3 dim surface plots

[David Marso]

Some time ago somebody posted syntax for generating 3 d surface plots.
Specifically say I have n,k,m (discrete normally 3..10), phi,rho, delta (continuous).
Resulting in a difficult to calculate value p .

I would like to obtain a plot of p with phi and rho as the other 2 axes and possible have the n k m as panels or indicated by color.
For the moment simply a surface plot of 3 continuous variables would be a start.

Visualize a 3 dimensional representation of 2 normally distributed variables (the 3rd dimension being the probability density ) if this helps to describe what I am after.

I have been through all the chart dialogs but nothing pops out.
I searched these archives for surface plot and contour plot but no joy!

I know it is possible because someone posted syntax.  I probably even have it somewhere but no luck finding it.
This is not the specific function I am after but represents the basic concept.

http://www.mathworks.com/help/matlab/visualize/surface_ex2.jpg

Any thoughts from the Graphics mavens in the room?

[Andy W]

See http://www.ibm.com/developerworks/forums/thread.jspa?threadID=344685

A google search for "spss surface plot" brings up an example using IGRAPH syntax from Raynald's site (I presume you could use that to convert to modern GGRAPH).

There is an example of a 3d histogram in the GGRAPH manual.

I haven't seen any paneled examples of 3d plots, so I can't say if it is possible (logically it should be given the grammer - but that doesn't necessarily mean it is).

[David Marso]

Thanks Andy!
Looks like RTFM time for me ;-)

Andy W wrote

See http://www.ibm.com/developerworks/forums/thread.jspa?threadID=344685

A google search for "spss surface plot" brings up an example using IGRAPH syntax from Raynald's site (I presume you could use that to convert to modern GGRAPH).

There is an example of a 3d histogram in the GGRAPH manual.

I haven't seen any paneled examples of 3d plots, so I can't say if it is possible (logically it should be given the grammer - but that doesn't necessarily mean it is).

[Jon K Peck]

Extending the example on the SPSS Community site for paneling just means adding another dimension to the ELEMENT algebra - it's a grammar.  For illustration purposes I just randomly divided up the points.
I also used a PAGE statement to improve the shape.

INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
compute random = trunc(rv.uniform(0,3)).
dataset name hat.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y random
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
PAGE: begin(scale(10.0in, 4.0in))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 DATA: random = col(source(s), name("random"), unit.category())
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(y*x*z*random), color(z))
 PAGE: end()
END GPL.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Andy W <[hidden email]>
To:        [hidden email],
Date:        03/21/2013 04:49 PM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

See http://www.ibm.com/developerworks/forums/thread.jspa?threadID=344685

A google search for " spss surface plot
<https://www.google.com/search?q=SPSS+surface+plot&aq=f&oq=SPSS+Surface+plot&aqs=chrome.0.59j60l2j62l3.4356&sourceid=chrome&ie=UTF-8>
" brings up an example using IGRAPH syntax from Raynald's site (I presume
you could use that to convert to modern GGRAPH).

There is an example of a 3d histogram in the  GGRAPH manual
<http://www.unt.edu/rss/class/Jon/SPSS_SC/Manuals/v19/GPL%20Reference%20Guide%20for%20IBM%20SPSS%20Statistics.pdf>
.

I haven't seen any paneled examples of 3d plots, so I can't say if it is
possible (logically it should be given the grammer - but that doesn't
necessarily mean it is).



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718954.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Thank you Jon,
That should get me going.
David

Jon K Peck wrote

Extending the example on the SPSS Community site for paneling just means
adding another dimension to the ELEMENT algebra - it's a grammar.  For
illustration purposes I just randomly divided up the points.
I also used a PAGE statement to improve the shape.

INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
compute random = trunc(rv.uniform(0,3)).
dataset name hat.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y random
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
PAGE: begin(scale(10.0in, 4.0in))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 DATA: random = col(source(s), name("random"), unit.category())
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(y*x*z*random), color(z))
 PAGE: end()
END GPL.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:   Andy W <[hidden email]>
To:     [hidden email],
Date:   03/21/2013 04:49 PM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>



See http://www.ibm.com/developerworks/forums/thread.jspa?threadID=344685

A google search for " spss surface plot
<
https://www.google.com/search?q=SPSS+surface+plot&aq=f&oq=SPSS+Surface+plot&aqs=chrome.0.59j60l2j62l3.4356&sourceid=chrome&ie=UTF-8
>
" brings up an example using IGRAPH syntax from Raynald's site (I presume
you could use that to convert to modern GGRAPH).

There is an example of a 3d histogram in the  GGRAPH manual
<
http://www.unt.edu/rss/class/Jon/SPSS_SC/Manuals/v19/GPL%20Reference%20Guide%20for%20IBM%20SPSS%20Statistics.pdf
>
.

I haven't seen any paneled examples of 3d plots, so I can't say if it is
possible (logically it should be given the grammer - but that doesn't
necessarily mean it is).



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718954.html

Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

For those interested in 2D paneled 3D plots, here is a very slightly modified version of Jon's code.
One final (probably not ;-)  Is there any way to get smoothing rather than the rough polygons?
I have hi-lighted my differences.

INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
compute random = trunc(rv.uniform(0,3)).
compute random2 = trunc(rv.uniform(0,3)).
dataset name hat.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y random random2
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
PAGE: begin(scale(20.0in, 20.0in))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 DATA: random = col(source(s), name("random"), unit.category())
 DATA: random2 = col(source(s), name("random2"), unit.category())

 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(y*x*z*random*random2), color(z))
 PAGE: end()
END GPL.

David Marso wrote

Thank you Jon,
That should get me going.
David

Jon K Peck wrote

Extending the example on the SPSS Community site for paneling just means
adding another dimension to the ELEMENT algebra - it's a grammar.  For
illustration purposes I just randomly divided up the points.
I also used a PAGE statement to improve the shape.

INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
compute random = trunc(rv.uniform(0,3)).
dataset name hat.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y random
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
PAGE: begin(scale(10.0in, 4.0in))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 DATA: random = col(source(s), name("random"), unit.category())
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(y*x*z*random), color(z))
 PAGE: end()
END GPL.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:   Andy W <[hidden email]>
To:     [hidden email],
Date:   03/21/2013 04:49 PM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>



See http://www.ibm.com/developerworks/forums/thread.jspa?threadID=344685

A google search for " spss surface plot
<
https://www.google.com/search?q=SPSS+surface+plot&aq=f&oq=SPSS+Surface+plot&aqs=chrome.0.59j60l2j62l3.4356&sourceid=chrome&ie=UTF-8
>
" brings up an example using IGRAPH syntax from Raynald's site (I presume
you could use that to convert to modern GGRAPH).

There is an example of a 3d histogram in the  GGRAPH manual
<
http://www.unt.edu/rss/class/Jon/SPSS_SC/Manuals/v19/GPL%20Reference%20Guide%20for%20IBM%20SPSS%20Statistics.pdf
>
.

I haven't seen any paneled examples of 3d plots, so I can't say if it is
possible (logically it should be given the grammer - but that doesn't
necessarily mean it is).



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718954.html

Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Andy W]

Check out the "smooth." functions in the GGRAPH manual. I'm on V15 at my work, so it is more limited than the current versions (e.g. your PAGE statements to control the height and width of the plot don't work in my version).

I tried this (with the 1 dimensional plot - not the small multiple plots) and my session of SPSS is currently frozen, perhaps you will have better luck with a newer version.

*******************************************************.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y MISSING=LISTWISE
  REPORTMISSING=NO
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(smooth.spline(y*x*z)), color(z))
END GPL.
*******************************************************.

[Jon K Peck]

smooth.linear, smooth.quadratic, smooth,cubic, smooth.mean, smooth.median, smooth.spline, smooth.step

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Andy W <[hidden email]>
To:        [hidden email],
Date:        03/22/2013 06:33 AM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Check out the "smooth." functions in the GGRAPH manual. I'm on V15 at my
work, so it is more limited than the current versions (e.g. your PAGE
statements to control the height and width of the plot don't work in my
version).

I tried this (with the 1 dimensional plot - not the small multiple plots)
and my session of SPSS is currently frozen, perhaps you will have better
luck with a newer version.

*******************************************************.
GGRAPH
/GRAPHDATASET NAME="graphdataset" VARIABLES=x z y MISSING=LISTWISE
 REPORTMISSING=NO
/GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
SOURCE: s=userSource(id("graphdataset"))
DATA: x=col(source(s), name("x"))
DATA: z=col(source(s), name("z"))
DATA: y=col(source(s), name("y"))
COORD: rect(dim(1,2,3))
GUIDE: axis(dim(1), label("y"))
GUIDE: axis(dim(2), label("x"))
GUIDE: axis(dim(3), label("z"))
ELEMENT: line(position(smooth.spline(y*x*z)), color(z))
END GPL.
*******************************************************.



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718967.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Andy W]

While we are on the subject, another cool way to represent 3d (or 2.5d) surface data like this when the sampling isn't so regular (so interpolation over surface plots doesn't look as nice), is to anchor the points to the floor with either an interval or an edge element. Here I use an edge element to make the dashed pattern, and some post-hoc editing in the output viewer results in the chart attached.

I believe I got this idea from the behavior of Mathematica. If you check it out with the full data (with values below zero) the lines end up looking like pin art (actually if you just use points instead of surface lines the shape of the function is readily apparent as well).  

*******************************************************.
INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
exe.
sample 300 from 2499.
temporary.
select if z >= 0.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y MISSING=LISTWISE
  REPORTMISSING=NO
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 TRANS: base=eval(0)
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: edge(position(y*x*(base + z)), transparency.interior(transparency."0.4"), shape(shape.dash))
 ELEMENT: point(position(y*x*z), transparency.exterior(transparency."0.5"), color.interior(z), transparency.interior(transparency."0.5"))
END GPL.
*******************************************************.

[David Marso]

"my session of SPSS is currently frozen"
Ditto ;-(
--

Andy W wrote

Check out the "smooth." functions in the GGRAPH manual. I'm on V15 at my work, so it is more limited than the current versions (e.g. your PAGE statements to control the height and width of the plot don't work in my version).

I tried this (with the 1 dimensional plot - not the small multiple plots) and my session of SPSS is currently frozen, perhaps you will have better luck with a newer version.

*******************************************************.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y MISSING=LISTWISE
  REPORTMISSING=NO
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: line(position(smooth.spline(y*x*z)), color(z))
END GPL.
*******************************************************.

[Patrick Estes]

Greetings,

Thank you for your message. I will be out of the office Friday, March 22 and will return on Monday, March 25th, and will respond to your email at that time.

Regards,

Pat Estes
Assessment Analyst
Office of Institutional Assessment and Research
Edgewood College

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Thanks Andy!
That might prove to be useful too.
I feel like a kid in a candy store!

Andy W wrote

While we are on the subject, another cool way to represent 3d (or 2.5d) surface data like this when the sampling isn't so regular (so interpolation over surface plots doesn't look as nice), is to anchor the points to the floor with either an interval or an edge element. Here I use an edge element to make the dashed pattern, and some post-hoc editing in the output viewer results in the chart attached.

I believe I got this idea from the behavior of Mathematica. If you check it out with the full data (with values below zero) the lines end up looking like pin art (actually if you just use points instead of surface lines the shape of the function is readily apparent as well).  

*******************************************************.
INPUT PROGRAM.
LOOP id=0 TO 2499.
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
EXE.
FORMATS id (F8.0).
compute #tmp1 = trunc(id/50).
compute x = -10 +(20/49)* #tmp1.
compute #tmp2 = mod(id, 50).
compute y = -10 +(20/49)* (#tmp2).
compute #tmp3 = (x**2+y**2)**0.5.
compute z =10* sin(#tmp3)/#tmp3.
exe.
sample 300 from 2499.
temporary.
select if z >= 0.
GGRAPH
 /GRAPHDATASET NAME="graphdataset" VARIABLES=x z y MISSING=LISTWISE
  REPORTMISSING=NO
 /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
 SOURCE: s=userSource(id("graphdataset"))
 DATA: x=col(source(s), name("x"))
 DATA: z=col(source(s), name("z"))
 DATA: y=col(source(s), name("y"))
 TRANS: base=eval(0)
 COORD: rect(dim(1,2,3))
 GUIDE: axis(dim(1), label("y"))
 GUIDE: axis(dim(2), label("x"))
 GUIDE: axis(dim(3), label("z"))
 ELEMENT: edge(position(y*x*(base + z)), transparency.interior(transparency."0.4"), shape(shape.dash))
 ELEMENT: point(position(y*x*z), transparency.exterior(transparency."0.5"), color.interior(z), transparency.interior(transparency."0.5"))
END GPL.
*******************************************************.

[Andy W]

Yes they are quite fun, and post back some examples to the list if you can get some of the smoothing functions to work with the surface plots.

The more you work with it you will find out that the GGRAPH manual does not do the system justice, and the examples section could be extended quite a bit, this being but one example.

(Of course the varieties are infinite, but most of the examples are things you can do in any program that makes stat graphs, there are many (that should be used more often) that you can really only do easily in a few statistical packages).

[Jon K Peck]

By the GGRAPH manual I think you mean the combination of GGRAPH in the CSR and the complete but all-to-terse GPL section of the Help system.  For a start, for the benefit of people not familiar with this, use the ChartBuilder.  It will generate all the boilerplate stuff, and it will illustrate a lot of features of GPL.  Also, you can use the Chart Editor to tweak a chart and save it as a template.

If you really want to get deeply into this, study the ViZml section in the help.  That's the assembly language of the graphics engine, and you can feed your own ViZml to GGRAPH if you are brave.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Andy W <[hidden email]>
To:        [hidden email],
Date:        03/22/2013 11:07 AM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Yes they are quite fun, and post back some examples to the list if you can
get some of the smoothing functions to work with the surface plots.

The more you work with it you will find out that the GGRAPH manual does not
do the system justice, and the examples section could be extended quite a
bit, this being but one example.

(Of course the varieties are infinite, but most of the examples are things
you can do in any program that makes stat graphs, there are many (that
should be used more often) that you can really only do easily in a few
statistical packages).





-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718979.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

"If you really want to get deeply into this, study the ViZml section in the
help.  That's the assembly language of the graphics engine, and you can
feed your own ViZml to GGRAPH if you are brave."

LOL:  When I first picked up Visual Basic around 1996, one of the first books I bought was Dan Appleman's book "Visual Basic Programmer's Guide to the Windows API"
http://www.amazon.com/Applemans-Visual-Basic-Programmers-Guide/dp/0672315904/ref=sr_1_2?ie=UTF8&qid=1363973521&sr=8-2&keywords=appleman+dan
Then I completely fried my brain when I got my mitts on Bruce McKinney's Hard Core Visual Basic.

Brave?  I'm bloody fearless!
Might have to defer this adventure until after I nail the essentials for some current projects and have time to go full bore brainiac into the core.
------------------------------------------

Jon K Peck wrote

By the GGRAPH manual I think you mean the combination of GGRAPH in the CSR
and the complete but all-to-terse GPL section of the Help system.  For a
start, for the benefit of people not familiar with this, use the
ChartBuilder.  It will generate all the boilerplate stuff, and it will
illustrate a lot of features of GPL.  Also, you can use the Chart Editor
to tweak a chart and save it as a template.

If you really want to get deeply into this, study the ViZml section in the
help.  That's the assembly language of the graphics engine, and you can
feed your own ViZml to GGRAPH if you are brave.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:   Andy W <[hidden email]>
To:     [hidden email],
Date:   03/22/2013 11:07 AM
Subject:        Re: [SPSSX-L] 3 dim surface plots
Sent by:        "SPSSX(r) Discussion" <[hidden email]>



Yes they are quite fun, and post back some examples to the list if you can
get some of the smoothing functions to work with the surface plots.

The more you work with it you will find out that the GGRAPH manual does
not
do the system justice, and the examples section could be extended quite a
bit, this being but one example.

(Of course the varieties are infinite, but most of the examples are things
you can do in any program that makes stat graphs, there are many (that
should be used more often) that you can really only do easily in a few
statistical packages).





-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/3-dim-surface-plots-tp5718953p5718979.html

Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD