►
From YouTube: Jitsi Live Stream
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
B
B
We
started
this
with
by
the
Panda
actual
quality
of
the
picture
that
is
transmitted
with
and
without
simulcast,
and
we
noticed
some
discrepancies
there,
namely
given
the
same
bit
rate
for
a
regular
video
stream
and
the
same
iterate
for
the
high
quality
of
the
simulcast
frame.
We
see
lower
quality
for
Samuels
castle,
so
we
are
trying
to
figure
out.
Why
happens.
A
C
A
Of
weeks
we
found
the
thing,
so
the
main,
the
main
discrepancy
that
we
are
observing
as
Boris
said,
was
that
simulcast
is
producing
a
lower
quality
high
stream
than
long
simulcast,
which
is
a
fully
fully
on
the
chrome
side
and
we've
been
if
anyone
has
any
ideas
about
that,
would
be
interested
to
hear
them.
D
Okay,
we
don't
get
comfy
couch,
that's.
D
Hardly
want
to
share
anything
we
launched
and
some
new
shards
of
meat
did
see
recently
with
some
of
the
latest
versions.
Everything
seems
to
be
working
well.
D
A
G
So
most
of
the
problems
are
just
a
very
happy
for
searching
for
posit
to
that.
You
can
java
education
and
ask
if
education
environment,
so
we
had
to
add
several
polyfills
first
hauser
API
or
something
like
that.
It's
missing
in
the
act
native
and
then
white
currently
was
becoming
of
powers.
That
was
using
a
project
which
integrate.
We
were
the
native
over
the
single
going
to
they
are
naked.
So
there's
some
differences
are
most
kind
of
behavior
third-party
project.
So
it's
those
references.
H
It
seems
some
stuff
well
like,
for
example,
motivating
reloads,
and
so
you
may
make
sure
that
the
recorder
is
not
the
life
it
as
a
moderator
in
conference
and
some
slow
for
you,
our
stuff,
from
the
recording
the
live
streaming
tag
and
now
shows
for
all
the
participants.
It
wasn't
the
case
before
we
have
known
I,
don't
remember
anything
else.
You
just
fix
me
on
the
front
recorder,
the
user
interface,
alright,.
E
F
It's
now
pushed
to
the
master
and
after
that
I
was
working
on
panting
automated
test.
For
that,
and
it's
almost
done
and
the
way
it
works
is
that
we
we're
sending
up
video,
which
has
a
green
dot
and
and
I'll
do.
It
is
a
bit
which
will
be
played
when
the
green
dot
appears
and
so
onion
we're
capturing
the
frames
on
the
video
and
looking
at
what
time
the
green
dot
appeared
and
checking
if
it's
in
sync
with
this
beep
sound.
F
F
There
was
a
problem
in
firefox,
which
term
Alton
two
separate
doors.
The
first
one
was
that
if
something
fails
became
the
bottom
is
basically
not
working.
That
was
caused
by
some
books
in
our
promises
when
we
are
adding
removing
tracks
leaving
the
conference.
So
the
promises
never
end
they,
they
was
never
completed,
so
I
fixed
up
and
now
those
from
those
those
methods
should
should
be
completed.
In
early
success.
Over
with
the
rejection,
there
are
fixed,
the
king,
you
put
them
where
there
isn't
some
kind
of
error
and
there
were
solutions
of
firefox.
A
B
B
F
A
Are
okay,
well
Debbie
ins
beyond
us
back
and
he
was
out
of
location
and
he'll,
be
what
might
be
interested
to
know
what
might
be
of
interest
to
many
is
that
he'll
get
back
to
fixing
GT
gassy,
so
that
was
broken
since
some
of
our
dtls
connection,
optimizations
and
I
know
that
many
people
have
asked
me
about
it,
so
we
should
be
fixing
that
that's
what
Damien
is
doing
all
right.
So
this
this
completes
the
update
on
our
site.
Open
mic.
Are
there
any
questions
recommends
desert.
C
C
C
Yeah
that
part
of
the
thing
was
eventually
to
push
it
to
call
stats,
but
there's
kind
of
the
call
stack
publishing
piece
and
then
there
was
also
the
gathering
piece
which
I
think
was
in
the
channel
because
we're
spending
some
time
getting
statistics
done
and
to
push
to
our
own
stuff
and
we
were
going
to
go
in
and
add
stuff
to
the
channel
we've
been
holding
off.
Since
I
saw
that
diff,
but
I
knows
hasn't
gotten
merged.
Yet
so
I
don't
know
if
it
was
because
Dame
is
on
vacation
or
what
yeah.
G
C
J
Hi
everybody
I,
guess
I
hope
you
guys
on
the
arm
area,
so
welcome.
Thank
you.
I
think
we
talked
to
George
a
little
bit
and
then
I
started
adding
some
of
their
logic.
You
guys
added
for
VP
for
issue
six
foot,
four
key
frame
depiction
and
all
that
for
simulcast.
So
that's
that's
like
it's
good
for
you
guys
to
review
something
that
they
helped
back
on.
Was
there?
That's
really
are.
A
Yeah,
so
this
is
we're
currently
at
Boris
and
George
mentioned
we're
currently
working
on
evaluating
getting
exact
members
of
the
gain
of
simulcast
and
are
having
a
full
picture.
What
exactly
what
we
are
achieving,
getting
extending
ramp
up
and
big
right
estimations
is
basically
the
immediate
task
after
that,
so
I
expect
we
would
start
next
week,
I'd
be
a
bit
optimistic
or
they're
not
impossible
and
the
week
after
for
sure,
okay.
H
A
A
And
by
the
way,
yeah
looking
forward
to
that
h.264
push
pull
request
because
we
gotta
do
pretty
much
the
same
short
stuff
or
for
VP
10
as
well.
I'm,
sorry,
yeah
well,
eventually
from
VP
teffon,
is
well
with
59.
First,
so
it'll
be
good
to
see
that
we
have
all
the
college
specific
stuff,
properly
isolated,
so
h.264
well,
but
then
we'd
also
like
to
start
playing
with
h.264.
Actually,
how
are
you
guys
doing
h.264?
Are
you
using
the
native
open,
h.264,
simulcasting
or
I
just
cloning,
a
stream,
so.
C
J
K
A
A
From
cisco
last
week-
and
he
said
that
that's
a
that
features
should
be
very
stable
and
working
at
this
point,
and
it
should
give
us
better
cpu
and
also
I
just
saying
in
case.
You
want
to
have
a
look
at
that.
Well,.
C
A
A
C
Not
sure
how
well
there
we
have
to
check
out
how
well
the
web
RTC
stack.
Wood
handle
that
as
opposed
to
they
all
they
do
rap,
like
kind
of
one
encoder
impul
wraps
all
three
individual
encoders,
but
I
do
think
that
they
have
this
expectation
around
spinning
up
the
multiple
versions
that
might
be
a
bigger
change,
but
I,
don't
know
we.
We
are
our
rate,
that's
how
they
do
it
in
vp8.
Already,
oh,
really
kind.
B
A
L
L
A
Ok,
I:
have
you
gotten
any
feedback
from
from
the
Google
guys
from
this?
No
we're
waiting
for
them.
Yeah
we
had
Marian
put
a
patch
up.
Ok
well,
with,
would
certainly
be
interested
in
that
as
well.
Actually,
Microsoft
is
going
to
come
out
with
h.264
support
in
the
next
month
on
their
better
problem,
the
public
better
program,
so
that
would
make
it
much
more
interesting
for
us,
as
well
as
a
at
least
as
a
fallback
option,
so
we
were
going
to
start
playing
with
it
as
well
as
soon
as
it
adds
an
edge
yeah.
C
Hardware,
acceleration
is
not
fully
baked,
yet
we're
probably
going
to
start
taking
a
look
at
that
soon
on
the
chrome
side
aren't
see
they
had
so.
The
Google
team
had
was
making
some
reasonable
progress
on
264
related
stuff
for
a
while
they're
in
it,
and
it
seems
to
have
kind
of
died
off
the
last
month.
So
I
don't
know
I'm
going
to
try
and
be
harassing
a
bit
and
see
what's
going
on
yeah.
A
A
Yeah
Lucy:
are
you
still
on
the
call.
C
D
C
It
was
in
the
comment
that
I
posted
the
cpu
usage
was
not
a
problem.
Wasn't
that
it
was
the
CPU
was
maxed
but
they're
quite
clients,
weren't
kind
of
ramping.
All
the
way
up
and
like
it
looks
like
I
didn't
get
enough
to
kind
of
dig
into
it.
Look
like
maybe
there's
not
a
raw
sea
view
problem,
but
maybe
contention
or
something
like
that
outer
ok,
yeah
that
I'm
not
sure
what
the
testing.
C
It
could
could
be
that
I
haven't
dug
into
that
yet
so
it's
the
the
bridge
is
not
be,
certainly
not
the
only
possibility
for
wrongdoing
there.
I
just
wanted
to
know
if
somebody
wants
I
didn't
report.
Anything
from
my
findings
file
a
bug,
but
somebody
else
filed
a
bug,
so
I
kind
of
added
what
I
saw
to
that
to
see?
If
maybe
you
know,
there
was
a
pattern
emerging
here
or
something
but
I,
don't
know
for
sure
yeah.
That's
interesting,
because.
A
So
it
would
basically
imply
that
we
stored
our
big
raid.
Estimator
starts
misbehaving
on
higher
loads,
which
is
of
course
not
impossible,
but.
C
C
C
If
you
have
time
to
look
at
it
a
little
bit
more
I'll
see,
if
I
can
get
I,
don't
know,
we'll
have
to
see
how
much
I
can
day
into
it,
but
I'd
like
to
at
least
get
maybe
like
a
profile
profile
of
snapshot
or
something
to
see.
If
that
reveals
anything
interesting
but
well,
I
think
that's
interesting
is
to
actually
see
what.
A
C
A
C
A
D
A
Well,
we
should
actually
publish
it
as
an
md
final
yeah
George.
Would
you
do
that?
Absolutely
okay,
so
you're,
not
a
bearded,
George,
Brian
and
I
decide
itself
so
by
the
way,
Brian
thanks
very
much
for
sending
for
sending
me
the
instructions.
On
the
example
on
setting
up
a
conference
room
rest,
I'm
going
to
publish
that.
I
owe
you
a
beer
for
that
abuse
on
campus
of
each
other
stole.
A
B
B
A
A
D
D
D
A
Yeah,
maybe
now
that
sounds
like
clear,
switching
problem
to
me:
maybe
not
yeah,
okay,
yeah,
sorry
for
sorry!
For
that
guys!
Ok,
this
do
we
have
anything
else.
I.
A
Guess
not
ok,.