00:00:00
Double-boosted Messi
00:00:01
With +4 purple boost and second boost
00:00:05
In-game individual ability values
00:00:06
Have been updated to a maximum of 104
00:00:09
Of course, this is without considering coach boosts
00:00:11
Now when players allocate points to top-tier double-boosted cards like Messi
00:00:16
They consider whether to boost advantageous abilities above 100
00:00:20
Even stacking second boost to reach 102 or 103...
00:00:24
...up to 104
00:00:25
Stats over 100 look impressive
00:00:28
But is it necessary to boost them this high?
00:00:30
Today we'll test various key ability values
00:00:33
And their benefits beyond 100
00:00:39
Let's start with an ability that Messi can easily push past 100
00:00:42
Dribbling - let's begin our research
00:00:45
First, we need to explain something
00:00:47
In this game, dribbling acceleration is about 10 meters
00:00:50
Before reaching maximum speed phase
00:00:52
The speed of acceleration phase and ball touch frequency depends on dribbling value
00:00:56
While speed and ball touch frequency at maximum speed
00:00:59
Depends on both dribbling and speed values
00:01:02
In importance
00:01:02
The ratio is one to one
00:01:03
Since dribbling value is used in both scenarios
00:01:06
Let's test them separately
00:01:08
First testing max speed dribbling scenario: using the old method - dribbling at top speed
00:01:13
From center line to penalty box is exactly 36 meters
00:01:17
Using double-boosted Messi and Neymar with different point allocations
00:01:20
Results are shown in the table before us
00:01:23
Two points need explanation
00:01:24
First, test data are averages from multiple measurements
00:01:28
Since high-speed dribbling performance is extremely consistent
00:01:31
You'll see the time column has no decimals
00:01:35
Because every run produces the same result
00:01:37
Second, using Neymar to assist testing
00:01:39
Is because Messi can't boost speed above 100
00:01:43
And although these players have different models
00:01:46
As mentioned before, after the v2.5 engine update
00:01:49
Various dribbling accelerations
00:01:50
And top speeds are no longer affected by player models
00:01:53
Or height and such factors
00:01:54
So these six rows of data can be compared together
00:01:57
Let's look at the table and analyze
00:01:59
When speed value is 95
00:02:01
Dribbling at 95
00:02:03
Comparing 100 and 104
00:02:05
We find dribbling from 95 to 100 makes a difference
00:02:08
Affecting both speed and ball touch frequency
00:02:12
But from 100 to 104
00:02:14
Has no impact on speed whatsoever
00:02:16
With dribbling at 104
00:02:18
The ball touch interval is actually longer
00:02:21
This slight difference is surely due to randomness
00:02:24
But this also shows
00:02:26
104 dribbling value
00:02:27
Provides no help to ball touch frequency
00:02:30
Then looking at when both speed and dribbling are 99
00:02:34
It's slightly better than
00:02:36
Speed 95, Dribbling 104
00:02:38
This means dribbling values above 100
00:02:41
Have no value for top speed dribbling
00:02:45
Now for dribbling acceleration scenario testing method
00:02:48
First test: timing of three ball touches during acceleration
00:02:51
Obviously, ball touch interval is a key indicator of acceleration ability
00:02:55
Affecting both speed and feel during acceleration
00:02:58
Faster ball touches
00:02:59
Often correspond to positional advantage
00:03:01
Test results as shown in table are from large data samples
00:03:05
Averaged
00:03:06
Found that first three touches during acceleration are very consistent
00:03:10
So results are all whole numbers
00:03:12
We can see that above dribbling 99
00:03:15
There's really no benefit
00:03:17
Specifically comparing these situations up and down
00:03:20
There's absolutely no time difference
00:03:23
And if we carefully compare
00:03:25
Every frame of movement is exactly the same
00:03:28
In comparison, increasing dribbling from 95 to 99
00:03:31
These four points are worth it
00:03:33
Actually improving acceleration phase performance
00:03:36
Similarly, dribbling 90 to 95 also has clear benefits
00:03:41
Next is testing method two
00:03:42
Comparing acceleration positioning
00:03:44
Using the game's built-in skill boost training
00:03:47
Because an important premise for testing is finding a stable initial state
00:03:50
After entering training
00:03:51
Players will start dribbling from the same initial state
00:03:55
Then testing with penalty box line as endpoint
00:03:58
We can see dribbling 104 and 99
00:04:02
Can't create any difference
00:04:04
Light circles reach penalty line simultaneously
00:04:06
And their animations at this moment are identical
00:04:09
In fact, throughout the entire acceleration process
00:04:12
Every frame of movement is the same
00:04:15
This shows it's not just about positioning
00:04:17
This five-point difference in dribbling
00:04:19
Hasn't caused any changes in the animation engine
00:04:22
As a comparison
00:04:23
Dribbling 95 to 99
00:04:24
These four points do make a difference
00:04:27
At 95 dribbling, hasn't completely crossed the line
00:04:29
While at 99, it has crossed
00:04:31
Their animations at this moment are also different
00:04:33
Combining all test results
00:04:35
We can say dribbling above 99
00:04:36
Has absolutely no benefit
00:04:38
Let's look at the next ability
00:04:40
Tight Possession
00:04:41
This ability affects dribbling turns over 90 degrees
00:04:44
Including low-speed dribbling circles and such movements
00:04:47
Test method one is timing sudden stop turns
00:04:52
From stepping on the ball to completing turn and starting reverse acceleration
00:04:58
Ending when foot touches ball
00:05:01
The advantage of this testing method is
00:05:03
The initial state is stable
00:05:05
And test scenario isn't affected by your operation
00:05:08
Because you're simply using d-pad to input opposite directions
00:05:12
Input commands are 100% identical each time
00:05:14
And this turning motion actually includes a period
00:05:17
Where player and ball separate
00:05:18
So there will be slight randomness
00:05:20
Test results must be averaged from many trials
00:05:23
Results will have decimals
00:05:26
But the results are puzzling
00:05:28
Tight Possession 104 actually turns slower than 96
00:05:32
Specific performance shown on screen
00:05:37
After investigation
00:05:38
It turns out the dribbling speed before sudden stop
00:05:40
Affects subsequent turning speed
00:05:42
Because when boosting Messi's Tight Possession
00:05:45
Dribbling value also increased
00:05:46
Making top speed faster
00:05:48
Making subsequent turns more difficult
00:05:51
This can be proven as follows
00:05:53
Same Tight Possession and Dribbling values
00:05:55
Taking another ability that affects dribbling speed
00:05:58
Speed reduced from 93 to 89
00:06:01
Result is this player indeed turns faster
00:06:04
Looking at this table makes it even clearer
00:06:07
So to control pre-turn max speed
00:06:10
This variable as shown in table
00:06:13
Speed value needs to decrease as dribbling increases
00:06:16
To achieve constant sum of speed and dribbling
00:06:20
Specifically
00:06:21
This sum is 196
00:06:23
Since double-boosted Messi can't boost speed too high
00:06:26
Last three rows were completed using created players
00:06:31
Under strict variable control
00:06:33
Tight Possession
00:06:34
Impact from 99 to 104
00:06:37
Less than 95 to 99
00:06:39
And 90 to 95 differences
00:06:41
Using sudden stop turns for testing
00:06:43
Has one drawback
00:06:45
Although creating stable initial state
00:06:47
Making this method very rigorous
00:06:49
But many scenarios caring about Tight Possession
00:06:52
Are low-speed dribbling turns
00:06:54
Like hip swings
00:06:56
Spinning circles and such movements
00:06:57
Specifically during low-speed dribbling after direction input
00:07:01
How long until player can touch ball
00:07:03
And turn body around
00:07:04
But testing this has one challenge
00:07:06
Timing of command input affects test results
00:07:10
For example, inputting direction change when ball is far vs just touched
00:07:14
Results will be different
00:07:15
This is actually inability to create
00:07:17
A stable initial state
00:07:20
Fortunately game's built-in skill improvement plan
00:07:23
Solved this problem
00:07:25
After entering training scenario
00:07:26
Players are in stable identical initial states
00:07:29
Immediately input backward turn command
00:07:31
Can test how long until ball can be touched
00:07:34
Test results as follows
00:07:36
Can see in this scenario
00:07:38
Tight Possession 103, 99 and 90
00:07:41
Indeed have thresholds making turns one frame faster
00:07:45
Screen shows difference between 104 and 102
00:07:48
This table's data conclusion I changed to other models
00:07:52
Same as Messi
00:07:53
Left-footed Maradona
00:07:54
Salah and Griezmann
00:07:56
Results still the same
00:07:58
Then this table's values developing downward
00:08:01
Are like this
00:08:02
You'll see at value 85 and below
00:08:05
Ball touch time suddenly becomes faster
00:08:07
Actually because animation modules are different
00:08:10
Tight Possession ≥86
00:08:12
First ball touch body has turned almost 90 degrees
00:08:16
While at Tight Possession 85 and below
00:08:18
First ball touch body only turned 30 degrees
00:08:21
Final result is
00:08:23
Tight Possession 86 above module
00:08:25
Will complete subsequent turning motion faster
00:08:28
This threshold's impact on turning time is 6/60 second
00:08:31
That is 6 frames
00:08:34
Back to this table
00:08:35
Does it mean Tight Possession can be boosted to 103
00:08:38
To reach fastest turning threshold?
00:08:40
Actually I think we can't draw this conclusion
00:08:43
Because this is only in specific scenario
00:08:45
Specific turning angle test
00:08:46
Result
00:08:47
If changed to like shown on screen
00:08:49
Turning from another angle with weak foot
00:08:51
Will have different thresholds
00:08:53
And in such turning tests
00:08:55
Difference between 104 and 99 Tight Possession disappears
00:08:59
So conclusion is Tight Possession above 99 for sudden stops
00:09:02
Almost no benefit
00:09:04
While low-speed turns have some benefits depending on situation
00:09:08
One thing is certain
00:09:10
Tight Possession above 86 triggers different animation modules
00:09:13
Causing qualitative change
00:09:14
Because in previous sudden stop turn tests
00:09:17
Also found Tight Possession 86
00:09:19
This threshold also corresponded to two different animation modules
00:09:23
Still Tight Possession 86 above animation module
00:09:25
Will turn body faster
00:09:27
And this was discovered last year
00:09:29
Showing in animation engine
00:09:31
Tight Possession 86 always had some important meaning
00:09:35
Combining above tests
00:09:36
We can say
00:09:37
Dribbling and Tight Possession from 99 to 104
00:09:40
Value for money is lower than imagined
00:09:43
If pursuing ultimate ball control ability
00:09:45
Main value of boosting above 100 isn't dribbling...
00:09:48
Quick acceleration and sudden stops
00:09:50
But low-speed spinning and turning
00:09:53
Why is this so?
00:09:54
This is unique to ball control three stats
00:09:56
Ball control three stats are highly related to player animations
00:09:59
So easily after exceeding 99
00:10:01
This original engine maximum value
00:10:03
Benefits don't increase linearly
00:10:06
For example
00:10:07
Speed and acceleration are moderately animation-related abilities
00:10:11
So also have non-linear benefits
00:10:14
As shown in graph
00:10:16
Speed benefits halve after 90
00:10:18
While 90 to 103 has linear growth benefits
00:10:23
Now boosting speed above 100 still has some value
00:10:28
While for acceleration above 95
00:10:31
Benefits halve
00:10:32
And more abilities aren't determining animation module speed
00:10:36
These abilities after breaking 100
00:10:38
Benefits maintain steady linear growth
00:10:40
Like stamina
00:10:42
Kicking power
00:10:43
Passing
00:10:44
Various awareness stats, etc
00:10:47
Most typical among these is stamina
00:10:50
It's unrelated to player animations
00:10:52
And unrelated to any engine physics
00:10:55
Through one-bar stamina test method we got following table
00:10:59
Can see stamina benefits above 99
00:11:03
Same as below 99
00:11:04
One point stamina equals one bar stamina gauge
00:11:08
Full sprint can last 7/60 seconds longer
00:11:11
This lets us glimpse programmers' design thinking
00:11:14
Didn't specially design for boost abilities above 99
00:11:16
Just simply linearly increased
00:11:18
The underlying parameters upward
00:11:21
While ball control three stats and dual speed
00:11:23
Are exceptions because they involve animation modules
00:11:27
Next let's test kicking power
00:11:28
To prove abilities not involving animation speed
00:11:31
All have nearly linear benefits
00:11:33
Using kicking power for testing is very suitable
00:11:35
Ball speed involves engine physics properties
00:11:39
More universal than pure backend stamina bar calculation
00:11:43
Test method is
00:11:45
Same charge time
00:11:46
After shooting test time for ball to travel five grass squares
00:11:49
To measure ball speed
00:11:51
Like shown on screen is normal shot situation
00:11:55
This is purple shot situation
00:11:58
Measured table shown as graph
00:12:00
To explain
00:12:01
For more precise testing
00:12:02
I not only measured time here
00:12:04
But also tested corresponding flight distance
00:12:07
Rather than always processing to same five grass squares length
00:12:10
This is because limited by engine running at 60 fps
00:12:13
Actually can't make every test have ball advance exactly
00:12:17
Precise five grass squares distance
00:12:20
And this table shows situation with lower ability values
00:12:24
Table data conclusion is kicking power benefits are very linear
00:12:29
Ball speed increase from 99 to 103
00:12:33
And 95 to 99
00:12:34
90 to 95 are about the same
00:12:37
Whether normal shot or purple shot
00:12:39
Both are like this
00:12:40
Additionally, compared to normal shots
00:12:43
Purple shots generally make ball speed about 6.3% faster
00:12:47
One thing worth noting is
00:12:49
The higher the kicking power
00:12:50
The less speed increase purple shots provide compared to normal shots
00:12:53
So from cost-effectiveness perspective, we can say
00:12:56
Players with insufficient kicking power should use purple shots more for long shots
00:12:59
While players who already have high kicking power
00:13:01
Normal shots should be most cost-effective
00:13:04
Speaking of which, previously tested long shot skill
00:13:06
Has no boost to kicking power
00:13:08
Adding linear non-diminishing kicking power benefits
00:13:11
I think kicking power is a very valuable ability stat
00:13:15
Power Messi's 103 kicking power is irreplaceable
00:13:19
Same for other players with kicking power above 99
00:13:22
Kicking power test proves what was said above
00:13:25
Most abilities after exceeding 99
00:13:27
Still maintain linear benefits
00:13:29
Except ball control three stats involving
00:13:30
Animation modules are exceptions
00:13:33
This point might need to wait until next engine version
00:13:35
When many players reach ball control three stats
00:13:38
Above 99 before it changes
00:13:41
So for super elite dribblers
00:13:45
Players who can boost ball control three stats above 100
00:13:48
Actually in v2.5 engine no need to pursue ball control three stats
00:13:51
Above 99
00:13:52
Allocating points to other abilities would be more worthwhile
00:13:56
Above is this video's content
00:13:58
See you next time
