forums.silverfrost.com Forum Index forums.silverfrost.com
Welcome to the Silverfrost forums
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Clearwin+ figures in Science Advances

 
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> ClearWin+
View previous topic :: View next topic  
Author Message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Wed Jan 18, 2017 11:13 am    Post subject: Clearwin+ figures in Science Advances Reply with quote

The plots for this publication were done using FTN95 and Clearwin+

http://advances.sciencemag.org/content/3/1/e1601558.full#F7

The only thing which was hard to implement is this damn tic numbering.
it is still done by hand. Anyone has good experience here and can help?



Hope new %pl will be eventually able to create such plots in one single line of Fortran text!

The new 64bit compiler i just moved to allows to load and plot MUCH higher resolution data (10000x at least) then in above figure made in 32bit CWP and OpenGL. And with almost a billion triangles displayed it is doing that really fast. It is almost unbelievable and fascinating that when plot is done you can move, scale, rotate all these B particles (and virtually travel inside this 3D space) almost in realtime with no visible lag.
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Thu Jan 19, 2017 11:18 am    Post subject: Reply with quote

Here are some press releases on this article

https://www.eurekalert.org/pub_releases/2017-01/csu-rci011017.php
Back to top
View user's profile Send private message
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Fri Jan 20, 2017 8:34 pm    Post subject: Reply with quote

Dan, in the plots of this type in several figures in the article I see some strange tilted 'contour bars', shouldn't they be vertical ? (like the one you posted above) - or is it more problems with SimplePlot ?
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Sat Jan 21, 2017 7:57 am    Post subject: Reply with quote

John,
Where specifically? I do not notice anything wrong. Can you post what you see using Postimage dot net for example or PM me directly?
Back to top
View user's profile Send private message
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Sun Jan 22, 2017 5:10 am    Post subject: Reply with quote

From the paper ... it may be intntional, but look at those contour scale 'bars' which appear to be running parallel to the edges of the scatter plat 'cubes'

Fig. 5


Fig, 6


Fig. 7
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Sun Jan 22, 2017 2:20 pm    Post subject: Reply with quote

John, I still do not get what you mean...and since I do not know exactly your terminology (here is no contours or scatter cubes, for example, here plotted individual particles, atoms and ions after laser irradiation, each with its color on top plot and little cubes on bottom plots which average these particles inside their little volumes. Specific particles with charge 25 and 26 are marked as red and blue ) can you highlight with your hand the problem exactly on these plots ?

By the way the plotting part for these figures is barely one page of Fortran text. The rest code is just the Clearwin niceties to make all to handle many other choices. All is written taking the OpenGL Examples supplied with the compiler. If Salford/Silverfrost would make 10-20 of them, and made looking nice, there would be few times more compiler users . Graphics becomes more and more important, you literally can not do anything without it in complex 2D and 3D codes generating TBs of data
Back to top
View user's profile Send private message
mecej4



Joined: 31 Oct 2006
Posts: 1885

PostPosted: Sun Jan 22, 2017 7:22 pm    Post subject: Reply with quote

Dan, correct me if I am giving the wrong interpretation below.

John, the plots display a scalar point function, namely, energy per unit volume, \rho E, which is a function of (x, y, z). The value of \rho E is indicated not by distance, but by color.

I agree with you that it is a bit misleading to show a color index bar parallel to one of the three axes. They could have given it in a label box, or outside the plot proper, and not slanted to make it parallel to anything other than the edges of the page or plot frame.

Had they plotted temperature distribution inside the core of a volcano, for example, they could have used a similar plot. In that case, it would be obvious that making the color scale rectangle parallel to anything would be impossible because of the irregular geometry. In the figure that Dan posted, the geometry is quite regular, hence the confusion.
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Sun Jan 22, 2017 8:48 pm    Post subject: Reply with quote

Right, Mecej4, thanks for useful clarification.

John, Here is some other analogy to what you see: just imagine that on the top picture you see how, for example, a fuse in your TV explodes by excessive current. You see its particles and vapors leaving its cylindrical volume and each piece of this vapor has its color depending on speed or temperature (the single wire on the top was cut like a sosige for clarity of seeing what's inside) . Bottom pictures shows 4 such fuses and are plotting the energy density inside exploded wire and its vapors

here are no any specific problems with the Clearwin related to these plots. And there is no meaning in the position, height or size of the color bar column. Color bars are often used with 2D plots, here is just 3D version of it.

The problem i have is that i can easily make tic marks in these color bars but i can not place numbers associated with them by some unknown to me reason (i never studied OpenGL. I love learning everything on examples, exercises and tasks which go after the theoretical topics, basically, i never RTFM Smile. So far I can do that easily on usual 2D XY plots but not using OpenGL. When i call text second time for example it does change the text font size or its position is hard to set etc. Any successful example would help, but no one uses OpenGL with Fortran, hell knows why, this is great thing, and this is the future.


Last edited by DanRRight on Sun Jan 22, 2017 9:46 pm; edited 1 time in total
Back to top
View user's profile Send private message
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Sun Jan 22, 2017 9:41 pm    Post subject: Reply with quote

no I don't think you understood, it's just visual.
On your example plot Dan at top of post there is a 'plane' indicated with 1um in each direction marked.
Your coloour bar in that plot is normal to that plane and to me appears 'vertical'
In the other plots I would have expected to see the colour bars orientated in the same way , but in fact they are slanted in a direction which is parallel with the edge of the scattr points zone (which is not a cube hnce the edge is not normal to the plotted plane.
Don't know if that explains better.
It doesn't really make any difference as the position of those bars is somewhat arbitrary, , except it just doesn't 'look right'.
As the plots were done in Clearwin+ (that's what the initial post implies, although Dan you go on later to talk about OpenGL) I wondered weather there was something amiss, but I don't know how the plots were actually don and how those contour 'bars' created

mecej4 - the colour bars are not parallel to ONE of the 3 axes !!! Look at fig. 5 plot avìbove. there ar 2 bars in different directions !! and neither are normal to the base 'plane' marked. They're drawn parallel to 2 of the 'tapering cube' (of scatter points) edges
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Sun Jan 22, 2017 9:52 pm    Post subject: Reply with quote

Ahhh, ok, well, our eyes just lie to us, John Smile. Do you feel that color bars look a bit "drunk"? That is true, but in reality the color bars here always stay strictly perpendicular to its base as well as do the exploding wires. Looks like flying in space in vicinity of these superhot objects we approached one of their edges way too closely Very Happy

In photography this can be often fixed. I actually wrote a program using Clearwin+ which is doing such "perspective fixing" with JPG etc files, i can send it to anyone. But here with these pictures above there was no time even to think about this
Back to top
View user's profile Send private message
mecej4



Joined: 31 Oct 2006
Posts: 1885

PostPosted: Mon Jan 23, 2017 2:36 am    Post subject: Re: Reply with quote

John-Silver wrote:
mecej4 - the colour bars are not parallel to ONE of the 3 axes !!! Look at fig. 5 plot above. there are 2 bars in different directions !! and neither are normal to the base 'plane' marked. They're drawn parallel to 2 of the 'tapering cube' (of scatter points) edges

John, they are using a perspective (non-orthogonal) projection. In a perspective projection, parallel lines meet at the "vanishing point". See, for example, <https://en.wikipedia.org/wiki/Perspective_(graphical)#/media/File:Railroad-Tracks-Perspective.jpg> . So, when we say that two lines in 3-D space are parallel, when we can only see a 2-D representation, we have to distinguish between "convey a sense of being parallel" and "their 2-D projections are parallel".
Back to top
View user's profile Send private message
John-Silver



Joined: 30 Jul 2013
Posts: 1520
Location: Aerospace Valley

PostPosted: Wed Jan 25, 2017 10:35 am    Post subject: Reply with quote

I thought it must be a question of perspective 'tapering'. I just got confused by talk of it all being done in Clearwin+ , it must have actually be done with specific square-one line-by-line drawing constructs for the bars etc ..... and not 'standardised' options for plotting axes and contour legends
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Wed Jan 25, 2017 11:55 pm    Post subject: Reply with quote

The %og is part of Clearwin+ and OpenGL.
Back to top
View user's profile Send private message
DanRRight



Joined: 10 Mar 2008
Posts: 2813
Location: South Pole, Antarctica

PostPosted: Sat Jan 28, 2017 8:56 pm    Post subject: Reply with quote

Here is what my perspective change program is doing. Top is original picture with falling buildings, bottom has its perspective fixed. Very handy. Makes picturing in the cities look professional by eliminating falling wall houses. Let me know if you need this easy FTN95 or even FTN77, i do not remember, source code which you totally control yourself. No programming needed just drag and drop and clicking.That will instantly save you $30 on some shareware Smile. And yes, this software would easily in seconds fix perspective for the Science Advances' pictures if there was no other way to do that natively.

Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> ClearWin+ All times are GMT + 1 Hour
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group