1 INFO-VAX	Tue, 11 Jul 2006	Volume 2006 : Issue 382       Contents: Re: Alpha remembrance day  Re: Any news of Sue ? + Re: DEC 2000 AXP fails system power up test + Re: DEC 2000 AXP fails system power up test + Re: DEC 2000 AXP fails system power up test + Re: DEC 2000 AXP fails system power up test  Re: Floating point questions Re: Floating point questions Re: Floating point questions Re: Floating point questions! Re: Google hires Alpha developers  Re: libxslt anywhere ?+ Re: Making LIB$*_VM_PAGE Caller's-mode safe $ Re: Multiple FTP servers - possible?$ Re: Multiple FTP servers - possible?2 Re: No OpenVMS RTLs supported for ASTs or Threads! Re: Office Friendly RX2620' Re: OT: Intel quad core X64 benchmarked & Re: OT: RS232 neutral sharing DC power& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?& Re: The possibility of vms opening up?P Re: Transitive Emulator Ports Sparc/Solaris Apps to Linux on Xeon, Itanium Itani- vax webserver outperforms linux/pc webserver!   F ----------------------------------------------------------------------    Date: 11 Jul 2006 06:39:45 +01002 From: "Dave Weatherall" <djw-nothere@nospam.nohow>" Subject: Re: Alpha remembrance day? Message-ID: <DTiotGxQ0bj6-pn2-W7L6joy8WjGC@dave2_os2.home.ours>   D On Mon, 10 Jul 2006 17:14:09 UTC, David Mathog <mathog@caltech.edu>  wrote:   > Andrew wrote:  > K > > Trying to re-write history by claiming that Alpha did not suffer from a L > > shortage of applications is rubbish, it was in fact a key contributor inK > > the decline even if the blame for the lack of software could be laid at # > > Digital and then Compaqs doors.  > K > Andrew is right on this one, the lack of native software was the primary  H > reason we eventually abandoned VMS.  This was true even though most ofI > the software we used came as source - it just didn't make sense to have J > to port every program we needed from Unix -> VMS.  OS and hardware costsH > were also factors, and I'd argue that these continued high costs were K > mostly to blame for the lack of software.  It was a vicious circle: high  C > OS and hardware cost -> small market -> repeat.  Not exactly the  9 > ecosystem that software developers wanted to invest in.   F The only trouble with the theory, and it may well be correct, is that B the same situation, a dearth of applications, was always going to  apply to Itaniuim too.   --   Cheers - Dave W.   ------------------------------  % Date: Mon, 10 Jul 2006 14:45:48 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: Any news of Sue ?9 Message-ID: <ucmdnXsI8_WCAi_ZnZ2dnUVZ_rqdnZ2d@libcom.com>    JF Mezei wrote:  > Got an email from Sue today. > G > She is still alive. The pain is starting to go down. (I think she was K > ina lot of pain before).  But she cannot spend much time at her desk yet.   A A knee replacement definitely isn't for sissies, and the pain is  I significant.  Things can go downhill rapidly from there.  There's danger  
 of infection.   I On the plus side, it works well for most people.  Regardless, there is 2   months of unpleasantness.    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  # Date: Mon, 10 Jul 2006 19:21:26 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>4 Subject: Re: DEC 2000 AXP fails system power up test0 Message-ID: <WMxsg.282$_J5.274@news.cpqcorp.net>  
 Tom wrote:L > I came into possesion of an Alpha system labeled as a DEC 2000 axp system.G > It had no video card so I stuck an old Paradise 8 bit VGA card in it, = > connected a PC keyboard to the PS2 port, and powered it up.   G    The Compaq Qvision is one of the few if not the only card supported  I by OpenVMS in this series of systems, and it runs at 1024 x 768 at 75 Hz.   E    You need to be exceedingly careful on the SCSI configuration with  C this series, as the supported Adaptec 1740 series SCSI controllers  D require the "G.1" ROM, and you can't extend the SCSI either into or I outside of the box; you can't extend the bus through the enclosure wall.  C   You also obviously need to be careful about the SCSI termination.   K > I get a screen telling me it has ROM version 1.3 and SRM firmware version E > 309 followed by system power up tests. All componets, including the B > keyboard, pass except the NVR which fails with a "00 0004" code.  I    As others have commented, the NVRAM storage (which houses the console  6 settings and the system clock) has very likely failed.   ------------------------------  # Date: Tue, 11 Jul 2006 00:56:04 GMT # From: Tom Dockray <dockray@acm.org> 4 Subject: Re: DEC 2000 AXP fails system power up test* Message-ID: <EGCsg.4025$bd4.1087@trnddc01>   david20@alpha2.mdx.ac.uk wrote:    [snip] >>> B >>> I think this means it is using a serial port because it hasn't? >>> detected the keyboard of the graphical console setup. Try a  >>> terminal...  >>>  >>[snip] >>K >>Well, the only VT device I have is a VT420 with an MMJ cable. Anyone know * >>where I can get an MMJ to DB9 converter? >> > B > Assuming you have a PC you could always use a serial cable and aI > terminal emulator rather than needing to use a real VT terminal - there A > are a number of terminal emulators for windows freely available  > ; > eg teraterm Pro 3.1.3 from http://www.ayera.com/teraterm/  [snip]  , Wouldn't I need a null modem cable for that?   ------------------------------  % Date: Mon, 10 Jul 2006 22:29:19 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 4 Subject: Re: DEC 2000 AXP fails system power up test9 Message-ID: <oLmdnQh3gpFeli7ZnZ2dnUVZ_tadnZ2d@libcom.com>    Tom Dockray wrote:! > david20@alpha2.mdx.ac.uk wrote:  >  > [snip]C >>>> I think this means it is using a serial port because it hasn't @ >>>> detected the keyboard of the graphical console setup. Try a >>>> terminal... >>>>
 >>> [snip] >>> M >>> Well, the only VT device I have is a VT420 with an MMJ cable. Anyone know , >>> where I can get an MMJ to DB9 converter? >>> C >> Assuming you have a PC you could always use a serial cable and a J >> terminal emulator rather than needing to use a real VT terminal - thereB >> are a number of terminal emulators for windows freely available >>< >> eg teraterm Pro 3.1.3 from http://www.ayera.com/teraterm/ > [snip] > . > Wouldn't I need a null modem cable for that? >  >   E Yes, but at the basic level, a serial line has ground, transmit, and  H receive.  A null modem has a line from the transmit pin at end A to the F receive pin at end B, and a line from the receive pin at end A to the = transmit pin at end B.  (Sorry if you already know all this.)   I The typical MMJ cables are a flat 6-wire cable.  The MMJ plugs have pins  I 1-6 going across the end of the plug.  When installed properly, with one   plug 'flipped', you get:   1 -> 6 2 -> 5 3 -> 4 4 -> 3 5 -> 2 6 -> 1  
 2 is transmit  5 is receive 3 and 4 are both ground(s) 1 is DTR (or such) 6 is DSR (or such)  B You could find converters, rather cheap, several dollars.  I have G extras, but the shipping would be too much.  Check with a cable supply  E place.  I'm not sure of the name of such.  You should be able to get  G them pre-wired, or not wired.  The latter allows you to build whatever   type of converter you wish.   C If you're going to use a PC, then you need a 9-pin to 9-pin cable,   configured as a null modem.    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 10 Jul 2006 23:24:35 -0500 / From: Chris Scheers <chris@applied-synergy.com> 4 Subject: Re: DEC 2000 AXP fails system power up test, Message-ID: <4hgnfmF1r3i3aU1@individual.net>   Hoff Hoffman wrote:  > Tom wrote:F >> I came into possesion of an Alpha system labeled as a DEC 2000 axp 
 >> system.H >> It had no video card so I stuck an old Paradise 8 bit VGA card in it,> >> connected a PC keyboard to the PS2 port, and powered it up. > K >   The Compaq Qvision is one of the few if not the only card supported by  H > OpenVMS in this series of systems, and it runs at 1024 x 768 at 75 Hz. > K >   You need to be exceedingly careful on the SCSI configuration with this  H > series, as the supported Adaptec 1740 series SCSI controllers require I > the "G.1" ROM, and you can't extend the SCSI either into or outside of  J > the box; you can't extend the bus through the enclosure wall.  You also : > obviously need to be careful about the SCSI termination.  H Actually, for reasonably current versions of VMS, you need the G.2 ROM. F   (No, I don't remember what "reasonably current" is in this context.)   --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------  % Date: Mon, 10 Jul 2006 16:29:01 -0700 0 From: glen herrmannsfeldt <gah@ugcs.caltech.edu>% Subject: Re: Floating point questions : Message-ID: <C5SdnUus-PEHfy_ZnZ2dnUVZ_oOdnZ2d@comcast.com>   Dave Froble wrote: (snip)  H > Uh... yeah... but, I've never seen a 'packed decimal' CPU.  Every one H > I've seen uses only ones and zeros.  Yeah, binary, that's what that's 	 > called.   9 Packed decimal is what IBM calls two BCD digits per byte.  Unpacked is one digit per byte.   - I believe IBM supplies hardware that does it.   ? Though most microprocessors support packed decimal add/subtract < by keeping track of the carry between nybbles, and supplying; an instruction to process that carry.  Intel has done it at > least from the 8080 through to x86.  Maybe for the 8008, also.   -- glen    ------------------------------  % Date: Mon, 10 Jul 2006 16:33:11 -0700 0 From: glen herrmannsfeldt <gah@ugcs.caltech.edu>% Subject: Re: Floating point questions : Message-ID: <jdOdncyjRp4Ifi_ZnZ2dnUVZ_vOdnZ2d@comcast.com>   JF Mezei wrote:    (snip)  J > IBM 360 (and beyond) architecture had instructions for packed decimal. I7 > remember being show how it worked but can't remember.   H > Basically, 1,234 is stored as 2 bytes 0x12 and 0x34  so you add 4 bitsI > at a time.  Can't remember how they dealt with the carry over (since it  > is base 10).   It would be X'01', X'23', X'4C'   A the low nybble is the sign, normally X'C' (plus) or X'D' (minus), ? though others are supported as input.  X'A', X'E', and X'F' are " also plus, and X'B' is also minus.   -- glen    ------------------------------  % Date: Mon, 10 Jul 2006 19:37:29 -0700 # From: "Tom Linden" <tom@kednos.com> % Subject: Re: Floating point questions ) Message-ID: <op.tch6wrdrzgicya@hyrrokkin>   / On Fri, 30 Jun 2006 16:55:09 -0700, JF Mezei  =   % <jfmezei.spamnot@teksavvy.com> wrote:    > Dave Froble wrote:I >> Uh... yeah... but, I've never seen a 'packed decimal' CPU.  Every one=   I >> I've seen uses only ones and zeros.  Yeah, binary, that's what that's=   
 >> called. >  > I > IBM 360 (and beyond) architecture had instructions for packed decimal.=   I7 > remember being show how it worked but can't remember.  > I > Basically, 1,234 is stored as 2 bytes 0x12 and 0x34  so you add 4 bits=   I > at a time.  Can't remember how they dealt with the carry over (since i=  t  > is base 10).  H Google for Excess-3 algorithm, that is what we use for packed decimal  =  
 arithmetic on Alpha to support PL/I   ------------------------------  % Date: Mon, 10 Jul 2006 23:21:01 -0400 ' From: Dave Froble <davef@tsoft-inc.com> % Subject: Re: Floating point questions 9 Message-ID: <itSdnbBC3rx8ii7ZnZ2dnUVZ_vWdnZ2d@libcom.com>    glen herrmannsfeldt wrote: > JF Mezei wrote:  >  > (snip) > K >> IBM 360 (and beyond) architecture had instructions for packed decimal. I 8 >> remember being show how it worked but can't remember. > I >> Basically, 1,234 is stored as 2 bytes 0x12 and 0x34  so you add 4 bits J >> at a time.  Can't remember how they dealt with the carry over (since it >> is base 10).  > ! > It would be X'01', X'23', X'4C'  > C > the low nybble is the sign, normally X'C' (plus) or X'D' (minus), A > though others are supported as input.  X'A', X'E', and X'F' are $ > also plus, and X'B' is also minus. > 	 > -- glen  >   G Yep!  Learned all that years ago in my Comp Sci classes.  But my point  F remains, it's a binary computer doing the work, with usually software F providing the logic.  Even if the task is performed in firmware, it's  still a binary computer.  H Many years ago I had to write a utility to convert data written to tape @ on an IBM machine to data usable on RSTS and later VMS systems. E Interesting writing the functions to convert the various data types,  H including packed decimal.  RSTS Basic Plus didn't have a packed decimal ) data type.  Integers, reals, and strings.   : Nibbles.  I always enjoyed that term for half a byte.  :-)   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Tue, 04 Jul 2006 09:27:17 +0200 ) From: "gl@decadence.it" <gl@decadence.it> * Subject: Re: Google hires Alpha developers9 Message-ID: <pan.2006.07.04.07.27.17.489967@decadence.it>   ; Il Mon, 03 Jul 2006 12:52:22 -0700, Rich Jordan ha scritto:  > Larry Kilgallen wrote:% >> That leads to a registration page.  >>K >> Please don't post URLs that require registration without mentioning that  >> so we don't waste our time. >  > Larry,I >      apologies, it worked for me when I clicked on it from Slashdot; no  > registration page came up.  ; For me, clicking on the link takes directly to the article. = No registration and I've never been registered on nytimes.com    bye  gl   ------------------------------  # Date: Mon, 10 Jul 2006 19:23:17 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com> Subject: Re: libxslt anywhere ? 0 Message-ID: <FOxsg.284$_J5.252@news.cpqcorp.net>  
 Pierre wrote:  > thanks a lot :)  >  > Jean-Franois Pironne wrote: 9 >> You may find PCSI kits for Libxslt/Libexslt V1.1.12 at / >> http://www.pi-net.dyndns.org/anonymous/kits/  >> >> JF  >   F    If you can get this libxslt port wrapped up and in within the next B few days, it might just make the Freeware distro for OpenVMS V8.3.   ------------------------------  # Date: Mon, 10 Jul 2006 17:57:07 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>4 Subject: Re: Making LIB$*_VM_PAGE Caller's-mode safe0 Message-ID: <Txwsg.271$vI5.200@news.cpqcorp.net>   Ian Miller wrote: H > I think the idea of having four heaps for four modes and the alternate9 > UWSS form of the lib$vm routines appears to be good to.   G    On the surface it's quite reasonable, and this matter was discussed  G (out-of-band; with other OpenVMS engineers) back in April 2006, when I  F first started discussing this particular issue with Mr. Maher, and in  various subsequent discussions.   F    The concerns with the engineers working in the RTL are around what B might break if/when this change is implemented -- a prototype was G prepared, but the risks to the then-current V8.3 release schedule were  H deemed comparatively unacceptable, given the release schedule and given 0 the unsupported nature of the sequence involved.  A    To ensure that my management doesn't become concerned with my  E statements here, please also realize that this change is also not an  D accepted project for a release after V8.3, and no formal commitment E exists to implement this sequence.  If this extension to what can be  D called from a UWSS is a requirement for a particular environment or F application, I can assist in making more formal contacts with OpenVMS F business management, or contact can be made via the support center or D via an OpenVMS Ambassador, or HP sales representative/reseller, etc.  F > Of course there is nothing stopping you from writing your own memoryH > allocation routines as a UWSS and calling from your own code but there, > is a case for these routines to be in VMS.  G    The "fun" is with any other stuff that assumes sharing; the current  F behaviour.  Additional "fun" around page ownerships, and what sort of G continued existence past or cleanup during process rundown is required.   I > I wonder if the overhead of calling a memory allocation routine that is > > a system service rather than a mode-of-caller rtl routine is > significant.  <    Barring cases of a gazillion calls in rapid sequence, no.  D    There are already heavily-used memory management calls available # within the OpenVMS kernel, as well.   I > In your proposed scheme, being able to mess up the supervisor mode heap  > seems a bad thing.  H    Any inner access mode is a privileged access mode, and it is best to   maintain a degree of separation.  F    I tend to use the kernel memory allocation routines here, should I H need memory within a UWSS, rather than the user-mode RTL routines.   Or F I allocate the memory in the user-mode wrapper to the UWSS call (note F that many UWSS calls have a user-mode shareable image for the API and F for user-mode processing, and an underlying exec- or kernel-mode UWSS F image), if I can trust that the memory will not be stomped on, or can I recover from instances of same.  These kernel memory management routines  G are fully documented in the device driver documentation, and should be  A quite familiar to anyone that's written an OpenVMS device driver.    ------------------------------  % Date: Mon, 10 Jul 2006 15:22:18 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Multiple FTP servers - possible? 9 Message-ID: <TKCdnQ6Wf9M2Oi_ZnZ2dnUVZ_tSdnZ2d@libcom.com>    hanblo {at} netscape.net wrote:  > Hello,A > we have a problem with ftp-clients setting up connections at an F > tremendous speed. This means our ftp-server isn't to happy about theG > situation and starts refusing connections. Would the proper remedy be F > to try to start a second ftp-server on another port? Anybody has any5 > experience on how to handle this kind of situation? G > OpenVMS 7.3-2, TCPIP 5.4 ECO 4 on AlphaServer ES47 7/1000 with 16 GB.  > 	 > Regards  > Hans >   I You might want to take a look at the TCP/IP manuals.  There is I think a  0 troubleshooting guide.  Also a management guide.  > There are parameters that can change things such as number of   connections, timeouts, and such.  F Is your system CPU and memory usage approaching the limits?  Possibly . you just aren't using what you have available.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 10 Jul 2006 16:37:51 -0700 0 From: glen herrmannsfeldt <gah@ugcs.caltech.edu>- Subject: Re: Multiple FTP servers - possible? : Message-ID: <jdOdnc-jRp4xeS_ZnZ2dnUVZ_vOdnZ2d@comcast.com>   hanblo {at} netscape.net wrote:   A > we have a problem with ftp-clients setting up connections at an F > tremendous speed. This means our ftp-server isn't to happy about theG > situation and starts refusing connections. Would the proper remedy be F > to try to start a second ftp-server on another port? Anybody has any5 > experience on how to handle this kind of situation? G > OpenVMS 7.3-2, TCPIP 5.4 ECO 4 on AlphaServer ES47 7/1000 with 16 GB.   C Traditional ftp uses one TCP connection for each file, which is not A so convenient for many small files.  The usual solution is to ftp ( an archive (tar, zip, etc) file instead.  @ There are ftp servers which will generate a tar file or tgz file@ on the fly when a request to ftp a directory with a .tar or .tgz suffix comes in.   -- glen    ------------------------------  # Date: Mon, 10 Jul 2006 18:25:11 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>; Subject: Re: No OpenVMS RTLs supported for ASTs or Threads! / Message-ID: <bYwsg.278$cK5.35@news.cpqcorp.net>    Steve Lionel wrote: D > As a (very) former VMS RTL developer, I find the statements in theE > original post baffling.  We put a lot of effort into making the RTL B > (LIB$, STR$, MTH$ and some more) AST-reentrant as of VMS 2.0 andH > thread-reentrant as of VMS 4.0.  I'd hate to think that our early workD > in this area had been thrown away.  I personally expended a lot ofF > effort in finding and removing reentrancy barriers in the RTL - thisB > was a task we took seriously.  There were even statements at theC > beginning of the various RTL reference manuals attesting to their  > reentrant status.   D    The post originated from Mr Maher (or from someone posting as Mr F Maher), and -- based on the obvious inference drawn by the widespread C and unattributed incorporation of text from various of my previous  G postings to the newsgroup -- intended to look like one of mine, and --  : again by inference -- to accordingly provoke a discussion.  E    The OpenVMS RTLs are expected to be user-mode AST-safe, and to be  H user-mode thread-safe, though certain of the calls calls may or may not 0 be safe from AST calls when threading is active.  F    Mixing AST routines and threads within the same application can be C somewhat "interesting."  The C RTL printf call is known to be (and  B documented to be) unsafe from an AST when threads are active, for F instance -- it mostly works, but it doesn't always work.    There's a G discussion of mixing ASTs and threads over in the Ask The Wizard area,  E as the documentation isn't (or wasn't) particularly available in the   OpenVMS manuals.  G    IIRC, there are also some non-reentrant C library calls that remain  D around, as well, and work to add reentrancy to various these calls. A IIRC, there were some host-info-related C calls that have become  I reentrant with V8.3, but I'd have to look up the details in my notes and  H in the release notes.  (Some C btree code has slated for inclusion into E the C RTL, but -- since the code used recursion -- the inclusion has  : been delayed as it ran afoul of matters of thread-safety.)  E    Calls to the OpenVMS RTLs are not supported from inner-mode code,   however.   ------------------------------    Date: 11 Jul 2006 06:39:46 +01002 From: "Dave Weatherall" <djw-nothere@nospam.nohow># Subject: Re: Office Friendly RX2620 ? Message-ID: <DTiotGxQ0bj6-pn2-3LvlUnbbHXww@dave2_os2.home.ours>   / On Mon, 10 Jul 2006 16:38:55 UTC, Keith Parris  % <keithparris_NOSPAM@yahoo.com> wrote:    > Larry Kilgallen wrote:J >  > can anyone compare (subjectively is fine) the noise levels of any of:8 >  > 	RX2620 - Office Friendly - at its quietest moments
 >  > 	DS10 >  > 	DS10L  >  > 	PWS433AU >  > 	AlphaStation 250 4/226 >  > 	Alpha ES40 > > > I have an rx2620, a DS10, and a couple of Alphastation 200s. > I > The Alphastation 200s are probably similar to your 250, and I consider  F > them reasonably tolerable for noise level for a desktop workstation. > I > I know it is in a desktop case, but the DS10 is what I consider just a  H > bit too loud to be comfortable to use in a typical office enfironment. > G > When I first turned on my rx2620, my first impression was that I had  I > used shop-vacs that were quieter. I first fired it up late on a Friday  J > afternoon and by Monday morning I already had questions asking how long K > I planned to have that thing in my office. At the moment I've lent it to  I > a co-worker and he put it 12 feet away in a separate cubicle, and even  3 > then only powers it up when it's actually in use.  > I > Some folks downstairs bought 5 rx2620s and they proved too loud to use  A > in the office environment. Two quickly got rack-mounted in the  H > datacenter. The remaining boxes have now had the Office-Friendly kits K > installed. They are physically situated close together, and when I first  H > encountered them recently, not being aware of the new kits, I thought G > they must all surely be powered off. They seem quieter now than many  
 > office PCs.   F My colleague and I have 3 As200's in our workspace. 2 have noisy fans 6 and 2 noisy disks. We don't hear them anymore tho' :-)    --   Cheers - Dave W.   ------------------------------  % Date: Mon, 10 Jul 2006 21:27:59 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 0 Subject: Re: OT: Intel quad core X64 benchmarked, Message-ID: <44B2FE9C.CC4A74D2@teksavvy.com>  
 Andrew wrote: H > fundamentally different because it is basically 2 Madison (Itanium II)I > cores packaged on to a single die. It uses the same FSB interface (400, I > 533 and 667 Mhz) it also cloclks at the same speed as Madison (1.6 Ghz)   C Question:  in terms of access to the outside world, does having two G single core CPUs make any difference relative to a single dual core CPU  ?   E In other words, does each chip have its own separate FSB, or is there  just one FSB per system ?   G IF there is one FSB per chip, then it is fair to state that a dual core F chip with a 400Mh< FSB is at a disadvantage versus 2 single core chips each with a 400mhz FSB ?   ------------------------------  % Date: Mon, 10 Jul 2006 16:42:23 -0700 0 From: glen herrmannsfeldt <gah@ugcs.caltech.edu>/ Subject: Re: OT: RS232 neutral sharing DC power : Message-ID: <jdOdnc6jRp4heC_ZnZ2dnUVZ_vOdnZ2d@comcast.com>   JF Mezei wrote:   " > I have a gizmo that has 4 leads: > ; > "+"      (8 to 40 volts DC, but I give it about 12 volts)  > "-"      (common ground) > Receive  (RS232) > Transmit (RS232)  : RS232 requires devices to accept +/- 25 volts on any input; or output.  This is usually done with a series resistor and  current limiting outputs.    -- glen    ------------------------------  % Date: Mon, 10 Jul 2006 14:20:42 -0400 ' From: Dave Froble <davef@tsoft-inc.com> / Subject: Re: The possibility of vms opening up? 9 Message-ID: <dY-dncHbrabdBC_ZnZ2dnUVZ_tOdnZ2d@libcom.com>    geletine wrote:  > David J. Dachtera wrote:G >> Then, let's answer a question with a question: how likely is it that B >> hackers/crackers/script-kiddies/etc. will launch an attack on a( >> vulnerability they do not know about? > ( > I cannot reasonably argue with that :) > J >> ...or, stated another way, if a system lacks common vulnerabilities, is; >> it wise to publish the measures taken to eliminate them?  >>I > from a closed source point of view no, where as open source software is E > known to tell its users and developers whenever a vulnerabilitie is  > found. >   B When VMS was initially produced, the prime reason was to sell VAX E computers.  At that time the only reason computer companies produced   software was to sell hardware.  B Today is a bit different, with the hardware in many cases being a F commodity.  A particular OS is not required in order to use commodity = hardware because there is a variety available to do that job.   G VMS exists today to provide some unique capabilities.  Possibly not as  I unique as some believe, but still it has it's values.  As a product, VMS    (should) produce profits for HP.  C 1) Why should HP give up such profits?  Because you'd like them to?   F 2) As pointed out many times, the source listings are available.  You F want them for free?  Fine, I'd like my next Cadillac to also be free. ) Actually, I'd like everything to be free.   E What you haven't done is define what you mean by 'open', which means  F entirely different things to different people, and you haven't made a E business case for HP to make an effort to do something that you want.   *  From HP's perspective, "why should they?"   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 10 Jul 2006 14:11:38 -0400 ' From: Dave Froble <davef@tsoft-inc.com> / Subject: Re: The possibility of vms opening up? 9 Message-ID: <beWdnYcDvOSACi_ZnZ2dnUVZ_q2dnZ2d@libcom.com>    Dave Weatherall wrote:@ > On Wed, 5 Jul 2006 00:36:02 UTC, Kilgallen@SpamCop.net (Larry  > Kilgallen) wrote:  > j >> In article <1152056817.147516.9460@75g2000cwc.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes: >>> Bill Gunshannon wrote:H >>>> In article <1152029086.974664.133530@a14g2000cwb.googlegroups.com>,, >>>> 	"AEF" <spamsink2001@yahoo.com> writes:K >>>>> The U.S. military keeps the true GPS error unavailable to the public. K >>>>> The public can't use GPS to its best accuarcy, and the military keeps J >>>>> just what that best accuracy is a secret. This way our enemies can'tM >>>>> build something just good enough to "sneak under the radar". This keeps 4 >>>>> them guessing and it makes it harder for them. >>>>>  >>>>L >>>> Your joking, right?  The most common Tactical GPS Reciever is the PLGR.O >>>> It is very long in the tooth and the only advantage it has over commercial J >>>> units (like those made by Garmin) is anti-spoofing.  A garden varietyN >>>> Garmin like all the hunters buy is just as accurate as the Army's.  Sorry >>>> to disappoint you.  >>>>	 >>>> bill  >>> A >>> Wow. I don't remember my source, but it made sense and stuck.  >>>  >>> Bummer.  >>> - >>> So why *don't* they do it as I described? J >> Originally they did, but they gave it up, presumable after figuring outI >> it would not help in any feasible scenario.  A rocket from North Korea I >> might blow up on the pad, a rocket from China might be successful with K >> some other navigation scheme.  A rocket from Al Queda, if such were ever I >> built, might be considered successful by the sender if it hit anyplace 6 >> in the country.  Hamas is doing that toward Israel. > ! > Leave the politics please Larry  >   C Mentioning current day activity is politics?  More like 'reality'.  ? Opps, prime time TV has even made that word rather meaningless.    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------    Date: 10 Jul 2006 15:27:16 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) / Subject: Re: The possibility of vms opening up? 3 Message-ID: <EREY+tduXpgj@eisner.encompasserve.org>   c In article <beWdnYcDvOSACi_ZnZ2dnUVZ_q2dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes:   K >>> Originally they did, but they gave it up, presumable after figuring out J >>> it would not help in any feasible scenario.  A rocket from North KoreaJ >>> might blow up on the pad, a rocket from China might be successful withL >>> some other navigation scheme.  A rocket from Al Queda, if such were everJ >>> built, might be considered successful by the sender if it hit anyplace7 >>> in the country.  Hamas is doing that toward Israel.  >>  " >> Leave the politics please Larry >>   > . > Mentioning current day activity is politics?  B I was not trying to get political, just to point out that for some0 applications GPS accuracy is not that important.   ------------------------------   Date: 4 Jul 2006 19:36:02 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) / Subject: Re: The possibility of vms opening up? 3 Message-ID: <Ifr67OJ23O6D@eisner.encompasserve.org>   g In article <1152056817.147516.9460@75g2000cwc.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  >  > Bill Gunshannon wrote:F >> In article <1152029086.974664.133530@a14g2000cwb.googlegroups.com>,* >> 	"AEF" <spamsink2001@yahoo.com> writes: >> >J >> > The U.S. military keeps the true GPS error unavailable to the public.J >> > The public can't use GPS to its best accuarcy, and the military keepsI >> > just what that best accuracy is a secret. This way our enemies can't L >> > build something just good enough to "sneak under the radar". This keeps3 >> > them guessing and it makes it harder for them.  >> > >> >>J >> Your joking, right?  The most common Tactical GPS Reciever is the PLGR.M >> It is very long in the tooth and the only advantage it has over commercial H >> units (like those made by Garmin) is anti-spoofing.  A garden varietyL >> Garmin like all the hunters buy is just as accurate as the Army's.  Sorry >> to disappoint you.  >> >> bill  >  > ? > Wow. I don't remember my source, but it made sense and stuck.  > 	 > Bummer.  > + > So why *don't* they do it as I described?   G Originally they did, but they gave it up, presumable after figuring out F it would not help in any feasible scenario.  A rocket from North KoreaF might blow up on the pad, a rocket from China might be successful withH some other navigation scheme.  A rocket from Al Queda, if such were everF built, might be considered successful by the sender if it hit anyplace3 in the country.  Hamas is doing that toward Israel.    ------------------------------    Date: 10 Jul 2006 15:02:57 -0700( From: "geletine" <adaviscg1@hotmail.com>/ Subject: Re: The possibility of vms opening up? C Message-ID: <1152568977.756411.238210@s13g2000cwa.googlegroups.com>    Dave Froble wrote: > C > When VMS was initially produced, the prime reason was to sell VAX F > computers.  At that time the only reason computer companies produced  > software was to sell hardware. >   E Microsoft continue to make an ever increasing bloated Os in order for @ hardware vendors to make money , of course there are open sourseG alternatives that run on older computers, apart from games and the ever E increasing multimedia market for faster computers. So the issue still 
 continues.  C > Today is a bit different, with the hardware in many cases being a G > commodity.  A particular OS is not required in order to use commodity ? > hardware because there is a variety available to do that job.    > G It all depends on what market your selling to, certain products work on  certain Os. B In the work enviroment for instance, the os with the best value inF support and that users know would mostly likely be chosen. Open source0 software and os support is on the ever increase.  H > VMS exists today to provide some unique capabilities.  Possibly not asJ > unique as some believe, but still it has it's values.  As a product, VMS" > (should) produce profits for HP. > F Hp surely can profit from support and selling more hardware and giving	 away vms.   E > 1) Why should HP give up such profits?  Because you'd like them to?  > D if this was just for me, this would be a dead forum a long time ago,C Maybe I have not said enough, but profit can be made in support, or E customised applications or patches to suit the client. Novell work to C this goal by selling a corporate version with all the source, and a E free version for anybody with all source. The two versions cater to a  different markets.  G > 2) As pointed out many times, the source listings are available.  You G > want them for free?  Fine, I'd like my next Cadillac to also be free. + > Actually, I'd like everything to be free.  > E source listings is not code source, if cars could be copied as simply G and cheaply as setting up storage space and bandwidth, which is already G part of the buisness and then freely available to the world to not just G take but to improve upon, without actually paying a wage then hell take E a car for all your friends and family. I am not sure if you are aware G that Sun's latest OpenSPARC T1 is open source, yes thats not a lie, the < schematics are truelly availabe to anybody for zero payment.E Lets not forget most of solaris is open source, are Sun closing their 	 buisness? D why not vms open at least some of the Os? Ok Sun have not yet openedB java, having said that there is a very good free version thats notC complete but progress is fast. And Sun have not complaind about it. & http://www.gnu.org/software/classpath/      F > What you haven't done is define what you mean by 'open', which meansG > entirely different things to different people, and you haven't made a G > business case for HP to make an effort to do something that you want.  >   G I mean open source , the source code being available, See above for the  buisness point of view. A I believe there will be a time in the future where not being open 5 source will kill a buisness, apart from a select few.    ------------------------------    Date: 05 Jul 2006 22:32:14 +0200( From: Andreas Davour <ante@update.uu.se>/ Subject: Re: The possibility of vms opening up? 4 Message-ID: <cs9zmfnsshd.fsf@Psilocybe.Update.UU.SE>  - Karsten Nyblad <nospam@nospam.nospam> writes:   G > SELinux was an addon, but it has been merged into the official kernel G > of Linux for some time now.  I would expect it to be included all new F > versions of commercial Linux from now on, at least versions intended  > for general purpose computing.  D FreeBSD 7 will have a lot of support for auditing and other securityE features from TrustedBSD. It is an heir to SELinux, I have been told.   , Not as good as VMS, but it's getting better.   /andreas   --  A A: Because it fouls the order in which people normally read text. ' Q: Why is top-posting such a bad thing?  A: Top-posting. ; Q: What is the most annoying thing on usenet and in e-mail?    ------------------------------  % Date: Mon, 10 Jul 2006 21:39:26 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: The possibility of vms opening up? , Message-ID: <44B3014B.B9FEDDFD@teksavvy.com>   Bob Koehler wrote:D >    That gets to the point.  He requested thrust at what would haveJ >    otherwise been an appropriate point, but the FBW setup added delay to@ >    the thrust increase, causing a delay in the speed increase.  H Do you have any evidence that the FBW system delayed increasing thurst ?6  There is no reasonf or it to delay increasing thurst.  B All the stuff I had read was that the pilot put the aircraft in anH unrecoverable situation without enough energy to start climbing and thatG the pilot did not increase thrust soon enough before the need to climb.    ------------------------------  % Date: Mon, 10 Jul 2006 23:00:56 -0400 ' From: Dave Froble <davef@tsoft-inc.com> / Subject: Re: The possibility of vms opening up? 9 Message-ID: <V8ednQwGYM22ji7ZnZ2dnUVZ_sednZ2d@libcom.com>    geletine wrote:  > Dave Froble wrote:D >> When VMS was initially produced, the prime reason was to sell VAXG >> computers.  At that time the only reason computer companies produced ! >> software was to sell hardware.  >> > G > Microsoft continue to make an ever increasing bloated Os in order for " > hardware vendors to make money ,  E Now that's just horseshit!  Microsoft could give a damn about the HW  H vendors, and has been smart enough not to compete with them.  Microsoft C keeps adding features hoping to sell a new version of the OS every  G couple of years.  To do so, they must provide reasons to upgrade, thus   the features/bloat.   ! > of course there are open sourse I > alternatives that run on older computers, apart from games and the ever G > increasing multimedia market for faster computers. So the issue still  > continues. > D >> Today is a bit different, with the hardware in many cases being aH >> commodity.  A particular OS is not required in order to use commodity@ >> hardware because there is a variety available to do that job. > I > It all depends on what market your selling to, certain products work on 
 > certain Os. D > In the work enviroment for instance, the os with the best value inH > support and that users know would mostly likely be chosen. Open source2 > software and os support is on the ever increase.  I Look, no matter what anyone wants to promote, Microsoft Office rules the  G office automation market.  Heck, even as the US Justice Department was  G suing Microsoft for being a monopoly, the US Government was specifying  J MS Office as the 'official' tools for office automation.  Figure that one.  I On a scale of 1 - 100, guess the market penetration of Microsoft Office,  I and guess the market penetration of any or all other products.  My guess  D is it's in the 90s for MS Office, and in single digits for all else.  I >> VMS exists today to provide some unique capabilities.  Possibly not as K >> unique as some believe, but still it has it's values.  As a product, VMS # >> (should) produce profits for HP.  >>H > Hp surely can profit from support and selling more hardware and giving > away vms.   G Well, most commercial VMS sites won't go without support.  The support  B is an ongoing money maker, vs a one-time amount for selling an OS 7 license.  I'd agree that support is where the money is.   G In the past, systems came with a 3-year warranty, which included basic  I support.  If the money wasn't made with the OS license cost, it wouldn't   be made at all.   F I'd think that support should be the prime target, but, for those who E won't purchase support, why should HP give away free something which   costs them money?   F Keep in mind that VMS has always been a high margin product, and that > doesn't bother those who need it.  Doesn't bother them at all.  F >> 1) Why should HP give up such profits?  Because you'd like them to? >>F > if this was just for me, this would be a dead forum a long time ago,E > Maybe I have not said enough, but profit can be made in support, or G > customised applications or patches to suit the client. Novell work to E > this goal by selling a corporate version with all the source, and a G > free version for anybody with all source. The two versions cater to a  > different markets.  E But profits are being made in support.  At least until HP completely  H guts the support organization.  With the exception of IBM, I'm guessing G nothing out there comes close to what VMS produces in support revenue.  # I really don't know, just guessing.   H >> 2) As pointed out many times, the source listings are available.  YouH >> want them for free?  Fine, I'd like my next Cadillac to also be free., >> Actually, I'd like everything to be free. >>G > source listings is not code source, if cars could be copied as simply I > and cheaply as setting up storage space and bandwidth, which is already I > part of the buisness and then freely available to the world to not just I > take but to improve upon, without actually paying a wage then hell take ( > a car for all your friends and family.  H That's part of the problem.  Because software is not seen as a tangible 1 property by many, they don't assign it any value.     > I am not sure if you are awareI > that Sun's latest OpenSPARC T1 is open source, yes thats not a lie, the > > schematics are truelly availabe to anybody for zero payment.  G But, is anyone free to produce product and sell it without Sun getting  A some type of cut?  It's most CPU designers dream to get multiple   producers of their product.   G > Lets not forget most of solaris is open source, are Sun closing their  > buisness?   F That's an experiment they were forced into, and how well it will work  out is not yet known.   + > why not vms open at least some of the Os?   F Cheap or free sources, yeah I'll go along with that.  Usage of the OS I however, I'd tend to stick with some type of license fee.  You don't see  F Microsoft giving away free usage of windows, do you?  I would like to I see more competition in price at the low end.  However, the license fees  H on itanic based systems are getting pretty good.  The system prices are  another matter.    > Ok Sun have not yet openedD > java, having said that there is a very good free version thats notE > complete but progress is fast. And Sun have not complaind about it. ( > http://www.gnu.org/software/classpath/ >  >  > G >> What you haven't done is define what you mean by 'open', which means H >> entirely different things to different people, and you haven't made aH >> business case for HP to make an effort to do something that you want. >> > I > I mean open source , the source code being available, See above for the  > buisness point of view.   E The source code is available now.  There are no/few secrets.  It may  I have a cost associated with it, but it IS AVAILABLE!  Usage of the OS is  + another matter.  See above about Microsoft.   C > I believe there will be a time in the future where not being open 7 > source will kill a buisness, apart from a select few.    Believe what you will.  @ It's not the cost that's hurting VMS the most.  It's visibility.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------    Date: 11 Jul 2006 06:39:49 +01002 From: "Dave Weatherall" <djw-nothere@nospam.nohow>/ Subject: Re: The possibility of vms opening up? ? Message-ID: <DTiotGxQ0bj6-pn2-rHFTjUPLyfUv@dave2_os2.home.ours>   ? On Mon, 10 Jul 2006 20:27:16 UTC, Kilgallen@SpamCop.net (Larry   Kilgallen) wrote:   e > In article <beWdnYcDvOSACi_ZnZ2dnUVZ_q2dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes:  > M > >>> Originally they did, but they gave it up, presumable after figuring out L > >>> it would not help in any feasible scenario.  A rocket from North KoreaL > >>> might blow up on the pad, a rocket from China might be successful withN > >>> some other navigation scheme.  A rocket from Al Queda, if such were everL > >>> built, might be considered successful by the sender if it hit anyplace9 > >>> in the country.  Hamas is doing that toward Israel.  > >>  $ > >> Leave the politics please Larry > >>   > > 0 > > Mentioning current day activity is politics? > D > I was not trying to get political, just to point out that for some2 > applications GPS accuracy is not that important.  D Accepted but then came the debatable statement about Hamas, which I  don't want to debate :-)  + I had decided to kill the post but forgot.!    --   Cheers - Dave W.   ------------------------------  # Date: Mon, 10 Jul 2006 18:06:40 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>Y Subject: Re: Transitive Emulator Ports Sparc/Solaris Apps to Linux on Xeon, Itanium Itani 0 Message-ID: <QGwsg.275$Bz5.171@news.cpqcorp.net>   Ian Miller wrote: 8 > http://www.itjungle.com/breaking/bn062806-story02.html > H > "In theory, there is no reason why the QuickTransit tools could not beC > used to port OpenVMS or OS/400 applications to new architectures, 
 > either."  F    Looks like an emulator with an interpreter-like midsection stacked E between the source instruction parsing code and the target execution  ? processing logic.  This layering very likely helps with faster  D implementations of instruction sets (what a compiler would call its @ front-end) and of run-time targets (a compiler's code generator I section), assuming that the overhead of the interpretation layer doesn't  = swamp the performance of the emulator, and assuming that the  F interpreter-like layer can deal with any "oddities" within the source 
 architecture.   I    Intermediate-language representations are quite common obviously, and  I I (personally) first ran into a pCode-like layer back around 1980 or so.  ?   (qv: the UCSD Pascal environment, and its pCode interpreter.)   G    The above is based on a quick reading of the cited material, and on  B various subsequent assumptions -- the text cited is obviously not C particularly detailed, nor would I expect such text to be detailed.    ------------------------------    Date: 10 Jul 2006 11:39:19 -0700 From: bob@instantwhip.com 6 Subject: vax webserver outperforms linux/pc webserver!B Message-ID: <1152556759.561109.57590@m79g2000cwm.googlegroups.com>  9 http://h71000.www7.hp.com/openvms/journal/v8/recipes.html    ------------------------------   End of INFO-VAX 2006.382 ************************                                                                                                                                                                                                                                                                                                                                LN˷ t9UPIlǖtPP}ɨVRNvPɈ^:GU,Tubj%W<Q<<jlS؛rv_*dLRP-y-XI^RP	bN]//iQol=eA(9GyG^`JK`Ԝ{q/~DĨ$9k?q|;FXF
S|Wc62yF{slS#fYDxiؽsͿn1[AǌA\m`VU0[Ս>]S6ؖ[>-@DGKOVsU/ZlCɽe
V].[ԻwjR<\*k:^5Zj5etk4B}iҖU5띕wim(5r$tCݸRd:*7WJeH* u5Z12ŕ\J%xx