Toshiba Camileo S20 product image (Source: data sheet)

I bought the Toshiba Camileo S20 for making brilliant Full HD Videos. You can have a look at Toshiba’s data sheet here.  In my opinion the main features look really nice and the price seems appropriate:

If you are more into visuals, you perhaps look at the feature pictograms:

Toshiba Camileo S20 feature pictograms (Source: data sheet)

Now, we switch to reality…

From the above information you might derive that you can make Full HD videos in 1080p with 4x digital zoom and video stabilisation, but unfortunately, this is absolutely wrong.

You can’t use the 4x digital zoom nor the video stabilisation in 1080p mode!

The missing video stabilisation makes sense. When the camera is at its sensor limits in 1080p, there are simply no border-pixels left for stabilisation. And the missing digital zoom? Perhaps the camera is too slow for that. And that it is extremely slow, do you experience when you try to make a still picture.

The product lie is bad, but what is the biggest problem of this camera?

The biggest problem of this camera is the bad optics. The noise when recording a video even with good indoor lighting is so extremely dominant, that it ruins every 1080p experience. A 5-year-old cheap digital still camera has a better light efficiency and makes brighter and cleaner videos.

Why is this camera not reasonably priced?

You can buy a really small digital camera (e.g. some Casio Exilim) with flash-light, brilliant optics, many MPs and high quality bright and clean 720p video recording  for the same price.  And the 720p format is much more “handable” in video editing software and video sharing websites than 1080p.

Summary

• Some features don’t work in 1080p mode.
• The quality is bad.
• It is unusable as still camera, because of the missing flash-light, the bad light efficiency and the slowness.
• The software is nearly unusable freeware / shareware garbage.
• The manual is not printed. Is it really “greener” if every customer has to print the manual? Or is it only cheaper for Toshiba?
• The usability is really bad.

Finally, it’s far away from being a good product for me!

Even the following minimalistic modalPanel will produce the error:

<rich:modalPanel id="ixpressor">
</rich:modalPanel>

Because the rendered output contains the following  javascript, which isn’t compatible with IE8:

[...] onmouseover="}" [...]

I found a workaround which is very strange:

<rich:modalPanel id="ixpressor" onmaskmouseout="//" >
</rich:modalPanel>

This is rendered to:

[...] onmouseout="//" onmouseover="//}" [...]

For me, this wasn’t expected but it works and you can put some Javascript before the “//”.

Perhaps, this helps you, if you can’t upgrade to a newer version of RichFaces.

19.01.2010  120.8
21.01.2010  118.8
10.02.2010  112.0
20.02.2010  108.3
24.03.2010  109.5
05.04.2010  105.0

In the first column is the date in the format “%d.%m.%Y” and in the second column is the weight in Kg. For plotting this data I created the following file called  ’weight.plt’‘:

#
# Personal Weight Plot
#
# size in m, weight in kg
#

# define your size
size=1.70

bmi2weight(size,bmi) = \
bmi * size**2

# bmi for adult males (WHO 2008)
bmi_best                   = bmi2weight(size, 22.5) # Define your personal target bmi

bmi_normal_start           = bmi2weight(size, 18.5)
bmi_normal_end             = bmi2weight(size, 25)

bmi_preadipositas_start    = bmi2weight(size, 25)
bmi_preadipositas_end      = bmi2weight(size, 30)

bmi_adipositas_I_start     = bmi2weight(size, 30)
bmi_adipositas_I_end       = bmi2weight(size, 35) 

bmi_adipositas_II_start    = bmi2weight(size, 35)
bmi_adipositas_II_end      = bmi2weight(size, 40)

bmi_adipositas_III_start   = bmi2weight(size, 40)

# begin plotting
set terminal png rounded size 1024, 768 x000000 xffffff xffffff
set output "weight.png"

set title "Weight over time\nFirst half of 2010"
set grid

set timefmt "%d.%m.%Y" # Define the format of your data

set xlabel "Date"
set xdata time
set xrange ["01.01.2010":"30.06.2010"] # Adjust
set format x "%b"

set ylabel "Weight/Kg"
set yrange [0:150] # Define the plotted range of data
set ytics nomirror 10 # Define the tics
set mytics 5 # Define the minor tics

#f(x) = m*x/300000000 + b
#fit f(x) "weight.data" using 1:2 via b, m

plot "weight.data" using 1:2 title "Meassured weight" with linespoints lt rgb '#1E90FF' pt 5,\
bmi_adipositas_III_start title "Begin of obese class III" with lines lt rgb '#FF0000',\
bmi_adipositas_II_start title "Begin of obese class II" with lines lt rgb '#FF4500',\
bmi_adipositas_I_start title "Begin of obese class I" with lines lt rgb '#FFA500',\
bmi_preadipositas_start title "Begin of overweight" with lines lt rgb '#F4A460',\
bmi_best title "Optimal weight" with lines lt rgb '#008000'#,\
f(x) title "Weight trend" lt rgb '#FFC0CB'

The following image shows the result of running the file with gnuplot.

I found two other posts dealing with this topic:

• http://www.stdlib.net/~colmmacc/2005/12/25/losing-weight-with-gnuplot/
• http://www.321blog.de/2007/06/05/uhufue-im-schaubild/ (in german)

Finally, I hope that your blue line goes down fast!