[AG-TECH] Re: Display Machine Performance

[Tom Coffin]

G-ware pegs the proc on the machine we use to run it.
Talk about an appliance machine! btw, it works great.

_______________________________________________
At 11:21 AM 3/15/2002 -0600, George Estes wrote:
   >   This story sounds a little fishy.  It's a little hard to understand
why 
   >an interface application, such as APWare, needs 100% of the CPU.
   >We tried the new version and then went back to the old version.
   >George
   >
   >
   >At 10:39 AM 3/15/2002 -0600, Art Ziarko wrote:
   >>I have spoken the Gentner tech support about apware in the past. They
said
   >>the software will use less processor power if other applications need
it. So
   >>in effect it  auto adjusts the processing power that it takes up
   >>Art
   >>
   >>----- Original Message -----
   >>From: "Osland, CD (Chris) " <C.D.Osland at rl.ac.uk>
   >>To: "Ag-Tech Mailing List (E-mail)" <ag-tech at mcs.anl.gov>;
"ETF-ACCESSGRID
   >>(E-mail)" <ETF-ACCESSGRID at JISCMAIL.AC.UK>
   >>Cc: "Cope, KP (Keith)" <K.P.Cope at dl.ac.uk>; "Kummer, PS (Paul)"
   >><P.S.Kummer at dl.ac.uk>; "'Art Ziarko'" <aziarko at insors.com>
   >>Sent: Friday, March 15, 2002 10:29 AM
   >>Subject: Display Machine Performance
   >>
   >>
   >> > If there's one thing worse than having to confess you've screwed
   >> > up, it's having to confess it to a world-wide audience!
   >> >
   >> > Here goes.
   >> >
   >> > All my comments about the latest Matrox drivers for G200
   >> > and G450 improving Display machine performance are probably
   >> > wrong.  Art Ziarko [aziarko at insors.com] asked me whether I had
   >> > looked at CPU utilization before and after.  I hadn't before but have
   >> > just spent a couple of hours doing so after.
   >> >
   >> > The first thing I spotted (not difficult) was that utilization was
100%
   >> > however many video frames were open (including zero!).  Second
   >> > thing was the APServe was taking all available cycles.  As I
mentioned,
   >> > we took the opportunity of AP-Ware running under W2000 to junk
   >> > the W98 machine and run AP-Ware in the Display machine.  Looks
   >> > like that little application thrashes around and is what is causing
   >> > us the problem.  If I kill off APserve, no sticky windows.  If I
restart
   >> > it, sticky windows again.
   >> >
   >> > If anyone knows whether there's a fix to this, do say.  I'll be
contacting
   >> > Gentner to see if they are aware of this behaviour, and whether they
   >> > have a fix.
   >> >
   >> > So I really don't know whether the new Matrox drivers are a good idea;
   >> > we always had the problem before (but were always running AP-Ware)
   >> > and when I did the tests on the Display machine (naturally, as I was
   >> > testing for video performance) I didn't bother starting AP-Ware.
   >> >
   >> > Just for the record, on our AMD Athlon 1.4 GHz
   >> >
   >> > 30 'small' video frames + Lobby thumbnails = 100%
   >> > 23 'small' video frames = 77%
   >> > 27 'small' + 1 'medium' + Lobby thumbnails = 93%
   >> > 4 'large' + Test Room thumbnails (about 16) = 40%
   >> >
   >> > [all the above with AP-Ware not running!]
   >> >
   >> > Cheers
   >> >
   >> > Chris O
   >> >
   >> > ____________________________________________________________________
   >> > Chris Osland                                               Office
tel: +44
   >> > (0) 1235 446565
   >> > Digital Media and Access Grid                          Medialab
tel: +44
   >>(0)
   >> > 1235 446459
   >> > BIT Department                                     Access Grid room
tel:
   >>+44
   >> > (0) 1235 445666
   >> > e-mail:   C.D.Osland at rl.ac.uk
Fax:
   >>+44
   >> > (0) 1235 445597
   >> > CLRC Rutherford Appleton Laboratory (Bldg. R18)
   >> > Chilton, DIDCOT, Oxon OX11 0QX, UK
   >> > [The contents of this email are confidential and are for the use of
the
   >> > intended recipient only.
   >> > If you are not the intended recipient do not take any action on it
or show
   >> > it to anyone else,
   >> > but return this email to the sender and delete your copy of it.]
   >> >
   >> >
   >> >
   >> >
   >> >
   >
   >


___________________________________________________________
Tom Coffin .......................... tcoffin at ncsa.uiuc.edu