McPIQ On-Line is based on a database containing two types of information on crop varieties:

• Attributes—descriptive variety characteristics such as a maturity rating, resistance to particular diseases, or novel traits such as herbicide tolerance. Most of the descriptive attributes come from information supplied to register varieties.
• Performance Attributes—performance information such yield, days to maturity, etc. Performance data comes from annual trials conducted in Manitoba by the Manitoba Crop Variety Evaluation Team. Most of the data in the database presently is for yield, but days to maturity, height, test weight, and other data is included if available. The number of trials and the specific locations available depend on the crop—for some crops, testing occurs at up to 14 sites annually, while others are tested at only a few sites. Not all varieties are tested in all years.

McPIQ is best described as a Variety Decision Support Tool, which growers use to identify varieties which suit their agronomic and management requirements. While growers can select varieties directly from a list of all varieties for a crop, McPIQ is best used to pre-select varieties which match agronomic or other descriptive criteria. Further, more detailed comparison is then conducted on these varieties. Thus the grower decides which variety characteristics are most important, and finds information to support a decision as to which variety (or varieties) is best suited for the grower's needs.

McPIQ involves the following basic steps:

The two options are:

1. From a list of varieties that match a particular agronomic or disease characteristic or attribute. This is the most powerful option, since a lengthy list of varieties in a crop like canola or peas can quickly be reduced to only the few which match your criteria.
2. Directly, from a list of all varieties for this crop. You may be approaching this with several varieties already in mind. If so, you can choose to select those varieties of interest directly from the checklist of all varieties available for that crop.

Begin with the attribute which is most important to you as a management consideration. In the selection box, attributes appear in order of common importance—for example, in oats, lodging and crown rust would be considered by many growers to be among the most important attributes, so they appear at the top of the list; in canola, blackleg resistance commonly would be considered to be of primary importance, so blackleg appears at the top of the list for canola. There are three types of descriptive attributes, each of which will result in the next screen step being slightly different (further explanation with next screen).

There are three types of descriptive attributes, each of which will result in the action of this screen being slightly different. They are: ratings, descriptions, or numeric ranges, which differ as follows:

1. Ratings. Examples include disease ratings on very good to very poor or highly resistant to highly susceptible scales, maturity on a very late to very early scale, size on a large-medium-small scale, etc. Selections are made on a "meets-or-exceeds" basis: for example, if for canola-blackleg you choose "moderately resistant" you will also automatically get varieties which exceed this level of resistance.
2. Descriptions. Examples include color attributes, barley awn characteristics, novel traits such as herbicide resistance, etc. . Selections are made on a "meets" basis: for example, if for canola-novel trait you choose "Roundup tolerant" you will get only varieties which meet this criterion.
3. Numeric ranges. Examples include some maturity information, protein content, test weight, etc. In this case, the screen will say something like:

The result of the selection above will be a screen that lists varieties which match the chosen criterion. If no varieties match, you will be given opportunity to select another attribute or attribute value.

Whether you chose the direct selection option or the attribute matching option, you will now have a list of varieties on the screen. Choose up to five for which you wish detailed comparative performance and/or descriptive information. You are currently restricted to five varieties so that data processing time does not become excessive.

A check (or reference variety) is a variety you wish to compare the performance of the others against. The best choice for a check variety would be one with which you have some experience and has been on the market for at least two or three previous seasons.

Performance measures or attributes are data which has been collected in field trials across Manitoba. Average yield in kg/ha is always determined. Where resources allowed, data such as test weight, height, days to maturity, protein content, pod height, etc. have also been collected, and if available were included in the database.

There are two types of output from which to choose:

• DETAILED - Which for the selected varieties provides the selected performance measure (vs. the check variety) plus other variety description attributes.
• BRIEF - Which provides only the selected performance measure (vs. the check variety).

You may also choose which trials (location-years) to include in the calculation of variety performance. The options are:

• Use all available locations. This provides the most meaningful result, because it covers performance in different seasons and locations, thus providing the best indicator of likely performance in the coming season, whose growing conditions cannot reliably be predicted. The variety that performed well in many different environments is one that is likely to perform well in the coming season. In most cases, we simply do not have the resources to conclusively test variety performance in specific environments or under specific management practices.
• Use all years for a selected location. Only a few locations have sufficent data for this option to be valuable, and these are indicated in the location list with a "*" at the end of the town name.
• Use all locations for a selected year.

Note that performance data will only be provided if there are at least two applicable trials (location-years) in the database.

The output will look something like:

Detailed Crop Information

A caution: only a few locations have sufficient data for this option to be valuable (i.e. multiple years of experience for varieties in question). Those most likely to have multiple years of data are indicated in the location list with a "*" at the end of the town name, but whether or not there is sufficient data will depend on the variety. Output will not be provided if there are less than two trials to include.

Output will not be provided if there are less than two trials to include.

The output table looks like:

An important feature of the reported data is that "all available trials" means all trials in which both the variety being calculated and the check variety appeared. The premise is that two varieties can be compared only if they were grown in the same trial.

Thus the "check average performance" is based only on the same trials as the variety being compared, even though the check variety may also have been grown in many other trials—this is why the check average frequently differs depending on the variety being compared.

The "number of trials" in the table is the number of trials in which both the variety and the check variety appeared.

"Performance compared to check" is given as a percent of check for yield, and as a difference from check for all other performance measures.