Difference between revisions of "User:Chekhov MacGuffin/Latest case"
(New numbers for Koumei at the bottom) |
(→Stacked charts by percentage: broke chart into two for better readability) |
||
Line 330: | Line 330: | ||
== Stacked charts by percentage == | == Stacked charts by percentage == | ||
+ | |||
+ | Raw percentages | ||
+ | |||
{| class="wikitable" | {| class="wikitable" | ||
|- | |- | ||
Line 345: | Line 348: | ||
! Other | ! Other | ||
! Total | ! Total | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|- | |- | ||
| Kansuke o_- | | Kansuke o_- | ||
Line 365: | Line 362: | ||
| 1.0% | | 1.0% | ||
| 100% | | 100% | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|- | |- | ||
| Koumei o_o | | Koumei o_o | ||
Line 385: | Line 376: | ||
| 3.6% | | 3.6% | ||
| 100% | | 100% | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|- | |- | ||
| Kuroda -_o | | Kuroda -_o | ||
Line 405: | Line 390: | ||
| 5.1% | | 5.1% | ||
| 100% | | 100% | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|- | |- | ||
| Rumi o_o | | Rumi o_o | ||
Line 425: | Line 404: | ||
| 7.5% | | 7.5% | ||
| 100% | | 100% | ||
+ | |} | ||
+ | |||
+ | Analysis | ||
+ | |||
+ | {| class="wikitable" | ||
+ | |- | ||
+ | ! Name | ||
+ | ! #Datapoints | ||
+ | ! SS (•_•) | ||
+ | ! SR:RS | ||
+ | ! SL:LS | ||
+ | ! RR:SR | ||
+ | ! LL:SL | ||
+ | ! SX:RX:LX | ||
+ | ! XS:XR:XL | ||
+ | |- | ||
+ | | Kansuke o_- | ||
+ | | 291 | ||
+ | | 56.4% | ||
+ | | 4.3 : 1 | ||
+ | | <span style="color:blue">'''1 : 1'''</span> | ||
+ | | 1 : 2 | ||
+ | | <span style="color:blue">'''2.6 : 1'''</span> | ||
+ | | 10.9 : 1 : <span style="color:blue">'''3.3'''</span> | ||
+ | | 4.8 : 1 : 1.6 | ||
+ | |- | ||
+ | | Koumei o_o | ||
+ | | 251 | ||
+ | | 52.6% | ||
+ | | 2.6 : 1 | ||
+ | | 9.3 : 1 | ||
+ | | 1.1 : 1 | ||
+ | | 1 : 1.4 | ||
+ | | 8.0 : 1.5 : 1 | ||
+ | | 3.0 : 1 : 1 | ||
+ | |- | ||
+ | | Kuroda -_o | ||
+ | | 78 | ||
+ | | <span style="color:black">'''78.2%'''</span> | ||
+ | | NaN | ||
+ | | 1 : 2 | ||
+ | | NaN | ||
+ | | 1 : 3 | ||
+ | | NaN | ||
+ | | 22.3 : 1 : 1.3 | ||
+ | |- | ||
+ | | Rumi o_o | ||
+ | | 80 | ||
+ | | 46.3% | ||
| 1 : 1.25 | | 1 : 1.25 | ||
| 3 : 1 | | 3 : 1 | ||
Line 433: | Line 461: | ||
|} | |} | ||
− | |||
− | |||
− | The | + | === Understanding the stats analysis === |
+ | * #Datapoints: I wouldn't put a whole lot of stock in any of the data until they include over 250 points per person. That's about 9 files worth of data. | ||
+ | |||
+ | * SS (•_•): The percent of panels in which a character is facing forward. I'm not sure what it means when it is abnormally high. Is it a feature of scary geezers, or could it be used to pinpoint someone who might have an eye out? | ||
− | + | * SR:RS / SL:LS : These two compare the ratio of instances where the character only turns their eyes and their heads remain straight (SR and SL) to the number of head turns to the right or left to look directly at something (RS and LS). A high ratio means the character is favoring more aside glances than headturns for direct looking. One of my assumptions is that characters with an eye out will turn their heads more directly to face someone on their blind slide. Kansuke seems to show this on his bad left, preferring more than other characters to turn his head to look straight at someone. | |
− | The second pair compares the ratio of the full head and eyes turn (LL and RR) to the eyeturn only (SL and SR). A character with their vision out should be forced to turn more completely on their bad side and therefore uses relatively fewer aside glances on their bad side. | + | * RR:SR / LL:SL : The second pair compares the ratio of the full head and eyes turn (LL and RR) to the eyeturn only (SL and SR). A character with their vision out should be forced to turn more completely on their bad side and therefore uses relatively fewer aside glances on their bad side. |
− | The second from the last looks at raw headturn ratio. | + | * SX:RX:LX : The second from the last looks at raw headturn ratio. |
− | The final one looks at raw eyeturn ratios. | + | * XS:XR:XL : The final one looks at raw eyeturn ratios. |
I expect some artistic bias shows up. Gosho seems to prefer drawing characters turning their heads to the left and looking straight rather than turning them to the right and looking straight. | I expect some artistic bias shows up. Gosho seems to prefer drawing characters turning their heads to the left and looking straight rather than turning them to the right and looking straight. |
Revision as of 05:22, 6 August 2016
Sandbox currently being used for peripheral vision analysis. I am currently researching to figure out what observations are high signal to noise for blind-in-one-eye chars - in short, quantifiable differences in depiction compared to full sight chars.
Contents
Summary and assumptions
This sheet is meant to document the depictions and behaviors of characters who are known to be blinded on one side so that other blinded characters can be identified, even if they are hiding that fact.
In real life, people with a nonfunctional eye will lack corresponding peripheral vision. Thus they will be required to turn their whole face in order to see things on their blinded side, and may miss observations if they are out of sight. Blinded people also have problems with depth perception. Some of these limitations may be carried over to the manga if Gosho has done appropriate research.
General observations
- Kansuke turns his head slightly further than other characters when looking with his bad side. Because it is rather subtle, the comparison is the least biased if compared within a single panel. To determine if a character is genuinely overturning, use the rule of thumb: "Once is chance, twice is coincidence, the third time is a pattern."
- At least two Gosho characters with blinded eyes wear glasses that are always dark over the bad eye. (Hyoue Kuroda, MK's Goudzu)
- Characters with bad eyes don't overfavor turning towards their good eye side to see people who are mostly behind them - they will turn with whichever side is closer, even if it is their bad one. (Kansuke)
- Characters with one eye still do certain quirky Gosho things like speak with their eye(s) shut for a moment and then open them for the dramatic punchline. (Kansuke)
Panel evaluation method description
The purpose of gathering these data is to evaluate if there any quantifiable differences between full-sighted and part-sighted character depiction. The directions are relative to the character's own perspective. Between 11 and 1 o'clock is assumed to be straight for the purposes of this evaluation.
- Head straight, eyes straight (•_•)
- Head straight, eyes right (<_<)
- Head straight, eyes left (>_>)
- Head turned right, eyes straight ('_' )
- Head turned right, eyes right (<_< )
- Head turned right, eyes left (•_• )
- Head turned left, eyes straight ( '_')
- Head turned left, eyes left ( >_>)
- Head turned left, eyes right ( •_•)
- Other (One eye closed. Something odd.)
- Total number clear panels with char looking or interacting.
Hyoue Kuroda
Kuroda is blinded in his right eye. Kuroda in his intro case very rarely spoke to anyone with his head turned. He either stood and looked at them directly or he didn't at all. This might be a hint in and of itself. People who don't turn heads could be more suspicious.
Case | SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|---|
913-917 | 61 | 3 | 3 | 0 | 0 | 0 | 6 | 1 | 0 | 4 | 78 |
913-917 | 78.2% | 3.8% | 3.8% | 0.0% | 0.0% | 0.0% | 7.7% | 1.3% | 0.0% | 5.1% | 100% |
920 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 |
Yamato Kansuke
Yamato is blinded in his left eye. While Yamaoto may give aside glances at first when his name is called, he tends to turn his head fully toward the speaker when being spoken to or holding a longer conversation (Yamato vs Koumei). Koumei in contrast seems to be more comfortable speaking without turning his head fully if the speaker is still in his peripheral vision. Yamato holds conversations with people in his blindspot, so long as the conversation doesn't require him to see the other.
Case | SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|---|
913-917 | 63 | 8 | 6 | 2 | 6 | 0 | 12 | 14 | 0 | 2 | 113 |
683-686 | 101 | 18 | 11 | 4 | 7 | 0 | 5 | 31 | 0 | 1 | 178 |
Totals | 164 | 26 | 17 | 6 | 13 | 0 | 17 | 45 | 0 | 3 | 291 |
Totals | 56.4% | 8.9% | 5.8% | 2.1% | 4.5% | 0.0% | 5.8% | 15.5% | 0.0% | 1.0% | 100% |
Some notable moments
Done 683-686, 913-914
- 685p3 - Didn't notice Koumei had been pulled from the fire because his body was on the bad left side. Conan had to inform him to keep from charging in.
- 913p6 - Kansuke catches Conan staring at him in the peripheral of his functional eye, turns to look at Conan full forward over the next two panels
- 913p13 - Kansuke didn't see a body he was looking for in his bad eye's blind spot. He had to turn his head fully 90 degrees to see where Koumei is squatting at his 8 o'clock with the body.
Koumei
Fully sighted control for Kansuke.
Case | SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|---|
913-917 | 82 | 15 | 15 | 3 | 17 | 0 | 2 | 8 | 0 | 4 | 146 |
683-686 | 50 | 9 | 13 | 6 | 9 | 0 | 1 | 12 | 0 | 5 | 105 |
Totals | 132 | 24 | 28 | 9 | 26 | 0 | 3 | 20 | 0 | 9 | 251 |
Totals | 52.6% | 9.6% | 11.2% | 3.6% | 10.4% | 0.0% | 1.2% | 8.0% | 0.0% | 3.6% | 100% |
Goudzu
Minor MK case villain. Presumed to be blind in right eye because of glasses.
SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|
30 | 1 | 2 | 1 | 2 | 0 | 2 | 2 | 0 | 1 | 41 |
Rumi
No apparent eye troubles
Case | SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|---|
966-968 | 37 | 4 | 12 | 5 | 4 | 0 | 4 | 8 | 0 | 6 | 80 |
Total | 46.3% | 5.0% | 15.0% | 6.3% | 5.0% | 0.0% | 5.0% | 10.0% | 0.0% | 7.5% | 100% |
Sakurako Yonehara
Chikara Katsumata
Mary
Conan
A control case, looking at cases since the start of the Rum arc.
Matsumoto
A control case for bulky geezers.
Stacked charts by percentage
Raw percentages
Name | #Datapoints | SS (•_•) | SR (<_<) | SL (>_>) | RS ('_' ) | RR (<_< ) | RL (•_• ) | LS ( '_') | LL ( >_>) | LR ( •_•) | Other | Total |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Kansuke o_- | 291 | 56.4% | 8.9% | 5.8% | 2.1% | 4.5% | 0.0% | 5.8% | 15.5% | 0.0% | 1.0% | 100% |
Koumei o_o | 251 | 52.6% | 9.6% | 11.2% | 3.6% | 10.4% | 0.0% | 1.2% | 8.0% | 0.0% | 3.6% | 100% |
Kuroda -_o | 78 | 78.2% | 3.8% | 3.8% | 0.0% | 0.0% | 0.0% | 7.7% | 1.3% | 0.0% | 5.1% | 100% |
Rumi o_o | 80 | 46.3% | 5.0% | 15.0% | 6.3% | 5.0% | 0.0% | 5.0% | 10.0% | 0.0% | 7.5% | 100% |
Analysis
Name | #Datapoints | SS (•_•) | SR:RS | SL:LS | RR:SR | LL:SL | SX:RX:LX | XS:XR:XL |
---|---|---|---|---|---|---|---|---|
Kansuke o_- | 291 | 56.4% | 4.3 : 1 | 1 : 1 | 1 : 2 | 2.6 : 1 | 10.9 : 1 : 3.3 | 4.8 : 1 : 1.6 |
Koumei o_o | 251 | 52.6% | 2.6 : 1 | 9.3 : 1 | 1.1 : 1 | 1 : 1.4 | 8.0 : 1.5 : 1 | 3.0 : 1 : 1 |
Kuroda -_o | 78 | 78.2% | NaN | 1 : 2 | NaN | 1 : 3 | NaN | 22.3 : 1 : 1.3 |
Rumi o_o | 80 | 46.3% | 1 : 1.25 | 3 : 1 | 1 : 1 | 1 : 1.5 | 5.9 : 1 : 1.3 | 5.8 : 1 : 2.5 |
Understanding the stats analysis
- #Datapoints: I wouldn't put a whole lot of stock in any of the data until they include over 250 points per person. That's about 9 files worth of data.
- SS (•_•): The percent of panels in which a character is facing forward. I'm not sure what it means when it is abnormally high. Is it a feature of scary geezers, or could it be used to pinpoint someone who might have an eye out?
- SR:RS / SL:LS : These two compare the ratio of instances where the character only turns their eyes and their heads remain straight (SR and SL) to the number of head turns to the right or left to look directly at something (RS and LS). A high ratio means the character is favoring more aside glances than headturns for direct looking. One of my assumptions is that characters with an eye out will turn their heads more directly to face someone on their blind slide. Kansuke seems to show this on his bad left, preferring more than other characters to turn his head to look straight at someone.
- RR:SR / LL:SL : The second pair compares the ratio of the full head and eyes turn (LL and RR) to the eyeturn only (SL and SR). A character with their vision out should be forced to turn more completely on their bad side and therefore uses relatively fewer aside glances on their bad side.
- SX:RX:LX : The second from the last looks at raw headturn ratio.
- XS:XR:XL : The final one looks at raw eyeturn ratios.
I expect some artistic bias shows up. Gosho seems to prefer drawing characters turning their heads to the left and looking straight rather than turning them to the right and looking straight.