[1887] in testers
7.3 speed
daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Wed Aug 14 10:25:42 1991
Date: Wed, 14 Aug 91 10:25:28 -0400
From: Mark Rosenstein <mar@MIT.EDU>
To: jhbrown@ATHENA.MIT.EDU
Cc: testers@ATHENA.MIT.EDU
In-Reply-To: jhbrown@ATHENA.MIT.EDU's message of Wed, 14 Aug 91 08:53:05 -0400 <9108141253.AA06945@W20-575-106.MIT.EDU>
I was not aware that this release ran slower at all. The only thing
we have intentionally done that should affect the login time is to
start dash, which should take only marginally longer than before since
dash was running in the previous release to put up the logout button.
A likely cause of the timing difference is how the two workstations
get their system software. W20-575-107 uses a local copy from the W20
machine room via the RVD protocol, while the test machines are getting
their software from E40 via AFS. If this accounts for most of the
difference, then logging out and immediately logging back in on the
same test machine should more closely match the times, since that will
cause any software that the workstation needs to already be copied
locally. When this release hits the field, all workstations will get
it the same way W20-575-107 is getting its software now.
In any case, I will investigate this further and see what we can do.
-Mark
Athena Systems Development
