• ▼MSDN Library Visual Studio 6.0
  • ▶Welcome to the MSDN Library
  • ▶Visual Studio Documentation
  • ▶Visual Basic Documentation
  • ▶Visual C++ Documentation
  • ▶Visual FoxPro Documentation
  • ▶Visual InterDev Documentation
  • ▶Visual J++ Documentation
  • ▶Visual SourceSafe Documentation
  • ▶Tools and Technologies
  • ▶Microsoft Office Development
  • ▶Platform SDK
  • ▶SDK Documentation
  • ▶DDK Documentation
  • ▶Windows Resource Kits
  • ▶Specifications
  • ▶Knowledge Base
  • ▶Technical Articles
  • ▶Backgrounders
  • ▶Books
  • ▶Partial Books
  • ▶Periodicals
  • ▼Conference Papers
    • ▶PDC 97 Conference Papers
    • ▶PDC 96 Conference Papers
    • ▶Tech*Ed 97 Conference Papers
    • ▶Tech*Ed 96 Conference Papers
    • ▼Tech*Ed 95 Conference Papers
      • ▼Microsoft Access
        • ▶Advanced Data Access Objects (DAO) for Client Server
        • ▶Advanced Microsoft Access Basic Code
        • ▶Client-Server Application Development Using Microsoft Access
        • ▼Forty-Two Ways to Make DAO Faster--Programming to DAO Using Microsoft Access
          • Introduction
          • Overview
          • Sample Code
          • ▼50 Ways to Speed Up DAO Code
            • Replace Find with Seek
            • Use table-type Recordsets instead of dynaset or snapshot-types
            • Open attached tables directly by opening the database directly
            • Use snapshots instead of dynasets for remote data (but beware of memos!)
            • Supply the DB_FORWARDONLY option on ODBC snapshots
            • With ODBC dynasets, use CacheStart, CacheFill, and CacheSize functionality
            • Store queries instead of using dynamic SQL text - especially low memory machines
            • Parameterize queries instead of using dynamic SQL text (especially for ODBC)
            • Open databases exclusive and read-only if for fastest single-user network performance
            • With ODBC SQL statements, use pass-through where possible
            • Replace DAO code loops with the equivalent SQL statements - but not always
            • Always wrap transactions around your DAO code
            • Return to your location using bookmarks instead of using Find
            • In inner loops, use Field objects instead of myRS!fieldname (Q112724)
            • Store infrequently updated tables in the local MDB
            • Reduce your ODBC time-outs to get faster return to DAO
            • Cache tabledef and field property collection references if used many times
            • Speed finds by creating a temporary indexed table and seeking
            • Replace repeated execution of dynamic SQL with a temporary query
            • Use the DB_APPENDONLY flag if you're adding records to dynasets, esp ODBC
            • Marginal: replace variants with specific data types
            • Replace short memo fields with long text fields
            • For snapshots, select just the fields you need instead of all fields
            • Split .mdb files that have lots of tables to minimize DAO collection initialization time
            • Try InStr() instead of Like if you're not concerned with foreign accent chars
            • Replace use of .Filter with Seek (or a query)
            • Re-query recordsets instead of reopening them
            • Dim objects and Set them rather than using lengthy references
            • Replace text collection lookup with ordinals (but field object is better)
            • Take advantage of default collections and drop the extra name
            • Replace old ListTables (etc.) code with collection based code
            • 'Posting' queries are faster than explicit code
            • Refresh current field values with Move 0 instead of MoveNext/Previous
            • Abandon an Edit or AddNew with Move 0
            • Clone dynaset recordsets instead of opening new ones
            • Compile modules and queries before distributing your application
            • For ODBC data with OLE or Memo fields use dynasets instead of snapshots
            • On ODBC data use find-only on indexed fields, otherwise open a new cursor
            • Replace Variant string functions with '$' functions if using strings
            • Replace use of .Sort with specifying an index (or a query)
            • Replace floating point numbers with integral numbers
            • Add the DB_DENYWRITE flag if no one else will be writing
            • Avoid refreshing collections unless you absolutely have to
          • Conclusion
        • ▶Microsoft Access to Microsoft Office: Applied Integration Techniques
        • ▶Writing Wizards for Microsoft Access
        • ▶Using Microsoft Access Forms Like the Masters
        • ▶Upsizing Microsoft Access-based Applications
      • ▶Microsoft BackOffice
      • ▶Microsoft Exchange
      • ▶Microsoft Mail
      • ▶Microsoft Office
      • ▶Microsoft SQL Server
      • ▶Microsoft Visual Basic
      • ▶Microsoft Visual FoxPro
      • ▶Microsoft Windows NT
      • ▶Multimedia