dev build 110506

[stan]

i don't think you do:)
if you look at your screen shot the most common difference in simulation results has occurrence of approximately 24000 (in level 2). so 1% is 240 so the minimum value transfered to estimates hints are simulated differences with occurrence of 240 or higher - so the result -28.

I need a more small reduction from 24 to 24,000 (1% to 1.000% or 0.1% to 100%) so (-48) - (-28) = 20 => 41.66% to become (-48) - (-46) = 2 => 4,166% if they choose 0.1% but I choose 0.4% and then I will be (-48) - (-40) = 8 => 16.66%.

the same applies to the maximum range transfered to estimates hints.

well, i can add decimal points to threshold fields if you insist. but i don't think it's wise to include differences with very low occurrences in your estimates

[teoman]

i don't think you do:)
if you look at your screen shot the most common difference in simulation results has occurrence of approximately 24000 (in level 2). so 1% is 240 so the minimum value transfered to estimates hints are simulated differences with occurrence of 240 or higher - so the result -28.

I need a more small reduction from 24 to 24,000 (1% to 1.000% or 0.1% to 100%) so (-48) - (-28) = 20 => 41.66% to become (-48) - (-46) = 2 => 4,166% if they choose 0.1% but I choose 0.4% and then I will be (-48) - (-40) = 8 => 16.66%.

well, i can add decimal points to threshold fields if you insist. but i don't think it's wise to include differences with very low occurrences in your estimates

Believe me it is wise, we made ​​many experiments and I know what I mean, it takes of percentage with decimal only to the left side, the minimum values with "-", the maximum values ​​in the right side with "+" is not needed, are OK.
More is need to put a request in Mantis issues?
Many thanks again.

[stan]

More is need to put a request in Mantis issues?

no, it's a simple change.

[teoman]

i don't think you do:)
if you look at your screen shot the most common difference in simulation results has occurrence of approximately 24000 (in level 2). so 1% is 240 so the minimum value transfered to estimates hints are simulated differences with occurrence of 240 or higher - so the result -28.

I need a more small reduction from 24 to 24,000 (1% to 1.000% or 0.1% to 100%) so (-48) - (-28) = 20 => 41.66% to become (-48) - (-46) = 2 => 4,166% if they choose 0.1% but I choose 0.4% and then I will be (-48) - (-40) = 8 => 16.66%.

the same applies to the maximum range transfered to estimates hints.

well, i can add decimal points to threshold fields if you insist. but i don't think it's wise to include differences with very low occurrences in your estimates

Stan is a error of operation? Why are not shown all decimals (0.10%, 0.20%, 0.30%, 0.40%, 0.60%, 0.70%, 0.80%, 0.90%)? Only 0.50% received permission to use? For values ​​with "+" 0.50% in the current build (110513) is greater than 1% of the build 110506, some values is in ​​decrease instead of increase. For values ​​with "-" 0,50% reduced differences with very least in compared with 1% like as when 0.50% would be by to 0.80% to 0.93% for build 110506.
Here is 1% for build 110506:

(-58) - (-43) = 15 => 25,86%
(-63) - (-43) = 20 => 31,75%
(-48) - (-28) = 20 => 41,66%
(-59) - (-44) = 15 => 25,42%
(-81) - (-56) = 25 => 30,86%
(-70) - (-48) = 22 => 31,43%
(-116)-(-80) = 36 => 31,03%
(-85) - (-57) = 28 => 32,94%
(-82) - (-53) = 29 => 35,37%
(-92) - (-70) = 22 => 23,91%
(-50) - (-34) = 16 => 32,00%
(-62) - (-41) = 21 => 33,87%
(-105)-(-75) = 30 => 28,57%

And here is 0,50% for build 110513

(-58) - (-46) = 12 => 20,69% => (0,80% build 110506)
(-63) - (-46) = 17 => 26,98% => (0,85% build 110506)
(-48) - (-31) = 17 => 35,42% => (0,83% build 110506)
(-59) - (-47) = 12 => 20,34% => (0,80% build 110506)
(-81) - (-58) = 23 => 28,39% => (0,92% build 110506)
(-70) - (-51) = 19 => 27,14% => (0,86% build 110506)
(-116)-(-85) = 31 => 26,72% => (0,86% build 110506)
(-85) - (-62) = 23 => 27,06% => (0,82% build 110506)
(-82) - (-55) = 27 => 32.93% => (0,93% build 110506)
(-92) - (-73) = 19 => 20,65% => (0,86% build 110506)
(-50) - (-36) = 14 => 28,00% => (0,87% build 110506)
(-62) - (-43) = 19 => 30,65% => (0,90% build 110506)
(-105)-(-81) = 24 => 22,86% => (0,80% build 110506)

It would be better to put all the decimals to be able to choose each user according to preferences.

[stan]

Stan is a error of operation? Why are not shown all decimals (0.10%, 0.20%, 0.30%, 0.40%, 0.60%, 0.70%, 0.80%, 0.90%)? Only 0.50% received permission to use? For values ​​with "+" 0.50% in the current build (110513) is greater than 1% of the build 110506, some values is in ​​decrease instead of increase. For values ​​with "-" 0,50% reduced differences with very least in compared with 1% like as when 0.50% would be by to 0.80% to 0.93% for build 110506.

you can put any value into the input fields. +/-0.5% is just an auto-increment when clicking up/down buttons in the spinner field.

[teoman]

Stan is a error of operation? Why are not shown all decimals (0.10%, 0.20%, 0.30%, 0.40%, 0.60%, 0.70%, 0.80%, 0.90%)? Only 0.50% received permission to use? For values ​​with "+" 0.50% in the current build (110513) is greater than 1% of the build 110506, some values is in ​​decrease instead of increase. For values ​​with "-" 0,50% reduced differences with very least in compared with 1% like as when 0.50% would be by to 0.80% to 0.93% for build 110506.

you can put any value into the input fields. +/-0.5% is just an auto-increment when clicking up/down buttons in the spinner field.

Thank you very very much! Great achievement!