diff --git a/tko/compose_query.cgi b/tko/compose_query.cgi
index 012c383..8b66145 100755
--- a/tko/compose_query.cgi
+++ b/tko/compose_query.cgi
@@ -66,7 +66,8 @@
     'user': 'tag',
     'status': 'tag',
    
-    'time_daily': 'tag',
+    'time': 'tag',
+    'time_daily': 'time',
 }
 
 
@@ -138,22 +139,6 @@
 def create_select_options(selected_val):
 	ret = ""
 	for option in sorted(frontend.test_view_field_dict.keys()):
-                ## exceptional handling of id and time :(
-                ## To do: when we have more then two such prohibitions,
-                ## something better should be implemented
-                ## e.g. let frontend.test_view_field_dict
-                ## have tuples as a values: (next_field,bAllowGrouping)
-		if option == "id": 
-			## do not allow to group by id 
-			## because it may result in jumbo data transacted
-			continue  
-		if option == "time": 
-			## we have time_daily and time_weekly 
-			## in both drop down menus
-			## They are two different clones of "time"
-			## just 'time' should be avoided due to big
-                        ## data transfer
-			continue  
 		if selected_val == option:
 			selected = " SELECTED"
 		else:
@@ -179,8 +164,6 @@
         ## - possibly, expect more various refinements for different fields
         if field_name == 'kernel':
                 return  map_kernel_base(header)
-        elif field_name.startswith('time') and header != None:
-                return datetime.date(header.year, header.month, header.day)
         else:
                 return header
 
@@ -289,7 +272,7 @@
 		if field_map.has_key(row):
 			dy = field_map[row](y)
 		y_header = header_tuneup(row, dy)
-		link = construct_link(None, y)                
+		link = construct_link(None, y)
 		cur_row = [display.box(y_header, header=True, link=link)]
 		for x in test_data.x_values:
 			## next 2 lines: temporary, until non timestamped
diff --git a/tko/frontend.py b/tko/frontend.py
index 046d8ff..0861bcb 100755
--- a/tko/frontend.py
+++ b/tko/frontend.py
@@ -1,7 +1,7 @@
 #!/usr/bin/python
 import os, re, db, sys, datetime
-MAX_RECORDS = 10000
-MAX_CELLS = 100000
+MAX_RECORDS = 20000L
+MAX_CELLS = 300000L
 
 tko = os.path.dirname(os.path.realpath(os.path.abspath(__file__)))
 client_bin = os.path.abspath(os.path.join(tko, '../client/bin'))
@@ -55,27 +55,12 @@
 		self.x_values = smart_sort(data.keys(), x_field)
 		# List of rows columns (y-values)
 		self.y_values = smart_sort(list(y_values), y_field)
-		if len(self.y_values)*len(self.x_values) > MAX_CELLS:
+		nCells = len(self.y_values)*len(self.x_values)
+		if nCells > MAX_CELLS:
 			msg = 'Exceeded allowed number of cells in a table'
 			raise db.MySQLTooManyRows(msg)
 			
 
-def truncateTimeFieldsInAllRecords(rows, pos):
-	## reduces hours:min:sec to 00:00:00 in time stamps
-	## in each record i.e. in each element of "rows"
-	## Argument pos is equal to position of x-field or y-field
-	## depending in which field we are going to make such truncation
-	def truncateTimeFieldInOneRecord(row):
-		altRow = list(row)
-		if altRow[pos]!=None:
-			altRow[pos] = datetime.datetime(
-				altRow[pos].year,
-				altRow[pos].month,
-				altRow[pos].day
-				)
-		return tuple(altRow)
-	return map(truncateTimeFieldInOneRecord, rows)
-	            
 def get_matrix_data(db_obj, x_axis, y_axis, where = None):
 	# Searches on the test_view table - x_axis and y_axis must both be
 	# column names in that table.
@@ -87,10 +72,6 @@
 	group_by = '%s, %s, status' % (x_field, y_field)
 	rows = db_obj.select(fields, 'test_view',
 			where=where, group_by=group_by, max_rows = MAX_RECORDS)
-	if x_field.endswith("time"):
-		rows = truncateTimeFieldsInAllRecords(rows, 0)
-	if y_field.endswith("time"):
-		rows = truncateTimeFieldsInAllRecords(rows, 1)
 	return status_data(rows, x_field, y_field)
 
 
@@ -107,8 +88,7 @@
 	'user'          : 'job_username',
 	'status'        : 'status_word',
 	'time'          : 'test_finished_time',
-	'time_daily'    : 'test_finished_time',
-	'id'            : 'test_idx',	    
+	'time_daily'    : 'DATE(test_finished_time)'
 }
 
 def smart_sort(list, field):