There seems to be some confusion regarding measurement systems used to qualify product versus those used to conduct a capability study. Depending on how tight the process is and where it is located with relationship to the specifications, a measurement system for determining whether the product is acceptable or not does not have to have the same discrimination or accuracy as a measurement system used to conduct a capability study. In fact, a go/no-go gage can be used to qualify product.
However, when a process capability is being conducted, the measurement system needs to have good accuracy and discrimination. The instrument must be able to measure to an additional decimal place over the product requirement. If the instrument used to qualify product is reading to the same number of decimal places as the product requires, an acceptable MSA and capability study will be very hard to accomplish.
The instrument used in the process capability must have an acceptable gage R&R. This is accomplished by selecting parts that are representative of the total process spread or tolerance. If the appraisers are well trained, product part to part variation will be the biggest contributor and repeatability and reproducibility will be relatively low contributors of variation.