►
From YouTube: ORI FPGA Stand-up Meeting 12 April 2022
Description
Greetings all! This is the FPGA Stand-up meeting for 12 April 2022.
A
Welcome
everyone
to
the
stand-up
meeting
for
april
12th.
This
is
open
research,
institute's
meeting
on
fpga
work
and
we
talk
about
what
we've
done
over
the
past
week,
what
we're
going
to
be
doing
over
the
next
week
if
we
have
any
roadblocks
or
if
we
need
any
resources
so
so
go
ahead.
Paul
you
have
the
floor.
First.
B
A
A
A
A
C
D
A
D
E
E
E
I
think
so
at
this
point
I
think
the
the
the
zc706
would
be
nice,
but
osho
is
he's
working
on
that,
like
I,
I
would
like
to
have
the
card
working
like
pluto
does
because
the
fpga
is
bigger,
so
it's
easier
to
add
ios
and
debug
some.
You
know
weird
stuff
that
I
don't
see
in
sim,
but
I
think
that's
going
to
come
with
time.
A
Okay,
yeah,
I
understand
yeah
the
transition
from
the
x,
so
this
is
I'd
like
to
make
sure
that
everybody
knows
that
this
is
not
an
either
or
thing.
This
is
both
and
like
the
the
work
done
on
the
pluto
actually
has
is
very
exciting
and
and
hugely
useful.
We
do
know
and
acknowledge
that
the
fpga
is
relatively
small
for
the
sorts
of
things
that
we're
trying
to
do,
but
it's
extremely
exciting
to
get
so
much
utility
out
of
this
platform
and
because
so
many
people
have
a
pluto.
D
A
Functions
with
this
code,
I
mean
it's
extremely
exciting
and
so
so
worth
doing
so
it's
not
a
it's
not
in
conflict,
but
we
also
want
to
take
what
we've
learned
with
the
pluto
and
get
working
what's
working
on
the
pluto
and
then
get
it
working
on
the
on
the
706
on
the
zc706
and
then
yeah.
It's
a
much
larger
fpga,
so
yeah.
E
If
there
is
set
up
a
fork
of
analog
devices,
the
hdl
repository
and
put
together,
essentially,
you
can
clone
the
repository
and
make
all
in
the
specific
folder
and
it
should
and
it
builds
the
you
know
the
bitstream
for
pluto
like
there
is
no
more.
You
know
passing
around
the
zipit.
The
compressed
project
with
you
know,
modif
files
that
are
not
on
the
repository,
and
you
know
things
like
that.
So.
A
Okay,
yeah,
I
know
we
have
had
so
on
the
particular.
The
pluto
is
attached
to
karapi
the
virtual
machine,
one
of
the
virtual
machines
that
we
have
in
remote
labs
west,
and
we
also
have
this
large
ultra
scale,
really
big
dev
board
the
the
zcu
106..
So
we
had
to
learn
over
the
past
week
that,
like
the
hardware
server,
doesn't
just
automatically
help
you.
So
what
we
might
do
is
separate
the
you
know
separate
the
two
or
we'll
figure
out
a
way
to
to
share
the
the
hardware
resources.
A
A
So
you
what
you
have
to
do
is
you
have
to
write
a
test
bench
and
you
have
to
write
a
you
know
your
ip
and
then
get
it
to
all
match
up.
Well
and
yeah.
You
do
have
to
still
use
vivato,
it's
called
as
a
command
line
option
and
that's
where
the
problem
kind
of
came
up
is
that
matlab
and
the
hdl
coder
there
doesn't
seem
to
be
any
way
to
do
anything
else,
other
than
use
the
default
hardware
server.
A
So
we
spent
some
time
trying
to
figure
that
out,
which
is
good
because,
as
a
remote
labs,
we
need
to
be
able
to
field
these
sorts
of
things
and
the
probably
the
best
solution
is
to
split
off
into
another
vm
and
have
each
vm
that
that
individuals
use
dedicated
to
a
particular
set
of
hardware.
But
we
did
actually
get
it
to
work.
You
know
by
shutting
down
hardware
servers
and
switching
over,
and
so
just
the
106
is
targeted
by
by
this
workflow,
and
now
we've
run
into
a
different
roadblock
in
that
from
vivado.
A
I
can
take
this
bitstream
that
matlab
created
and
program
the
zcu
106,
just
fine,
it
programs
great,
but
from
hdl
coder
from
matlab
it
doesn't.
It
throws
a
weird
error,
so
there's
something
still
wrong,
but
we've
eliminated
the
contention.
The
hardware
contention
problem,
so
this
is
properly
you
know,
should
be-
should
be
classed
as
a
remote
labs
access
problem
and
not
an
fpga
ip
development
problem.
A
A
The
there
is
another
issue
that
I
should
bring
up
with
the
the
matlab
script
for
m17
for
our
them
17
protocols.
What
we're
using
on
the
uplink
we're,
not
convinced
that
it's
right
yet
so
we're
looking
at
the
spectrum
results
from
matlab
and
they
seem
to
be
way
wide
much
wider
than
the
than
the
bandwidth
expected.
A
The
bit
rate's
9
600
right
now,
and
we
are
going
to
expand
that
for
space
for
five
gigahertz.
But
what
we're
using
right
now
is
the
vhf
uhf
version
of
of
m17,
and
the
script
is
not
really
looking
like
what
we
expected
when
we
really
drill
down.
So
so
there
may
be
some
some
work
yet
on
on
correctly
translating
the
protocol
into
matlab,
which
then
is
you
know,
carried
through
to
hdl
anyway.
D
D
D
B
B
B
A
A
Okay,
so
the
next
couple
of
opportunities
for
us
to
demonstrate
work,
which
is
which
is
really
our
strength
to
to
actually
show
working
hardware
and
software
and
firmware
and
then
publish
everything
as
we
go.
We
have
a
talk
with
ieee
coming
up
on
the
4th
of
may,
and
this
is
in
san,
diego,
it's
in
person
and
virtual,
and
it
focuses
on
the
m17
forward.
Air
correction,
but
also
error,
correction
and
and
techniques
enter
fpga
progress.
A
So
this
is
a
sort
of
a
floor
show,
along
with
a
talk
and
also
citing
work
done
by
andre
and
everest
and
potentially
leonard
and
and
all
of
us
that
have
been
working
on
this
all
along.
So
so
it's
one
of
the
many
times
where
we
just
simply
show
what
we
do,
and
that
has
been
a
very
good
sort
of
technique
for
us
to
to
simply
show
what
we
do.
So,
if
you
can
possibly
help
with
getting
something
working
over
the
air
for
the
for
may
4th
here
in
san
diego.
A
That
would
be
wonderful.
Now.
This
is
a
a
relatively
small
group.
So
it's
a
chapter
meeting
of
ieee
and
not
a
not
something
like
a
conference
like
ims,
2023,
which
we're
we're
also
looking
at
participating
in
the
big
demonstration
where
we
really
want
to
have
as
much
working
as
possible
is
in
august
mid-august
in
las
vegas
nevada
in
the
united
states
at
defcon,
and
I
believe
we
also
have
a
variety
of
people
that
are
going
to
be
able
to
represent
us
at
a
very
large
event
in
friedrichshafen.
A
That
is
a
large
amateur
radio
event
in
europe
and
we
have
a
a
booth,
a
an
exhibit
for
m17
and
for
open
rtx
related
project
that
we
help
and
support,
and
they
are
going
to
talk
about
the
the
m17
protocol,
which
it's
two
things
it's.
It
is
our
our
chosen
digital
uplink
protocol
for
for
space
and
for
the
transponder,
but
it's
also
a
vhf
uhf
voice
protocol
for
two
meter:
ham,
radio,
pfbhf,
uhf
ham,
radio,
so
anything
that
anyone
can
do
in
order
to
participate
at
friedrichshafen.
A
A
B
Getting
back
to
the
question
of
getting
gse
into
wireshark,
there
actually
is
a
way
that
we
already
have.
We
have
the
ayeka
sr1
receiver,
which
does
gse
over
dbs2,
and
it
already
knows
how
to
output.
These
udp
frames
that
wireshark
takes
in
fact
that's
how
we
developed
the
or
how
I
made
the
site
small
changes
to
the
wireshark.
Okay.
B
The
plug-in
I
forget
what
they
call
them,
so
that
could
be
put
up
in
the
remote
lab
and
used
very
quickly
to
to
receive
gse
and
and
get
it
out
to
where
we
can
look
at
the
bits
that
might
be.
It
would
certainly
be
easier
and
would
probably
be
quicker
and
would
be
an
at
least
another
reference
point
for
correctness
of
gsa.
D
Yeah,
that's.
That
sounds
good.
I
think
that
the
tech
is
really
needed
when
we
want
to
test
the
dbs
2x
extension
until
that,
if
there
is
other
cards
which
are
already
validated
receiving,
the
dvdc
is
very
well.
It
could
be
very
convenient
to
to
have
a
reference
on
that,
and
I
don't
know
if
the
the
card
you
mentioned
is
dvbs2x,
but
it's
a
very
good
starting
point.
A
I'd
be
hesitant
to
say,
I
think
that
it
that's
the
why
we
picked
it
was
or
actually
I
think
we
picked
it
mainly
because
of
gse,
and
we
should.
We
should
go
look
and
then
talk
about
it
on
onslack
and
the
mailing
list,
so
yeah
we'll
report
back.
I
think
that
it
does
and
that's
why
we
got
it,
but
it's
been
a
while.
B
A
A
A
A
B
A
Okay,
yeah:
that's
a
very
understandable
approach:
okay,
yeah!
Let's:
let's
go
ahead
and
try
to
do
that
then
we'll
we'll
write
it
up,
it'll
be
what
you
know
and
we'll
we'll
post
it
to
see.
If
there's
any,
if
anyone
else
has
that's
on
the
project
has
some
input.
But
that
sounds
like
a
pretty
good
division
of
of
validation
to
me
and
it's
equipment.
D
A
B
A
E
E
A
B
A
You
you
always
end
up,
even
if
you
don't
know
it
or
if
you
aren't
aware
of
it,
there's
there's
always
something.
That's
that's
giving
you
leverage
when
you
do
it
over
the
air.
It
forces
you
to
do
everything
synchronization
all
the
way
down.
You
know
everything
has
to
be
right
and
it's
a
much
bigger
win,
and
that's
that's
one
of
the
reasons
why
I
pushed
so
hard
for
over-the-air
demos
and
and
working
over
the
air.
A
A
E
B
C
A
D
D
D
D
A
A
So
in
terms
of
like
a
point-to-point,
communications
like
leonard,
is
interested
in
doing
for
for
a
demo,
a
point-to-point
demo
with
pluto's
that
that
he
he
is
is
thinking
about.
Then
it
doesn't
make
a
lot
of
sense
to
have
all
the
complexity
of
a
polyphase
filter
bank.
I
mean,
if
you're,
going
to
do
to
receive
a
large
number
of
frequency
division,
multiple
access,
communications.
A
You
know
if
you're
thinking
about
the
uplink
being
all
of
these
stations
on
the
earth
that
all
are
assigned
a
slot
in
frequency,
then
you're
going
to
need
a
polyphase
filter
bank
in
your
satellite
and
and
so
for
for
you
as
a
pluto,
you
know.
If
you
don't
need
to
have
a
whole
bunch
of
channels
coming
at
you,
then
you
don't
need
the
polyphase
or
multi-rate
processing,
so
don't
feel
like
you
have
to
use
that
technique.
A
Right,
yes,
you
don't
need
that
on
the
correct
you're,
exactly
right,
yeah,
the
the
our
polyphase
or
our
multi-rate
techniques
are
for
manipulating
a
10
megahertz.
Hopefully
a
10
megahertz
wide
uplink
band
at
5
gigahertz
and
all
of
these
signals
might
be
coming
at
you.
So
you
need
to
have
something
flexible,
remarkable
and-
and
you
know,
multi-rate
techniques
and
the
polyphase
filter
can
be.
That
is,
we
think
the
best
solution
for
that
problem,
but
that's
not
something
that
you
have
to
have
with
the
pluto
at
all,
so
so
you're
on
the
right
track.
A
A
A
Because
one
of
the
goals
is
to
to
make
sure
that
everybody
gets
lots
of
credit
and
is
ish
and
their
work
gets
gets
shown
and
shared
we've
gotten
with
we've
been
able
to
to
take
advantage
of
some
excellent
opportunities,
and
their
feedback
has
been
very,
very
positive.
So
the
I'd
like
to
to
make
sure
that
everybody
has
the
opportunity
to
publish
and
to
present
wherever
they
like,
so
so
just
keep
it
in
mind
if
there's
anything
that
anybody
needs
help
with.
Let
me
know.
E
A
Yeah,
I
know
I
get
it
yeah
if
it
wasn't
for
google
calendar
there's.
No,
I
would
do
nothing.
Yes,
I
will
it's
it
I'll
I'll.
Add
it
to
the
our
little.
Our
weekly
report
and
I'll
also
put
it
in
slack
the
the
big
events
that
are
that
are
coming
up
for
us,
where
we
have
the
best.
What
I
think
might
be
the
best
opportunity
and
longer
term
the
ims.
The
international
microwave
symposium
from
ieee
is
coming
up
in
2023.
A
This
will
be
the
summer
of
2023,
so
quite
a
while,
but
as
you
know,
ieee
has
long
lead
times
for
for
submitting,
since
it's
in
san
diego,
I'm
part
of
the
so
so
I'm
part
of
the
volunteer
team
for
that
and
amateur
radio
is
a
prominent
part
of
ims
in
ieee.
So
we
have
some
opportunities
to
show
present
and
publish.
So
if
you're,
if
you
have
any
sort
of
ambition
to
present
at
an
ieee
conference
or
be
published
in
ieee
journals,
and
no,
they
aren't
especially
open
access.
A
Yes,
there's
lots
of
paywalls,
but
it
is
ieee.
So
we
have
lots
of
opportunities
to
to
put
the
work,
open,
source
work
and
stick
up
for
open
source
work
at
that
particular
conference,
and
since
it
happens
to
be
in
my
hometown,
I
can
help
out
a
lot
more
than
than
I
would
if
it
was
for
say,
for
example,
if
it
was
in
hawaii
or
jakarta,
or
you
know,
berlin
or
new
york,
so
so
anyway,
we're
we
are
all
invited
to
to
apply
to
submit
work
for
ims
2023..
A
A
C
Basically,
what
I
have
is
a
face
for
ground
repository
here.
You
see
here,
here's
the
main
repository
for
the
dvb
fpga
and
then
I
created
a
sandbox
basically
over
here.
Essentially
what
this
is
is
we
can
figure
out
what
to
do
with
it
or
where
to
put
it
later,
but
my
vision
of
it
was
and
doesn't
necessarily
need
to
be
the
correct
one.
Is
it.
A
C
C
You
go
into
the
scripts
thing
here.
We
have
a
couple
of
scripts,
one
sets
up
the
environment
and
then
the
other
one
produces
the
user
ip,
which
is
different
than
the
xilinx
ip
right.
So
I've
already
got
a
tickle
screen
here,
it's
initialized,
so
it's
yeah,
so
it's
in
that
scripts
directory
already,
so
this
directory
right
here.
C
C
C
So
the
first
thing
is:
is
you
give
it
the
part
that
you're
working
on
that
would
be
the
user's
part,
whatever
they're
using
or
doesn't
necessarily
need
to
be
because
once
it's
an
ip,
you
can
suck
it
into
whatever
and
it
goes
and
creates
the
ip.
So
now
it's
done
and
if
we
go
into
the
dvb
encoder
has
created
the
xilinx
artifacts
these
three
folders
right
here.
C
C
C
That's
what
the
whole
reason
is,
but
this
would
be
the
customer's
design
anyways
whatever
it
is,
it
doesn't
have
to
be
the
pluto
so
now
that
you've
created
that
piece
of
ip
it's
in
a
repository,
you
go
into
the
settings
thing
here:
I'll
open
up
the
settings,
dialog
box
and
under
ip
it
says
repository
this
was
already
in
there,
but
normally
that
wouldn't
be.
It
would
look
like
this.
Whatever
your
repository
is,
and
then
you
just
you-
you
say
you
add
the
repository
and
so
you
fit.
C
If
we
remember
it
was
all
the
way
over
here
and
this
guy
here,
which
contains
your
gu
and
the
other
file.
So
you
just
click
on
that
select
and
then
it
adds
it.
You
get
a
dialog
box
here
to
say:
okay,
you
want
to
add
this
in
there.
It
recognizes
that
there's
a
encoder,
you
said:
okay
and
then
okay
and
then
now,
when
you
hit
the
plus
sign
here
to
add
you
know
all
your
ip
cores
show
up,
including
what
other
user
ip
is.
C
C
C
Here,
whatever
makes
sense,
I
don't
I
don't
know
what
what
would
make
sense,
but
the
idea
is
that
the
way
I'm
doing
it
right
now
is
it's
just
I'm
playing
in
my
own
sandbox
and
then
I
I've
commented
this
section
out
here,
which
is
just
an
example
of
how
to
you
know,
go
to
a
directory
and
copy.
You
know
make
a
directory
locally
and
copy
it
all
over.
So
we
can
do
that.
C
C
E
C
Ipx
package
project,
you
tell
it
what
it
is
and
and
then
it'll
go
off
and
create
those
things.
So
there's
an
exact
couple
examples
that
I
was
using.
I
was
it's
funny
how
simple
this
can
be
when,
when
you
look
in
the
right
places,
anyways
I'll
stop
sharing
now
and
that's
it.
Hopefully
we
can
figure
out
if
we
need
to
adjust
that
or
use
it
or,
but
the
flow
is
basically
somebody
wants
to
use
a
core.
A
Yeah,
that's
that's
exactly
where,
where
we
we
want
to
go,
we
want
to
make
this
as
easy
to
use
as
possible,
and
I
think
you've
hit
you've
hit
right
on
the
nail
on
the
head
about
about
what
we
need
to
do
in
order
to
enable
that,
and
it's
it's
not
it's
not
easy.
I
mean
you
did
say
it's
as
easy
as
this,
but
there's
a
there's
an
awful
lot
of
things
that
need
to
be
prepared.
It's
a
in
order
to
package
ip
and
to
make
it
available
to
people
that
are
using
these
tools.
A
There's
certain
things
that
you
need
to
do
and
it's
a
it's
a
definitely
a
high-end
sort
of
sort
of
function.
So
so
thank
you.
This
looks
like
it's
really,
not
that
far
off
from
from
being
something
that
can
be
delivered,
some
open
source
work
that
can
be
delivered
through
vivado
to
to
somebody
that
wants
to
use
the
work
in
their
design.
C
Yeah,
I
think,
from
a
let's
say
or
a
face
for
ground
perspective.
We
just
need
to
figure
out.
Do
we
want
to
have
a
repo
that
you
can
get
clone?
That's
like
the
face
for
ground
ip
and
then
that
acts
as
a
library
of
a
bunch
of
different
ips
that
you
put
in
there
or
you
want
each
individual
project
such
as
dbb
fpga,
to
have
their
each
individual
piece
of
ip
that
that
you
do
that
to
so,
you
know,
there's
pluses
and
minuses
to
both
things.
You
just
need
to
decide
on.
C
A
Okay
yeah,
it
sounds
like
something
we
should
start
thinking
about
now
and
and
talking
about
and
see.
If
there's
a
see,
if
there's
a
right
solution
for
us
because
it
it
it
may
be,
to
our
benefit,
to
have
this
to
be
very
modular
and
to
have
each
individual
chunk
available
or-
and
it
may
not
be
in
either
or
maybe
at
both
end-
to
have
each
individual
piece
available,
but
also
have
the
entire
design
from.
E
C
A
C
So
if
you
look
at
that,
so
I
was
thinking
that
if
we
did
something
similar
yeah
that
that,
in
our
for
our
example,
the
projects
would
be
equivalent
to
the
dvd
fpga
and
the
library
here,
I'm
calling
it
p4g
underscore
ip.
That
would
be
like
the
just
the
ip
library
and
then
when
you're
in
the
project
you
just
pull
in
you
know
the
source
or
whatever.