Swiss-Sys versus WinTD
by Mark Kaprielian
01-14-01
Here are my thoughts comparing WinTD and SwissSys.
SwissSys (which shall be referred to as SS from this point forward )initially came out as a DOS only program. A while back, perhaps 2 years ago, it came out as a Windows version.
Initially it had some serious bugs which seemed to be scaring long time DOS SS users away from
moving to it. WinTD came out in 1997 many years after the DOS SS and came out as a Windows only version written from scratch.
The MetroWest Chess Club (MCC) used WinTD when it first went to computer pairings back in 1977 and used it faithfully for 3 years.
We started using it shortly after it became available.
We decided to try out the SS Windows version as we had three particular items we did not like about the way the WinTD creator chose to do things and
to date we have not been able to convince him to change. In
addition, we had recently run into a clear pairing bug regarding odd man
pairings so it seemed a good time to give SS a good looking over.
Let me say that for the three years we dealt with WinTD, they made many, many improvements that we
asked for. Also, except for the bug found at the end of the three years, each and every case that we questioned the pairings,
it turned out to be either our having set an option incorrectly or not a bug.
In many cases we didn't understand what was being done but upon close examination and much debate then finally actually re-reading the
manual and discussions with WinTD, we concurred that the program did as it was told to do.
WinTD only does Look Ahead pairings which is the USCF recommendation. Few people are aware of the subtle differences of Look Ahead versus Top Down pairing and hence, many
times the pairing would be questions but a look at the wall chart could dispel
the doubts. Pairings will indeed often come out differently. SS lets you pick which type of pairing to use
but, the MCC remains with Look-Ahead. A TD who has good experience doing
manual pairing will, I believe end up doing Look-Ahead pairing as they know they
need to manage future rounds. In my own case, I discovered that the
refinements and judgment calls I was making when I was doing Top-Down pairings
were in effect making me do Look-Ahead.
Both programs necessitate careful understanding and selection of the options.
Neither program would keep an experienced Windows user from complaining about the interface and how the authors decided to do things.
Our decision to evaluate the Windows version of SS was based on two points we could not get WinTD to budge on:
1. If all other things are absolutely equal after going through all sorts of evaluations, it uses a randomization for the final selection criteria. This came to light when people in the 3 point group were paired differently because of a change in a much lower point group. To generate the random seed, it looked at the whole section.
If you check variation 29I4 on page 107 of the rules, the TD can let the players choose by lot in those situations.
We prefer to have the option to not have it be random or at least not use the
method it is using for generating the random seed.
2. Printing of pre and post ratings quickly and easily on standing reports went from being just a check box option to a many step, not simple process and hence, to us constituted a loss of much desired functionality.
Another shortcoming that does not affect the MCC as we are a weeknight club only, is the ability to run several sections at the same time that are on different rounds.
Some actions, most notably undo pairings, apply to more than one section. This is not good.
Also, I can not speak to how well it does team events. I do know that SS
DOS has been doing it for years and hence the experience should be part of SS.
We started our modern experience with SS at version 3.2 Within a few uses of the program, I had accumulated a list of 20 or more
usability issues and submitted them to SS. After three months of struggling with the
usability issues, we were resolved that if the (at the time) just announced new release 4.0 did not solve many of the issues, we would switch back to WinTD.
Version 4.0 addressed almost all of the issues we had raised and we have since decided to stick with it instead of going back. Version 4 was much, much better than 3.2.
After going through the learning curve of SS, we find it to be much more
powerful, robust and flexible.
WinTD is, in my opinion, easier to get going with. If you need to make manual adjustments to pairings, the user needs more than basic experience with the program.
Also, it is easier to make irreversible mistakes with WinTD than SS. The reason for this is that each round, both pre and post paired is kept in a
separate file for SS. This means that old info is still around and you can go back as far as you need.
On the other hand, this creates a lot of files along the way that at the end you will want to "clean up" and in fact, there is such a function built in to the program. WinTD on the other hand is one file for all your sections.
With SS, you truly need to do the built in tutorial to understand what's going on and how to effectively use it.
If you don't, you will struggle a lot longer and harder.
One area of difficulty which may or may not apply is that, if you plan on doing clean up work on the event by moving the files from one computer to another, you will have little difficulty with WinTD but with SS, you will have many files to move and all the paths and most of the remembered settings will be invalid. SS uses mostly absolute file paths instead of relative ones and hence the source of the problems. I've discussed this with SS. They were surprised at our use and our
difficulties. I don't think that SS has been used much if at all in a Weeknight type of
scenario. There are different considerations for weeknight versus a weekend. In a weekend, one round usually happens right after another. Files aren't being moved around, lots of late round byes being taken on the fly, late round entries, etc.
SS has been, so far, pretty responsive to our complaints and as mentioned above, address many of them in going to version 4.
WinTD has been extremely responsive to all of my concerns and questions. They've also incorporated a good share of my suggestions. We've had maybe a dozen serious go rounds in the last year among the Club's TDs on whether a pairing was in error or not. In each case we discovered that we had set an option incorrectly (it wasn't incorrect until we didn't like what it produced) or that it did the pairings correctly. Most of our doubts we realized later was because we didn't realize that it was doing Look Ahead pairing and not Top Down. The two methods produce subtly different pairings. There is no option to select this but the USCF recommends Look Ahead.
Another important factor is that SS seems to be more actively worked on for improvements though at perhaps a slower pace than WinTD. The makers of WinTD have gone a long stretch now since their last release and no date for a new release is scheduled, I asked. This concerns me as I'm not sure how much drive they have do continuing improvement.
The program is very easy to use but some of the steps are not always obvious. If you go through the tutorial and the examples, you will probably have no trouble at all.
One very nice feature of WinTD that makes it flexible for doing many things with the information that can be generated is that you can enable it to print to a special text window that opens up within the program. If you enable printing to the text window, every time you print something, and there is a print button usually associated with each window, it opens a window within WinTD and outputs to it as plain, formatted text. You can output other things to the same window if you choose thereby appending to it. You can edit anything in the window. When you have what you want in the window the way you want it you can hit either the save to file button or the printer button.
This is how I generate the cross table, wall chart, pairings, pairing log file (that explains how pairings were done!!) and upset lists for the web site.
Regarding reports and posting to the web. WinTD has by far the better output to HTML but is less flexible on the printed report. SS 4.0 made great improvements on their output to HTML but clearly has a way to go. I'm pushing for these improvements and am hopeful that we'll get them eventually. I had SS look at what WinTD produced for our site and SS was impressed at what it did but couldn't get to it just yet. With WinTD I could get the complete set of files generated and posted up in less than 20 minutes. With SS, I took me a lot of practice and the fixes in version 4.0 to get them up in only half an hour.
To compare HTML outputs, go to the Tournament Archive button on the home page of the MCC to
view the outputs of both WinTD and SS. Everything before June of 2000 was WinTD. Everything after has been SS. I'd recommend looking at November of 1999 and then November of 2000 for comparison purposes. Please note that color added to the 1999 pages were done by me at posting time and not done by the program.
SS does not currently do an HTML output of cross tables and I just can't seem to get it to output text such that it converts over nicely, hence the mess on the web site. SS does not create nice point score groupings like WinTD does in its HTML output. Currently SS creates
separate files for each section so that you have to merge them yourself into one page. Not a lot of work but an
inconvenience. SS did at least make a batch file create for us in version 4 which made life easier.
You should be aware that WinTD has been used for some National Level events with very good results. SS has been used for years also but, primarily the DOS version. As of last years World open, I
believe Goichberg of CCA still uses the DOS version.
My overall conclusion comparing the two are:
A WinTD is currently more friendly for Weeknight club use, especially if transporting files and posting results to the web.
B SS is certainly the more powerful, robust and allows the most
flexibility in making changes, especially to pairings.
C If you try to learn without the tutorials, you'll be able to figure out WinTD sooner than SS.
SS requires a greater one time investment in learning to use it but you will be able to solve any problem situations that arise a lot quicker than WinTD.
With WinTD you will struggle to do those same adjustments.
D If you run a multi section weekend event, SS is the program that will serve you better. You can more easily have things split between machines should you desire and it will go back
together better. I think that SS Windows version though not out all that long, is probably more battle ready than WinTD.
For a while it looked like the world was heading towards WinTD. I think WinTD has stalled a bit and SS Windows version has caught up if not passed it due to the deeper level of functionality present.
I would, at this time, recommend SS over WinTD. Either program will take you some time to get comfortable with it, SS a little longer than WinTD.
After that learning period, SS will prove to be more powerful and easier to use especially when you have to make any adjustments.
I think the extra effort will pay off in the long run.
I would purchase the product of choice directly from the maker and not from the USCF.
No need to have someone in the middle.