (OpenVMS V8.1), an Alpha DS20E (OpenVMS V7.3-2) and a F VAX 4000-100 (OpenVMS V7.3). This was built by Colin using 100Mbit/secB Fast Ethernet with a Digital Networks managed switch and a NemonixB Fast Ethernet / SCSI adapter for the VAX. Please note that VAX andF Integrity mixed architecture clusters will not be supported. Alpha and= Integrity mixed architecture clusters will be supported. This F unsupported three architecture cluster configuration was quickly builtB to demonstrate some of the possibilities. It worked enough for the@ purposes of the day. You own experience might well be different.  C The day continued with an explanation of some OpenVMS fundamentals, D especially partitioning and Galaxy, leading into a discussion of howD OpenVMS boots and detects hardware devices on the different hardwareE platforms. The discussion highlighted the significant effort required > of the OpenVMS Engineering group to understand the Intel basedB hardware / firmware environment and be able to make it work in the@ context of OpenVMS. The final part of the demonstrations was theA installation of OpenVMS onto the different hardware platforms. As ? expected the main interest was in seeing the RX2600 systems and ; getting a feel for the console level commands and behavior.   < Attendees then had time to experience the RX2600 systems forF themselves, Ray Turner having been able to provide 2x single processor? RX2600s and a dual processor RX2600 for demonstration purposes.   F Martin Riley concluded the day with a discussion of the future for theA BCS division and included an announcement for ES45/47 purchasers.   A The slides are available at http://www.xdelta.co.uk/training.html    ------------------------------  $ Date: Thu, 5 Feb 2004 15:10:15 -0500* From: "Syltrem" <syltremzulu@videotron.ca>L Subject: Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ?????2 Message-ID: <svxUb.1175$l4.5832@tor-nn1.netcom.ca>  I I do hot backups on my databases every day. I started with Oracle version 
 7.3, now 9.2. 7 It was never a problem and I can't see why it would be. K BTW - backup is one thing, but I have succesfully recovered from them too !    --   Syltrem    OpenVMS 7.3-1 + Oracle 8.1.7.4H http://pages.infinit.net/syltrem (OpenVMS related web site, en franais)% ---zulu is not in my email address--- A "Paul Sture" <nospam@sture.homeip.net> a crit dans le message de * news:40225182.1B1EA40D@sture.homeip.net... > Dr. Dweeb wrote: > >  > > Daryl Jones wrote: > > > Dear Fabio Cardoso:  > > > K > > > Oracle Rdb will run on VMS, OpenVMS, and Unix (maybe NT?). Therefore, H > > > making Rdb apart of VMS would not be cost worthy. Yes, once upon aH > > > time, VMS and Rdb run-time kernel was tied together(DTM?). at thatK > > > time, Rdb name was Rdb/VMS. This was when the VMS license was $30,000 I > > > and so was(I believe) the price of the Rdb development license. Rdb E > > > back then was bringing in about $100 million a year in for DEC. J > > > However, Palmer started to sell off DEC assets. Therefore, Rdb, CDD,G > > > and related products were sold to Oracle. I always thought by now I > > > Oracle would have move everyone over to its databases, Oracle RDBMS L > > > 7,8,8i,9i, and now Oracle 10g. As you can see, Oracle Rdb still exists& > > > with its current version of 7.x. > >  > > Rdb Current Version 7.1.x  > > I > > While Rdb has been under a cloud recently, it has historically been a  veryL > > good performer for the Oracle Corporation according to heresay evidence.H > > There are a plethora of reasons why Rdb users do not want to move to/ > > OracleClassic database.  The main ones are;  > > G > > * Administration - Rdb is an order og magnitude easier to setup and  > > administer. L > > * Performance - For a large class of TP applications Rdb is a measurably > > better solution I > > * 3rd. party apps. - TP applications you have probably never heard of G > > running very critical, 24*7 systems with zero downtime requirements  providedK > > by 3rd party providers - it is doubtful whether these applications will  evenF > > run on Oracle classic, so the chances of them being ported is low.F > > * The products are in some ways mutually exclusive as a deployment	 database, > > > so flipping from Rdb to Oracle Classic is not a no-brainerK > > * Bad experience - There is a body of emprical evidence suggesting that  Rdb F > > to Oracle classic conversions are unlikely to be the success their > > proponents hope./ > > * OracleRdb is a better product technically  > >  > H > * Hot backups too. I know of one large application where the main show9 > stopper for migration away from VMS is that very issue.  >  >  > > >  > > > Regards, > > > Daryl Jones  > > > ? > > > fabiopenvms@yahoo.com.br (Fabio Cardoso) wrote in message ? > > > news:<f30679fb.0402041522.4c5cee0c@posting.google.com>... 5 > > >> "John Smith" <a@nonymous.com> wrote in message  > > >>F news:<xVaUb.130724$9Ce1.54975@news04.bloor.is.net.cable.rogers.com>... > > >>> Malcolm Dunnett wrote:E > > >>>> In article <f30679fb.0402040128.4659871@posting.google.com>, 9 > > >>>> fabiopenvms@yahoo.com.br (Fabio Cardoso) writes:  > > >>>>> Click 	 > > >>>>> C > > >>>>> http://news.com.com/2100-7344_3-5152672.html?tag=nefd_top 	 > > >>>>> 	 > > >>>>> I > > >>>>> Oracle announced the availability of its Oracle 10g database on K > > >>>>> Tuesday and cut prices, in an effort to gain more customers among  > > >>>>> midsize businesses. 	 > > >>>>> K > > >>>>> As previously reported by CNET News.com, Oracle released the Unix G > > >>>>> and Linux versions of its Oracle 10g database and dropped the J > > >>>>> price of its entry-level database to about $5,000 per processor,F > > >>>>> matching the cost of Microsoft's SQL Server 2000 database. AJ > > >>>>> Windows version of Oracle 10g is slated for completion in a "few2 > > >>>>> weeks," according to company executives. > > >>>>> (...) 	 > > >>>>>  > > >>>>J > > >>>>    Do they say somewhere what the difference is between "Standard > > >>>> Edition One" K > > >>>> and "Standard Edition"? The only thing I can see is that "Standard  > > >>>> Edition One" J > > >>>> supports a maximum of 2 processors in a server ( vs unlimited for > > >>>> "SE" ). > > >>>>L > > >>>>    Any indication they will release "Standard Edition One" for VMS? > > >>>> Right nowG > > >>>> it only says "Windows, Linux and Unix". None of our VMS Oracle  > > >>>> servers have H > > >>>> more than 2 processors so I'd hate to be spending an additional > > >>>> $10k per - > > >>>> processor just to run Oracle on VMS.  > > >>>>L > > >>>>   (note to Sue et al - it would be very supportive for the "low-end
 > > >>>> VMS" L > > >>>> market if Oracle could be encouraged to make this product available > > >>>> on VMS )  > > >>> F > > >>> A far cry from the days of Rdb run-time included with NAS 200,7 > > >>> which IIRC was included with every VMS licence.  > > >>> > > >> What Oracle gains retaining RDB ?  It should be bundled: > > >> with OpenVMS autmatically and Oracle could sell the4 > > >> other products for the customer side ! Etc...9 > > >> How is the cost of  RDB developemnt for Oracle and # > > >> how much it worth for them ?  > > >> > > >> Regards > > >>	 > > >> FC  >  > --   >  > --   > Paul Sture   ------------------------------  # Date: Thu, 05 Feb 2004 19:57:13 GMT # From: "John Smith" <a@nonymous.com> L Subject: Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ?????H Message-ID: <texUb.4508$UMB.860@twister01.bloor.is.net.cable.rogers.com>   Fabio Cardoso wrote:7 > Paul Sture <nospam@sture.homeip.net> wrote in message . > news:<40225182.1B1EA40D@sture.homeip.net>... >> Dr. Dweeb wrote:  >>>  >>> Daryl Jones wrote: >>>> Dear Fabio Cardoso: >>>>? >>>> Oracle Rdb will run on VMS, OpenVMS, and Unix (maybe NT?). F >>>> Therefore, making Rdb apart of VMS would not be cost worthy. Yes,; >>>> once upon a time, VMS and Rdb run-time kernel was tied F >>>> together(DTM?). at that time, Rdb name was Rdb/VMS. This was whenG >>>> the VMS license was $30,000 and so was(I believe) the price of the F >>>> Rdb development license. Rdb back then was bringing in about $100G >>>> million a year in for DEC. However, Palmer started to sell off DEC C >>>> assets. Therefore, Rdb, CDD, and related products were sold to D >>>> Oracle. I always thought by now Oracle would have move everyoneG >>>> over to its databases, Oracle RDBMS 7,8,8i,9i, and now Oracle 10g. E >>>> As you can see, Oracle Rdb still exists with its current version  >>>> of 7.x. >>>  >>> Rdb Current Version 7.1.x  >>> G >>> While Rdb has been under a cloud recently, it has historically been A >>> a very good performer for the Oracle Corporation according to F >>> heresay evidence. There are a plethora of reasons why Rdb users doC >>> not want to move to OracleClassic database.  The main ones are;  >>> G >>> * Administration - Rdb is an order og magnitude easier to setup and  >>> administer. A >>> * Performance - For a large class of TP applications Rdb is a  >>> measurably better solutionF >>> * 3rd. party apps. - TP applications you have probably never heard= >>> of running very critical, 24*7 systems with zero downtime A >>> requirements provided by 3rd party providers - it is doubtful F >>> whether these applications will even run on Oracle classic, so theC >>> chances of them being ported is low. * The products are in some F >>> ways mutually exclusive as a deployment database, so flipping fromF >>> Rdb to Oracle Classic is not a no-brainer * Bad experience - There@ >>> is a body of emprical evidence suggesting that Rdb to OracleG >>> classic conversions are unlikely to be the success their proponents 5 >>> hope. * OracleRdb is a better product technically  >>>  >>D >> * Hot backups too. I know of one large application where the main> >> show stopper for migration away from VMS is that very issue >  > = > We are using Hot Stanby and its working fine too ! But ...  > @ > When was the last Ad / News of Oracle RDB in Oracle Magazine ?    I About the same time the last ad for OpenVMS was done by Digital  - 1992 I  think   :-(    ------------------------------  # Date: Thu, 05 Feb 2004 23:58:45 GMT # From: Jonas Lindholm <jlhm@usa.net> L Subject: Re: Oracle ships 10g database, cuts price  <-- Oracle RDB too ?????: Message-ID: <VMAUb.121938$cM1.22583032@twister.nyc.rr.com>  D Rdb was running on NT (around 97). I've one copy at home on CD that  works well. C You could take an RMU backup on VMS, transfer the backup to NT and   restore it and run. Very cool.   /Jonas   Daryl Jones wrote: > Dear Fabio Cardoso:  > G > Oracle Rdb will run on VMS, OpenVMS, and Unix (maybe NT?). Therefore, D > making Rdb apart of VMS would not be cost worthy. Yes, once upon aD > time, VMS and Rdb run-time kernel was tied together(DTM?). at thatG > time, Rdb name was Rdb/VMS. This was when the VMS license was $30,000rE > and so was(I believe) the price of the Rdb development license. RdbfA > back then was bringing in about $100 million a year in for DEC.2F > However, Palmer started to sell off DEC assets. Therefore, Rdb, CDD,C > and related products were sold to Oracle. I always thought by nowCE > Oracle would have move everyone over to its databases, Oracle RDBMShH > 7,8,8i,9i, and now Oracle 10g. As you can see, Oracle Rdb still exists" > with its current version of 7.x. > 
 > Regards, > Daryl Jones  D > u > fabiopenvms@yahoo.com.br (Fabio Cardoso) wrote in message news:<f30679fb.0402041522.4c5cee0c@posting.google.com>...e > w >>"John Smith" <a@nonymous.com> wrote in message news:<xVaUb.130724$9Ce1.54975@news04.bloor.is.net.cable.rogers.com>...i >> >>>Malcolm Dunnett wrote:a >>>s@ >>>>In article <f30679fb.0402040128.4659871@posting.google.com>,4 >>>>fabiopenvms@yahoo.com.br (Fabio Cardoso) writes: >>>>
 >>>>>Click >>>>>A> >>>>>http://news.com.com/2100-7344_3-5152672.html?tag=nefd_top >>>>>d >>>>> D >>>>>Oracle announced the availability of its Oracle 10g database onF >>>>>Tuesday and cut prices, in an effort to gain more customers among >>>>>midsize businesses. >>>>> J >>>>>As previously reported by CNET News.com, Oracle released the Unix andG >>>>>Linux versions of its Oracle 10g database and dropped the price ofpI >>>>>its entry-level database to about $5,000 per processor, matching thetG >>>>>cost of Microsoft's SQL Server 2000 database. A Windows version ofaG >>>>>Oracle 10g is slated for completion in a "few weeks," according tos >>>>>company executives.
 >>>>>(...) >>>>>s >>>>E >>>>   Do they say somewhere what the difference is between "Standarda >>>>Edition One"F >>>>and "Standard Edition"? The only thing I can see is that "Standard >>>>Edition One"E >>>>supports a maximum of 2 processors in a server ( vs unlimited forr >>>>"SE" ).o >>>>G >>>>   Any indication they will release "Standard Edition One" for VMS?i
 >>>>Right nowcB >>>>it only says "Windows, Linux and Unix". None of our VMS Oracle >>>>servers haveH >>>>more than 2 processors so I'd hate to be spending an additional $10k >>>>per-( >>>>processor just to run Oracle on VMS. >>>>G >>>>  (note to Sue et al - it would be very supportive for the "low-endu >>>>VMS"G >>>>market if Oracle could be encouraged to make this product availablec >>>>on VMS ) >>>dL >>>A far cry from the days of Rdb run-time included with NAS 200, which IIRC' >>>was included with every VMS licence.e >>: >>What Oracle gains retaining RDB ?  It should be bundled 6 >>with OpenVMS autmatically and Oracle could sell the / >>other products for the customer side ! Etc....5 >>How is the cost of  RDB developemnt for Oracle and   >>how much it worth for them ? >>	 >>Regards  >> >>FC   ------------------------------  % Date: Fri, 06 Feb 2004 00:27:56 +0100p* From: Paul Sture <nospam@sture.homeip.net>, Subject: Re: Problems with File Access Dates0 Message-ID: <4022DF8C.274D4BD6@sture.homeip.net>   Andreas Gruhl wrote: >  > Paul Sture wrote:  > >c  
 <big snip> > $ > The machine is running standalone.K > And yes, we have a support contract and we will approach HP 'officially'.c >   ! Please let us know how that goes.e   --  
 Paul Sture   ------------------------------  * Date: Thu, 5 Feb 2004 19:08:51 +0000 (UTC), From: lewis@mazda.mitre.org (Keith A. Lewis)L Subject: Re: SYS$STARTUP:MMOV$SHUTDOWN.COM -- Insufficiently discriminating?. Message-ID: <bvu4c3$emc$1@newslocal.mitre.org>  l sms@antinode.org writes in article <04020501124811@antinode.org> dated Thu, 5 Feb 2004 01:12:48 -0600 (CST):F >   So, every time I shut down ALP2, it whacks the MMOV$SERVER on ALP,@ >instead of its own.  Is there any good reason for this behavior> >somewhere else?  (It certainly makes no sense in my cluster.)  H I can't think of a reason you'd need to shut MMOV down even on ALP2 whenD you're already shutting down the system.  The only reason I ever runJ MMOV$SHUTDOWN is when my unsupported PWS audio device needs a reset ('bout4 once a month) and I don't want to reboot the system.  G While we're talking about MMOV, I wonder if anybody has dealt with this* startup issue:  G After MMOV$STARTUP, if I invoke MMOV$AUDIOSETVOLUME.EXE, I get an errormK saying "protected images must be installed".  I think it needs a decwindowsoH image, because that's pretty much the only thing that starts after MMOV.  G What's the right way around this?  My current workaround is to submit ayJ batch job with a delay, but that's dirty because then I also have to delayK starting the audio queue or else I get a sudden volume change a few seconds  into the audio program.r  @ Is there a way to execute just the image-installation portion of
 DECW$STARTUP?r  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  * Date: Thu, 5 Feb 2004 17:11:32 -0600 (CST) From: sms@antinode.orgL Subject: Re: SYS$STARTUP:MMOV$SHUTDOWN.COM -- Insufficiently discriminating?) Message-ID: <04020517113236@antinode.org>.  , From: lewis@mazda.mitre.org (Keith A. Lewis)  H > >   So, every time I shut down ALP2, it whacks the MMOV$SERVER on ALP,B > >instead of its own.  Is there any good reason for this behavior@ > >somewhere else?  (It certainly makes no sense in my cluster.) > J > I can't think of a reason you'd need to shut MMOV down even on ALP2 whenF > you're already shutting down the system.  The only reason I ever runL > MMOV$SHUTDOWN is when my unsupported PWS audio device needs a reset ('bout6 > once a month) and I don't want to reboot the system.  E    I can't either, but I believe that I was just following orders.  A0E quick Google search for "SYSHUTDWN MMOV$SHUTDOWN" found a copy of theb@ Installation Guide (for "MMS Version 2.2", where "MMS" refers to5 "Multimedia Services"), which includes the following:   !      3.2 Editing the System Filese  A            You must edit the system startup and shutdown files to8?            provide for automatic startup and shutdown of Compaq A            Multimedia Services for OpenVMS Alpha when your systemr            is rebooted.n  =            Add the command line that starts Compaq MultimediaoA            Services for OpenVMS Alpha to the system startup file,g)            SYS$MANAGER:SYSTARTUP_VMS.COM.   A            You must position this new command line after the lineuB            that invokes the network startup command procedure. TheC            following example shows the network startup command lineeA            followed by the Compaq Multimedia Services for OpenVMSI&            Alpha startup command line:  &            $ @SYS$MANAGER:STARTNET.COM                 .s                 .o                 . *            $ @SYS$STARTUP:MMOV$STARTUP.COM  @            Add the following command line to the system shutdown            file,%            SYS$MANAGER:SYSHUTDWN.COM:.  +            $ @SYS$STARTUP:MMOV$SHUTDOWN.COM,  <    I also made a SYS$MANAGER:MMOV_RESTART.COM which does theA MMOV$SHUTDOWN.COM and MMOV$STARTUP.COM to deal with miscellaneouss5 problems, and, of course, it caused the same problem.f  <    I have no info on your MMOV$AUDIOSETVOLUME.EXE complaint.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-orgr    Saint Paul  MN  55105-2547t   ------------------------------  % Date: Fri, 06 Feb 2004 00:06:34 +01001" From: Didier Morandi <no@spam.com>) Subject: Re: VMS ODS-5 and Macintosh OS X 2 Message-ID: <4022cc7b$0$281$636a15ce@news.free.fr>  ! Did u try the X version of Fetch?h http://fetchsoftworks.com/   D.     Elliott Roper wrote:  E > Can anybody recommend a pointy-clicky ftp client for Mac that dealsc' > elegantly with ODS-5 /style=extended?o > E > (Fetch and Transmit's drag and drop both overdo the ^_ quoting whenmH > files walk from VMS to Mac. They make it into ^^_ and VMS declares the > file is not there.)i > I > It would also be pleasant if it dealt with version numbers and reporteds > the file size. > F > It is no big deal, since I can operate with the command line ftp OK.H > e.g if ODS-5 says Doofus^_Picture.jpg then get "Doofus Picture.jpg" is > fine.e >    -- n2 VAXUS - Your new helpful friend in the DEC Family!2 EHQ: 19 chemin de la Butte, 31400 Toulouse, France/       Phone: +336 7983 6418 Fax: +335 6154 1928s$                 http://www.vaxus.org   ------------------------------  % Date: Thu, 05 Feb 2004 23:35:33 +0000 & From: Elliott Roper <nospam@yrl.co.uk>) Subject: Re: VMS ODS-5 and Macintosh OS Xd1 Message-ID: <050220042335338041%nospam@yrl.co.uk>3  A In article <4022cc7b$0$281$636a15ce@news.free.fr>, Didier Morandig <no@spam.com> wrote:  # > Did u try the X version of Fetch?1 > http://fetchsoftworks.com/ > 7 Thanks Didier. I knew there were some Macheads on here.dA Yes, of course. OS X Fetch happens to be my favourite ftp client. G Long ago Jim Matthews said he was going to add VMS syntax but so far hee0 has not done so. It is stuck at 4.0.3 for years.  A What is happening is that a file called "Sample Picture.jpg" whenn5 dragged to ODS-5 appears after refreshing the list asa4 "Sample^_Picture.jpg;1" (sans quotes in both cases.)  D On attempting to drag it back toward the Mac, it alerts with "ServerF response: Failed to open XXX$DKA200:[BACKUP.MAC]Sample^^_Picture.jpg;1 for input. file not found"  D Pressing the Get... button in Fetch and typing Sample Picture.jpg inG the text bar works fine, as does the Mac's unixy ftp command line thingo0 as long as I put the file name in double quotes.  + I was hoping for one that worked perfectly.i   > D. >  >  > Elliott Roper wrote: > G > > Can anybody recommend a pointy-clicky ftp client for Mac that deals-) > > elegantly with ODS-5 /style=extended?1 > > G > > (Fetch and Transmit's drag and drop both overdo the ^_ quoting whencJ > > files walk from VMS to Mac. They make it into ^^_ and VMS declares the > > file is not there.)n > > K > > It would also be pleasant if it dealt with version numbers and reported  > > the file size. > > H > > It is no big deal, since I can operate with the command line ftp OK.J > > e.g if ODS-5 says Doofus^_Picture.jpg then get "Doofus Picture.jpg" is	 > > fine.  > >n   -- 4N Swen has got to me. I thought I would be the last on earth to mangle my e-mail address. fsnospam$elliott$$s   ------------------------------   Date: 5 Feb 2004 14:54:06 -0800a! From: tom@kednos.com (Tom Linden) % Subject: Re: Why was VAX abandonned ?h= Message-ID: <ef893e89.0402051454.24aa7762@posting.google.com>   F All wrong.  Translation Systems the predecessor to Kednos Provide PL/II Fortran and Pascal.  Prime themselves built Cobol and RPG to interface to O the common PL/I backend.  Conboy did the C, but so did we.  Trnslation  Systemss5 was BTW founded by Bob Freiburghouse of Multics fame.s  x "Robert A. Matern" <ramatern@SEND.MEuninetsNO.SPAM.net> wrote in message news:<bvtt53$24k$1@ngspool-d02.news.aol.com>...I > And Prime's Multics heritage didn't hurt, either...   I miss Multics sodN > much...   it makes most newer OSes look like garbage...   the last U.S. NavyL > Multics was shutdown in '96, the last Multics in the world bit the dust in. > early 2000...    RIP, but live on forever... > A > Bob (Multician from CSC & NWGS - see http://www.multicians.org)w > ; > "Bill Gunshannon" <bill@gw5.cs.uofs.edu> wrote in message 6 > news:bvtof3$10t5c5$2@ID-135708.news.uni-berlin.de...= > > In article <NDEMLKKEBOIFBMJLCECIAEHGCMAA.tom@kednos.com>, ) > > "Tom Linden" <tom@kednos.com> writes:h > > >aJ > > > I had seveal 50 series as well.  Guess who provided their compilers? > > >- > >5@ > > Hmmmm......  University of Sheffield.(F77)  Garth Conboy.(C) > >mE > > All the rest came directly from Pr1me and all my support contacts H > > were directly with them so I assumed they did most of it themselves.- > > Boy do I miss working with SPL and PL1/g.n > >s > > bill > >  > > -- =N > > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesH > > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > > University of Scranton   |B > > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Thu, 05 Feb 2004 18:24:00 -0500o3 From: "Richard B. Gilbert" <rgilbert88@comcast.net>i% Subject: Re: Why was VAX abandonned ?o0 Message-ID: <T7adnS9yzI0aTb_dRVn-tA@comcast.com>  , This is a multi-part message in MIME format.& --------------0501000000000103000906009 Content-Type: text/plain; charset=us-ascii; format=flowed  Content-Transfer-Encoding: 7bit   G DEC could not have lowered its prices.   It was hopelessly addicted to IL 2000%  markups!  They weren't forced to control their costs, so they didn't!  H Anyone remember the sixteen or so variants of the RZ26?  The difference H was the mounting hardware!!!!  Every machine seemed to be designed from G scratch, requiring a new case, new disk mounting hardware, a new power gH supply, new memory that was not compatible with anything else on earth, 	 etc, etc.a  E Their costs pretty much required that they charge ~$700 for a set of PI 256K memory chips that you could buy at a computer show for $33.   Ditto cH for the 20Mb MFM hard drive that you could buy for $300 anywhere except F Digital who wanted $2200+.  Then there were the Postscript books with F cover prices in the $35-$50 range; Digital offered them for something 7 line $150 per copy; I suspect that they had few takers!B  D I recall a DECUS Symposium, ca. 1995 at which someone stood up in a G general session and explained that he used to have an all DEC shop but  C now Sun hardware was appearing because it cost a lot less than DEC rI hardware.   He wanted to know "When was DEC going to roll up its sleeves t6 and compete?"   His question was ruled "out of order!"   JF Mezei wrote:l   >Bob Koehler wrote:p >  x >UC >>   While the performance was gasping it's last breath, cheap RISCrF >>   workstations were close to VAX 9000 performance levels.  Sure theG >>   price/performance was poor, but peak performance was also missing.t >>     >> > I >gasping last breath ? Seems to me that Digital was able to significantlytP >increase the VAX performance after the start of the decline in the late 1980s.  >nJ >Secondly, even if vax wasn't a world leader in performance at that stage,L >proper pricing would still have made it a ideal choice for workstations. ItM >was still far more performant than the 8086 desktops. And yes, it would havetL >have to be sold cheaper that the higher performance new kids on the block.  >SM >And with clustering which others didn't have, you could have setup a cluster"G >of cheap less performant VAXes to get the computing power you needed.   >XO >History has shown that the money was in the large volume low value market, not ( >the low volume high performance market. >w >gH >Look at wintel: with performance far worse than VAX, but with the right >pricing, they succeeded.r >  n >B  & --------------050100000000010300090600) Content-Type: text/html; charset=us-asciix Content-Transfer-Encoding: 7bitA  ? <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">P <html> <head>I   <meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">(   <title></title>  </head>e' <body text="#000000" bgcolor="#ffffff">,P DEC could not have lowered its prices.&nbsp;&nbsp; It was hopelessly addicted toN 2000%&nbsp; markups!&nbsp; They weren't forced to control their costs, so they didn't!<br>  <br>L Anyone remember the sixteen or so variants of the RZ26?&nbsp; The differenceL was the mounting hardware!!!!&nbsp; Every machine seemed to be designed fromF scratch, requiring a new case, new disk mounting hardware, a new powerG supply, new memory that was not compatible with anything else on earth,P
 etc, etc.<br>t <br>D Their costs pretty much required that they charge ~$700 for a set ofL 256K memory chips that you could buy at a computer show for $33.&nbsp;&nbsp;F Ditto for the 20Mb MFM hard drive that you could buy for $300 anywhereL except Digital who wanted $2200+.&nbsp; Then there were the Postscript books@ with cover prices in the $35-$50 range; Digital offered them forE something line $150 per copy; I suspect that they had few takers!<br>  <br>C I recall a DECUS Symposium, ca. 1995 at which someone stood up in a-F general session and explained that he used to have an all DEC shop butB now Sun hardware was appearing because it cost a lot less than DECJ hardware.&nbsp;&nbsp; He wanted to know "When was DEC going to roll up itsL sleeves and compete?"&nbsp;&nbsp; His question was ruled "out of order!"<br> <br> JF Mezei wrote:<br>h> <blockquote type="cite" cite="mid40225927.6600674B@istop.com">!   <pre wrap="">Bob Koehler wrote:    </pre>   <blockquote type="cite">R     <pre wrap="">   While the performance was gasping it's last breath, cheap RISCD    workstations were close to VAX 9000 performance levels.  Sure theE    price/performance was poor, but peak performance was also missing.a
     </pre>   </blockquote>t   <pre wrap=""><!---->H gasping last breath ? Seems to me that Digital was able to significantlyO increase the VAX performance after the start of the decline in the late 1980s.    I Secondly, even if vax wasn't a world leader in performance at that stage,oK proper pricing would still have made it a ideal choice for workstations. ItOL was still far more performant than the 8086 desktops. And yes, it would haveK have to be sold cheaper that the higher performance new kids on the block. e  L And with clustering which others didn't have, you could have setup a clusterF of cheap less performant VAXes to get the computing power you needed.   N History has shown that the money was in the large volume low value market, not' the low volume high performance market.n    G Look at wintel: with performance far worse than VAX, but with the rightL pricing, they succeeded.   </pre>
 </blockquote>o </body>o </html>n  ( --------------050100000000010300090600--   ------------------------------  # Date: Fri, 06 Feb 2004 03:55:37 GMT # From: "John Smith" <a@nonymous.com>f% Subject: Re: Why was VAX abandonned ? G Message-ID: <ZeEUb.2803$o6t1.1466@news04.bloor.is.net.cable.rogers.com>o   Bob Koehler wrote:= > In article <40222b3e$0$5224$afc38c87@news.optusnet.com.au>, + > "alphadoc" <all.the.fun@the.fair> writes:p >>, >> What lessons can be learnt from the past? >iH >    The average consumer has been taught that computers are unreliable.G >    You can sell them anything that will keep their attention for fived
 >    minutes.i >cE >    When the same systems are applied to serious business the TCO isn8 >    bigger, but the purchase cost still blinds the CIO.    F One only has to look at the average Mercedes Benz automobile....a hugeL number of them are still on the road after 15 years (see mid-80's body styleK E series), with the mechanical parts in good working order and many without L rust. How many average North American automobiles are still running after 15 years?  J What metric would the average CIO use to decide which is a better purchaseE value - a permium priced product that runs well year after year, or alE cheaper product that has to be replaced several times during the samew1 timeframe (with its attendant change-over costs)?   ) The answer is: the cheaper one of course.eG Why? Because the average CIO is going to move to another job at anotherf# company before the next changeover.vI Why? Because by keeping his expendtitures low at his current employer, heo0 looks better to the next company that hires him.G The Result: His personal bottom line is bolstered at the expense of theP9 shareholder of the company which previously employed him.d   ------------------------------  # Date: Fri, 06 Feb 2004 03:57:56 GMTt# From: "John Smith" <a@nonymous.com>L% Subject: Re: Why was VAX abandonned ?aF Message-ID: <8hEUb.2836$o6t1.766@news04.bloor.is.net.cable.rogers.com>   Fred Kleinsorge wrote:G > But the PC really arrived when DEC "thought" that the competition foroD > it was the PDP11.  The Pro-300 was the ill-fated response, and didG > reflect the opinion that it was so much better technology that people E > would just flock to it.  The VAX was big-iron in comparison, and itHC > was years before it made it down to workstations.  If there was atG > real DEC mistake, the mistake was to not pursue the merger with AppleD2 > and adopt the Lisa (which evolved into the Mac).    I That potential merger/acquisition (which I believe would have been a goodeD one) would have only worked had the correct people been permitted toE remain/ascend to positions of power. The wrong people would have only-0 resulted in the Curly scenario 15 years earlier.   ------------------------------  # Date: Fri, 06 Feb 2004 03:58:51 GMT # From: "John Smith" <a@nonymous.com>i% Subject: Re: Why was VAX abandonned ?4F Message-ID: <%hEUb.2850$o6t1.438@news04.bloor.is.net.cable.rogers.com>   Tom Linden wrote:f >   -----Original Message-----B >   From: Fred Kleinsorge [mailto:my-last-name@stardotzko.dec.com]- >   Sent: Thursday, February 05, 2004 5:45 AM  >   To: Info-VAX@Mvb.Saic.Com_) >   Subject: Re: Why was VAX abandonned ?P >b >B >t; >   "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagei' >   news:40217DAA.F7493BDC@istop.com...i >   > Bob Koehler wrote:F >   > >    The compelling reason was that DEC did a nose dive when theH >   VAX > >    simply couldn't keep up with everyone else's performance. >   >e >   >sF >   > No. The real reason was that Digital did not lower the prices of >   VAXes quicklyeB >   > enough to maintauin price-performance leadership against the >   newcomers such as7 >   > Apollo and Sun.C >nF >   Apollo was never a threat.  I agree that there were times *before*F >   the VAX was running on vapors that aggressive pricing (on say, the >   VAXstation 2000)D >   may have changed the long term outcomes of what were at the time >   very small= >   upstarts like Sun.  However, DEC found itself desperatelyt >   lowering prices toD >   be in the "price/performance" curve - but low on it... while the> >   next generation of VAX processors never arrived (in time). > A > BTW, the same happened to Prime and DG.  Intel was able to keep 	 > pumpimgdD > the X86 thanks to Microsoft, until they were finally able to match > (and exceed)$ > the performance of the RISC chips.    A HP wasn't a particularly great threat until they acquired Apollo.i   ------------------------------  # Date: Fri, 06 Feb 2004 04:10:07 GMTd# From: "John Smith" <a@nonymous.com> % Subject: Re: Why was VAX abandonned ? G Message-ID: <zsEUb.3009$o6t1.1777@news04.bloor.is.net.cable.rogers.com>    Bob Koehler wrote:4 > In article <40217DAA.F7493BDC@istop.com>, JF Mezei% > <jfmezei.spamnot@istop.com> writes:2 >> Bob Koehler wrote:nF >>>    The compelling reason was that DEC did a nose dive when the VAX@ >>>    simply couldn't keep up with everyone else's performance. >> >>C >> No. The real reason was that Digital did not lower the prices ofeA >> VAXes quickly enough to maintauin price-performance leadershipt0 >> against the newcomers such as Apollo and Sun. >iC >    While the performance was gasping it's last breath, cheap RISClF >    workstations were close to VAX 9000 performance levels.  Sure theG >    price/performance was poor, but peak performance was also missing.l    I Agreed. However all it would have taken to maintain the customer base wasp to: G a) Re-price the product to 'close' to price/performance parity. 'Close'cL would have been good enough given the costs of porting - not forgetting thatK all unix products of that era weren't worth a warm pitcher of spit in terms ' of reliability vs. VMS of the same era.c  H b) Learn one thing that one of Digital's competitors did well - IBM - byG introducing FUD in the customer's mind about migrating. IBM did that bynI saying the next generation System 38 or AS400 or whatever would offer theuF performance the customer needed, and that IBM would be there for them.% Digital never learned how to do that.    ------------------------------   Date: 5 Feb 2004 21:03:18 -0800r7 From: jones.computer.srv@worldnet.att.net (Daryl Jones)T% Subject: Re: Why was VAX abandonned ?M= Message-ID: <8a646952.0402052103.6555591e@posting.google.com>n   Dear Richard B. Gilbert:  B In 1999, I experience at a client site the moving a large ACMS-RdbB application from a max out VAX 78xx system to a Alpha 8400 where IC only saw saw drop of cpu utilization go from 100% to 66%. I thoughtiA this was strange since the Interger test comparison was 1800 to 1 E faster for the Alpha. Therefore, I find your statement that the Alpha ? blows off the doors of the VAX processor not true at that time.c   Regards, Daryl Jones   k "Richard B. Gilbert" <rgilbert88@comcast.net> wrote in message news:<svidnWLeioqV773dRVn-ug@comcast.com>...xE > Well, it was many years ago, but I seem to recall that "faster and sG > cheaper" competing systems were selling like hotcakes and VAXen were iI > not.  No matter how great the O/S was, the hardware was overpriced and o > underpowered!e > E > There may have been technical considerations as well but there was mK > definitely a financial reason; they needed faster hardware to run VMS on tI > in order to sell VMS and the VAX architecture simply could not do it.   K > Even using today's technology, the best Alpha chips would blow the doors y, > off the best VAX chips that could be made.K > I has been three or four years since I last booted a VAX;  I still own a aJ > VAXstation 4000/VLC and a MicroVAX 3100 but I may never boot either one  > agains >  > JF Mezei wrote:e > H > >>>I personally think a 64-bit extended VAX architecture incorporating7 > >>>current processor technologies would be dreamy :-)e > >>>      y > >>>i > > O > >At the time the decision was made to ditch VAX and develop Alpha, were therebQ > >compelling technical reasons to do so, or was there a strong marketing urge to * > >adopt the then buzzword-du-jour: RISC ? > >iQ > >The one argument I had heard was the need to have fixed length instruction setdQ > >in order to make pipelining etc work better. Does the 8086 have a fixed lengthe > >instruction set ? > > Q > >When one looks at what Intel was able to do with the 8086, it makes one wonder - > >if the same could have been done with VAX.o > >.( > >What was the fastest VAX chip made ?  > >s > >6P > >Also, if, for my birthday, Sue were to give me the rights and all designs forL > >VAX architecture, could I go to TI, IBM or Intel and ask them to FAB me a  > >couple thousands VAX chips ?  > >lM > >If, during the last fab, they clocked the VAX at say 200mhz,  if I were tolJ > >provide the same designs today, but have it fabbed using the latest andP > >greatest process, could the mhz be cranked up significantly because the chip,G > >albeit the same phsyical size as before, would be manufacturerd withb/ > >significantly better precision than before ?u > >  n > >  >  > --   ------------------------------   End of INFO-VAX 2004.072 ************************