Hamidreza

191 Reputation

7 Badges

15 years, 193 days

MaplePrimes Activity


These are replies submitted by Hamidreza

Thanks. May I know what Windows you are using?

To upgrade to 64bits, I installed the version labeled as "Upgrade Only", which means no uninstall, just upgrade (Am I right?). After upgrading, I can see two folders in the installation directory, "bin.win" and "bin.X86_64_WINDOWS", which both of them contain maple executaion file, but the shortcut is refering to the one in the former, i,e, "bin.win". I don't know which one is the 64bits one, but when I run the one in "bin.X86_64_WINDOWS", it gives an error regarding fail in starting mserver. Please let me know how to handle this.

Thank you!

To upgrade to 64bits, I installed the version labeled as "Upgrade Only", which means no uninstall, just upgrade (Am I right?). After upgrading, I can see two folders in the installation directory, "bin.win" and "bin.X86_64_WINDOWS", which both of them contain maple executaion file, but the shortcut is refering to the one in the former, i,e, "bin.win". I don't know which one is the 64bits one, but when I run the one in "bin.X86_64_WINDOWS", it gives an error regarding fail in starting mserver. Please let me know how to handle this.

Thank you!

I am using windows 7 64 bits, and I upgraded my maple to 64 bits version as well. But my wonder is although I upgraded it to 64 bits, in the task manager of windows it is still tagged as a 32 bits application (e.g. maple.exe *32  -  mserver.exe *32). Then if what is shown is true, the memory allocation issue comes from that. Right? Does it have anything to do with Java version? I mean, do I need to install Java 64bits along with maple 64 bits as well?

Thanx!

I am using windows 7 64 bits, and I upgraded my maple to 64 bits version as well. But my wonder is although I upgraded it to 64 bits, in the task manager of windows it is still tagged as a 32 bits application (e.g. maple.exe *32  -  mserver.exe *32). Then if what is shown is true, the memory allocation issue comes from that. Right? Does it have anything to do with Java version? I mean, do I need to install Java 64bits along with maple 64 bits as well?

Thanx!

Thank you! Your second example guided me to search for the same idea (color=...) in plot3d, too. In fact, I switched from plot3d to contourplot3d because the default coloring option of plot3d is set to be "x", and I changed it to my function, i.e. color=f(x,y). Therefore it works as I want to and even more, as I have much more control on the color of the contour. The only conceptual problem with this, which I think should be considered in maple engine, is that it starts with red and ends with red, too, which makes it a little confusing. Because in most engineering apps, the color corresponding to the lowest values stops at dark blue, like as shown in the sample of the link in my first post.

Thank you!

Thank you! Your second example guided me to search for the same idea (color=...) in plot3d, too. In fact, I switched from plot3d to contourplot3d because the default coloring option of plot3d is set to be "x", and I changed it to my function, i.e. color=f(x,y). Therefore it works as I want to and even more, as I have much more control on the color of the contour. The only conceptual problem with this, which I think should be considered in maple engine, is that it starts with red and ends with red, too, which makes it a little confusing. Because in most engineering apps, the color corresponding to the lowest values stops at dark blue, like as shown in the sample of the link in my first post.

Thank you!

Thank you  hirnyk!

Since the trend of my function is unknown to me, unfortunately the first approach is not applicable to my case. But the second answer was helpful. However, still I do not know how I can change the font-size of axes without replotting it, I mean using the option font=[TIMES, ROMAN, 16] for example. Because each function needs in some cases 5 mins to be plotted, and I want to change the font size based on my convenience, so I have to change it by trial and error and see what looks good. So I guess it should be another way of handling this.

Thank you@!

Thank you  hirnyk!

Since the trend of my function is unknown to me, unfortunately the first approach is not applicable to my case. But the second answer was helpful. However, still I do not know how I can change the font-size of axes without replotting it, I mean using the option font=[TIMES, ROMAN, 16] for example. Because each function needs in some cases 5 mins to be plotted, and I want to change the font size based on my convenience, so I have to change it by trial and error and see what looks good. So I guess it should be another way of handling this.

Thank you@!

Thank you acer! Your example helped me solve the problem!

Thank you acer! Your example helped me solve the problem!

Thank you all for the idea. I got the cause. The problem was where I didn't expect to be there. In fact, my final function is:

f_total:=simplify(evalf(add(add(f[i, j](i, j, x, y), i = 1 .. 50), j = 1 .. 100)))

Apparently, "simplify" made the form of the total function even more complicated. Now it calculates it n less than a minute!! Still I can not believe "simplify" caused the problem.


Thank you so much!

Thank you all for the idea. I got the cause. The problem was where I didn't expect to be there. In fact, my final function is:

f_total:=simplify(evalf(add(add(f[i, j](i, j, x, y), i = 1 .. 50), j = 1 .. 100)))

Apparently, "simplify" made the form of the total function even more complicated. Now it calculates it n less than a minute!! Still I can not believe "simplify" caused the problem.


Thank you so much!

My functions (array of functions) are defined like:

 

f[i,j](x,y)=-D1()*(y-i*line_sp)*(3*(x-j*dis_sp)^2+(y-i*line_sp)^2)/((x-j*dis_sp)^2+(y-i*line_sp)^2)^2

 

and what I am trying to evaluate is:

evalf(add(add(f[i, j](i, j, x, y), i = 1 .. 50), j = 1 .. 100))

And I am plotting the summation by plot3d and takes about 5 hours; But, when I decrease intervals and evaluate this summation, e.g.:

evalf(add(add(f[i, j](i, j, x, y), i = 1 .. 50), j = 1 .. 5))

plot3d plots it in 2':30''. But the time increases exponentially as the interval of i and j increases, and for 100  intervals, it takes more than 5 hours. So I think it makes sense if I calculate each function values numerically, and add them up. the number of subfuncs are 50*100=5000 against 5*50=250. meaning by summing them up numerically, it should take 5000/250*(2':30'') = 50', not 5 hours!

If there is no alternative, then SurfacePlot seems to be the best way. But still I don't have any idea about why it is taking much longer time strangely. Am I mistaken somewhere?

Thank you Preben!

1 2 3 Page 1 of 3