From: Michael Huth (enigma_at_mail.lipsia.de)
Date: 2008-05-22 20:15:00
Wolfgang Moser schrieb:
> Hello Michael
>
> Michael Huth schrieb:
>
>> Sorry some correction to my previous posting, the subdir name in the
>> link is 'vic2r1',
>> so the correct link addresses are:
>>
>> http://mail.lipsia.de/~enigma/vic2r1/fullvic_lowquality.jpg
>>
>
> great job with the stitching I would say, there are no
> deformations recognizable. The blending for the vertical
> rows is very smooth. Between each of the rows some
> changes in luminance can be seen. I wonder if a stronger
> blending would make this less recognizable.
>
In fact I'am no photoshop expert. The rows (x-direction) were stitched
together with ptgui and the columns (y-direction)
was put together with photoshop since ptgui had serious problems. Mainly
because the x-length of the resulting rows varied a little bit.
For the fitting I used Multiply Blending and for putting it together
Dissolve. I'am not sure if this is suitable, but it looked reasonable well.
If there are differences in X-direction it is due to the sharpness
adjustment, if the sharpness was readjusted for the next column the
image plane might have shifted which means it gets a bit more
green/brown f.e..
This probably has happened at some column and also ptgui had problems
there finding control points.
Though it doesn't kills information.
> Thanks for the explanation elsewhere that the shadow
> pictures from some of the structures result from the
> microscopes optics (e.g. the S from MOS). I wondered
> if these were hidden layers.
>
>
>> http://mail.lipsia.de/~enigma/vic2r1/1.png
>> http://mail.lipsia.de/~enigma/vic2r1/2.png
>> http://mail.lipsia.de/~enigma/vic2r1/3.png
>>
>
> In the structures to the right, I believe there are
> some gates recognizable. These "L-JL"-pairs (upside down)
> cold be some drivers stages. I thought on CMOS first
> (PMOS coupled with NMOS along with a pre-driver), but
> all these old chips are NMOS only.
>
I think in the right lower corner are some circuits for the analog output.
Just a guess.
>
>> http://mail.lipsia.de/~enigma/vic2r1/4.png
>> http://mail.lipsia.de/~enigma/vic2r1/5.png
>> http://mail.lipsia.de/~enigma/vic2r1/6.png
>> http://mail.lipsia.de/~enigma/vic2r1/7.png
>> http://mail.lipsia.de/~enigma/vic2r1/8.png
>> http://mail.lipsia.de/~enigma/vic2r1/9.png
>> http://mail.lipsia.de/~enigma/vic2r1/10.png
>>
>
> Nevertheless I see the darkfield images same as you,
> they give a big plus in recognizing lower layer
> structures.
>
>
However recording the whole die in darkfield mode would require more
time, since the exposure time rises from 18 ms to about 3800 ms for one
image plus several images for fine adjustment. So if someone thinks a
darkfield image of a certain area would help him to resolve the circuit,
I can take again some in a quick shot.
(But best would be nearer future, i.e. the next months, you might want
to take the opportunity).
Ciao...
...Micha
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.