Bug 1431 - auto-scale of analog trace may not reliably determine values' range
Summary: auto-scale of analog trace may not reliably determine values' range
Status: RESOLVED FIXED
Alias: None
Product: PulseView
Classification: Unclassified
Component: Data display (show other bugs)
Version: unreleased development snapshot
Hardware: All All
: Normal normal
Target Milestone: ---
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-10-19 09:32 CEST by Gerhard Sittig
Modified: 2019-11-24 21:06 CET (History)
1 user (show)



Attachments
analog channels, V/div auto scale out of range (176.75 KB, image/png)
2019-10-19 09:32 CEST, Gerhard Sittig
Details
srzip file generated from CSV input file (2.16 KB, application/zip)
2019-10-19 09:41 CEST, Gerhard Sittig
Details
csv file which triggers the issue but results differ (207.17 KB, text/csv)
2019-10-19 09:55 CEST, Gerhard Sittig
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Gerhard Sittig 2019-10-19 09:32:58 CEST
Created attachment 552 [details]
analog channels, V/div auto scale out of range

How to reproduce: 

Got an input file with four analog channels which repeat a ramp signal that 
spans 100 samples. The channels have a phase difference of a quarter period 
each. Voltage ranges from 0V to 100V (Python code was used to generate the 
data, used integers and printed them with .3f, so the usual FP inaccuracy 
may apply).

When the file gets imported, then the application automatically picks a scale 
of 100V/div for the first channel, but won't detect the range of the other 
channels, which makes them exceed their trace height. Manual adjustment of 
the scaling works of course.
Comment 1 Gerhard Sittig 2019-10-19 09:41:23 CEST
Created attachment 553 [details]
srzip file generated from CSV input file

Also happens with the attached .sr file which results from importing the .csv 
file. Same ranges, same voltages at the same offsets, but different scaling 
results (this time channel 1 also exceeds its trace height).
Comment 2 Gerhard Sittig 2019-10-19 09:55:18 CEST
Created attachment 555 [details]
csv file which triggers the issue but results differ

Am attaching the CSV file as well since identical data provided in different 
file formats appears to always trigger the issue but resulting scales differ.
Comment 3 Uwe Hermann 2019-11-24 21:06:56 CET
Fixed in 0822e33a2eb713815ae2cd5d6fcca0865643bb00, thanks!