Skip to content

Commit a19b08b

Browse files
apradhanaapradhana
authored
Merge pull request #1501 from apradhana/TSC-2022-10-04
Add meeting notes.
2 parents d5076f6 + d8978da commit a19b08b

File tree

3 files changed

+238
-0
lines changed

3 files changed

+238
-0
lines changed

‎tsc/meetings/2022-06-21.md‎

Lines changed: 81 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,81 @@
1+
Minutes from 137-th OpenVDB TSC meeting, June 21st, 2022.
2+
3+
Attendees: Jeff L., Nick A, Dan B., Ken M., Rich J., Andre P.
4+
5+
Additional Attendees: Sebastian Gaida, Peter Cheng (DW), Greg Hurst
6+
(United Therapeutics), Peyton Murray, Sergio Rojas, Karl Marret, Jamie
7+
8+
Regrets: None
9+
10+
Agenda:
11+
12+
1) Confirm quorum
13+
2) Secretary
14+
3) Forum
15+
4) TAC Update
16+
5) Half
17+
6) OpenVDB Mathematica Link
18+
7) VFX Reference Platform
19+
8) Value Accessor on LeafNode
20+
9) Next meeting
21+
22+
--------------------
23+
24+
1) Confirm quorum
25+
26+
Quorum is present.
27+
28+
2) Secretary
29+
30+
Secretary is Andre Pradhana.
31+
32+
3) Forum
33+
34+
Nothing to discuss from the Forum.
35+
4) TAC Update
36+
37+
a) Emily Ollin wants to setup a process for releasing major versions for all
38+
projects. Regarding the 9.1 release, Ken will send an email to the larger
39+
OpenVDB ASWF mailing list. Dan will make an announcement on twitter.
40+
41+
b) PyPi We should create our own account. This is what OpenTimeIO has done.
42+
We will follow their example.
43+
44+
c) CI Amazon has hardwares that we should be able to use. It is being used by
45+
OpenColorIO. The process still works through Github Actions, but the host
46+
machines are different. It has to be approved by the ASWF governance board.
47+
Once this is approved, we can follow up to change the CI machines.
48+
49+
There is a slack channel hosting this ASWF CI topic.
50+
51+
5) Half
52+
53+
Greg met with an Autodesk representative. Greg's plan is to use half as a
54+
first-class citizen in OpenVDB and to do computations in floats, but to store
55+
the results in halfs. He plans half for the OpenVDB 10 release in October.
56+
57+
Autodesk cannot use NanoVDB because they need to perform dilation before
58+
performing motion blur.
59+
60+
6) OpenVDB Mathematica Link
61+
62+
Greg is optimistic that we can find a working solution to have a Wolfram Engine
63+
for the purpose of contributing to OpenVDB.
64+
65+
7) VFX Reference Platform
66+
67+
Jeff Lait said that VFX Reference Platform now lists OpenVDB 10, unless we fail
68+
to release version 10 by October 2022, under the CY2023 draft.
69+
70+
8) Value Accessor on LeafNode
71+
72+
Nick improved the VolumeToMesh tool in OpenVDB by making it up-to-date with
73+
C++11 and C++14. He also improved the use of Leaf Node Value Accessor by creating
74+
a new wrapper that pulls the buffer data in order to avoid atomic access with
75+
delayed-loading. The whole effort makes the tool 10-15% faster.
76+
77+
Nick will do a PR on changing the Leaf Value Iterator of the Leaf Node.
78+
79+
9) Next meeting
80+
81+
Next meeting is on June 28th, 2022. 1pm-2pm EDT (GMT-4).

‎tsc/meetings/2022-09-06.md‎

Lines changed: 88 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,88 @@
1+
Minutes from 145th OpenVDB TSC meeting, September 6th, 2022, (EDT)
2+
3+
Attendees: *Ken* M., *Dan* B., *Rich* J. *Andre* P.
4+
5+
Regrets: *Nick* A., *Jeff* L.
6+
7+
Additional Attendees: Peter Cheng (DW), Karl Marett, Sebastian Gaida,
8+
JT Nelson (Blender), Greg Hurst (United Therapeutics), Bruce Cherniak (Intel)
9+
10+
Agenda:
11+
12+
1) Confirm quorum
13+
2) Secretary
14+
3) Forum and Discussion
15+
a) Request to expose 4x4 matrix in Python bindings
16+
b) Update build instruction
17+
4) Failure in NanoVDB CI
18+
5) VDB in medical imaging (OpenVDB Slack Channel)
19+
6) Mathematica Link Update
20+
7) Update from DreamWorks
21+
8) CI Effort
22+
9) Root Node Offset
23+
10) Next Meeting
24+
25+
--------------------
26+
27+
1) Confirm quorum
28+
29+
Quorum is present.
30+
31+
2) Secretary
32+
33+
Secretary is Andre Pradhana
34+
35+
3) Forum and Discussion
36+
a) Request to expose 4x4 matrix in Python bindings.
37+
This should be done through PyBind 11. A few people expressed an interest
38+
to help with this task.
39+
40+
b) Update build instruction to reflect the correct version of TBB.
41+
42+
4) Failure in NanoVDB CI
43+
44+
We look at failure NanoVDB CI failure in MacOS. Ken has an idea that it may
45+
be caused by the new allocation that is implemented.
46+
47+
5) VDB in medical imaging (OpenVDB Slack Channel)
48+
49+
Somebody was asking about the potential of using OpenVDB or NanoVDB for medical
50+
imaging purposes. The answer is: yes, VDB can be used for this purpose.
51+
52+
6) Mathematica Link Update
53+
54+
Greg is still waiting for a reply from Wolfram.
55+
56+
7) Update from DreamWorks
57+
58+
DreamWorks said that they will go through with a more thorough investigation of
59+
existing old VDB files that uses the dense table in the root node. They will
60+
probably convert old VDB files to new VDBs.
61+
62+
8) CI Effort
63+
64+
Dan gives an update on the CI effort. We have an access to two machines provided
65+
by AWS.
66+
67+
Dan claims that the billable hours for all CI to finish is 1 hr and 43 minutes.
68+
69+
9) Root Node Offset
70+
71+
Ken revisited his work on the root node offset. He did not find any performance
72+
issues with the new implementation. Several things to note:
73+
a) The way that Ken solves the problem is by adding a CoordT origin in
74+
the root node to offset the position of the children of the root node.
75+
b) The offset should coincide with the grand-child of the root. In a regular
76+
tree configuration the offset should be 2048 = 16 * 16 * 16 / 2.
77+
c) Ken proposed the origin offset on the root-node to be dynamic, because
78+
in the future, we can read an old VDB file with the root node origin
79+
set to be (0, 0, 0).
80+
81+
He proposed that we would go through with the ABI change, i.e. adding the
82+
origin member data in the root. He also proposed that in the future, doing
83+
operations on two or more different grids should be restricted on grids
84+
with the same origin.
85+
86+
10) Next meeting
87+
88+
Next meeting is on September 13th, 2022. 1pm-2pm EDT (GMT-4).

‎tsc/meetings/2022-10-04.md‎

Lines changed: 69 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,69 @@
1+
Minutes from 149th OpenVDB TSC meeting, September 27th, 2022
2+
3+
Attendees: *Jeff* L., *Andre* P, *Nick* A., *Dan* B.,
4+
*Rich* J., *Ken* M.,
5+
6+
Additional Attendees: JT Nelson (Blender), Sebastian Gaida, Greg Hurst,
7+
Barry Dempsey
8+
9+
Regrets: none
10+
11+
Agenda:
12+
13+
1) Confirm quorum
14+
2) Secretary
15+
3) Forum
16+
4) Mathematica Link
17+
5) CI
18+
6) PR-1469
19+
7) Root Node Offset
20+
8) VDB Tool
21+
9) Meeting link
22+
10) Next meeting
23+
24+
--------------------
25+
26+
1) Confirm quorum
27+
28+
Quorum is present.
29+
30+
2) Secretary
31+
32+
Secretary is Andre Pradhana
33+
34+
3) Forum
35+
36+
No issues.
37+
38+
4) Mathematica Link
39+
40+
Greg compiled the Mathematica Link on a Windows machine. He will write
41+
a README and a documentation.
42+
43+
5) CI
44+
45+
Nick fixed CI issues related to the way vcpkg deals with tbb. The Houdini 19.5
46+
CI is currently failing because of a problem related to either zlibc or boost.
47+
48+
6) PR-1469
49+
50+
Nick will take a look at PR-1469 from Dan.
51+
52+
7) Root Node Offset
53+
54+
It is recommended to remove some additional methods from the root node class and
55+
move them outside. It was agreed to make the get-origin and set-origin methods to
56+
be public, but throw an error if a user calls it.
57+
58+
8) vdb_tool
59+
60+
It was clarified that python support through VDB tool is not supported by VDB tool
61+
for the VDB 10 release.
62+
63+
9) Meeting link
64+
65+
We will still use the new meeting link for next week.
66+
67+
10) Next meeting
68+
69+
Next meeting is on October 11th, 2022. 2pm-3pm EDT (GMT-4).

0 commit comments

Comments
 (0)