Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 15:00:00 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.46642400000000000526(real)/0.44000000000000000222(user) seconds
1000000 times empty loop took 0.47663400000000000212(real)/0.46000000000000001998(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1049022 times empty loop took 0.48865399999999997727(real)/0.46999999999999997335(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1073379 times empty loop took 0.50004000000000004(real)/0.48999999999999999112(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Addition of 2 [1073379] arrays in a scalar-addition loop, without support for simple array arithmetic: 6.1593390000000001194(real)/6.1399999999999996803(user)s
Addition of 2 [1073379] arrays in a scalar-addition loop, with support for simple array arithmetic: 6.2312240000000000961(real)/6.2099999999999999645(user)s
10x array[1073379] addition without support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.02037599999999999828(real)/0(user)s
10x array[1073379] addition with support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.9735869999999999802(real)/0.94999999999999995559(user)s

Comparing the special 2-D indexing function 2Dindex to nDindex

1073379 times 2Dindex[1,2,3,4]: 2.647101999999999844(real)/2.6200000000000001066(user)s
1073379 times nDindex[1,2,3,4]: 2.8510249999999999204(real)/2.8300000000000000711(user)s

Some loops to time the effect of $UseConstantsLists (and the '?' last-value operator):

1073379 times 'progn[2,add[2,kk]]' took 3.6231759999999999522(real)/3.6099999999999998757(user) seconds with $UseConstantsLists=0
1073379 times 'progn[2,?add,?ran]' took 2.2138110000000001953(real)/2.2000000000000001776(user) seconds with $UseConstantsLists=0
1073379 times 'progn[2,add[2,kk]]' took 3.4334069999999998757(real)/3.4100000000000001421(user) seconds with $UseConstantsLists=1
1073379 times 'progn[2,?add,?ran]' took 1.9895400000000000862(real)/1.9699999999999999734(user) seconds with $UseConstantsLists=1
1073379 times 'var[foo[?ran]]'=0.1766426425429159508 took 3.1011769999999998504(real)/3.0800000000000000711(user) seconds with $UseConstantsLists=0
1073379 times 'var[foo[?ran]]'=0.1766426425429159508 took 1.8659259999999999735(real)/1.8400000000000000799(user) seconds with $UseConstantsLists=1
1073379 times 'var'=0 took 0.80767199999999994553(real)/0.78000000000000002665(user) seconds with $UseConstantsLists=0
1073379 times 'var'=0 took 0.80751400000000006507(real)/0.79000000000000003553(user) seconds with $UseConstantsLists=1
1073379 times 'var[$loop]'=1073378 took 1.9573760000000000048(real)/1.9399999999999999467(user) seconds with $UseConstantsLists=0
1073379 times 'var[$loop]'=1073378 took 1.7245249999999998636(real)/1.6999999999999999556(user) seconds with $UseConstantsLists=1
1073379 times 'var[foo[$loop]]'=1073378 took 3.120886000000000049(real)/3.1000000000000000888(user) seconds with $UseConstantsLists=0
1073379 times 'var[foo[$loop]]'=1073378 took 1.870085999999999915(real)/1.8500000000000000888(user) seconds with $UseConstantsLists=1
1073379 times 'var[foo[bar[$loop]]]'=1073378 took 4.2752080000000001192(real)/4.2599999999999997868(user) seconds with $UseConstantsLists=0
1073379 times 'var[foo[bar[$loop]]]'=1073378 took 3.0376120000000002008(real)/3.0099999999999997868(user) seconds with $UseConstantsLists=1
expression-loops per second:
	1073379/2.2767633749999998116=471449.51986940676579 (real time)
	1073379/2.2562500000000000888=475735.84487534622895 (user time)
Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 15:02:04 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.46616400000000002279(real)/0.4500000000000000111(user) seconds
1000000 times empty loop took 0.46610099999999998754(real)/0.4500000000000000111(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1072728 times empty loop took 0.49965100000000001179(real)/0.47999999999999998224(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 15:02:10 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.46705700000000005545(real)/0.4500000000000000111(user) seconds
1000000 times empty loop took 0.47156799999999998718(real)/0.4500000000000000111(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1060292 times empty loop took 0.49333000000000004626(real)/0.47999999999999998224(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1074627 times empty loop took 0.4999239999999999795(real)/0.47999999999999998224(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Addition of 2 [1074627] arrays in a scalar-addition loop, without support for simple array arithmetic: 6.1390130000000002752(real)/6.1200000000000001066(user)s
Addition of 2 [1074627] arrays in a scalar-addition loop, with support for simple array arithmetic: 6.2224769999999995918(real)/6.2000000000000001776(user)s
10x array[1074627] addition without support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.020364000000000000157(real)/0(user)s
10x array[1074627] addition with support for simple array arithmetic, array addition (add[&z,&x,&y]): 1.127302999999999944(real)/1.1100000000000000977(user)s

Comparing the special 2-D indexing function 2Dindex to nDindex

1074627 times 2Dindex[1,2,3,4]: 2.6430649999999999977(real)/2.6200000000000001066(user)s
1074627 times nDindex[1,2,3,4]: 2.8713630000000001097(real)/2.8500000000000000888(user)s

Some loops to time the effect of $UseConstantsLists (and the '?' last-value operator):

1074627 times 'progn[2,add[2,kk]]' took 3.6387789999999999857(real)/3.6200000000000001066(user) seconds with $UseConstantsLists=0
1074627 times 'progn[2,?add,?ran]' took 2.2074920000000002318(real)/2.1800000000000001599(user) seconds with $UseConstantsLists=0
1074627 times 'progn[2,add[2,kk]]' took 3.4491860000000000852(real)/3.4199999999999999289(user) seconds with $UseConstantsLists=1
1074627 times 'progn[2,?add,?ran]' took 1.9851419999999999622(real)/1.9699999999999999734(user) seconds with $UseConstantsLists=1
1074627 times 'var[foo[?ran]]'=0.1766426425429159508 took 3.102485999999999855(real)/3.0899999999999998579(user) seconds with $UseConstantsLists=0
1074627 times 'var[foo[?ran]]'=0.1766426425429159508 took 1.86617999999999995(real)/1.8400000000000000799(user) seconds with $UseConstantsLists=1
1074627 times 'var'=0 took 0.80828199999999994496(real)/0.79000000000000003553(user) seconds with $UseConstantsLists=0
1074627 times 'var'=0 took 0.8092749999999999666(real)/0.79000000000000003553(user) seconds with $UseConstantsLists=1
1074627 times 'var[$loop]'=1074626 took 1.9711220000000000407(real)/1.9499999999999999556(user) seconds with $UseConstantsLists=0
1074627 times 'var[$loop]'=1074626 took 1.7195709999999999607(real)/1.6899999999999999467(user) seconds with $UseConstantsLists=1
1074627 times 'var[foo[$loop]]'=1074626 took 3.133556000000000008(real)/3.1099999999999998757(user) seconds with $UseConstantsLists=0
1074627 times 'var[foo[$loop]]'=1074626 took 1.8693480000000000096(real)/1.8500000000000000888(user) seconds with $UseConstantsLists=1
1074627 times 'var[foo[bar[$loop]]]'=1074626 took 4.3040490000000000137(real)/4.2900000000000000355(user) seconds with $UseConstantsLists=0
1074627 times 'var[foo[bar[$loop]]]'=1074626 took 3.0420780000000000598(real)/3.0099999999999997868(user) seconds with $UseConstantsLists=1
expression-loops per second:
	1074627/2.2859170833333330997=470107.60269264655653 (real time)
	1074627/2.2650000000000001243=474449.00662251655012 (user time)
Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 15:07:59 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.44784699999999999509(real)/0.4500000000000000111(user) seconds
1000000 times empty loop took 0.44794000000000000483(real)/0.4500000000000000111(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1116220 times empty loop took 0.50030300000000005323(real)/0.5(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Addition of 2 [1116220] arrays in a scalar-addition loop, without support for simple array arithmetic: 6.368399000000000143(real)/6.3700000000000001066(user)s
Addition of 2 [1116220] arrays in a scalar-addition loop, with support for simple array arithmetic: 6.4347750000000001336(real)/6.4400000000000003908(user)s
10x array[1116220] addition without support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.00037899999999999999661(real)/0(user)s
10x array[1116220] addition with support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.98370199999999996532(real)/0.97999999999999998224(user)s

Comparing the special 2-D indexing function 2Dindex to nDindex

1116220 times 2Dindex[1,2,3,4]: 2.7248009999999998065(real)/2.7200000000000001954(user)s
1116220 times nDindex[1,2,3,4]: 2.9442599999999998772(real)/2.9500000000000001776(user)s

Some loops to time the effect of $UseConstantsLists (and the '?' last-value operator):

1116220 times 'progn[2,add[2,kk]]' took 3.7551070000000001947(real)/3.7599999999999997868(user) seconds with $UseConstantsLists=0
1116220 times 'progn[2,?add,?ran]' took 2.2740179999999998728(real)/2.2700000000000000178(user) seconds with $UseConstantsLists=0
1116220 times 'progn[2,add[2,kk]]' took 3.5506890000000002061(real)/3.5499999999999998224(user) seconds with $UseConstantsLists=1
1116220 times 'progn[2,?add,?ran]' took 2.0412409999999998611(real)/2.0400000000000000355(user) seconds with $UseConstantsLists=1
1116220 times 'var[foo[?ran]]'=0.1766426425429159508 took 3.198763000000000023(real)/3.2000000000000001776(user) seconds with $UseConstantsLists=0
1116220 times 'var[foo[?ran]]'=0.1766426425429159508 took 1.9167980000000000018(real)/1.9199999999999999289(user) seconds with $UseConstantsLists=1
1116220 times 'var'=0 took 0.81832499999999996909(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=0
1116220 times 'var'=0 took 0.81866499999999997605(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=1
1116220 times 'var[$loop]'=1116219 took 2.0262820000000001386(real)/2.0299999999999998046(user) seconds with $UseConstantsLists=0
1116220 times 'var[$loop]'=1116219 took 1.7647740000000000649(real)/1.7600000000000000089(user) seconds with $UseConstantsLists=1
1116220 times 'var[foo[$loop]]'=1116219 took 3.2459150000000001057(real)/3.2400000000000002132(user) seconds with $UseConstantsLists=0
1116220 times 'var[foo[$loop]]'=1116219 took 1.921526999999999985(real)/1.9299999999999999378(user) seconds with $UseConstantsLists=1
1116220 times 'var[foo[bar[$loop]]]'=1116219 took 4.4491129999999996514(real)/4.4500000000000001776(user) seconds with $UseConstantsLists=0
1116220 times 'var[foo[bar[$loop]]]'=1116219 took 3.1654469999999998997(real)/3.1699999999999999289(user) seconds with $UseConstantsLists=1
expression-loops per second:
	1116220/2.4260465217391304904=460098.34930940606864 (real time)
	1116220/2.4269565217391306788=459925.83303475449793 (user time)
Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 17:18:49 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.44788899999999998158(real)/0.4500000000000000111(user) seconds
1000000 times empty loop took 0.44750500000000004164(real)/0.4500000000000000111(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1117305 times empty loop took 0.50066100000000002268(real)/0.5(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Addition of 2 [1117305] arrays in a scalar-addition loop, without support for simple array arithmetic: 6.3595829999999997639(real)/6.3399999999999998579(user)s
Addition of 2 [1117305] arrays in a scalar-addition loop, with support for simple array arithmetic: 6.4265590000000001325(real)/6.4299999999999997158(user)s
10x array[1117305] addition without support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.00036200000000000001708(real)/0(user)s
10x array[1117305] addition with support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.97735000000000005205(real)/0.97999999999999998224(user)s

Comparing the special 2-D indexing function 2Dindex to nDindex

1117305 times 2Dindex[1,2,3,4]: 2.7317160000000000331(real)/2.7299999999999999822(user)s
1117305 times nDindex[1,2,3,4]: 2.9471349999999998381(real)/2.9500000000000001776(user)s

Some loops to time the effect of $UseConstantsLists (and the '?' last-value operator):

1117305 times 'progn[2,add[2,kk]]' took 3.7592110000000000802(real)/3.7599999999999997868(user) seconds with $UseConstantsLists=0
1117305 times 'progn[2,?add,?ran]' took 2.2759040000000001491(real)/2.2799999999999998046(user) seconds with $UseConstantsLists=0
1117305 times 'progn[2,add[2,kk]]' took 3.5553140000000000853(real)/3.5600000000000000533(user) seconds with $UseConstantsLists=1
1117305 times 'progn[2,?add,?ran]' took 2.0436920000000000641(real)/2.0400000000000000355(user) seconds with $UseConstantsLists=1
1117305 times 'var[foo[?ran]]'=0.1766426425429159508 took 3.2018480000000000274(real)/3.2000000000000001776(user) seconds with $UseConstantsLists=0
1117305 times 'var[foo[?ran]]'=0.1766426425429159508 took 1.9201489999999998837(real)/1.9199999999999999289(user) seconds with $UseConstantsLists=1
1117305 times 'var'=0 took 0.81937099999999996047(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=0
1117305 times 'var'=0 took 0.8196539999999999937(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=1
1117305 times 'var[$loop]'=1117304 took 2.0277569999999998096(real)/2.0200000000000000178(user) seconds with $UseConstantsLists=0
1117305 times 'var[$loop]'=1117304 took 1.7664279999999998871(real)/1.7600000000000000089(user) seconds with $UseConstantsLists=1
1117305 times 'var[foo[$loop]]'=1117304 took 3.2492849999999999788(real)/3.25(user) seconds with $UseConstantsLists=0
1117305 times 'var[foo[$loop]]'=1117304 took 1.9207670000000001131(real)/1.9199999999999999289(user) seconds with $UseConstantsLists=1
1117305 times 'var[foo[bar[$loop]]]'=1117304 took 4.4532160000000002853(real)/4.4500000000000001776(user) seconds with $UseConstantsLists=0
1117305 times 'var[foo[bar[$loop]]]'=1117304 took 3.1681080000000001462(real)/3.1600000000000001421(user) seconds with $UseConstantsLists=1
expression-loops per second:
	1117305/2.4269332173913040762=460377.31569762120489 (real time)
	1117305/2.4256521739130438853=460620.45169385185 (user time)
Current file: examples/benching.xg: Sat Apr 13 14:57:56 2002
A collection of timing benchmarks. Can take quite a while to complete, and doesn't generate
overly nice output! Performance measured in bogoflops, <err> I mean seconds :)
Current xgraph info:
 xgraph.new -progname xgraph examples/benching.xg
 XGraph build gcc[Apr 13 2002 01:50:38], version 200204013
 Executed d.d. Sat Apr 13 20:06:40 2002
 Input files: 'examples/benching.xg' 
 Working directory: /home/bertin/work/src/xgraph
 Last opened on display :0.0/-1/-1 (visual 0x24, 16 planes TrueColor) by bertin


1000000 times empty loop took 0.45549399999999995448(real)/0.46000000000000001998(user) seconds
1000000 times empty loop took 0.44760699999999997711(real)/0.44000000000000000222(user) seconds
Calibrating adapted loop size (should be 0.5s)... 1117051 times empty loop took 0.49982400000000004603(real)/0.5(user) seconds

timing of addition of 2 arrays, either with a for-toMAX[] loop construct (scalar addition), or
with (internal) array addition. The latter is done 10x to increase the reliability of the measured
duration. Procedures are used to ensure that the code is executed from compiled representations
(*EVAL* would not automatically compile the array-addition test procedure [test2]).
On a PIII 450Mhz (linux), test1 completes in around 5.8s, test2 under 1s (meaning that the array
addition is some 60x faster than the looped scalar addition).

Addition of 2 [1117051] arrays in a scalar-addition loop, without support for simple array arithmetic: 6.357604000000000255(real)/6.3600000000000003197(user)s
Addition of 2 [1117051] arrays in a scalar-addition loop, with support for simple array arithmetic: 6.4248729999999998341(real)/6.4299999999999997158(user)s
10x array[1117051] addition without support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.00037500000000000000781(real)/0(user)s
10x array[1117051] addition with support for simple array arithmetic, array addition (add[&z,&x,&y]): 0.89950499999999999901(real)/0.9000000000000000222(user)s

Comparing the special 2-D indexing function 2Dindex to nDindex

1117051 times 2Dindex[1,2,3,4]: 2.7277010000000001533(real)/2.7299999999999999822(user)s
1117051 times nDindex[1,2,3,4]: 2.9460820000000000896(real)/2.9500000000000001776(user)s

Some loops to time the effect of $UseConstantsLists (and the '?' last-value operator):

1117051 times 'progn[2,add[2,kk]]' took 3.7578409999999999869(real)/3.7599999999999997868(user) seconds with $UseConstantsLists=0
1117051 times 'progn[2,?add,?ran]' took 2.2756690000000001639(real)/2.2799999999999998046(user) seconds with $UseConstantsLists=0
1117051 times 'progn[2,add[2,kk]]' took 3.5529760000000001341(real)/3.5499999999999998224(user) seconds with $UseConstantsLists=1
1117051 times 'progn[2,?add,?ran]' took 2.0432199999999998141(real)/2.0499999999999998224(user) seconds with $UseConstantsLists=1
1117051 times 'var[foo[?ran]]'=0.1766426425429159508 took 3.2006280000000000285(real)/3.2000000000000001776(user) seconds with $UseConstantsLists=0
1117051 times 'var[foo[?ran]]'=0.1766426425429159508 took 1.9187950000000000284(real)/1.9199999999999999289(user) seconds with $UseConstantsLists=1
1117051 times 'var'=0 took 0.819108999999999976(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=0
1117051 times 'var'=0 took 0.81921299999999996899(real)/0.81999999999999995115(user) seconds with $UseConstantsLists=1
1117051 times 'var[$loop]'=1117050 took 2.027292000000000094(real)/2.0200000000000000178(user) seconds with $UseConstantsLists=0
1117051 times 'var[$loop]'=1117050 took 1.7664420000000000677(real)/1.7600000000000000089(user) seconds with $UseConstantsLists=1
1117051 times 'var[foo[$loop]]'=1117050 took 3.248048999999999964(real)/3.25(user) seconds with $UseConstantsLists=0
1117051 times 'var[foo[$loop]]'=1117050 took 1.9218729999999999425(real)/1.9299999999999999378(user) seconds with $UseConstantsLists=1
1117051 times 'var[foo[bar[$loop]]]'=1117050 took 4.4521829999999997796(real)/4.4500000000000001776(user) seconds with $UseConstantsLists=0
1117051 times 'var[foo[bar[$loop]]]'=1117050 took 3.1668750000000001066(real)/3.1699999999999999289(user) seconds with $UseConstantsLists=1
expression-loops per second:
	1117051/2.4230099999999996641=461017.90747871453641 (real time)
	1117051/2.4239130434782616419=460846.15246636758093 (user time)
