Glass cockpit hard to read

Phil wrote in news:1191732077.235895.295410
@w3g2000hsg.googlegroups.com:

On Oct 6, 1:51 pm, Bertie the Bunyip wrote:
Phil wrote in news:1191696116.820241.83540@
19g2000hsx.googlegroups.com:







On Oct 6, 10:21 am, Arno wrote:
Hello,

I am computer scientist and usually really like fancy technology. But
I just had my first flight with a "glass" PFD (Avidyne) and must say
I
am not impressed. In particular reading altitude and airpeed from
these scrolling bands requires a lot more attention than with regular
gauges, just like reading a digital clock takes longer than reading
an
analog one. Glancing at it and checking against a known picture, like
"speed at 3 o'clock is fine on final" or "altitude at 20 minutes past
midnight is minimum", just does not work anymore, instead I end up
reading the actual numbers every time I look. Does anyone feel the
same? Am I missing a particular technique?

Arno

I am a fellow computer geek, and a student pilot. I usually fly a
plane with steam gauges, but a couple of times now I have flown with
digital displays. Like you, I found it a little hard to adjust to the
digital displays. The digital displays I was using presented altitude
and tachometer values simply as numbers. The analog displays I am
used to present these values as positions on a dial, showing the
current value in its context of a spectrum of values. With the analog
displays, I am used to adjusting the position of the pointer. With
the digital display, I need to simply set the correct numerical
value. It's a little mental adjustment, and given that I am a newbie
to all this it is an extra distraction.

But, I do think that it is mostly a matter of what you are used to.

Flying is a right hand brain activity. At least the handling portion is.
The right hand side of the brain dosn't do abstractions like numbers, at
least not until the left hand side (which can't fly worth a ****) sends
it over to the right side in a readily digestable form which enables the
right brain to chew it into a picture.

An analogue display cuts the left hand side out of the loop and enables
the calcualtion rate to increase the right sides "frame rate" so that
corrections can be made more frequently thus enabling the pilot to fly
the airplane more smooothly and with more authority.
Caorse rule of thumb math can be laid over this for descent angles,
interceptin angles and wo on, but generally, the fewer numbers involved,
the better.
People who prefer the numbers usualy don't fly very well at all.

You don't do trig while you're shooting pool and expect to win the game.

Bertie- Hide quoted text -

- Show quoted text -

As I understand it, the whole right-brain left-brain thing has pretty
much been discredited. The more recent research shows that any
complex task engages both sides of the brain. I don't think there
were too many neuroscientists who ever embraced the idea that one side
does "art" and the other side does "math" anyway.



Well, it's a simplification, but I have had cause to chat with some
neurosurgeons in the recent past over a protracted period and they still
seem to think it's pretty much still the way it works.

anyone who has ever known a stroke victim know that if the left side gets
zapped they can lose the power of speech altogether, for instance.

(no, my own brain was not invovlved)


Found this anyway http://www.bbc.co.uk/dna/h2g2/alabaster/A659874

I think the biggest difference between steam gauges and digital
displays is that steam gauges give you a value in context of a
spectrum of values. When you look at a analog tachometer, you can see
the where the needle is compared to the minimum and maximum values.
If you are trying to set 75% power, it might be useful to have the
information presented that way, because you can see where 3/4 of the
range is. Same with airspeed. You can see how close you are to the
red arc, where you are in the green arc, etc.


Yes, I agree and is pretty much what I was trying to say. Your brain picks
this up in the same language it's doing the flying job in. You don't have
to convert a bunch of numbers to a mental picture of what's going on, The
rate the needles move, their angle, the way they quiver and dance as they
pick up what's going on, all give subtle information in a form that's easy
to digest

On the other hand, why would you want to see altitude displayed with
needles? If I want to fly at a specific altitude, I am looking for a
number, not a position in a range. I think the traditional altimeter
was made the way it was because it was the easiest way to get a
mechanical instrument to display the information that way, not because
it was the easiest way to digest the information.

Yeah, another god point. trad altimiters aren't great that way, but you do
get used to them.
Sill, if you have a tape only, you have to read the numbers...


Ideally, every instrument should be designed to convey the appropriate
information in a way that is conducive to how that information is
going to be used. If you just need a value, then a digital display of
a number make sense. If you need to have a sense of where you are
relative to minimum and maximum, then a display showing relative
position in a range should be used. The nice thing about a
computerized display is it can be set up to display the information
either way.


Well, not in the things I fly! They pretty much give you what has been
decided for you.

One thing I positivley loath on some modern EFIS displays is the "track
up" option on the nav screen. For those not fmailiar, basicallly the DG
function puts our actual ground track on the lubber line and has a bug to
indicate your heading. This is the reverse of the traditional display where
the heading is on the lubber line and if you have some sort of RNAV
comouter giving info your track is displayed by a bug. It's OK when you are
motoring along enroute, but if you arent used to it, shooting an approach
is a nightmare. Adherents of track up say that it is actually easier, but
it isn't if you are used to the old way. doing an NDB appraoch, for
instance, all you need to do is put the track on the lubber line and the
ADF needle should stay glued to the inbound track, but if you have to
constantly remind yourself to do the oppostie of what you are used to it
can be an absolute nightmare to do by hand. If we did NDB appraoches every
day of the week it wouldn't be a problem, but I've done one in anger in the
last five years and the rest are all in the sim.
Even weirder, the newer airbusses use a speed reference system that uses
groundspeed on the approach. so, you set your Vref and the airplane
automatically raises it to accomodate a headwind by flying a constant
ground speed. (or advising you to fly faster by pushing the speed bug up)
It's simple, but interferes with the pilot's direct communications with the
wing. I suppose I'm trying to say it's translating for you and somethng is
always lost in the translation..

But don't mind me, I wish the 75 had flying wires so I could hear them sing
to me..


Bertie

On Oct 7, 4:55 am, Bertie the Bunyip wrote:
Phil wrote in news:1191732077.235895.295410
@w3g2000hsg.googlegroups.com:



Ideally, every instrument should be designed to convey the appropriate
information in a way that is conducive to how that information is
going to be used. If you just need a value, then a digital display of
a number make sense. If you need to have a sense of where you are
relative to minimum and maximum, then a display showing relative
position in a range should be used. The nice thing about a
computerized display is it can be set up to display the information
either way.

Well, not in the things I fly! They pretty much give you what has been
decided for you.


Sorry, I wasn't very clear there. What I meant was the designer of
the computer display can program the pixels to convey the information
either way. The display can show simply a number, or it can show a
graphical representation of a range of values with a pointer to show
the current position in the range. Right now you are stuck with
whatever the designer chose for you, but as EFIS becomes more common
and more sophisticated, it is very possible that it will be re-
configurable by the pilot. I doubt that it will ever be completely
flexible where you can get it to display any way you want, but I can
imagine that there might be a selection of four or five different ways
to display an instrument, and you can pick which one you like.

One thing I positivley loath on some modern EFIS displays is the "track
up" option on the nav screen. For those not fmailiar, basicallly the DG
function puts our actual ground track on the lubber line and has a bug to
indicate your heading. This is the reverse of the traditional display where
the heading is on the lubber line and if you have some sort of RNAV
comouter giving info your track is displayed by a bug. It's OK when you are
motoring along enroute, but if you arent used to it, shooting an approach
is a nightmare. Adherents of track up say that it is actually easier, but
it isn't if you are used to the old way. doing an NDB appraoch, for
instance, all you need to do is put the track on the lubber line and the
ADF needle should stay glued to the inbound track, but if you have to
constantly remind yourself to do the oppostie of what you are used to it
can be an absolute nightmare to do by hand. If we did NDB appraoches every
day of the week it wouldn't be a problem, but I've done one in anger in the
last five years and the rest are all in the sim.

That makes me wonder how the designers of EFIS displays choose their
designs. Are the designs based on research or are they just the
personal preferences of the designer? Hopefully there is some kind of
objective research used to choose a display pattern that is easiest to
use. Of course, that raises the whole issue of testing displays. Who
do you get to test them? If you use pilots who have been flying
analog gauges for years, you are probably going to find that they want
EFIS displays that are like the analog instruments. But those
displays may not actually be the optimal way for the brain to digest
the information. It makes me think about the keyboard I am using to
type this. It was designed back in the days of manual typewriters.
Because manual typewriters tended to jam if you tried to type too
fast, the keyboard was arranged to slow the typist down. So we are
all using a keyboard that forces us to type more slowly than we could
with an optimal keyboard. The Dvorak keyboard was designed to
eliminate this problem. But hardly anyone uses it because we have all
been taught to type on the old QWERTY keyboard.

Even weirder, the newer airbusses use a speed reference system that uses
groundspeed on the approach. so, you set your Vref and the airplane
automatically raises it to accomodate a headwind by flying a constant
ground speed. (or advising you to fly faster by pushing the speed bug up)
It's simple, but interferes with the pilot's direct communications with the
wing. I suppose I'm trying to say it's translating for you and somethng is
always lost in the translation..


I understand what you are saying. In the software world, designers
try to make each new release more "helpful" than the last. Sometimes
this is good, but mostly I just find it annoying. The software tries
to guess what you intend to do, and do it for you. If it guesses
right, that's fine. But it seems like it mostly guesses wrong, and
then you just have to un-do what it did. That's not an improvement.
I wish they would put more time into designing simple, intuitive user
interfaces so I can more easily tell the program what I want it to
do. That way the program doesn't have to guess.

But don't mind me, I wish the 75 had flying wires so I could hear them sing
to me..

Maybe some EFIS designer can set up the option to have the airspeed
converted to a flying wire sound and played in your headset. :-)

Phil

Phil writes:

That makes me wonder how the designers of EFIS displays choose their
designs. Are the designs based on research or are they just the
personal preferences of the designer?

My guess is that the expensive commercial stuff in airliners is the product of
fairly extensive studies into ergonomy, whereas the inexpensive stuff sold for
small aircraft has not been subjected to that kind of study, since it's not
required for certification and it's very expensive. That's why the
inexpensive stuff looks so much more like a video game and so much less like a
cockpit.

So we are all using a keyboard that forces us to type more slowly than we could
with an optimal keyboard. The Dvorak keyboard was designed to
eliminate this problem. But hardly anyone uses it because we have all
been taught to type on the old QWERTY keyboard.

The most recent studies I've seen on the Dvorak indicate that it actually
isn't any faster than a QWERTY keyboard. It turns out that the brain adapts
very well to whatever layout is used, and quickly gets up to speed.

More evidence of this can be seen in the way some people type on their
Blackberries or cell phones.

I understand what you are saying. In the software world, designers
try to make each new release more "helpful" than the last. Sometimes
this is good, but mostly I just find it annoying.

The problem is that it can be deadly in aviation, and not just annoying.
Airbus is a classic example of the software-developer syndrome.

The software tries
to guess what you intend to do, and do it for you. If it guesses
right, that's fine. But it seems like it mostly guesses wrong, and
then you just have to un-do what it did. That's not an improvement.

Especially if it doesn't allow you to un-do anything.

This is a serious problem even in ordinary office automation software, but
it's much worse in safety-of-life software.

I recall a study done by Microsoft that showed that a great many people who
ask for new features for the Office product are actually asking for things
that are already there ... they just don't have any way of finding them in the
bloated mess that Office has become.

For this reason, I don't use Office--I spend more time trying to prevent it
from doing things I don't want it to do than I spend accomplishing anything
productive.

I wish they would put more time into designing simple, intuitive user
interfaces so I can more easily tell the program what I want it to
do. That way the program doesn't have to guess.

It's extremely difficult and expensive to design such interfaces. And often
the goal is simply to add features to encourage sales and upgrades, and nobody
really cares about the ergonomy.

As I've said, I see signs of this in the low-end glass cockpits.
Unfortunately it diminishes safety.

Mxsmanic wrote in
:

Phil writes:

That makes me wonder how the designers of EFIS displays choose their
designs. Are the designs based on research or are they just the
personal preferences of the designer?

My guess is that the expensive commercial stuff in airliners is the
product of fairly extensive studies into ergonomy, whereas the
inexpensive stuff sold for small aircraft has not been subjected to
that kind of study, since it's not required for certification and it's
very expensive. That's why the inexpensive stuff looks so much more
like a video game and so much less like a cockpit.

So we are all using a keyboard that forces us to type more slowly
than we could with an optimal keyboard. The Dvorak keyboard was
designed to eliminate this problem. But hardly anyone uses it
because we have all been taught to type on the old QWERTY keyboard.

The most recent studies I've seen on the Dvorak indicate that it
actually isn't any faster than a QWERTY keyboard. It turns out that
the brain adapts very well to whatever layout is used, and quickly
gets up to speed.

More evidence of this can be seen in the way some people type on their
Blackberries or cell phones.

I understand what you are saying. In the software world, designers
try to make each new release more "helpful" than the last. Sometimes
this is good, but mostly I just find it annoying.

The problem is that it can be deadly in aviation, and not just
annoying. Airbus is a classic example of the software-developer
syndrome.

The software tries
to guess what you intend to do, and do it for you. If it guesses
right, that's fine. But it seems like it mostly guesses wrong, and
then you just have to un-do what it did. That's not an improvement.

Especially if it doesn't allow you to un-do anything.

This is a serious problem even in ordinary office automation software,
but it's much worse in safety-of-life software.

I recall a study done by Microsoft that showed that a great many
people who ask for new features for the Office product are actually
asking for things that are already there ... they just don't have any
way of finding them in the bloated mess that Office has become.

For this reason, I don't use Office--I spend more time trying to
prevent it from doing things I don't want it to do than I spend
accomplishing anything productive.

I wish they would put more time into designing simple, intuitive user
interfaces so I can more easily tell the program what I want it to
do. That way the program doesn't have to guess.

It's extremely difficult and expensive to design such interfaces. And
often the goal is simply to add features to encourage sales and
upgrades, and nobody really cares about the ergonomy.

As I've said, I see signs of this in the low-end glass cockpits.
Unfortunately it diminishes safety.


You have no idea what it does.

you don't fly, fjukkwit


Bertie

To all: Be advised that mxsmanic (Anthony Atkielski) is not a pilot and
never has been
one. In fact he has never flown in a small plane at all, or been at the
controls of anything other than a game.

He certainly doesn't know anything about avionics.

To be clear, my original complaint was not about screens in general,
but rather the way they are being used. I think the holy grail of how
to represent flight information has not yet been found. Designing a
good user interface is more art than science, as any computer
programmer knows.

Imagine we had open source cockpit software. It could run on actual
plane hardware and also on simulators, so you would have the whole sim
community writing cockpit software. And aircraft owners could download
whatever new software is out there and try it out (on a sunny day at
an uncontrolled airport:-))The FAA would probably have a fit, but the
EAA somehow managed to convince us that homebuilts are safe, so merely
writing software for perfectly airworthy planes can't be so bad.

Arno

On Sun, 7 Oct 2007 12:54:18 -0500, "Viperdoc"
wrote:

To all: Be advised that mxsmanic (Anthony Atkielski) is not a pilot and
never has been

HIWAS for USENET. G

Phil wrote in
oups.com:

On Oct 7, 4:55 am, Bertie the Bunyip wrote:
Phil wrote in news:1191732077.235895.295410
@w3g2000hsg.googlegroups.com:



Ideally, every instrument should be designed to convey the
appropriate information in a way that is conducive to how that
information is going to be used. If you just need a value, then a
digital display of a number make sense. If you need to have a
sense of where you are relative to minimum and maximum, then a
display showing relative position in a range should be used. The
nice thing about a computerized display is it can be set up to
display the information either way.

Well, not in the things I fly! They pretty much give you what has
been decided for you.


Sorry, I wasn't very clear there. What I meant was the designer of
the computer display can program the pixels to convey the information
either way. The display can show simply a number, or it can show a
graphical representation of a range of values with a pointer to show
the current position in the range. Right now you are stuck with
whatever the designer chose for you, but as EFIS becomes more common
and more sophisticated, it is very possible that it will be re-
configurable by the pilot. I doubt that it will ever be completely
flexible where you can get it to display any way you want, but I can
imagine that there might be a selection of four or five different ways
to display an instrument, and you can pick which one you like.


Nah, i knew what you meant, I just meant that means nothng to us. we get
what we get!
The exception is th nav screen where we can select traditional HSI type
dispays which is useful sometimes, but mostly we stay in map mode.
The operator can choose what kind of display is used to some extent. The
track up thing is an operator choice (by operator I mean the airline)
mine chose track up but the traditional display is available as well.



That makes me wonder how the designers of EFIS displays choose their
designs. Are the designs based on research or are they just the
personal preferences of the designer? Hopefully there is some kind of
objective research used to choose a display pattern that is easiest to
use. Of course, that raises the whole issue of testing displays. Who
do you get to test them? If you use pilots who have been flying
analog gauges for years, you are probably going to find that they want
EFIS displays that are like the analog instruments. But those
displays may not actually be the optimal way for the brain to digest
the information. It makes me think about the keyboard I am using to
type this. It was designed back in the days of manual typewriters.
Because manual typewriters tended to jam if you tried to type too
fast, the keyboard was arranged to slow the typist down. So we are
all using a keyboard that forces us to type more slowly than we could
with an optimal keyboard. The Dvorak keyboard was designed to
eliminate this problem. But hardly anyone uses it because we have all
been taught to type on the old QWERTY keyboard.

Exactly

Even weirder, the newer airbusses use a speed reference system that
uses groundspeed on the approach. so, you set your Vref and the
airplane automatically raises it to accomodate a headwind by flying a
constant ground speed. (or advising you to fly faster by pushing the
speed bug up) It's simple, but interferes with the pilot's direct
communications with the wing. I suppose I'm trying to say it's
translating for you and somethng is always lost in the translation..


I understand what you are saying. In the software world, designers
try to make each new release more "helpful" than the last. Sometimes
this is good, but mostly I just find it annoying. The software tries
to guess what you intend to do, and do it for you. If it guesses
right, that's fine. But it seems like it mostly guesses wrong, and
then you just have to un-do what it did. That's not an improvement.
I wish they would put more time into designing simple, intuitive user
interfaces so I can more easily tell the program what I want it to
do. That way the program doesn't have to guess.

But don't mind me, I wish the 75 had flying wires so I could hear
them sing to me..

Maybe some EFIS designer can set up the option to have the airspeed
converted to a flying wire sound and played in your headset. :-)


It's been done!

Well, almost. Some MD80 types have an engine noise generator to aid the
crew in hand flying approaches.


Bertie