In a message dated 16-Sep-99 13:20:36, Andrew Bell([EMAIL PROTECTED])
said the following about [voyager] Re: Opening new browser windows is very
slow,
>Hi Mitch,
>On 16-Sep-99, you wrote:
>> It's VERY slow here, although I'm sure my machine speed has
>> something to do with it ('030@40MHz). With Pre2 I have noticed
>> delays of up to 3 minutes or more from whenever I click on the "OK"
>> button of the "This is a beta..." window and a V window actually
>> appears. Of course, not being prepared for this resulted in several
>> V windows due to my having clicked the icon a few times...:^/.
>There is something very wrong with your Voyager setup, Mitch. I use a similar
>system to yours (030@50Mhz) and yet it takes less than a second for Voyager
>to display it's first window, after I click the OK button. Strange :/
Well, I think I have found the problem, and it appears to be a combination of
things.
First of all is the relatively slow CPU (well, compared to others with PPC and
060's). Second, I have been running the RC5DES cracking program for months
now in the background, and usually forget it is there. It detects when I go
online and updates its blocks on its own.
Also, I run Executive, and by default, V is set to be scheduled, which means
that its priority is constantly changing, if I am doing something in a shell,
or whatever. I set V to be not scheduled, and its priority put above
Executive's catch range, and it took about 10 seconds to get the "This is a
snapshot..." requestor, and 40 seconds to get the first window. Better, but
with the NOschedule set, things seem to lock up. I can push windows around,
but my clock freezes, I couldn't type this message, until the V window finally
came up.
I ran snoopdos, and aside from some failures finding libs at first (see
below) everything loaded ok.
I'm running on a PicassoII with P96, 256 color Workbench. During the load
there is a long period of time where nothing happens, no disk activity, or
anything. Well, guess I need to look more closely at my system. Strange that
V is the only program giving me this problem, but I realize it /is/ a beta.
>(I use 3.0.8, prerelease 2, NC2 keyfile)
Same version/prelease, Voyager 2 keyfile.
>Perhaps V3 is searching for some missing files before opening the
>window?
Yes, well that is where I have a question. After running SnoopDOS to try to
determine a misconfiguration, as you were suggesting above, I noticed that
every attempt by V to open a *.mcc file showed a search path like this:
06:02:53 V 166D520 OpenLib Textinput.mcc Ver 0 Fail
06:02:53 ramlib 107BE20 Load LIBS:Textinput.mcc Fail
06:02:53 ramlib 107BE20 Load �king/Voyager/Textinput.mcc Fail
06:02:53 V 166D520 OpenLib mui/Textinput.mcc Ver 0
06:02:53 ramlib 107BE20 Load LIBS:mui/Textinput.mcc OK
This is just an example of something that happened several times while V was
loading. As you can see, V eventually found the file, /where it was supposed
to be/, in MUI:libs/mui. This is how MUI is installed, as I recall it, and a
set of assigns are put in your user-startup to point to where the .mcc files
go. So why does V go searching in three other places first? (For all I know,
without testing it right now, all MUI-based programs do the exact same thing).
So, anyway, apologies to all for shooting my mouth off without checking the
facts first.
However, pre1 #was much faster# at loading and running than pre2, even with
RC5DES running.
--
For Sale: Positronic Brain- Found near S.F.- Needs Work
____________________________________________________________
Voyager Mailing List - Info & Archive: http://www.vapor.com/
For Listserver Help: <[EMAIL PROTECTED]>, "HELP"
To Unsubscribe: <[EMAIL PROTECTED]>, "UNSUBSCRIBE"
