Electron beam exposure system with increased efficiency of exposure operation

A method for exposing a pattern on an object by an electron beam having a selectively shaped cross section, one shot by one shot, comprising the steps of:

• generating pattern data that specifies the cross section of the electron beam as a time sequential mixture of first pattern data specifying the number of repetitions of the shots to be made and second pattern data specifying the cross sectional shape of the electron beam, said time sequential mixture lacking the first pattern data when the shot of a given cross section is to be made without repetition;
• sorting the first and second pattern data such that said first and second pattern data form parallel data wherein said first pattern data and said second pattern data have the same timing;
• said step of sorting comprising a step of reading data contained in said time sequential mixture of said first pattern data and said second pattern data, discriminating whether the data that has been read represents the first pattern data or not, outputting the data that has been read as the first pattern data and simultaneously the data that follows immediately behind the data that has been read in the time sequential mixture as the second pattern data when it is discriminated that the data that has been read is the first pattern data;
• and outputting dummy data as the first pattern data and outputting the data that has been read as the second pattern data when it is discriminated that the data that has been read is the second pattern data;
• and
• conducting the shot by the electron beam with the cross sectional shape according to the second pattern data for a number of times that is determined by the first pattern data.

A method for writing a pattern on an object by an electron beam having a selectively shaped cross section based upon a first data train, said first data train being a time sequential mixture of first type and second type data pairs, each first type data pair comprising first type data and second type data, said first type data specifying the desired cross section of the electron beam and said second type data specifying the number of repetitions of writing the pattern on the object by the electron beam of which the cross section thereof is specified by said first type data, and each second type data pair comprising the first type data and a third type data that is different from said second type data, comprising the steps of:

• reading a data pair in the first data train;
• discriminating as to whether said read data pair is the first type data pair or the second type data pair;
• and
• deleting the data pair that follows immediately behind said read data pair when said read data pair is the first type data pair, to form a second data train;
• and
• writing the pattern on the object according to the second data train.

A method of sorting data that is supplied as a time sequential mixture of first type data and second type data, comprising the steps of:

• receiving data included in said time sequential mixture;
• discriminating whether the received data is the first type data or the second type data;
• transferring the received data to a first output bus when it is discriminated in said step of discrimination that said received data is the first type data;
• transferring the received data to a second, different output bus when it is discriminated in said step of discrimination that said received data is the second type data;
• and
• transferring data that follows immediately behind said received data, when it is discriminated in said step of discrimination that said received data is the first type data, to said second output bus, said step of transferring said data that follows immediately behind said received data being achieved simultaneously with said step of transferring the first type data to said first output bus.

A method for compressing data of a data train comprising a time sequential mixture of first type data pairs and second type data pairs, each of said first type data pairs comprising first type data specifying a desired operation and associated, second type data specifying the number of repetitions of said operation specified by the associated first type data, said second type data pair comprising third type data and said second type data, said method comprises the steps of:

• reading a data pair in said data train;
• discriminating whether said read data pair is the first type data pair or the second type data pair;
• and
• deleting the data pair that follows immediately behind said read data pair when said read data pair is the first type data pair, to form a second data train that is substantially identical with the first data train except that the data in the first data train that follows immediately behind the first data pair is eliminated.