First of all, you need to decide which type of input method to use. This depends on what you are going to do, select arrows or tune your equipment by analyzing group sizes. Furthermore, what size of targets you use and whether you have a digital camera or not etc.
There are (so far) four different ways to feed data into ArrowMaster:
The grid-coordinate method is no way obsolete, because it's very cost-effective needing no extra equipment (except the computer, of course). You can draw the grid on the targets yourself, but you get them (especially the smaller indoor targets) more easily by printing them from ArrowMaster. Select File/Print target or press the corresponding toolbar button. See chapter xx for details.
The second method is perhaps the fastest, at least when you are not interested in the arrow numbers, and do not care of the lost accuracy.
In serious testing, the third method is a must. If you want to use arrow identification, you can do it in different ways. When you take a photo of the target, you must include the arrow numbers somehow. One solution is to draw the number close to the hit mark. A felt tip pen is usually better than a ballpoint pen. You can also use thumb tacks, which you have marked with arrow numbers. When you draw the arrow off the butt, put a thumb tack of the same number in the arrow hole.
When you look at the application, on the left there is a three-tab pane. The Tests tab is selected by default. You can add a new test or edit an existing test by pressing the Add and Edit buttons, respectively. You can examine the standard deviation and distance differences between the tests easier by checking the Show graphs box. The program calculates the total score of the tests. The score may not be accurate, depending on the grid resolution.
Sometimes it may happen that you have shot the test using a certain grid setting, but enter the values with the default grid. For example, when shooting indoors, your grid may cover only the gold and be very tight with 5mm cell size. If you type in the results using the default grid, the groups seem to be on the seven or eight in the lower left corner. You can correct the situation without re-entering the values by changing the grid settings and then pressing the Use current grid in the Tests panel.
You can add and edit bow and arrow types by clicking the Bow types and Arrow types buttons. The names of the types must be unique (naturally). All other data is optional, but I recommend filling them in, because then it's much easier to compare the results later on.
The Bow type contains currently some free text and directly measurable data:
The Arrow type contains currently some arrow specific data:
When you selected the test, you noticed that a bunch of black dots appeared on the target. By selecting the Hits tab you see a list of ends, and by selecting one of these, there appears a list of hits. At the same time the corresponding dots on the target turn red. The green dot represents the center of the group.
You can add new hits by clicking the Add button. If you have written the hit data in the order x, y, sector, arrow number, you should be able to key them in quite easily. Each time you press the Enter-key, the input focus moves to the next field. When in the arrow field, the hit data is recorded and the fields are cleared for a new hit.
If the current end already contains the given arrow number, the program asks, if you want to start a new end. You can also click the Done button, which ends the input for the current end. This is usually necessary only for the last end or when recording hits without arrow identification.
If some end seems to be especially unsuccessful (difficult wind conditions, for example), you can Ignore it. The data is not deleted, but the hits are not included in the calculations.
You can add or edit the comment by selecting an end and clicking the Edit comment button.
The order of the ends can be changed using the up and down buttons. Only one end at a time can be moved up or down.
About the sample data: in 'test 1' I really shot ten arrows at a time, and pretty quickly for that matter. There were some really bad shots, but I included them also to test the filter feature. In the second test I shot them in groups of five, because the other archers were waiting on the shooting line, while I scribbled the numbers on a piece of paper.. Even if it takes some time to write down the arrow position, it is relatively easy.
I looked the x and y coordinates, and if the arrow had hit on the edge of a cell, the sector. I kept these in mind while I pulled the arrow and checked the number from the shaft. Then I wrote the numbers in respective columns on a sheet of paper. It takes about one minute to do this with ten arrows, if you are quick. If you happen to have an assistant, it takes almost no time at all. Anyway, it is much quicker than the classic method. There you have to check the shaft number before pulling out the arrow to be able to decide, in which target picture you are to draw the mark.
Another and even quicker way is to use a new target for each end. If you print the target using ArrowMaster and use a photocopier, you get very cheap targets. When one end has been shot, just mark the arrow numbers at each hole. Back home, you can easily enter the cell coordinates into the program. This method is especially efficient, if you are doing an equipment test and do not need care about arrow identification.
You may find working with coordinate and sector numbers a little clumsy. If you have the punctured target at hand, you can try entering the hits using mouse clicks. Select Use mouse and mark the hits in appropriate locations. For each hit you can type in the arrow number in the Arrow field.
You can also use the mouse to edit the hit data afterwards. When you put the cursor on a hit, after a short delay a 'tool tip' box appears below the hit, containing hit info. When you click the hit, the hit data dialog opens up in the left panel. You can also ignore and reset individual hits using this dialog.
If you have a digital camera or can scan the target faces to picture files, you can use the perhaps most convenient and accurate way to enter data: click Use picture file. One good example would be an outdoor FITA competition. There you cannot use a custom target face with a grid drawn on it, nor can you take it home with you. But you can leave your arrows in the target when the other archers have taken theirs, and take a quick photo. Digital photos cost virtually nothing at all. You can experiment with the picture resolution, but very probably even a low resolution (640*480) is quite sufficient.
When using this feature you should set the zoom ring setting so that it matches the outermost ring in the picture. In this example it's the six ring, but especially with larger target faces you can use the zoom of the camera, as long as all arrows fit in.
When the picture appears in the window, you must first mark the center of the target, then the outermost ring; the program needs to know the resolution. I strongly recommend that you maximize the application window before doing this, because then working with the mouse is much easier.
Usually the picture files loaded from digital camera are named according to date and a running number, or something like that. If you have many files, it's often hard to keep in mind which files have been processed and which are not. To avoid this confusion, the program moves the file into .\handled folder in your ArrowMaster folder. This happens when you press Done. If you happen to load a wrong picture file, press Cancel, and the file is not moved.
 |  |
The Arrows tab shows the arrows sorted by standard deviation or by distance. The values are calculated with or without normalization, depending on the view selection. The group center is calculated as an average of all hits of an arrow. The distance between this point and the center of the target is displayed in the table. So, a good arrow must have both the standard deviation and the distance as small as possible. The distance is expressed in centimeters.
When you can select a row, the hits of the arrow are highlighted as red dots, or if View/All hits is not active, only the selected arrow hits are displayed. The center of the hits is shown as a green dot.
You can ignore the hits of arrows by selecting the arrows and clicking Ignore sel. arrows. All the arrows' hits are then ignored when performing the calculations. This way you can examine the behaviour of the other arrows without the disturbance of the selected arrows. The button text changes to Reset sel. arrows, so you can restore the ignored hits, if you want.
Sometimes arrows break. Usually there is no point keeping the data anymore, so just select the arrow and press Remove sel. arrows.
You can change the order of the arrows by clicking the column headers. The first click sorts in ascending order, the second click in descending order etc. You can examine the differences visually by selecting the Show arrow graphs box:
Instead of selecting the best arrows manually, you can let the program do it. Select the number of 'good' arrows you need. There are two different methods to use: select the best set or ignore the wrost arrows.
If you choose the ignore-option, the program removes the worst arrow, recalculates the values and repeats this, until the selected amount of arrows is left.
The definition of a 'good' arrow is simple, but in the real world the selection of a set of good arrows is not so straightforward. The set must be a good match, i.e. the arrows must hit the same spot of the target. There may be some arrows with a small deviation, but if their path is different from the others, they are no good for competitions (unless there are enough 'good' arrows going in the same direction).
The best set selection algorithm would be trying all possible arrow combinations and select the ones giving the smallest group. When you have a dozen arrows, selecting three best out of these would require calculating group sizes of 220 combinations. This poses no problem to any computer. But if you are preparing for World Championships, for example, and want to select six best out of, say, four dozen arrows, you should choose from over 12 million combinations. This might take a while.
The program uses a compromise. The first priority for the selected arrows is that the centers of their groups are as close each other as possible. The groups with the smallest distance between their centers can then be compared against each other to find the smallest group standard deviation.
One arrow is taken as the centerpoint, and the distances to the other arrow group centers is calculated. The closest arrows are selected in the set. This calculation is repeated for each arrow. The result usually contains many identical sets of arrows. The duplicates are ignored. Finally, the remaining sets are sorted by their standard deviation. The 3 out of 12 arrows selection takes 132 calculations, and the 6 out of 48 arrows takes 2256 calculations, which is not very bad.
There is yet another option to use, when selecting the best set: either sort the sets by overall group size or by the mutual distance between the group centers. These options may give a little different arrow sets. I suggest you try them both, scan through the sets by using the buttons on both sides of the St.dev field, use the Show set graphs, and then decide, which is the optimal set.
If you have many arrows to choose from, you probably haven't the time or endurance to test them all a one time. Then it's a good practise to test them in parts, say, six to ten arrows at at time. If there are significant differences between the behaviour of the arrows (there usually are), you should use one arrow as a reference arrow. When you include the refrence arrow in all test sessions, you can safely combine the different tests by using Normalization/By one arrow (selected from the tool bar, normalization settings). Then you choose the arrow from the list, and the program moves the hits of all tests so that that the reference arrow is located at the center of the target. Then you should press Save current locations of the hits.
The toolbar contains a bunch of controls. They consist of:
Load from file, clear existing data
'Normal' way of opening a data file. All existing data is cleared, and data is loaded from the selected file. ArrowMaster data files are xml-format files, and the default directory is .\arrowmaster\data.
Load from file, add to existing data
Enables combinating data from several data files.
Save all data to a file
'Normal' way of saving data to a file. Data of all tests, bow and arrow types and most settings are stored.
Save selected tests to a file
Only the selected tests are stored.
Print target
Opens the target print dialog, where you can select the target size and coloring. From here you can print the target face and avoid the task of drawing the grid on an existing target. The current grid settings are used when printing the grid.
There are a few target color choices, and a fine adjust option. The adjustment covers +-3%, which means +-6mm for a 40cm spot target (which has a 20cm diameter). Each printer produces a slightly different result, so if you want a perfectly right-sized target, you must do some experimenting. The target is printed using the options given on the Options pane. Only spot targets are printed, however.
If you choose the custom target face, you can define the colors freely by clicking the Colors button.
There are several ways you can use these self-made targets. Especially using a copier the costs should not be a big problem against using the targets quite freely.
The normalization can be turned on and off by using the check box. If you are comparing the arrow characteristics, it may be useful to turn it on, while when interested in other equipment tuning, it's usually better examine the true arrow hit positions.
Clicking the 'wrench' icon opens a dialog, where the normalization method can be chosen from By tests, By ends or By arrows. The Filter feature uses this setting, too. When experimenting with these options, you should always be very honest to yourself, there is no point in fiddling with the settings just to make the results show better!
The results can be examined 'normal' mode or projected to the vertical or horizontal line. The vertical projection in intended for nocking point/tiller adjustment tests, the horizontal projection for button tension and center shot adjustment etc.
The grid can be visible or hidden. When typing in the hits, it's easier to check the input when the grid is visible. Otherwise, it is better to keep it hidden.
The size and density can be adjusted quite freely. The grid settings are stored along the hit coordinate data, so that the grid changes do not affect the existing data.
All hits
When you want to examine the hits one particular end or arrow, it's better to switch the other hits off. On the other hand, when you want to see the big picture, it's easier to see the hit pattern relative to the other hits when 'All hits' is on.
Ignored hitsYou can show or hide the ignored hits with this option. If there are a lot of ignored arrows, for example, they tend to clobber the display. On the other hand, the effect of the filter shows more clearly when the ignored hits are visible.
Hit sizes
These settings are there to satisfy different opinions of how the hits should be displayed. Using fixed size the dot diameter can be given in pixels. Proportional size uses the given arrow diameter in millimetres. This setting is used in score calculations, too. In theory, you can check how many points you would gain, if you would have used the maximum allowed arrow diameter :)
You can filter out unsuccessful shots with this feature. If you feel during the test session, that some shot goes really bad, the arrow drops from rest for example, do not record it. All others should be included, because you may not know, whether the bad score depends on the properties of the arrow, at least partially.
I'm not good in mathematics, but I've been told that 1.96 * standard deviation gives 95% confidence interval, and is a commonly used limit for purposes like this. Here you can experiment by changing the factor and turning the filter on and off. The ignored hits are displayed as grey spots. The standard deviation calculation is made for one end at a time.
Group center
The center of the group is displayed as a green spot when a end or an arrow is selected. If you do not want to see it, turn it off.
All group centers
All group centers are displayed (tests, ends or arrows).
Group size
The size of the group is displayed as a gray circle when a end or an arrow is selected. The circle denotes the 95% probability area of the group.
Most program controls are both in the menu and the toolbar. Some controls, like export and color settings can be used through the menu only:
Export
Hit data can be exported to a text file to be used in external programs like Excel. So far there has been no need to it, but if need arises, this feature may be developed further.
Colors
Most of the colors of the user interface can be modified, like the dot, grid and bar graph colors.
All user interface texts are modifiable. The default, built-in UI texts are written into the default.lng file in the ArrowMaster directory every time the program starts. If you want to change some or all of the texts, copy this file to another name and edit the texts. Each line of the file format contains one display text. The first field in the line is the name of the text, the following field is the text to show in the display, the third is the optional mnemonic character of the text. For example:
pb_cancel Cancel C
The fields are separated by tabulators. The mnemonic characters are not used so far, except in the menu items. Some of the longest texts (usually in messageboxes) are divided on multiple lines. The linefeed character is "\n".
If you are not changing all texts, you can delete lines which are not modified. The default values are used for other texts.
If you go through the trouble of translating all the texts (there are about 270 of them), maybe you would like to share the result with others? Send the file to me, so I can put it here to be downloaded. For example, here are the first files (right-click the link and select 'Save Target As...'):
| File | Language | Author |
| fin.lng | finnish | Jukka Ojala |
| french.lng | french | Patrick Canler |
To use the modified UI texts, click Change language-button and load the customized language file.
Personally, I find the default font a bit difficult to see, especially when using small font sizes. That's why I added a possibility to use bold font in most display texts. To set bold font on, select the Bold font checkbox.