# Development

**URL:** https://discourse.obspy.org/c/development/12.md

[Latest](https://discourse.obspy.org/latest.md) · [Categories](https://discourse.obspy.org/categories.md) · [Tags](https://discourse.obspy.org/tags.md)

---

## [About the Development category](https://discourse.obspy.org/t/about-the-development-category/948)

<div class="topic-metadata">

**Author:** [@megies](https://discourse.obspy.org/u/megies)\
**Replies:** 0

</div>

Discussions about ObsPy development. How to contribute, make pull requests on github, discussions about development decisions, feature requests, … Please try to properly format your posts, see e.g. here. Most importantl…

---

## [Steim-2 vs a modern entropy coder: 26.8% smaller on the same IU samples. Is my methodology sound?](https://discourse.obspy.org/t/steim-2-vs-a-modern-entropy-coder-26-8-smaller-on-the-same-iu-samples-is-my-methodology-sound/2407)

<div class="topic-metadata">

**Author:** [@xolacompression-bit](https://discourse.obspy.org/u/xolacompression-bit)\
**Replies:** 2\
**Last updated:** [September 21, 2026, 8:11am UTC](https://discourse.obspy.org/t/steim-2-vs-a-modern-entropy-coder-26-8-smaller-on-the-same-iu-samples-is-my-methodology-sound/2407 "2026-09-21T08:11:09Z")

</div>

Steim-2 has been the standard since 1991 and I could not find a single published comparison against a modern entropy coder. So I ran one, and the result was large enough that I think I must be measuring something wrong. …

---

## [AI generated code](https://discourse.obspy.org/t/ai-generated-code/2324)

<div class="topic-metadata">

**Author:** [@megies](https://discourse.obspy.org/u/megies)\
**Replies:** 0\
**Last updated:** [November 5, 2025, 8:34am UTC](https://discourse.obspy.org/t/ai-generated-code/2324 "2025-11-05T08:34:35Z")

</div>

Do we need an AI policy in obspy? Personally, I don’t wanna spend (waste?) my time looking at (partially) auto-generated AI slop, at least not unless it’s clearly pointed out beforehand and has a valid point to it.

---

## [Maintenance branch](https://discourse.obspy.org/t/maintenance-branch/1523)

<div class="topic-metadata">

**Author:** [@trichter](https://discourse.obspy.org/u/trichter)\
**Replies:** 6\
**Last updated:** [September 19, 2025, 9:12am UTC](https://discourse.obspy.org/t/maintenance-branch/1523 "2025-09-19T09:12:48Z")

</div>

Often, it is difficult to understand into which branch bugfixes should go. E.g. at the moment, after 1.3.1 is released, and we are working on 1.4.0, bugfixes should go into master to be included in the upcoming 1.4.0 rel…

---

## [write traces to seg2](https://discourse.obspy.org/t/write-traces-to-seg2/2235)

<div class="topic-metadata">

**Author:** [@xiansch](https://discourse.obspy.org/u/xiansch)\
**Replies:** 1\
**Last updated:** [May 27, 2025, 7:29am UTC](https://discourse.obspy.org/t/write-traces-to-seg2/2235 "2025-05-27T07:29:05Z")

</div>

hello, I am working with a program that can only read seg2 files. The original data I am working with is also in seg2, and obspy reads this just fine. However, when I try to exported some processed clips back into seg…

---

## [Response stage numbering with gaps](https://discourse.obspy.org/t/response-stage-numbering-with-gaps/2094)

<div class="topic-metadata">

**Author:** [@TerryBarnaby](https://discourse.obspy.org/u/TerryBarnaby)\
**Replies:** 2\
**Last updated:** [November 19, 2024, 8:46am UTC](https://discourse.obspy.org/t/response-stage-numbering-with-gaps/2094 "2024-11-19T08:46:41Z")

</div>

Hi, I am a developer working with Blacknest (https://www.blacknest.gov.uk) on the Blacknest Data System (BDS), https://portal.beam.ltd.uk/bdspublic. The BDS stores seismic data and Metadata from various sources includin…

---

## [NonLinLoc hyp metadata](https://discourse.obspy.org/t/nonlinloc-hyp-metadata/2009)

<div class="topic-metadata">

**Author:** [@jwellik](https://discourse.obspy.org/u/jwellik)\
**Replies:** 3\
**Last updated:** [September 27, 2024, 7:51am UTC](https://discourse.obspy.org/t/nonlinloc-hyp-metadata/2009 "2024-09-27T07:51:59Z")

</div>

I’m hoping to modify ObsPy’s NonLinLoc \_read\_single\_hypocenter() function so that more metadata is added. I can do the coding, but I’m unsure how developers would prefer it to be done. The specific lines from the NonLinL…

---

## [Trace data from seedlink too slow](https://discourse.obspy.org/t/trace-data-from-seedlink-too-slow/1760)

<div class="topic-metadata">

**Author:** [@Wrath\_Iions](https://discourse.obspy.org/u/Wrath_Iions)\
**Replies:** 1\
**Last updated:** [August 22, 2023, 8:24am UTC](https://discourse.obspy.org/t/trace-data-from-seedlink-too-slow/1760 "2023-08-22T08:24:46Z")

</div>

I’m using handle data trace but the response is about 1 sec or more. So how to get realtime data or faster data trace for examlpe per 200 milisecond ?

---

## [Geosig data format](https://discourse.obspy.org/t/geosig-data-format/1703)

<div class="topic-metadata">

**Author:** [@BoDoNgAhAi](https://discourse.obspy.org/u/BoDoNgAhAi)\
**Replies:** 4\
**Last updated:** [May 15, 2023, 7:55am UTC](https://discourse.obspy.org/t/geosig-data-format/1703 "2023-05-15T07:55:06Z")

</div>

Hello, I have 4 geodas seismometers and I can’t read data (Geosig) using obspy. Does anyone know how to read the data and convert it to mseed? Thank you for your time and consideration.

---

## [Script to Automatically Download Seismic Data for the Last 30 Minutes](https://discourse.obspy.org/t/script-to-automatically-download-seismic-data-for-the-last-30-minutes/1710)

<div class="topic-metadata">

**Author:** [@Tsoghik\_Petrosyan](https://discourse.obspy.org/u/Tsoghik_Petrosyan)\
**Replies:** 1\
**Last updated:** [May 19, 2023, 7:44am UTC](https://discourse.obspy.org/t/script-to-automatically-download-seismic-data-for-the-last-30-minutes/1710 "2023-05-19T07:44:36Z")

</div>

Hey there, Obspy members! I hope everyone’s doing well and staying curious. I have a question for you all today, and I’m excited to hear your thoughts on it. I’ve been tinkering with ObsPy and exploring ways to automat…

---

## [Xseed parser does not allow reading mutiple blockette 55 into ResponseListResponseStage object](https://discourse.obspy.org/t/xseed-parser-does-not-allow-reading-mutiple-blockette-55-into-responselistresponsestage-object/1642)

<div class="topic-metadata">

**Author:** [@andreas\_nor](https://discourse.obspy.org/u/andreas_nor)\
**Replies:** 1\
**Last updated:** [March 6, 2023, 8:57am UTC](https://discourse.obspy.org/t/xseed-parser-does-not-allow-reading-mutiple-blockette-55-into-responselistresponsestage-object/1642 "2023-03-06T08:57:42Z")

</div>

I stumbled over a (legacy) dataless seed response file where the responses are defined as frequency, amplitude, phase list (FAP) values. The issue: Obspy read\_inventory() reads only the first part of the FAP list. It tu…

---

## [EIDA moving to StationXML 1.2](https://discourse.obspy.org/t/eida-moving-to-stationxml-1-2/1507)

<div class="topic-metadata">

**Author:** [@jbnkwsk](https://discourse.obspy.org/u/jbnkwsk)\
**Replies:** 2\
**Last updated:** [October 3, 2022, 8:38am UTC](https://discourse.obspy.org/t/eida-moving-to-stationxml-1-2/1507 "2022-10-03T08:38:28Z")

</div>

Hi ObsPy team, My name is Jarek Bienkowski, I’m currently chairing the EIDA IDG. I would like to ask you a question related to the developments of the new StationXML scheme (v1.2). The 12 data centers comprising EIDA fe…

---

## [Issue with water level in instrumental correction (when working with non-flat responses, e.g. additional integration/derivation on a flat response)](https://discourse.obspy.org/t/issue-with-water-level-in-instrumental-correction-when-working-with-non-flat-responses-e-g-additional-integration-derivation-on-a-flat-response/1487)

<div class="topic-metadata">

**Author:** [@vpdmtb](https://discourse.obspy.org/u/vpdmtb)\
**Replies:** 2\
**Last updated:** [September 5, 2022, 8:19pm UTC](https://discourse.obspy.org/t/issue-with-water-level-in-instrumental-correction-when-working-with-non-flat-responses-e-g-additional-integration-derivation-on-a-flat-response/1487 "2022-09-05T20:19:40Z")

</div>

We found a bug with the instrumental correction tool of Obspy. The bug occur only with low waterlevel values (\<80 dB) and when a change from velocity to acceleration or inversely is requested. The waterlevel problems is…

---

## [Feature request: Add list style wildcard to bulk pull](https://discourse.obspy.org/t/feature-request-add-list-style-wildcard-to-bulk-pull/1365)

<div class="topic-metadata">

**Author:** [@jschmidt](https://discourse.obspy.org/u/jschmidt)\
**Replies:** 1\
**Last updated:** [February 23, 2022, 8:24am UTC](https://discourse.obspy.org/t/feature-request-add-list-style-wildcard-to-bulk-pull/1365 "2022-02-23T08:24:55Z")

</div>

The bulk pull for waveforms works decently well, but there are times in which I have a list of stations but I want to pull only a selection of stations from that station (i.e., stations may have an accelerometer and high…

---

## [Associate Stream with Event and/or Channel](https://discourse.obspy.org/t/associate-stream-with-event-and-or-channel/1359)

<div class="topic-metadata">

**Author:** [@crotwell](https://discourse.obspy.org/u/crotwell)\
**Replies:** 1\
**Last updated:** [February 7, 2022, 9:42am UTC](https://discourse.obspy.org/t/associate-stream-with-event-and-or-channel/1359 "2022-02-07T09:42:22Z")

</div>

Issue #1842 (closed) mentions development of a DataSet, but I can’t find anything more about any development in that direction. Is there an issue to track this, or any discussions about it? Specifically, a lot of seismo…

---

## [Get\_stations not supporting multiple station codes](https://discourse.obspy.org/t/get-stations-not-supporting-multiple-station-codes/1219)

<div class="topic-metadata">

**Author:** [@blaz](https://discourse.obspy.org/u/blaz)\
**Replies:** 3\
**Last updated:** [April 13, 2021, 3:52pm UTC](https://discourse.obspy.org/t/get-stations-not-supporting-multiple-station-codes/1219 "2021-04-13T15:52:19Z")

</div>

I am sure this should be brought to the INGV people, but maybe there is a simple workaround? The obspy manual states: Select one or more SEED station codes. Multiple codes are comma-separated (e.g. “ANMO,PFO”). In cas…

---

## [Catch Warning message](https://discourse.obspy.org/t/catch-warning-message/1186)

<div class="topic-metadata">

**Author:** [@Nikolaos\_Triantafyll](https://discourse.obspy.org/u/Nikolaos_Triantafyll)\
**Replies:** 4\
**Last updated:** [February 23, 2021, 6:06pm UTC](https://discourse.obspy.org/t/catch-warning-message/1186 "2021-02-23T18:06:28Z")

</div>

Hi, How can I catch a warning message, for example: /usr/local/lib/python3.8/dist-packages/obspy/core/inventory/response.py:1895: UserWarning: More than one PolesZerosResponseStage encountered. Returning first one foun…

---

## [Spectrogram - Fix min-max dB scale](https://discourse.obspy.org/t/spectrogram-fix-min-max-db-scale/1182)

<div class="topic-metadata">

**Author:** [@jschmidt](https://discourse.obspy.org/u/jschmidt)\
**Replies:** 1\
**Last updated:** [February 11, 2021, 5:23pm UTC](https://discourse.obspy.org/t/spectrogram-fix-min-max-db-scale/1182 "2021-02-11T17:23:03Z")

</div>

I am currently trying to make a spectrogram for a station that spans a day, however, it is hard to visualize a full 24 hour block. I have made hour long segments of the time series and created spectrograms for each chann…

---

## [Make TauP (Arrival) API more modular (and efficient?)](https://discourse.obspy.org/t/make-taup-arrival-api-more-modular-and-efficient/1121)

<div class="topic-metadata">

**Author:** [@adigitoleo](https://discourse.obspy.org/u/adigitoleo)\
**Replies:** 2\
**Last updated:** [November 13, 2020, 1:25am UTC](https://discourse.obspy.org/t/make-taup-arrival-api-more-modular-and-efficient/1121 "2020-11-13T01:25:55Z")

</div>

This is a follow up on my previous post, but it’s now in the proper category and a broader question. I find the TauP API a little awkward in that we can’t mutate Arrival objects very easily. Consider the documented ways …

---

## [Run CI on a big endian system?](https://discourse.obspy.org/t/run-ci-on-a-big-endian-system/971)

<div class="topic-metadata">

**Author:** [@megies](https://discourse.obspy.org/u/megies)\
**Replies:** 5\
**Last updated:** [August 10, 2020, 8:42pm UTC](https://discourse.obspy.org/t/run-ci-on-a-big-endian-system/971 "2020-08-10T20:42:00Z")

</div>

So, is it worth a thought to eventually run CI/tests on a system with big endian? I have a feeling our code has never been tested on big endian, to be honest. At least reftek tests would have failed then (see comments in…