Tip #1 What do the numbers mean? When you start your integration wait and examine the numbers you see on the screen carefully. Particularly take note of the "Correlation" column on the far right of the output. Ideally these numbers should average to more than 0.5. If they are significantly lower you may have a problem. Check the column "percent less than two sigma" and "number of sigma". If the former is consistently ABOVE fifty percent even when there are a large number of data the chances are that you are integrating a centered cell as a primitive cell.
If that is the case you should stop integration ( <CTRL/C> ) and delete all of the saint output files ( rm *._ls and rm *.raw ). If you have a monoclinic cell then the only possible centering condition is C centering; check out the matrix0.p4p file to see if you have it. If you have a higher symmetry cell you should consult the matrix0.p4p file (and a texbook if need be) to decide what type of centering you have based on the systematic absences in your hkl values. Then you should change the value for centering absences with the "p" command within SAINT from 0 to the correct number before re-running SAINT (review the SAINT help file).
If you would like to understand what the other numbers on the screen are telling you consult the cryptic SAINT manual.
Tip #2 What should I look at in my SAINT output? The SAINT output (filenamem._ls) is large and contains a ton of information that for the most part is uninteresting to the average user. The SAINT manual discusses the output if you are interested in every detail. The most important things that you should be aware of are covered here.
About one third down the first page of the output is the number of reflections integrated (number written to filenamem.raw). You should write this number down as it may be useful to know when you are writing up.
The last page of the output has most of the useful information on it. Under the heading "Coverage Statistics" are several columns of numbers. The top of each column corresponds to the strong low angle data and the bottom of each collumn corresponds to the weaker high angle data.
The column labled "Rsym" is an "internal R-value" of your data before any corrections have been done on it. (corrections are done in XPREP and SADABS) If this is very high it is an indication that either your statistics are bad, your data is bad or some problem has occured such as crystal motion during collection. If, for instance, you told SAINT that the cell was monoclinic and it is really triclinic then your Rsym will reflect a horrible lack of agreement. If this is the case don't worry. Your integration is probably fine you just got statistics on the wrong cell. The only time you will have to rerun SAINT is if you didn't put in the correct crystal system (ie you missed the centering) or the crystal moved. The former problem can be identified by looking at the last column "percent less than two sigma" [ %<2(sigma) ] Look at the numbers in this column near the top (the strong data). If they are consistently close to 0.5 then you may need to reintegrate in a centered cell (see tip #1).
If, after looking at your data closely, you are sure that the centering was correct you should consult Dr. Hollander about the possibility that your crystal has moved. This is not a terminal problem.
About half way down the last page of the output are the refined unit cell parameters. You should write these down and enter them into teXsan later. You should also take note of how many reflections were used to calculate this cell and the angle range for the reflections used in obtaining the final cell parameters. This number is just above the cell dimensions in parentheses.