Tweet
I'm a die-hard Unix person from the start, so my opinion may be somewhat
biased, but I think my reasoning behind it is all well-founded.
1. Unix is an inherently more stable operating system than NT. You have
much more control over running processes and multiple processes in Unix
than in NT. ie, if you are running multiple windows doing different
calculations, and one of them crashes, you won't lose the other processes
(as opposed to NT which would probably die on you).
2. Unix machines (ie, SGI's, SunSparc's, DEC's) are designed to be
calculation/mathematically intensive processors. Even though you may not
be doing much graphics work (of which, Unix workstations are often used
for), graphics manipulation is essentially all mathematical calculations.
When I did Cardiac MRI and Brain PET/MRI analysis, we always used SGI's or
SunSparcs because when it came down to it, they were more reliable and
faster.
3. Unix is an ideal multi-user/networking platform. Though NT is
"designed for multi-user access", Unix truly is. It's significantly more
secure, your file sharing and permissions can be managed alot better, and
remote access is more simple. The workstation would also make an ideal
server if you wanted your own web-server (or had other machines running
linux that you want to rhost to).
I'd say the only advantage of using an NT station over a Unix based one is
cost. Another slight advantage is the availability of word processors or
other"regular" application software, but I'm sure you already have other
systems that satisfy that task.
Hope this helps.
Ed
Edward Auyang || auyanged@email.uc.edu
Biomechanics and Ergonomics Laboratories
Department of Environmental Health
University of Cincinnati Medical Center
On Wed, 27 Oct 1999, Curt G. DeGroff wrote:
> Dear BIOMCH subscribers:
> To those of you who do heavy computational work.
> Where are we (collectively if I may) going in regards to NT machines versus
> UNIX machines?
> I need to buy a new workstation with lots of computational horse power. I am
> not concerned so much with graphics. Most of the CFD or FEA software packages
> that we use support NT or UNIX (SGI, SUN, etc) and do not yet support things
> like LINUX. I hate Windows NT but like many it may be time to see the writing on
> the wall.
> Any thoughts?
> Thanks.
>
> --
>
> Curt G. DeGroff, M.D.
> degroff.curt@tchden.org
> University of Colorado - Health Science Center
> The Children's Hospital, Denver
> Pediatric Cardiology
> Cardiovascular Flow Laboratory
> (303) 764-8473
> (303) 837-2595 FAX
>
> ---------------------------------------------------------------
> To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
> For information and archives: http://isb.ri.ccf.org/biomch-l
> ---------------------------------------------------------------
>
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------
biased, but I think my reasoning behind it is all well-founded.
1. Unix is an inherently more stable operating system than NT. You have
much more control over running processes and multiple processes in Unix
than in NT. ie, if you are running multiple windows doing different
calculations, and one of them crashes, you won't lose the other processes
(as opposed to NT which would probably die on you).
2. Unix machines (ie, SGI's, SunSparc's, DEC's) are designed to be
calculation/mathematically intensive processors. Even though you may not
be doing much graphics work (of which, Unix workstations are often used
for), graphics manipulation is essentially all mathematical calculations.
When I did Cardiac MRI and Brain PET/MRI analysis, we always used SGI's or
SunSparcs because when it came down to it, they were more reliable and
faster.
3. Unix is an ideal multi-user/networking platform. Though NT is
"designed for multi-user access", Unix truly is. It's significantly more
secure, your file sharing and permissions can be managed alot better, and
remote access is more simple. The workstation would also make an ideal
server if you wanted your own web-server (or had other machines running
linux that you want to rhost to).
I'd say the only advantage of using an NT station over a Unix based one is
cost. Another slight advantage is the availability of word processors or
other"regular" application software, but I'm sure you already have other
systems that satisfy that task.
Hope this helps.
Ed
Edward Auyang || auyanged@email.uc.edu
Biomechanics and Ergonomics Laboratories
Department of Environmental Health
University of Cincinnati Medical Center
On Wed, 27 Oct 1999, Curt G. DeGroff wrote:
> Dear BIOMCH subscribers:
> To those of you who do heavy computational work.
> Where are we (collectively if I may) going in regards to NT machines versus
> UNIX machines?
> I need to buy a new workstation with lots of computational horse power. I am
> not concerned so much with graphics. Most of the CFD or FEA software packages
> that we use support NT or UNIX (SGI, SUN, etc) and do not yet support things
> like LINUX. I hate Windows NT but like many it may be time to see the writing on
> the wall.
> Any thoughts?
> Thanks.
>
> --
>
> Curt G. DeGroff, M.D.
> degroff.curt@tchden.org
> University of Colorado - Health Science Center
> The Children's Hospital, Denver
> Pediatric Cardiology
> Cardiovascular Flow Laboratory
> (303) 764-8473
> (303) 837-2595 FAX
>
> ---------------------------------------------------------------
> To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
> For information and archives: http://isb.ri.ccf.org/biomch-l
> ---------------------------------------------------------------
>
---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------