TRLog
[Top] [All Lists]

[TRLog] Minimum PC requirements for TRLOG

Subject: [TRLog] Minimum PC requirements for TRLOG
From: geoiii@kkn.net (George Fremin III - K5TR)
Date: Mon, 26 Nov 2001 08:42:27 -0800
On Mon, Nov 26, 2001 at 09:04:19AM -0700, D. R. Evans wrote:
> That is theory. The practice is that you'll need quite a bit more than a 
> 286, although your machine should be fine.

I will confess that I have not run TR-Log on a 286 in some time now but
I have used it on a 386 some in the last year and depending on 
what sort of features you have turned on you might see some things 
that you will find annoying. 


> I spent this past weekend using the stock TR build on a 33MHz 486 and a 
> 50MHz 486 at a fun multi-multi. Both machines displayed an annoying 
> hesitancy at certain points in a QSO, but nothing like as bad as my poor 
> old 20MHz 386.

I use a 486 66Mhz machine for all my contesting and it works fine.
I do have SMARTDRIVE turned on for caching of reads - I have write 
cache turned off. 

> 
> The guess about slow disks may have something to it. My 386 certainly has 
> an especially slow drive, so swapping bits of the program in and out leads 
> to especially poor performance.


The real killer in performance on slower machines is the disk IO.
Since most of my contesting with TR-Log has been on 286/386/486 machines
over the years I have learned a few things to make performance better on 
slower machines.  I am sure I have posted it here a few times but not so 
good at making suggestions for the manual.

Solving the disk IO problems will get you working much better. 

Turn on Smartdrive.  Some suggest you should turn on the write caching 
as well but I dont feel comfortable with that option personally.
This will speed things up quite a bit.
Turning on write caching will speed things up even more.

Smartdrive is not installed on WIN9x systems - but you can 
install it and use it in DOS mode on those systems.  I have not 
looked on a Win98 disk but smartdrive (I think) was on the 
Win95 install disk under the directory named OLD DOS or something
similar.

I perfer to turn off some things in TR that result in less disk activity.

After the first 5 QSOs are logged in TR the program performs the following 
each time you log a QSO.

- Writes a QSO to the .TMP log file
- Writes the QSO to the .RST file (restart.bin)
- Writes QSO from the .TMP file to the .LOG file.

One thing that will speed this process is to do away with one of 
the disk writes:

UPDATE RESTART FILE ENABLE = FALSE

This will turn off the updating of the restart file and get rid of one of the 
disk wirtes.  The only pitfall to this is that if your computer should crash or
you should lose power you will not have a restart file and you will
lose all of your exchange memory data.  In CQ WW or Sweepstakes this is 
not a problem but in other contests you might find it annoying that you 
have to copy the exchange again on other bands when you work folks again.

You can also turn off this feature:

POSSIBLE CALL = FALSE

As this will get rid of more disk activity.

Although I never use the Super Check Partial (SCP) feature, it too will
create more disk activity and should not be used if you are really 
trying to squeeze alot out of your machine. 

There may be a few more things you could do to make TR-Log run
better on slower machines. 

I have found that on my 486 machine I do not need to turn off
the update restart file writes - I just need smart drive running.

--

George Fremin III                 
Johnson City, Texas             "Experiment trumps theory." 
K5TR (ex.WB5VZL)                            -- Dave Leeson W6NL
geoiii@kkn.net                             
830-868-2510                      
http://www.kkn.net/~k5tr                   


--
FAQ on WWW:               http://www.contesting.com/FAQ/trlog
Submissions:              trlog@contesting.com
Administrative requests:  trlog-REQUEST@contesting.com
Problems:                 owner-trlog@contesting.com
Feature Wishlist:         http://web.jzap.com/n6tr/trwish.html


<Prev in Thread] Current Thread [Next in Thread>