larry@pocus.uucp (Larry Williamson) (03/09/90)
We run a number of Unix machines on a network. We've been
using a variety of devices for terminals. Some Wyse 60s,
some ms-dos machines with ethernet cards. Some of the unix
machines have X display hardware built in (Sony News, Mips,
etc).
I manage this motley mess, and I would like to replace some
of the character terminals with X terminals.
But I need some quantitative analysis of their benefits.
What measurable value do they provide *you* with.
I know the users would love to see this change. The few who
use the X displays, love them. Those who don't, envy them.
We've seen some value in the X displays, but it is difficult
to justify them (given their cost!) when the old character
based displays have done us so well.
Arguing that with X one gets more than one window to work in
does not cut it. We use a variety of multi-windowing
packages already (Screen on BSD, Facet/Term on SYS-V). With
these, users can hot key to any one of a number of active
login sessions.
The big advantage with X is that you can have all these
sessions visible at once (or parts of them, as you wish).
But what's that worth?
Cost issues are (amoung others I'm sure) (prices are CDN):
. X terminal >$3K
. Wyse 60 terminal ~1200
. terminal ~$900
. transerver line ~$300 (10 line transerver ~3K)
. X will impose a heavier load
. more memory ($??K)
. more cpus ($??K)
. more disk space (for swap?)
. what else?
I think it pretty obvious that X is expensive. But is the
cost justifiable?
-Larry
devin@boulder.Colorado.EDU (Yampalimardilor) (03/10/90)
In article <LARRY.90Mar8212600@pocus.uucp> larry@pocus.uucp (Larry Williamson) writes:
!I know the users would love to see this change. The few who
!use the X displays, love them. Those who don't, envy them.
!We've seen some value in the X displays, but it is difficult
!to justify them (given their cost!) when the old character
!based displays have done us so well.
!
!I think it pretty obvious that X is expensive. But is the
!cost justifiable?
!
!-Larry
A) Given that your users like them so much, I would guess that X
terminals would result in higher productivity - I know I work much faster when
I an working on hardware/software I like...
B) I don't know if this will help you but we need X because we
do a lot of image processing work and the graphics capabilities are wonderful-
we also have several different types of machine and one graphics program can
run on all of them. This is nice.
-Devin
Hope this helps.
Yampalimardilor * a.k.a Devin Hooker
Mage School, Glantri * University of Colorado-Boulder
(Grad Student) * (Undergrad Student)gjc@mitech.com (03/14/90)
In article <LARRY.90Mar8212600@pocus.uucp>, larry@pocus.uucp (Larry Williamson) writes: > But I need some quantitative analysis of their benefits. > What measurable value do they provide *you* with. > The best way to justify hardware is that you need it to run some software that you need to run. Some desk-top publishing software is currently X-based. Some spreadsheet products are now X-based, and you could look to those vendors for a comparison of the increased functionality and benefits of the X-based mode over the character-cell mode of operation. Then there are various graphics-intensive and data-visualization packages. Other than sort of thing you are probably going to have to look at the COST parameters. e.g. * Next time somebody says they need a workstation to run Program-XYZ, just like Joe is running Program-XYZ on *his* workstation, Suggest to management: No, get an X-terminal instead, and since Joe has a 10-Mip workstation it should be able to support *two* users. Unfortunately then Joe loses his status-symbol of having his own machine. * Start things in the back-door by first purchasing some of the lower cost X-window terminals, e.g. the small-screen stuff. -gjc