I have looked at this issue. There are definitely bugs in the waveform viewer, which I believe I have fixed. But there are also issues with your data.

You claim that Electric is using the row number instead of a proper value for v(in). But when I look at the .RAW file, I see that the v(in) and v(sweep) values are exactly the same as the row numbers (divided by 1000). They both run from 0 to 5 over the 5000 data points. So doing an Eye plot is no different than simply plotting v(out).

But there were also bugs in Electric that prevented zooming, panning, and displaying of all data points. These bugs are fixed in the source code repository, and will be part of the release of version 9.05 which will be happening in the next week.

   -Steven Rubin

At 01:32 AM 6/22/2014, you wrote:
Hello,
I'm using Electric 9.04 on OS X 10.9.3 and ngspice 26 as engine. I'm able to write spice deck and have Electric probe the raw output file, so far so good, but I'm not able to get an eye plot in the waveform window.

If I run a DC analysis on an inverter sweeping Vin from 0 to 5V with 1mV steps I get a time-based plot with a very wide scale, up to 5ks in this case. A quick inspection to at RawSpiceOut.java confirmed that Electric uses the row number as time variable for analysis that doesn't have a time value. See attached figure 01.

Dragging vin or v-sweep (they are the same in this analysis) produces an empty plot, see attached figure 02.

Panning the plot a bit to the right shows a vertical line close to the 0, see attached figure 03.

This could be the [0, 5] range on a 5k scale but adjusting the horizontal scale to [0, 5] doesn't help at all and the plot is still black. See attached figure 04.

Panning again the plot revelas a much smaller presence of data, vin near vertical 0 and vout near vertical 5V, see attached figure 05.

Zooming near the vout data and displaying just vertex point reveals there are only two points in the dataset. As you can see in figure 06 their distance is 1e-3 so the correct step in vin domain.

Where are the other data points?
What's the correct way to have an eye plot in Electric?
Can someone reproduce the issue?
Can someone provide an example of a working eye plot in Electric?


[1] http://www.staticfreesoft.com/jmanual/mchap04-11.html


Regards

--
Andrea

--
You received this message because you are subscribed to the Google Groups "Electric VLSI Editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to <mailto:[email protected]>[email protected]. For more options, visit <https://groups.google.com/d/optout>https://groups.google.com/d/optout.
Content-Type: image/png; name=figure_01.png
Content-Disposition: attachment; filename=figure_01.png
X-Attachment-Id: 506b28fd-ec88-4f75-9fad-271464d9e32c
Content-ID: <506b28fd-ec88-4f75-9fad-271464d9e32c>

Content-Type: image/png; name=figure_06.png
Content-Disposition: attachment; filename=figure_06.png
X-Attachment-Id: ea58947e-2c10-41ba-83e9-35d81ed05e0c
Content-ID: <ea58947e-2c10-41ba-83e9-35d81ed05e0c>

Content-Type: image/png; name=figure_05.png
Content-Disposition: attachment; filename=figure_05.png
X-Attachment-Id: 084f7c6c-aba0-4f1b-a464-e52126605ad7
Content-ID: <084f7c6c-aba0-4f1b-a464-e52126605ad7>

Content-Type: image/png; name=figure_03.png
Content-Disposition: attachment; filename=figure_03.png
X-Attachment-Id: 0f01057b-c64d-4235-8ea6-d96164d600c6
Content-ID: <0f01057b-c64d-4235-8ea6-d96164d600c6>

Content-Type: image/png; name=figure_02.png
Content-Disposition: attachment; filename=figure_02.png
X-Attachment-Id: 30cf93b5-8739-4bb2-8aba-8f4c948d3755
Content-ID: <30cf93b5-8739-4bb2-8aba-8f4c948d3755>

Content-Type: image/png; name=figure_04.png
Content-Disposition: attachment; filename=figure_04.png
X-Attachment-Id: 26a1056a-924d-4fd9-8eb9-e61ff3c696f4
Content-ID: <26a1056a-924d-4fd9-8eb9-e61ff3c696f4>

--
You received this message because you are subscribed to the Google Groups "Electric 
VLSI Editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to