Skip to content

Commit

Permalink
SpikeGLX 20231207
Browse files Browse the repository at this point in the history
  • Loading branch information
billkarsh committed Jan 30, 2024
1 parent 529a4c2 commit 462cfda
Show file tree
Hide file tree
Showing 20 changed files with 887 additions and 51 deletions.
Binary file added App/Release_v20231207-phase30.zip
Binary file not shown.
784 changes: 784 additions & 0 deletions Readme/Readme_v20231207-phase30.txt

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion Sgl_help/Metadata_30.html
Original file line number Diff line number Diff line change
Expand Up @@ -285,7 +285,7 @@ <h2 id="imec">Imec</h2>
<p>This is the probe serial number.</p>
<pre><code>imDatPrb_type=21</code></pre>
<p>This is the probe type {0=NP1.0, 21=NP2.0(1-shank), 24=NP2.0(4-shank)}.</p>
<pre><code>imErrFlags_IS_CT_SR_LK_PP_SY=0 0 0 0 0 0</code></pre>
<pre><code>imErrFlags0_IS_CT_SR_LK_PP_SY=0 0 0 0 0 0</code></pre>
<p>For each imec stream we monitor the cumulative count of several error flags. These are labeled {COUNT, SERDES, LOCK, POP, SYNC}. The metadata field 'IS' = 1 if any error occurred, 0 otherwise. I.e., &quot;is an error.&quot;</p>
<p>Any instances of these errors implies that samples have been dropped. It is not possible to tell how many samples are dropped from these counts. All you can tell is that some of the data being transmitted from the device are corrupt or missing.</p>
<p>These flags correspond to bits of the status/SYNC word that is visible as the last channel in the graphs and in your recorded data:</p>
Expand Down
2 changes: 1 addition & 1 deletion firmware/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -159,7 +159,7 @@ <h2 id="fix-imec-module-doesnt-work-in-my-chassis">FIX: Imec Module Doesn't Work
designed the module firmware for X4 and higher slots.</p>
<p>Modules have two copies of the firmware. The normal/working copy is loaded,
unless you power-up with the ISP button depressed, in which case the boot
image is loaded. Note the the boot image is a backup copy in case the working
image is loaded. Note the boot image is a backup copy in case the working
copy is corrupted.</p>
<p>In roughly Q2 2022, imec began shipping modules with an X1-compatible
working image, but not an X1-compatible golden image. By Q2 2023, both
Expand Down
2 changes: 1 addition & 1 deletion help/NIDriverInstall/NI_driver_installation/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -163,7 +163,7 @@ <h1 id="ni-chassis-and-io-card-setup">NI chassis and I/O card setup</h1>
<hr />
<p>This document covers the steps to install the necessary software for an NI
chassis, NI remote control module, and an NI acquisition device for
non-neural signals. We've selected a PXI-6133 module in these example.</p>
non-neural signals. We've selected a PXI-6133 module in these examples.</p>
<blockquote>
<p>SpikeGLX will work with multifunction IO S-series 61XX, M-series 62XX,
X-series 63XX devices, and with digital-only 653X devices. Moreover, you
Expand Down
53 changes: 39 additions & 14 deletions help/catgt_tshift/catgt_tshift/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -140,11 +140,11 @@
<ul class="nav flex-column">
</ul>
</li>
<li class="nav-item" data-level="2"><a href="#what-happened-to-global-demux" class="nav-link">What Happened to Global Demux?</a>
<li class="nav-item" data-level="2"><a href="#what-is-global-demux-gbldmx" class="nav-link">What is global demux (gbldmx)?</a>
<ul class="nav flex-column">
</ul>
</li>
<li class="nav-item" data-level="2"><a href="#what-about-catgt-loccar" class="nav-link">What about CatGT loccar?</a>
<li class="nav-item" data-level="2"><a href="#what-is-local-car-loccar" class="nav-link">What is local CAR (loccar)?</a>
<ul class="nav flex-column">
</ul>
</li>
Expand Down Expand Up @@ -189,7 +189,7 @@ <h2 id="artifact-corrections-car-gfix">Artifact Corrections {CAR, Gfix}</h2>
channels and that haven't saturated on any channels.</p>
<p><strong>Zeroing (CatGT -gfix option)</strong></p>
<p>Sometimes the hardware experiences a large electrical insult that affects
all channels in a nonunifom way. This may arise from licking, chewing,
all channels in a nonuniform way. This may arise from licking, chewing,
head banging or laser stimulation events. In these cases CAR won't
work as well. CatGT can detect these events and replace them with zeros.</p>
<p>This option is specified, for example, as follows: <code>-gfix=0.40,0.10,0.02</code>.
Expand All @@ -215,17 +215,42 @@ <h2 id="artifact-corrections-car-gfix">Artifact Corrections {CAR, Gfix}</h2>
<li>Right: -gblcar -gfix; All artifacts removed.</li>
</ul>
<p><img alt="&lt;BR/&gt;" src="../car.jpg" /></p>
<h2 id="what-happened-to-global-demux">What Happened to Global Demux?</h2>
<p>Before version 2.2, and before tshift was implemented, CatGT offered a
-gbldmx option. This separated the channels into groups that are sampled
concurrently so suffer no time shifts within the group. CAR was performed
on a group by group basis which often outperformed CAR on all channels.
Because the number of channels in a multiplex group was small the averages
were noisier than we can achieve by tshifting all channels into one large
<h2 id="what-is-global-demux-gbldmx">What is global demux (gbldmx)?</h2>
<p>The CatGT gbldmx option separates the channels into groups that are sampled
concurrently so suffer no time shifts within the group. CAR is then performed
on a group by group basis which has advantages and disadvantages compared to
doing (tshift + gblcar).</p>
<p>Disadvantages:</p>
<ul>
<li>
<p>Because the number of channels in a multiplex group is small the averages
are noisier than we can achieve by tshifting all channels into one large
group and applying CAR after that.</p>
<p>Moreover, tshifting the data improves the results of other downstream
operations like whitening and average waveform spread.</p>
<h2 id="what-about-catgt-loccar">What about CatGT loccar?</h2>
</li>
<li>
<p>Outlier channels have a stronger biasing effect with smaller sample sizes.</p>
</li>
<li>
<p>The correction factors tend to be larger (small channel count in denominator)
than with CAR, and that can produce overcorrection artifacts that look like
spikes (often small inverted spikes).</p>
</li>
</ul>
<p>When to use gbldmx:</p>
<ul>
<li>
<p>The probe sampling rate is 30 kHz, so noise faster than 15 kHz is not
adequately sampled, and therefore, not accurately propagated to other
channel groups by the tshift operation. So for very fast noise artifacts,
gbldmx may produce better cancellation within each subsampling group.</p>
</li>
<li>
<p>gbldmx usually outperforms gblcar when tshift is omitted. You may want
to omit tshift in CatGT, using the -no_tshift option, if tshift will be
applied by some later component in your analysis pipeline.</p>
</li>
</ul>
<h2 id="what-is-local-car-loccar">What is local CAR (loccar)?</h2>
<p>CatGT offers the option to do local CAR over a ring of channels. The idea
is that for each channel a disk is used to average the signal of neighbor
channels, but an inner disk (containing a potential spike footprint) is
Expand Down Expand Up @@ -253,7 +278,7 @@ <h2 id="what-about-catgt-loccar">What about CatGT loccar?</h2>
</ul>
<h2 id="what-about-lfp">What about LFP?</h2>
<p>Tshift is valuable for aligning LFP channels. CAR is not recommended when
analyzing low freqencies: LFP varies slowly over the whole shank, making
analyzing low frequencies: LFP varies slowly over the whole shank, making
distant channels a bad reference for correcting noise.</p>
<p><em>fin</em></p></div>
</div>
Expand Down
2 changes: 1 addition & 1 deletion help/syncEdges/Sync_edges/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -566,7 +566,7 @@ <h4 id="extractors-inverted-pulse">Extractors inverted pulse</h4>
<h4 id="extractors-bf-bit-field">Extractors bf (bit-field)</h4>
<p>The -xd and -xid options treat each bit of a digital word as an individual
line. In contrast, the -bf option interprets a contiguous group of bits
as a non-negative n-bit binary number. The -bf extactor reports value
as a non-negative n-bit binary number. The -bf extractor reports value
transitions: the newest value and the time it changed, in two separate files.
Following <strong>-xa=js,ip,word</strong>, the parameters are:</p>
<ul>
Expand Down
8 changes: 4 additions & 4 deletions index.html
Original file line number Diff line number Diff line change
Expand Up @@ -265,7 +265,7 @@ <h2 id="latest-application-downloads">Latest Application Downloads</h2>
firmware versions.</p>
</blockquote>
<ul>
<li><a href="App/Release_v20230905-phase30.zip">Release 20230905-phase30</a>...<a href="Readme/Readme_v20230905-phase30.txt">Readme</a> : Spike Viewer, bug fixes, imec v3.62</li>
<li><a href="App/Release_v20231207-phase30.zip">Release 20231207-phase30</a>...<a href="Readme/Readme_v20231207-phase30.txt">Readme</a> : NP1016, imec v3.62.1</li>
</ul>
<p><strong>Xilinx Kintex versions not covered by 3.0</strong>:</p>
<ul>
Expand All @@ -278,9 +278,9 @@ <h2 id="latest-application-downloads">Latest Application Downloads</h2>
versions there as long as you run one at a time:</p>
</blockquote>
<pre><code>SpikeGLX\
Release_v20230905-phase30.zip
Release_v20231207-phase30.zip
Release_v20200520-phase3B2.zip
Release_v20230905-phase30\
Release_v20231207-phase30\
Release_v20200520-phase3B2\
Drivers\
Enclustra_Windows_10&amp;11\
Expand Down Expand Up @@ -592,5 +592,5 @@ <h4 class="modal-title" id="keyboardModalLabel">Keyboard Shortcuts</h4>

<!--
MkDocs version : 1.1.2
Build Date UTC : 2023-12-18 19:07:47.654976+00:00
Build Date UTC : 2024-01-30 02:38:16.586664+00:00
-->
1 change: 1 addition & 0 deletions legacy/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -201,6 +201,7 @@ <h2 id="imec-project-phases">Imec Project Phases</h2>
<h2 id="older-versions">Older Versions</h2>
<p><strong>30</strong>:</p>
<ul>
<li><a href="../App/Release_v20230905-phase30.zip">Release 20230905-phase30</a>...<a href="../Readme/Readme_v20230905-phase30.txt">Readme</a> : Spike Viewer, bug fixes, imec v3.62</li>
<li><a href="../App/Release_v20230815-phase30.zip">Release 20230815-phase30</a>...<a href="../Readme/Readme_v20230815-phase30.txt">Readme</a> : NP2014, low latency mode, imec v3.62</li>
<li><a href="../App/Release_v20230425-phase30.zip">Release 20230425-phase30</a>...<a href="../Readme/Readme_v20230425-phase30.txt">Readme</a> : Anatomy data, better SNR, imec v3.60</li>
<li><a href="../App/Release_v20230411-phase30.zip">Release 20230411-phase30</a>...<a href="../Readme/Readme_v20230411-phase30.txt">Readme</a> : 2.0 probes, Fix PH2B headstage, imec v3.60</li>
Expand Down
4 changes: 4 additions & 0 deletions new/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -140,6 +140,10 @@
<div class="col-md-9" role="main">

<h1 id="whats-new-on-the-site">What's New On The Site?</h1>
<p>January 29, 2024</p>
<ul>
<li>SpikeGLX 20231207-phase30 (NP1016, imec 3.62.1).</li>
</ul>
<p>December 18, 2023</p>
<ul>
<li>CatGT 4.3 (Supercat can join runs with varied sample rates).</li>
Expand Down
2 changes: 1 addition & 1 deletion search/search_index.json

Large diffs are not rendered by default.

Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@

This document covers the steps to install the necessary software for an NI
chassis, NI remote control module, and an NI acquisition device for
non-neural signals. We've selected a PXI-6133 module in these example.
non-neural signals. We've selected a PXI-6133 module in these examples.

>SpikeGLX will work with multifunction IO S-series 61XX, M-series 62XX,
X-series 63XX devices, and with digital-only 653X devices. Moreover, you
Expand Down
2 changes: 1 addition & 1 deletion sglhome/docs/Help/SyncEdges/Sync_edges.md
Original file line number Diff line number Diff line change
Expand Up @@ -375,7 +375,7 @@ level and deflect toward lower values.

The -xd and -xid options treat each bit of a digital word as an individual
line. In contrast, the -bf option interprets a contiguous group of bits
as a non-negative n-bit binary number. The -bf extactor reports value
as a non-negative n-bit binary number. The -bf extractor reports value
transitions: the newest value and the time it changed, in two separate files.
Following **-xa=js,ip,word**, the parameters are:

Expand Down
2 changes: 1 addition & 1 deletion sglhome/docs/firmware.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ designed the module firmware for X4 and higher slots.

Modules have two copies of the firmware. The normal/working copy is loaded,
unless you power-up with the ISP button depressed, in which case the boot
image is loaded. Note the the boot image is a backup copy in case the working
image is loaded. Note the boot image is a backup copy in case the working
copy is corrupted.

In roughly Q2 2022, imec began shipping modules with an X1-compatible
Expand Down
41 changes: 29 additions & 12 deletions sglhome/docs/help/catgt_tshift/catgt_tshift.md
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ channels and that haven't saturated on any channels.
**Zeroing (CatGT -gfix option)**

Sometimes the hardware experiences a large electrical insult that affects
all channels in a nonunifom way. This may arise from licking, chewing,
all channels in a nonuniform way. This may arise from licking, chewing,
head banging or laser stimulation events. In these cases CAR won't
work as well. CatGT can detect these events and replace them with zeros.

Expand Down Expand Up @@ -69,20 +69,37 @@ Figure panels (from SpikeGLX FileViewer):

![<BR/>](car.jpg)

## What Happened to Global Demux?
## What is global demux (gbldmx)?

Before version 2.2, and before tshift was implemented, CatGT offered a
-gbldmx option. This separated the channels into groups that are sampled
concurrently so suffer no time shifts within the group. CAR was performed
on a group by group basis which often outperformed CAR on all channels.
Because the number of channels in a multiplex group was small the averages
were noisier than we can achieve by tshifting all channels into one large
The CatGT gbldmx option separates the channels into groups that are sampled
concurrently so suffer no time shifts within the group. CAR is then performed
on a group by group basis which has advantages and disadvantages compared to
doing (tshift + gblcar).

Disadvantages:

- Because the number of channels in a multiplex group is small the averages
are noisier than we can achieve by tshifting all channels into one large
group and applying CAR after that.

Moreover, tshifting the data improves the results of other downstream
operations like whitening and average waveform spread.
- Outlier channels have a stronger biasing effect with smaller sample sizes.

- The correction factors tend to be larger (small channel count in denominator)
than with CAR, and that can produce overcorrection artifacts that look like
spikes (often small inverted spikes).

When to use gbldmx:

- The probe sampling rate is 30 kHz, so noise faster than 15 kHz is not
adequately sampled, and therefore, not accurately propagated to other
channel groups by the tshift operation. So for very fast noise artifacts,
gbldmx may produce better cancellation within each subsampling group.

- gbldmx usually outperforms gblcar when tshift is omitted. You may want
to omit tshift in CatGT, using the -no_tshift option, if tshift will be
applied by some later component in your analysis pipeline.

## What about CatGT loccar?
## What is local CAR (loccar)?

CatGT offers the option to do local CAR over a ring of channels. The idea
is that for each channel a disk is used to average the signal of neighbor
Expand Down Expand Up @@ -110,7 +127,7 @@ setting a fairly large outer radius to balance locality vs smoothness.
## What about LFP?

Tshift is valuable for aligning LFP channels. CAR is not recommended when
analyzing low freqencies: LFP varies slowly over the whole shank, making
analyzing low frequencies: LFP varies slowly over the whole shank, making
distant channels a bad reference for correcting noise.


Expand Down
6 changes: 3 additions & 3 deletions sglhome/docs/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,7 @@ Latest release highlights:
>Important: Please read the `README` file for instructions regarding imec
firmware versions.

* [Release 20230905-phase30](App/Release_v20230905-phase30.zip)...[Readme](Readme/Readme_v20230905-phase30.txt) : Spike Viewer, bug fixes, imec v3.62
* [Release 20231207-phase30](App/Release_v20231207-phase30.zip)...[Readme](Readme/Readme_v20231207-phase30.txt) : NP1016, imec v3.62.1

**Xilinx Kintex versions not covered by 3.0**:

Expand All @@ -77,9 +77,9 @@ versions there as long as you run one at a time:

```
SpikeGLX\
Release_v20230905-phase30.zip
Release_v20231207-phase30.zip
Release_v20200520-phase3B2.zip
Release_v20230905-phase30\
Release_v20231207-phase30\
Release_v20200520-phase3B2\
Drivers\
Enclustra_Windows_10&11\
Expand Down
1 change: 1 addition & 0 deletions sglhome/docs/legacy.md
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,7 @@ can be used after updating their firmware, which can be done via SpikeGLX.

**30**:

* [Release 20230905-phase30](../App/Release_v20230905-phase30.zip)...[Readme](../Readme/Readme_v20230905-phase30.txt) : Spike Viewer, bug fixes, imec v3.62
* [Release 20230815-phase30](../App/Release_v20230815-phase30.zip)...[Readme](../Readme/Readme_v20230815-phase30.txt) : NP2014, low latency mode, imec v3.62
* [Release 20230425-phase30](../App/Release_v20230425-phase30.zip)...[Readme](../Readme/Readme_v20230425-phase30.txt) : Anatomy data, better SNR, imec v3.60
* [Release 20230411-phase30](../App/Release_v20230411-phase30.zip)...[Readme](../Readme/Readme_v20230411-phase30.txt) : 2.0 probes, Fix PH2B headstage, imec v3.60
Expand Down
4 changes: 4 additions & 0 deletions sglhome/docs/new.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,9 @@
# What's New On The Site?

January 29, 2024

* SpikeGLX 20231207-phase30 (NP1016, imec 3.62.1).

December 18, 2023

* CatGT 4.3 (Supercat can join runs with varied sample rates).
Expand Down
20 changes: 10 additions & 10 deletions sitemap.xml
Original file line number Diff line number Diff line change
@@ -1,43 +1,43 @@
<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url><url>
<loc>None</loc>
<lastmod>2023-12-18</lastmod>
<lastmod>2024-01-30</lastmod>
<changefreq>daily</changefreq>
</url>
</urlset>
Binary file modified sitemap.xml.gz
Binary file not shown.

0 comments on commit 462cfda

Please sign in to comment.