|
|
|
|
INNARDS LEE's Content Development Journal for Project Luminance
|
winter quarter 3/5/05
VALUABLE TESTS
Quite useful results from evening of tests! Users very patient, beer and snacks helped. Variety of responses VERY interesting, some really getting into movement, especially with the bubbles. All did a good job of writing down suggestions.
Users see all FoggyWindows imagery as a puzzle. But there was not enough room for them to get a good angle on the underlying images. Most perceived this interactivity as a “Guess the Picture” process, and so were frustrated when they never were shown the whole movie or image involved.
Projected images’ resolution needs to be higher to enable clarity of image within screen’s proportions and to scale illusion of virtual depth.
Users oohed and ahhed over bubbles and Foggy Windows, but lack of closure was bothersome.
Shadows of users messed up their ability to see what the screen held. After test, Scott agreed that switching the user to the side with the data capture will work just as well. For the illusions, this is even better: there will be no shadow blocking screen images - it will seem even more like magic.
Ghost data really screwed up users' perceptions of exactly WHAT they were affecting on the screen AFTER a while. Their initial impressions of their FW actions was one of direct interaction. But as the ghost data kept squares on screen that they had not touched, disbelief set in. Since the piece is intended to go on for several minutes, we gotta fix this.
2/22/05
ROADWAY CONTENT
Unknown whether illusion of movement forward can be calibrated to operate slow enough for users to feel like they are moving into space but also have the time to explore, "touch" objects, etc. Problem of cycling environment which doesn't seem like a treadmill is usually solved by sporadic variety of graphic elements. No idea yet how well our Flash program will handle large files.
J.J.'s current tests are at 640 X 480, default 12 fps. Will set user test material @ those specs to see of speed & resolution work OK.
Reinforce z axis with underwater roadway: bubbles that scale FORWARD, not just straight up & same with touched animals in the piece's latter level. They can lunge forward by scaling heads up to create more of the 3D illusion.
Flash loops are glitchy. Tendency as ever to very very briefly hesitate at 12 fps.
Contacted all possibles for test users. Revisited the test docs I had created. Staying with few questions, but have given them plenty of opportunity for them to volunteer feedback.
2/16/05
3RD CLUBHOUSE TESTS Met with Phil. Need to send him reminders re our site, blogs regularly. He's interested in more narrative story, discussed sound. Steam bath oppressive.
J.J.: how
about taking roadway underwater? Could use fish > eye of fish. Me:
catfish! dogfish! Outta bubbles. PROJECTOR to SCREEN = 9.5'. SCREEN to CAMERA = 8'2". Discussions
re install. work results: White
outlines around each fog.window square does temp job on masking problems. Scott: will
try matching refresh rate of camera with projector re jitter effect. Performance impact. Display points' name assignments, points coming over in order, and ghost behavior. Halfway through
converting content chart to storyboard+ activities chart a la game production
> all relevant info in one doc. Doc will be in Word, so any can enter. I need to
contact victims, uh, test users re day & time.
2/13/05
To team for list: We need to test in a measured way the greyscale range for
front projection, rear projection w/ Otis screen, and iStory screen re: 1. wide columns of 10% & up in B&W (like video bars) 2. "
"
"
"
"
" " of saturation
check out SAS > LEE > graphics > Greyscale
Saturation.psd Otis mentioned the importance of saturation response to
interactive installations he'd done. Good to see what affect just saturation
variables Eyesweb responds to, even if it only "sees" in B&W. Also need to fix teeny
squares' outlines on expanded masks
2/12/05
2nd CLUBHOUSE TESTS - REAR CAPTURE Clubhouse = goddam steambath. Screen is only SLIGHTLY narrower than room > we're crawling under. Rear capture > much sharper definition for data points generation. Persistence of ghost data on screen, doesn't clear. PC-ish J.J. ready to toss Mac out window. Showed team the content/activity charts I'd developed based on the boards. Much amusement re pencil on paper. Clearer discussion re tasks & needed technologies. In conjunction with storyboards> checkered response & good suggestions: first pass lacks unifying theme, too big a jump to animal creation. Showed my new Surreal Highway idea, then ending back with eye. All liked it. Scott liked animals too, how about fish? TO DO LIST Scott: proximity algorithm density motion sensor control panel ghost data on screen squares latency J.J.: big circle into bubbles transitions, so mebbe for later, but to not forget: creating
a grid of solid circles that wipe on & grow in their spots grid
of solid circles into grid of squares spread of unmasked squares outward from shadow edge do squares fade test from one still to another still instead
of back to white " "
"
"
"
a light color to movie ( very light orange to foggywindows.red?) Lee: board "Circular" (version of scenes w/ Surreal
Roadway to Eye as end) do up content chart as spreadsheet send out test invite to mmhive & civilians, demanding
RSVPs create look for eye, eye bkgd. and roadway that works with
bubbles create art for above as well as first pass movie loop for
roadway contact iStoryites re screen board "Playmates" for thesis update at end of
quarter
2/11/05
TO TEAM The following
is a stream of consciousness flow. So some parts may contradict others. Emphasize
Z axis throughout. Reinforce moving forward or through to another level, both as a "reward" for completing current level and as a "start" for next level. The appearance of a transition image
conveys that the user is now poised to begin a new activity. Possible
transition images/actions = storyboard trans1.psd (torn textures) Curtain (cycles
slightly) "split" in screen that is flat color Universal
characteristic of all transitions = sliver of next level seen is a lighter
color for attract mode. In foggy
windows pattern, instead of fading back to white after wipe/touch, a
changed square fades to something else another image, or another movie
clip. A constant in every level is a fleeing,
gleeful figure, who
occasionally quickly glances back. Interactivity
progression: 1.) first appearance of a shape = teaches
user what to do with that shape, its behavior if you touch it, push it OR
what the shape does in reaction to the nearness/touch of a shadow 2.) next time the same shape appears, the
user knows its behavior THEN shape can become a tool to make something else
happen > a progression in interactivity. Example: User finds
that pushing an abstract shape next to another abstract shape makes the two
stick together, blend. Then, later on when the user reaches storyboard trans.1,
panel 5 he/she will know to push parts together. Each level
reflects a change in complexity: TEST CONTENT
1 Attract
level: tracking eye Action
of storyboard Eye 3page1 ends in eyewipe to a differently colored bkgd.
where: First level:
simple shapes to move around shapes
become more numerous shapes
that are made to bump, then stick together shapes
become too dense for any other actions Second level:
torn textures takes user to next level Storyboards' shapes into animating dogs and cats sequence, which ends in eye image again. TEST CONTENT
2 Foggy Windows Attract level: Random, scattered blocks blink on and fade off, following the pattern of user's cast shadow (hopefully encouraging user to do more on purpose) Beneath = image of still eye Continued
movement brings eye forward (as though user is moving into eye) Second level: eye image is progressively replaced by curtain, cycling slightly. When"pulled" apart, movement reveals a moving scene, with the fleeing figure. Scene moves on the z axis, so as the user wipes, he/she is moving deeper into the imaginary space. Third level:
certain zones of squares become more malleable , distorting as they are
touched by the cast shadows
(perspective lines and chiaroscuro can continue to give the z axis
illusion) The fleeing figure shifts among them, perhaps disappearing
altogether for brief moments, like a deer running in a dense forest. Fourth level:
movies in different zones change into new movies another form of
manipulation? Final: fleeing
figure runs into blurred center that comes into focus to be Š the eye. OR
Maybe
final eye of content 1 begins to blink into blocks as a segue into Foggy
Windows FLEEING
FIGURE Could perhaps
be manipulated like the dog and cat sequence to do jumps or body rolls,
while always returning to fleeing. Fleeing = default pattern, but each bump
activates a different short action. Flee jump - flee forward roll flee,
etc.
2/9/05
TEST DOCS Finished 90%
of testing documents for user testing and uploaded to SAS: Consent form, Observer's evaluation form, User's evaluation form and a Monitoring advise sheet: MONITORING TESTS LIVE Videoing and
monitoring notes will give us best bang for buck re time & schedule. Most
instructive or all videos can go up on site. Each user
plays without having observed others. No revealing of project's content/usage by finished users to upcoming ones. Any
instructions at all to users at start?
Completely virginal start? I
say no instructions. But have places in participant listing for notes re
confusing bits, boring bits, etc. Any answering
of questions during play or do they have to tough it out? No voluntary
comments to users while they are playing. USER
EVALUATION ELEMENTS: Pleasure Puzzlement Frustration Feedback
response/ understanding Assertiveness Immersion/interest INSTALLATION
EVALUATION FACTORS Errors Response time Visual clarity
2/8/05
1st CLUBHOUSE TESTS Front
projection screen. Awkward with capture on one side & projection on
other. Must communicate verbally alone. Scott
captures data off just blank wall behind screen: can see effect clearly of
uneven lighting on data capture surface > points cluster and/or disappear. Obvious
latency , get down to 2 seconds (acceptablr range?) After reset,
still severe lag despite monkeying w/ # of squares, etc. Messages stacked in
gateway? Memory use
readout not wholly useful. Need to add logging, readout to text file + time
stamp. Expanded
Foggywindows create a perhaps sub-pixel/pixel gap in mask, so can see through
to image below.
2/4/05
OTIS MEETING NOTES Screen: 6X8 ,
quick to put up, legs @ 4 ft. Other artists use cotton weave. If use just fabric: 1/4" thick, approx. $5K - $35K alone. Projector :
will have to check our brand for lumens level. Rear projection needs 5K 8K,
but if rear capture, maybe 2K-3K.
Moderate darkness, not full darkness necessary. Much on
set-up re environment. Pipe and drape to create a zone? Screen breaks down to 5' X 1' carrying case. He will lend
us one!! Yeaaa!! Downside = can only be 1 month. Info from
J.J.: data from flosc = too fast! Baf w/ Scott: control panel> local
shared objects vs remote shared objects.
1/30/05
IT MOVES! Played around with J.J.'s impressive FoggyWindows programming, substituting animated .swfs for his still image. VERY cool illusion he's created with fading rectangles. Perceived concerns to be evaluated: 1.)
Scale of squares & placement of content beneath the squares > serious impact on user's perception of what's there. 2.)
Pace of each squares' fade the slow dissolve gives great weight to sophisticated
feel of experience. 3.)
Moving
images beneath should have slow pace for wipe-off pattern to reveal in a
clear and pleasing style, fast pace = annoyingly confusing for user. find the screamy face: http://sasweb.csuhayward.edu/tests/foggywindows.red cruisin' down the river: http://sasweb.csuhayward.edu/tests/foggywindows.blue
1/26/05
STORYBOARDS Uploaded 7
very rough boards to SAS>LEE>graphics>STORYBOARDS with braindump of possible visuals
and interaction, using both the trailing effect (bubbles) and FoggyWindows
effect. Hard to keep
tone light and make it deep, most ideas coming out dark (function of own head
space. Made interactions deeper and deeper as progression, but tale not there
yet.
1/20/05
PERMANENT SPACE Brainstorm of new Year's will work. We will be able to use my ex-office in AE275 for our permanent installation. Raquel has come through like a champ & is getting the key. So, after all this time . . . and a shrunk team THE CLUBHOUSE
LIVES! Narrow space
and tons of furniture > awkward, doable.
1/18/05
SECOND TAKE ON INSTALLATION CHARACTERISTICS In
addition to the Principles of interactivity outlined in the Pause & Effect book:
1) Inside / Outside 2) Input / Output 3) Closed / Open our project seems to follow 3 outta 4 of Murray's take on online environments. However, our task will be to create #3. To activate an illusion of z axis. Janet
Murray, Hamlet on the Holodeck. 1997: MIT Press, Cambridge.
Online environments are:
1.)
procedural. They are structured according to the
capabilities and limitations of the technology.
2.)
participatory. They respond to reader/user input.
3.)
spatial. They simulate a space which can be
navigated.
4.) encyclopedic. Because of vast
storage potential, online environments can be wide in scope and intricately
detailed.
1/12/05
EMAIL TO SAS TEAM RE DA SCREEN Finally heard back from Otis Brown of Billingsly & Brown
A/V. He is totally jazzed and down with art installations &
bent my ear with tales of his favorites, curved tunnels, etc. etc. Signed on
as display guru due to connection with Mike, his "good buddy". Hmm. His info: Commercial translucent screens come only in one vinyl-based
fabric density, which has a slight reflectivity in order to prevent blur.
Sizes from 5' to 8', with 8'er = 6' X 8'. They have an aluminum frame,
attached legs & sit as low as 4 feet off the ground. Folks do run around to fabric stores with large flashlights to
find other screens, which they stretch over all sorts of frames. To buy the
commercial grade stuff raw goes for approx. $300 a sq. yd. When I described our project, he said that the key factor was
the projector. How the projector receives & transmits images would be as
crucial as screen translucency. I hadn't paid attention to the make of ours.
I do recall we had talked about the difference of the experience in AE 160 to
our grad lab demos re projectors. He gave a fast lecture on LCD, CRT & Digital Laser
projectors re grey scale handling, and brightness. Highlights: LCDs good for
computer generated graphics - rather than architectural drawings. (I think he
was referring to saturation) CRT projector is more sensitive to grey scale,
but can only go to 1000 lumens. Digital Laser projectors have most excellent
grey scale range, but super expensive. Nothing really shows true black,
although DLP reaches human perception of it. And much more . . . I concluded that our trio needs to visit him in San Leandro to
maybe even do a lil' show and tell. He's glad to do that, but Jan. is very
busy time for them due to all clients start-of-year presentations. But he
seemed eager to "play", even after I mentioned our No Budget. So he
gave me his personal cel phone #.
1/11/05
EYE ON THE PRIZE Whipped up several storyboards of content possibilities - faces and close-ups of an eye - for the project's attract mode.
J.J. and I agreed that the lone eye seems to have most potential. J.J.'s Flash masking demos indicate a good way to go for the write-on illusion, either for brush stroke user activity or even some patchwork quilt reveals. J.J. on Flash sound: Can't change pitch. Possible = load a different tone. Use different sound samples. Can change in volume and balance. J.J. found
excellent book I instantly ordered: Pause
& Effect: The Art of Interactive Narrative by Mark S. Meadows
seems exactly to our points!
1/03/05
TIME FACTOR Another
variable we could use to enable changes in interactive results would be
time. That is, having one
gesture when repeated or sustained past some initial amount of time make
a different effect than the initial one. Circling your
hand would make an object on the screen move in a circle for a while, then
(perhaps accompanied by a sound) continuing to circle would make the object
smaller. Or make its perspective
seem to change by distorting one side of the object to be smaller. If the object were a rectangle > a
door opening. This could
give the impression of moving inward to another level. Time for drawings. The aim is to create more complex experiences WITHOUT demanding that the user do more complicated gestures. As in other activities we've discussed, the pacing of these alterations would be key in their comprehensibility and fun quotient.
|
