1 INFO-VAX	Tue, 30 Aug 2005	Volume 2005 : Issue 484       Contents: Re: Another PHP solution... 
 Re: Gamers
 Re: Gamers
 Re: Gamers
 Re: Gamers
 Re: Gamers
 Re: Gamers
 Re: Gamers: Re: How far can you  cluster (SCS) unsupported over a WAN?: Re: How far can you  cluster (SCS) unsupported over a WAN?: Re: How far can you  cluster (SCS) unsupported over a WAN?B Re: Multiheaded ATI Radeon 7500 using both VGA and DVI connections2 Re: OT: .com spending no longer and porting issues  Re: simh with logical networking RE: VMS performance measuring   F ----------------------------------------------------------------------  % Date: Tue, 30 Aug 2005 13:24:30 +0200 9 From: "Robert Trawiski" <robert.trawinski@softax.com.pl> $ Subject: Re: Another PHP solution.../ Message-ID: <df1flm$53q$1@atlantis.news.tpi.pl>   C Uytkownik "Michael Austin" <maustin@firstdbasource.com> napisa w  L wiadomoci news:638d3df1d56bbe398d6f1cf4f5d706be$1@www.firstdbasource.com... > This is cool!! > H > One of the problems I have from work is being able to read newsgroups M > because of the potential that they could contain virii or worse -- the "p"  J > word.  I need access to this ng as well as a few Oracle related groups. L > So, I found TR Newsportal WebNewsReader and installed it on my VMS server J > at home.  I made the appropriate edits to the config file to add my ISP M > news server, authentication etc... and made a few minor changes to some of  H > the function include files to strip  multi-dot filenames.  I am using M > Apache 1.3-1 (as of last night) and trying to set parse = extended did not  A > fix a problem reading/opening/writing multi-dot filenames like  2 > comp.os.vms-data.dat - so I "fixed" the program. >   L It works without changes on ODS-5 disks (at least on Itanium/VMS 8.2). Look  at  6 http://www.issinoho.com:8080/phpbb2/viewtopic.php?t=31   Robert    G > here is the result of a diff on the file that works and the original.  >  > ************ > File [.LIB]thread_inc.php;5 5 >   91           $group = str_replace(".","",$group); & >   92     if ($compress_spoolfiles) { > ****** > File [.LIB]thread_inc.php;1 & >   91     if ($compress_spoolfiles) { > ************ > ************ > File [.LIB]thread_inc.php;5 5 >   97           $group = str_replace(".","",$group); : >   98       $file=fopen("$spooldir/$group-data.dat","w"); > ****** > File [.LIB]thread_inc.php;1 : >   96       $file=fopen("$spooldir/$group-data.dat","w"); > ************ > ************ > File [.LIB]thread_inc.php;5 C >  398           $infofilename = str_replace(".","",$infofilename); 4 >  399           $infofile=fopen($infofilename,"w"); > ****** > File [.LIB]thread_inc.php;1 4 >  396           $infofile=fopen($infofilename,"w"); > ************ > ************ > File [.LIB]thread_inc.php;5 = >  486           $cachefile = str_replace(".","",$cachefile); 3 >  487         $fp_cachefile=fopen($cachefile,"w");  > ****** > File [.LIB]thread_inc.php;1 3 >  483         $fp_cachefile=fopen($cachefile,"w");  > ************ > ************ > File [.LIB]thread_inc.php;5 C >  958                   $filename = str_replace(".","",$filename); * >  959         $file=fopen($filename,"w"); > ****** > File [.LIB]thread_inc.php;1 * >  954         $file=fopen($filename,"w"); > ************ > > > So, you can see that I only need to remove the extra "dots". >  >  > Happy news-reading.... >  >  > Michael Austin >    ------------------------------    Date: 30 Aug 2005 01:18:58 -0700 From: icerq4a@spray.se Subject: Re: Gamers C Message-ID: <1125389938.400669.232850@g44g2000cwa.googlegroups.com>    JF Mezei wrote:  > Dave Froble wrote:C > > Games are pretty much tied to the x86 architecture and windows.  >  > 3 > Early on, Alpha had a shot for gamers running NT.    Really?      The window of ) > opportunity was squandered (of course).   A Yes, but not in the beginning. That window though was very small.   " > Had there been enough of a push,E > Alpha's speed edge would have gotten it the gaming crowd. But Alpha G > remained just a wet dream for gamers because the lack of marketing by H > Digital resulted in their favourite games not being available on their > dream platform.   F Digital never had a CPU that was competitive in that market. Alpha was@ fast, but it had very large caches compared to x86 and typically# consumed twice the amount of power.    ------------------------------    Date: 30 Aug 2005 01:21:39 -0700 From: icerq4a@spray.se Subject: Re: Gamers C Message-ID: <1125390099.569301.180140@g43g2000cwa.googlegroups.com>   E X-Box is far from the most popular gaming console. Current generation E is on x86, the next is based on the PowerPC ISA, but that CPU is very , much different to what the Apple boxes have.   ------------------------------    Date: 30 Aug 2005 01:22:36 -0700 From: icerq4a@spray.se Subject: Re: Gamers C Message-ID: <1125390156.468027.200950@f14g2000cwb.googlegroups.com>   
 Yes, why not.    ------------------------------  # Date: Tue, 30 Aug 2005 12:35:57 GMT   From: CJT <abujlehc@prodigy.net> Subject: Re: Gamers * Message-ID: <431452A4.1090709@prodigy.net>   Jeff Cameron wrote:   J > On 8/29/05 1:03 PM, in article 11h6q7tb15t4qaf@corp.supernews.com, "Dave& > Froble" <davef@tsoft-inc.com> wrote: >  > I >>While consulting my favorite rumor rag, The Inquirer, I read this post:  >>+ >>http://www.theinquirer.net/?article=25776  >>* >>Of particular interest is the following: >>G >>Intel estimateth that there are 300 million gamers worldwide, and the H >>industry is worth $31 billion - a sum which includes all revenues from >>handhelds and hardware.  >>D >>I've always felt that people tended to ignore games.  Perhaps this >>would be a mistake.  >>H >>Games are pretty much tied to the x86 architecture and windows.  OlderG >>games wanted to get directly to the HW, which doesn't work so well on H >>NT, 2000, XP, and such.  Another drag on trying to introduce somethingM >>new, such as the itanic.  Another reason AMD and the Athlon are prospering.  >>+ >>That's a significant number, 300 million.  >  > N > The most popular Gaming Console (Microsoft's X-BOX) is based on the Power-PC- > G series Processors, same as the MACintosh.  > + I thought the PS2 was outselling the X-Box.    --  D The e-mail address in our reply-to line is reversed in an attempt toC minimize spam.  Our true address is of the form che...@prodigy.net.    ------------------------------  # Date: Tue, 30 Aug 2005 15:26:44 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: Gamers 0 Message-ID: <00A490DE.2A2DE2FB@SendSpamHere.ORG>  [ In article <BF392F12.133D8%roktsci@comcast.net>, Jeff Cameron <roktsci@comcast.net> writes: M >The most popular Gaming Console (Microsoft's X-BOX) is based on the Power-PC , >G series Processors, same as the MACintosh.  ? ...and I thought that Weendoze itself was *the* gaming console.    --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Tue, 30 Aug 2005 11:57:19 -0400 - From: William Webb <william.w.webb@gmail.com>  Subject: Re: Gamers 6 Message-ID: <8660a3a105083008571cdd5b8@mail.gmail.com>  ( ------=_Part_3832_27147398.1125417439375, Content-Type: text/plain; charset=ISO-8859-1+ Content-Transfer-Encoding: quoted-printable  Content-Disposition: inline   I On 8/30/05, VAXman-@sendspamhere.org <VAXman-@sendspamhere.org> wrote:=20  >=20A > In article <BF392F12.133D8%roktsci@comcast.net>, Jeff Cameron <  > roktsci@comcast.net> writes:I > >The most popular Gaming Console (Microsoft's X-BOX) is based on the=20 
 > Power-PC. > >G series Processors, same as the MACintosh. >=20A > ...and I thought that Weendoze itself was *the* gaming console.  >=20 > --K > VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM  >=204 > "Well my son, life is like a beanstalk, isn't it?" >=20  L I don't know about gaming but there's a certain amount of gambling involved= =20 ' every time one hits CONTROL-ALT-DELETE.    WWWebb   --=20 F NOTE: This email address is only used for noncommerical VMS-related=20 correspondence. L All unsolicited commercial email will be deemed to be a request for service= s=202 pursuant to the terms and conditions located at=20# http://bellsouthpwp.net/w/e/webbww/   ( ------=_Part_3832_27147398.1125417439375+ Content-Type: text/html; charset=ISO-8859-1 + Content-Transfer-Encoding: quoted-printable  Content-Disposition: inline    <br><br>L <div><span class=3D"gmail_quote">On 8/30/05, <b class=3D"gmail_sendername">=L <a href=3D"mailto:VAXman-@sendspamhere.org">VAXman-@sendspamhere.org</a></b=L > &lt;<a href=3D"mailto:VAXman-@sendspamhere.org">VAXman-@sendspamhere.org<= /a>  &gt; wrote:</span>L <blockquote class=3D"gmail_quote" style=3D"PADDING-LEFT: 1ex; MARGIN: 0px 0=L px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">In article &lt;<a href=3D"mailto=L :BF392F12.133D8%roktsci@comcast.net">BF392F12.133D8%roktsci@comcast.net</a>= &gt;, Jeff Cameron &lt; L <a href=3D"mailto:roktsci@comcast.net">roktsci@comcast.net</a>&gt; writes:<=L br>&gt;The most popular Gaming Console (Microsoft's X-BOX) is based on the =L Power-PC<br>&gt;G series Processors, same as the MACintosh.<br><br>...and I=7  thought that Weendoze itself was *the* gaming console. L <br><br>--<br>VAXman- A Bored Certified VMS Kernel Mode Hacker&nbsp;&nbsp; =L VAXman(at)TMESIS(dot)COM<br><br>&quot;Well my son, life is like a beanstalk=( , isn't it?&quot;<br></blockquote></div>L <div><br>I don't know about gaming but there's a certain amount of gambling=7  involved every time one hits CONTROL-ALT-DELETE.</div> L <div><br>WWWebb<br clear=3D"all"><br>-- <br>NOTE: This email address is onl=L y used for noncommerical VMS-related correspondence.<br>All unsolicited com=L mercial email will be deemed to be a request for services pursuant to the t=! erms and conditions located at=20 L <a href=3D"http://bellsouthpwp.net/w/e/webbww/">http://bellsouthpwp.net/w/e= /webbww/</a> </div>   * ------=_Part_3832_27147398.1125417439375--   ------------------------------  # Date: Tue, 30 Aug 2005 17:15:54 GMT   From: CJT <abujlehc@prodigy.net> Subject: Re: Gamers ( Message-ID: <4314943D.80705@prodigy.net>   Bill Gunshannon wrote:  , > In article <43147662.9000701@prodigy.net>,% > 	CJT <abujlehc@prodigy.net> writes:  > ? >>... and it should also be noted that Apple is moving to Intel  >> >  > B > And I would be willing to bet that is more a cost thing than any% > admission of technical superiority.  >  > bill > - Perhaps, but that's not the reason they give.    --  D The e-mail address in our reply-to line is reversed in an attempt toC minimize spam.  Our true address is of the form che...@prodigy.net.    ------------------------------  % Date: Tue, 30 Aug 2005 07:57:49 -0400 " From: "Col. Panic" <blip@spam.org>C Subject: Re: How far can you  cluster (SCS) unsupported over a WAN? 2 Message-ID: <-PadnUNH8-ut1IneRVn-2A@warpdrive.net>   Larry Kilgallen wrote:  J > In article <HJoQe.30588$5m3.21418@fe1.news.blueyonder.co.uk>, Alan Greig! > <greigaln@netscape.net> writes:  >>  K >> After reading through the OpenVPN docs I see I can use it to bridge (all I >> protocols not just IP) a simulated LAN over IP. This immediately makes K >> me think of attempting to bring a cluster up between two simulated VAXes  >> across the Internet.  >>  I >> Now I know people have done this before with physical network bridging G >> over high speed WANs and I am sure I did know the limitations at one C >> time but have forgotten. So the question is what sort of rtt and K >> throughput will I need before something really bad happens? Any ballpark  >> figures?  > 5 > Throughput should be less of an issue than latency. G > If you can make the effective latency match that of actual 10 megabit  > Ethernet things should be ok. 3  How would one 'make the effective latency match' ?  --   --   ------------------------------    Date: 30 Aug 2005 07:25:50 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) C Subject: Re: How far can you  cluster (SCS) unsupported over a WAN? 3 Message-ID: <MnRiJyPsFVMH@eisner.encompasserve.org>   W In article <-PadnUNH8-ut1IneRVn-2A@warpdrive.net>, "Col. Panic" <blip@spam.org> writes:  > Larry Kilgallen wrote: > K >> In article <HJoQe.30588$5m3.21418@fe1.news.blueyonder.co.uk>, Alan Greig " >> <greigaln@netscape.net> writes: >>> L >>> After reading through the OpenVPN docs I see I can use it to bridge (allJ >>> protocols not just IP) a simulated LAN over IP. This immediately makesL >>> me think of attempting to bring a cluster up between two simulated VAXes >>> across the Internet. >>> J >>> Now I know people have done this before with physical network bridgingH >>> over high speed WANs and I am sure I did know the limitations at oneD >>> time but have forgotten. So the question is what sort of rtt andL >>> throughput will I need before something really bad happens? Any ballpark >>> figures? >>  6 >> Throughput should be less of an issue than latency.H >> If you can make the effective latency match that of actual 10 megabit  >> Ethernet things should be ok.5 >  How would one 'make the effective latency match' ?   G Keep looking at vendors until you find one who can provide the required  (lack of) latency.  J Obviously the latency does not have to "match" - it has to be "equal to orI faster than" latency on the hardware for which the software was designed. $ I trust that did not confuse anyone.   ------------------------------    Date: 30 Aug 2005 08:58:50 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) C Subject: Re: How far can you  cluster (SCS) unsupported over a WAN? 3 Message-ID: <0fpqfjSjdZ2H@eisner.encompasserve.org>   h In article <4xYQe.34822$5m3.33437@fe1.news.blueyonder.co.uk>, Alan Greig <greigaln@netscape.net> writes: >  >  > Larry Kilgallen wrote: >  >>  M >> Obviously the latency does not have to "match" - it has to be "equal to or L >> faster than" latency on the hardware for which the software was designed.  K > Not that you will ever manage that on a home ADSL connection! I've heard  D > from the author of the simh ethernet code that he has succesfully J > clustered over home broadband. It works but is unstable under load  and . > he wouldn't recommend it but it is possible.  M VMS Development does not declare things "unsupported" just for the fun of it. G It takes effort to ensure particular configurations work, and home ADSL $ is hardly the target market for VMS.   ------------------------------  # Date: Tue, 30 Aug 2005 12:15:51 GMT * From: "FredK" <fred.nospam@nospam.dec.com>K Subject: Re: Multiheaded ATI Radeon 7500 using both VGA and DVI connections 3 Message-ID: <X5YQe.11386$0u6.2003@news.cpqcorp.net>   F This is from the VMS section of the installation guide and respresents tested resolutions.            Resolution      Color Depths        (Bits per Pixel)      Refresh Rates
       (Hertz)          640 x 480       8, 16, 24      60, 72, 75, 85          800 x 600       8, 16, 24      60, 72, 75, 85          1024 x 768      8, 16, 24      60, 70, 75, 85          1024 x 864      8, 16, 24      60, 70, 75, 85          1152 x 864      8, 16, 24      60          1280 x 1024       8, 16, 24      60, 72, 75, 85          1600 x 1200       8, 16, 24      60, 65, 75, 85          1792 x 1344       8, 16, 24      60, 75          1856 x 1392       8, 16, 24      60, 75          1920 x 1080       8, 16, 24      60, 75          1920 x 1200       8, 16, 24      60, 75          1920 x 1440       8, 16, 24      60, 75          2048 x 1536       8, 16, 24      60, 65, 70, 75          2048 x 2048       8, 16, 24      57         ( <elementyl@hotmail.com> wrote in message< news:1125358434.243377.63470@g44g2000cwa.googlegroups.com...D > According to the Radeon 7500 release notes, August 2005, there are$ > still no "wide" geometries listed. >    ------------------------------    Date: 30 Aug 2005 05:46:54 -0700 From: bob@instantwhip.com ; Subject: Re: OT: .com spending no longer and porting issues B Message-ID: <1125406014.219611.27560@f14g2000cwb.googlegroups.com>  F when we ported form vax to alpha, RMS was RMS and all we had to do wasG port our synergy dibol apps and recompile ... tcpware is tcpware and so D was everything else we ported ... I do not expect anything different going to itanium form alpha ...    ------------------------------  % Date: Tue, 30 Aug 2005 14:00:15 +0200   From: S <soterroAT@yahooDOT.com>) Subject: Re: simh with logical networking 1 Message-ID: <df1ht7$9nh$1@klatschtante.init7.net>    Dan Foster wrote: # > TAP is a virtual ethernet device.  > J > In a SIMH context, it makes it appear there is a second Ethernet adapterJ > installed in the system, so that you can connect to the virtual ethernetH > network interface from within the host system instead of having to put/ > in a second actual physical Ethernet adapter.   ; VMWare (a x86 emulator) works with such a beast by default.    S    ------------------------------  % Date: Tue, 30 Aug 2005 09:52:54 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> & Subject: RE: VMS performance measuringR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B22D8@tayexc19.americas.cpqcorp.net>   > -----Original Message-----3 > From: Jim McCullars [mailto:jim@info2.uah.edu]=20  > Sent: August 29, 2005 5:54 PM  > To: Info-VAX@Mvb.Saic.Com ( > Subject: Re: VMS performance measuring >=209 > Hi Kerry, thanks for the feedback.  Let me take your=20  > questions one at a > time:  >=20A > : - What versions and patch levels of OpenVMS and Oracle are=20  > you using? >=20H >    VMS version is 7.3-1.  Oracle version is 9.2.0.4.0.  The version of! > application server is 1.0.2.2.2  >=20  @ I would highly recommend getting OS to VMS V7.3-2 (+latest maint@ patches) as this version has a number of significant performanceE enhancements over V7.3-1. It might not be *the* answer, but it should  help in some areas.   : Note also that VMS V7.3-1 is no longer supported by HP.=20  
 Reference:; http://h71000.www7.hp.com/openvms/openvms_supportchart.html     + > : - has autogen (with feedback) been run?  >=20	 >    Yes.  >=20< > : - please determine if the disk drives are excessively=20 > fragmented or not.G > : If the drives have been in use for a long time, then this is a real  > : possibility.=3D20  >=20? >    Fragmentation should not be an issue.  In an attempt to=20  > squeeze betterB > performance, the tablespace files for our production database=20 > were put on their ? > own raidset (more on that later) and spread across all six=20  > busses of the HSZ70.$ > This was done just a few days ago. >=20> > : - do a show device /full on each disk. If the highwater=20 > marking feature A > : is on and if there is a lot of writing, then this security=20 
 > feature may 8 > : be contributing to some of the performance issues=20 > (especially if disks > : are fragmented). >=20> >    It was enabled and we did turn it off, but I wonder if=20 > this is really an < > issue with an OS file that is a container for an Oracle=20 > database.  Doesn'tB > highwater marking have to do with when a file gets extended? =20 > An Oracle fileH > usually gets allocated all in one operation, and then writes performed@ > within the filespace that was originally allocated.  That's=20 > my understanding, 	 > anyway.  >=20> > : - Can your vendor determine that the Oracle indexes are=20 > being used or & > : are the acceses mostly sequential? >=20B >    This is the big unknown.  I agree that the symptoms we are=20
 > seeing look B > suspiciously like large sequential searches.  The problem is,=20 > this is a=20H > delivered system and we are still learning our way around it.  We have? > some technical assistance from them and I will bring this up.  >=20: > : - If running VMS V7.3-2, a reasonably recent Oracle=20 > suggestion has been A > : shown to increase Oracle performance on OpenVMS by roughly=20 
 > 20-30% (not ; > : sure if these were actual numbers) is to install the=20  > Oracle.exe file. >=20= >    We are going to look at that, but will have to get on=20  > 7.3-2 first.  In> > our case, however, it looks like the hangups are with the=20 > datafiles rather1 > than the Oracle image.  Still worth a try, tho.  >=20? >    Back to our raidsets.  We have six drives configured in=20  > the HSZ70 as aG > Raid 5 set.  One of my DBAs seems to recall a warning against running @ > Oracle databases as Raid 5 sets (apparently something about=20 > the way Oracle= > does its I/O is less efficient on Raid 5).  Should we be=20  > running something A > like a striped mirror instead?  Also, do you have any advice=20  > on the chunk7 > size?  We are running the default of 256, but some=20  > documentation indicated A > that that setting is better when the OS is Unix and that 128=20  > would be better , > for VMS.  Do you know anything about that? >=20  F RAID5 is always worse for applications that do a significant amount ofH writing. However, if write back cache is turned on (need cache batteries8 installed), then this is likely not the main culprit.=20  F Fwiw, I usually prefer RAID0+1 for any database files that may at someG point be performance sensitive. Costs more from a MB/$ perspective, but 9 imho, disk space is cheap relative to performance issues.   E If I was a betting person, I would be betting that the DB indexes are 8 for some reason not being used as they were designed.=20   Other comments: > - ensure that the Oracle.exe file gets installed as per recent recommendations.H - review Oracle 9.2.0.5 on OpenVMS release notes to see if there are anyB performance fixes that might be applicable to your environment.=20     =20   % >    Thanks again for the feedback...  >=20 > Jim McCullars  >=20 >=20   ------------------------------   End of INFO-VAX 2005.484 ************************