This is a **preview release** — Lumiera is not usable yet.
Publication was again motivated by a series of major upgrades and clean-up.
Compilation now requires C++23 and Debian/Trixie is considered the reference
platform. The GUI is connected to the core over an asynchronous messaging
bus and the timeline display has been reshaped to accommodate flexibly
nested structures of media content, instructed by the session through
a series of _diff messages_. A scheduler has been implemented to drive
the Render Engine, but the connection to the player and UI for output
is yet unfinished.
- comb through the Website, starting at the frontpage
- add a **news** entry to confirm this major upgrade step (C++23)
- improve the wording in various overview pages
- adapt the ''release checklist'' to align it with **git-flow**
- reorganise the image folder(s) on the website
- the animated beating heart is back ;-)
Some recherche regarding the requirements and conditions of the GPL
made it clear that we are obliged to have at least an summary statement
of all authors and the years of their contributions embedded somewhere
in the source itself. The Git history alone is not sufficient to fulfil
this requirement, since a code base could be packaged as Tarball
and passed-on outside of any Git repository.
Thus I propose the following policy
- add yourself to the copyright header whenever you made
some ''significant'' contribution to a code file
- maintain a list of relevant authors at top-level,
indicating a time range of their contribution
- use the same information also in the debian/copyright (DEB package)
- clarify in the LICENSE that authorship can be extracted from Git
__Remark__: Anton Yakovlev made a single code contribution in 2009 (87e528bd),
which was never used anywhere and removed as part of a general clean-up in 2023.
Thus I do not list him as a code author, but he is mentioned in credits.txt
Debian-Docbase allows to register some HTML documentation;
My old package definition added placeholder config, which renders
the documentation configuration invalid (as pointed out by Lintian).
However, I still think it is a good idea to have the anchor point
already defined, and thus I came up with the idea of in fact
providing some usable placeholder content...
As it turns out, we also have a placeholder page at the Lumiera website,
where the User Manual is assumed to be located later — so why not extend
this one and then provide the HTML-rendering for the DEB package?
To allow for this setup
* I have now extended the placeholder page for the Website
to include some generic description about Lumiera (from the 'about' page)
* Furthermore, I added the screenshot (from the »Outer Space« page)
* and I use this a an opportunity to document the various test / demo
facilities currently available in the GUI, since these are rather obscure.
While only intended for the developer, it seems still worthwhile
to describe the possible effects — it may well be that we retain
some of that test/demo functionality and in that case, we have
now already some starting point for a documentation
* Then, to include that page as stand-alone HTML, I used the 'Print Edit WE'-plugin
from Firefox, to encode the images as inline-base64 URLs (which are restored
by a tiny JavaScript embedded into that page)
* and last but not least, our SCons buildsystem needs the ability
to install such a documentation file, since it seems most adequate
to handle this requirement as part of the generic installation (and
not hidden in some Debian scripting)
- reorganise the navigation tab structure
- place all further index lists into a common "Index" tab
- include a list of concepts
- remove the various "member" sub-tabs, these are not helpful
- the "modules" tab is now called "topics" (since C++ has now Modules!)
- generally re-sync DoxygenLayout.xml with a current pristine template
- comb though the Doxygen Warnings and fix a lot of small problems
The RfC documents were written to complement discussions of the Lumiera developers;
yet since the time where ''Ichthyo'' is working basically alone on the project,
this kind of discussions have ceased. During the following years, some ideas
promoted by the existing RfC documents became rather detached from the
actual state of development in the code base.
Many of the existing RfC documents require some commentary to place them
into context, and some of the decisions taken in the early stage of the
project should be **re-assessed**. This includes the decision to reject
some proposals, which initially might have seemed desirable, yet could not
be reconciled with the understanding of the matter and topic in question,
as was gained through the ongoing analysis and development.
The purpose of RfCs is to channel and document discussions
among a group of developers regarding questions of design.
While there is unfortunately no longer a group of developers
discussing matters and working together on the code base,
some questions can be considered more or less settled,
by being implemented and validated.
- the Scheduler
- structure of the Developer Documentation
- Application installation as relocatable bundle
Some sections of the Lumiera website document meeting minutes,
discussion protocols and design proposals from the early days
of the project; these pages were initially authored in the
»Moin Moin Wiki« operated by Cehteh on pipapo.org at that time;
this wiki backed the first publications of the »Cinelerra-3«
initiative, which turned into the Lumiera project eventually.
Some years later, those pages were transliterated into Asciidoc
semi-automatically, resulting in a lot of broken markup and links.
This is a long standing maintenance problem problem plaguing the
Lumiera website, since those breakages cause a lot of warnings
and flood the logs of any linkchecker run.
Create a new subcategory "design/architecture/time"
and rearrange several pages related to time handling and time codes.
NOTE: starting with this changeset, a ''Link-Farm'' is required for cross-links;
since we don't have an automatic solution for this task yet, I have created
the necessary forwarding pages manually in the website repository.
Starting with ''preview release'' `v0.pre.04`, branch and version tags
will be handled in accordance to the **Git-flow** naming scheme.
Notably this implies that from now on the version in-tree will indicate
the ''next expected release,'' adorned by a suffix to mark the preview.
To accommodate this transition to Git-flow
- the new branch `integration` will be introduced
- the version number will once (and the last time for this release)
be adjusted ''before'' forking the release branch
- branch `master` will transition to reflect the latest released state
- several existing branches will be discontinued, notably
`gui`, `steam`, `vault`, `release`, `play`
Indeed — this change set is kind of sad.
Because I still admire the design of the GAVL library,
and would love to use it for processing of raw video.
However, up to now, we never got to the point of actually
doing so. For the future, I am not sure if there remains
room to rely on lib-GAVL, since FFmpeg roughly covers
a similar ground (and a lot beyond that). And providing
a plug-in for FFmpeg is unavoidable, practically speaking.
So I still retain the nominal dependency on lib-GAVL
in the Build system (since it is still packaged in Debian).
But it is pointless to rely on this library just for an
external type-def `gavl_time_t`. We owe much to this
inspiration, but it can be expected that we'll wrap
these raw time-values into a dedicated marker type
soon, and we certainly won't be exposing any C-style
interface for time calculations in future, since
we do not want anyone to side-step the Lumiera
time handling framework in favour of working
„just with plain numbers“
NOTE: lib-GAVL hompage has moved to Github:
https://github.com/bplaum/gavl
- remove obsolete configuration settings
- walk through all settings according to the documentation
https://www.doxygen.nl/manual/config.html
- now try to use the new feature to rely on Clang for C++ parsing
- walk through the doxygen-warnings.txt and fix some obvious misspellings
and structural problems in the documentation comments.
With Debian-Trixie, we are now using Doxygen 1.9.8 —
which produces massively better results in various fine points.
However, there are still problems with automatic cross links,
especially from implementation to the corresponding test classes.
With the ability to invoke a Render Node graph,
the development on branch `play` reached some kind of milestone
regarding the »Playback Vertical Slice«.
This is a good opportunity to update the reference platform
and upgrade the preview releases and packaging setup accordingly.
This will include adjustments to compile on recent compilers and
upgrade the build system to support Python-3.
Remove left-overs from the preceding prototypical implementation,
which is now obliterated by the change to a flexibly configured `FeedManifold`
with structured, typed storage for buffers and for parameter data.
The Render Node invocation sequence, as rearranged and reworked for the »Playback Vertical Slice«, now seems reasonably clear and settled.
Adding extensive documentation to describe the conventions and structures worked out thus far;
moreover, start makeover of old documentation in the !TiddlyWiki to remove concepts obviously obsoleted now...
The behaviour seems consistent and the schedule breaks at the expected point.
At first sight, concurrency seems slightly to low; detailed investigation
however shows that this is due to the structure of the load graph,
and in fact the run time comes close to optimal values.
the `BreakingPoint` tool conducts a binary search to find the ''stress factor''
where a given schedule breaks. There are some known deviations related to the
measurement setup, which unfortunately impact the interpretation of the
''stress factor'' scale. Earlier, an attempt was made, to watch those factors
empirically and work a ''form factor'' into the ''effective stress factor''
used to guide this measurement method.
Closer investigation with extended and elastic load patters now revealed
a strong tendency of the Scheduler to scale down the work resources when not
fully loaded. This may be mistaken by the above mentioned adjustments as a sign
of a structural limiation of the possible concurrency.
Thus, as a mitigation, those adjustments are now only performed at the
beginning of the measurement series, and also only when the stress factor
is high (implying that the scheduler is actually overloaded and thus has
no incentive for scaling down).
These observations indicate that the »Breaking Point« search must be taken
with a grain of salt: Especially when the test load does ''not'' contain
a high degree of inter dependencies, it will be ''stretched elastically''
rather than outright broken. And under such circumstances, this measurement
actually gauges the Scheduler's ability to comply to an established
load and computation goal.
- use parameters known to produce a clean linear model
- assert on properties of this linear model
Add extended documentation into the !TiddlyWiki,
with a textual account of the various findings,
also including some of the images and diagrams,
rendered as SVG
Investigate the behaviour over a wider range of job loads,
job count and worker pool sizes. Seemingly the processing
can not fully utilise the available worker pool capacity.
By inspection of trace-dumps, one impeding mechanism could
be identified: the »stickiness« of the contention mitigation.
Whenever a worker encounters repeated contention, it steps up
and adds more and more wait cycles to remove pressure from the
schedule coordination. As such this is fine and prevents further
degradation of performance by repeated atomic synchronisation.
However, this throttling was kept up needlessly after further
successful work-pulls. Since job times of several milliseconds
can be expected on average in media processing, such a long
retention would spread a performance degradation over a duration
of several frames. Thus, the scheme for step-down was changed
to decrease the throttling by a power series rather than just
documenting the level.
- repeated invocations of the same test setup for statistics
- the usual nasty 64-node graph with massive fork out
- limit concurrency to 4 cores
- tabulate data to look for clues regarding a trigger criteria
Hypothesis: The Scheduler slips off schedule when all of the
following three criteria are met:
- more than 55% glitches with Δ > 2ms
- σ > 2ms
- ∅Δ > 4ms
...the idea is to use the sum of node weights per level
to create a schedule, which more closely reflects the distribution
of actual computation time. Hopefully such a schedule can then be
squeezed or stretched by a time factor to find out a ''breaking point'',
at which the Scheduler is no longer able to keep up.
The last round of refactorings yielded significant improvements
- parallelisation now works as expected
- processing progresses closer to the schedule
- run time was reduced
The processing load for this test is tuned in a way to overload the
scheduler massively at the end -- the result must be correct non the less.
There was one notable glitch with an assertion failure from the memory manager.
Hopefully I can reproduce this by pressing and overloading the Scheduler more...
* added benchmark over synchronous execution as point of reference
* verified running times and execution pattern
* Scheduler **behaves as expected** for this example
Some test-runs performed excitingly smooth,
but in one case the processing was was drastically delayed,
due to heavy contention. The relevance of this incident is not clear yet,
since this test run uses a rather atypical load with very short actual work jobs.
Anyway, the dump-logs are documented with this commit.
At various places, concepts and drafts from the early stage of the
Lumiera Project are still reflected in the online documentation pages.
During the last months, development focussed on the Render Engine,
causing a shift in some parts of the design, and obsoleting other
parts altogether (notably we consider to use IO_URING for async IO)
- introduce a new entity: RenderDrive
- it supersedes the CalcPlanCalculation, but is managed by CalcStream
- moreover, the RenderDrive will house a IterTreeExplorer-Pipeline
- define the concerns and relationships more clearly (see Drawing)
- prerequisite to disentangle the Job-planning "mechanics"
- decision: the Monad-style iteration framework will be abandoned
- the job-planning will be recast in terms of the iter-tree-explorer
- job-planning and frame dispatch will be disentangled
- the Scheduler will deliberately offer a high-level interface
- on this high-level, Scheduler will support dependency management
- the low-level implementation of the Scheduler will be based on Activity verbs
The drawing code extracts style information from some "virtual"
widgets, which serve as logical placeholder for the actual nested
structure of tracks.
For sake of demonstration, I used rather obvious colours and
also all kinds of margin and padding; a screenshot was added
with annotations to indicate where some specific style settings
are utilised from the drawing code
- pick up all relevant values from CSS
- also control the width of the StaveBracket
- observe the given overall height
Moreover, complete documentation drawing in Inkscape
and add a page to the TiddlyWiki, describing the principles
underlying this design and construction.
Identify the elements of the construction geometry in the "Sketch"
object in the FreeCAD document and paste the corresponding coordinate
values into the SVG drawing prepared for documentation.
The arc segment parameters seemingly are given in radians;
and while FreeCAD uses the common mathematical right-handed orientation,
the orientation in SVG is applied clockwise rather.
