gnuplot development

if you are plotting data just add the offset to all x-values, e.g.
plot "..." using ($1+offset):2


hxk257 wrote:
> 
> Hi,
> 
> Is there a way to adjust the offset of the numbers labeling an axis?  I'm
> not talking about the label for the axis, e.g., "Age of subjects".  I'm
> talking about the actual numbers indicating the variable, e.g., age. 
> Thanks,
> 

-- 
View this message in context: http://old.nabble.com/range-offset-tp30321712p30325142.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Dear friend,

You have to give actual file name "marathon.txt"

Vinod


saurav_bioinfo wrote:
> 
> Dear all,
> 
> I am very new to both linux and gnuplot. some how from different forums, i
> have installed gnuplot 4.4.2 on my ubuntu system. Now whenever i write
> gnuplot on my terminal one interative gnuplot terminal opens. As an
> tutorial i have started with the gnuplot in action documentation. But when
> i tried the very first example of it, i failed to do so. i wrote (plot
> "marathon" using 1:2 with boxes) on gnuplot terminal after creating a data
> txt file named marathon.txt.
> 
> It does not show any error too but along with that i could not file any
> plot file. 
> please help.
> 
> Thank you in advance.
> 
> Saurav
> 

-- 
View this message in context: http://old.nabble.com/new-bie-to-linux-and-gnuplot-tp30276130p30278517.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Hi,

Did you specify an output device?  If you want to write out a file of your
plot, you can specify a PostScript output file with the command:

gnuplot > set terminal postscript landscape enhanced mono dashed lw 1
"Helvetica" 14
gnuplot > set output "nameoffile.ps"
gnuplot > plot

You should see nameoffile.ps in the local directory.  Good luck,



saurav_bioinfo wrote:
> 
> Dear all,
> 
> I am very new to both linux and gnuplot. some how from different forums, i
> have installed gnuplot 4.4.2 on my ubuntu system. Now whenever i write
> gnuplot on my terminal one interative gnuplot terminal opens. As an
> tutorial i have started with the gnuplot in action documentation. But when
> i tried the very first example of it, i failed to do so. i wrote (plot
> "marathon" using 1:2 with boxes) on gnuplot terminal after creating a data
> txt file named marathon.txt.
> 
> It does not show any error too but along with that i could not file any
> plot file. 
> please help.
> 
> Thank you in advance.
> 
> Saurav
> 

-- 
View this message in context: http://old.nabble.com/new-bie-to-linux-and-gnuplot-tp30276130p30321698.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

try this:

y1=y2=y3=0
plot './eva.dat1' using 1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC")
?(y1=$2+$3,y1) : y1) with linespoints title "ARRAY01/B/1",''using
1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(y2=$2+$3,y2) : y2) with
linespoints title "ARRAY01/B/2",'' using
1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(y3=$2+$3,y3) : y3) with
linespoints title "ARRAY01/B/3"

explanation:

stringcolumn(.) == "string"
doesn't work, but
strstr(stringcolumn(.),"string")
does.

using 'NaN' sets the datapoint to 'undefined' and this interrupts
the plot - this way you don't get lines between the points.
so it's better to just plot the last valid point again.

(and omit the 'multiplot', you don't need it for a single plot.)


mbm329 wrote:
> 
> Hello,
> 
> I'm trying to plot read and write requests from a storage array.  I
> have an application that dumps data with each array controller's read
> and write requests on a separate line.  I'd rather not mangle the data
> from the original application's dump if possible.  Within gnuplot, I'm
> trying to conditionally say, "If the 4th column is <predetermined
> name>, then plot the sum of columns 2 and 3."  It's not working out
> for me, so far.  Instead of three lines, I get a single line with
> multiple points on the same y axis.  Below is my config and data file.
>  Is this even possible with gnuplot?  Any help is certainly
> appreciated.
> 
> Thanks,
> 
> Mike
> 
> ############################
> A sample of the gnuplot config that is not working properly:
> 
> set title "Array I/O"
> set terminal png size 1024,768
> set output "/var/www/html/eva.png"
> set key below reverse height 1 width -1 box
> set grid x y
> set datafile separator ","
> set multiplot
>    set origin 0.0,0.0
>    set size 1.0,1.0
>    set xlabel "Time (MM/DD:HH:MM:SS)"
>       set xdata time
>       set xtics rotate by -70
>       set timefmt "%d/%b/%Y %H:%M:%S"
>       set format x "%m/%d:%H:%M:%S"
>       set xrange [ "5/Nov/2010 00:56:29" : "5/Nov/2010 00:57:25" ]
>    set ylabel "Req/sec"
>       set format y "%10.0f"
>       set yrange [ 0 : 5000 ]
>    plot\
>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDC" ?
> ($2+$3) : NaN) with linespoints title "ARRAY01/B/1",\
>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDD" ?
> ($2+$3) : NaN) with linespoints title "ARRAY01/B/2",\
>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDE" ?
> ($2+$3) : NaN) with linespoints title "ARRAY01/B/3"
> set nomultiplot
> 
> 
> ############################
> A sample of the data file:
> 
> 5/Nov/2010 00:56:30,3043,71,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:56:30,1043,71,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:56:30,2023,71,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:56:34,3371,61,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:56:34,3173,60,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:56:34,1139,42,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:56:44,2905,110,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:56:44,3074,88,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:56:44,1352,76,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:56:54,3355,58,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:56:54,2932,170,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:56:54,2015,38,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:57:04,3115,120,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:57:04,3197,85,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:57:04,2546,92,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:57:14,1163,121,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:57:14,1408,51,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:57:14,3521,33,5000-1FE1-5000-BDDE
> 5/Nov/2010 00:57:24,1412,58,5000-1FE1-5000-BDDC
> 5/Nov/2010 00:57:24,2612,85,5000-1FE1-5000-BDDD
> 5/Nov/2010 00:57:24,1589,52,5000-1FE1-5000-BDDE
> 
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> gnuplot-info mailing list
> gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 

-- 
View this message in context: http://old.nabble.com/Plotting-row-oriented-data-tp30282816p30283789.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Thank you for the reply, Thomas.

Unfortunately, it still did not plot properly.  However, it is much
closer than before; at least I get three lines now :)

With the change you suggested, I have 2 issues.
The first item, "5000-1FE1-5000-BDDC", plots correctly all the way
through, (1) but the second and third lines (strings ending with D and
E) plot starting at 0.

Example:
http://imagebin.ca/view/f19vOIc.html

I fixed this by setting y1=y2=y3=NaN.  So that gets it starting at the
right values.

(2) The rest of those two lines are plotting multiple values on the y
axis still.  I added a "set table 'table.out' " to the config and I
can see what's going on, but still unsure of how to fix it.

As you can see from the table output below, when the next timestamp (X
value) is approached, the second and third items are setting their Y
values on it making a second plot on the same timestamp.

Mike

#################################
Here's the table.out to assist:

# Curve 0 of 3, 21 points
# Curve title: "ARRAY01/B/1"
# x y type
"11/05:00:56:30"       3114  i
"11/05:00:56:30"       3114  i
"11/05:00:56:30"       3114  i
"11/05:00:56:34"       3432  i
"11/05:00:56:34"       3432  i
"11/05:00:56:34"       3432  i
"11/05:00:56:44"       3015  i
"11/05:00:56:44"       3015  i
"11/05:00:56:44"       3015  i
"11/05:00:56:54"       3413  i
"11/05:00:56:54"       3413  i
"11/05:00:56:54"       3413  i
"11/05:00:57:04"       3235  i
"11/05:00:57:04"       3235  i
"11/05:00:57:04"       3235  i
"11/05:00:57:14"       1284  i
"11/05:00:57:14"       1284  i
"11/05:00:57:14"       1284  i
"11/05:00:57:24"       1470  i
"11/05:00:57:24"       1470  i
"11/05:00:57:24"       1470  i


# Curve 1 of 3, 21 points
# Curve title: "ARRAY01/B/2"
# x y type
"11/05:00:56:30"          0  u
"11/05:00:56:30"       1114  i
"11/05:00:56:30"       1114  i
"11/05:00:56:34"       1114  i
"11/05:00:56:34"       3233  i
"11/05:00:56:34"       3233  i
"11/05:00:56:44"       3233  i
"11/05:00:56:44"       3162  i
"11/05:00:56:44"       3162  i
"11/05:00:56:54"       3162  i
"11/05:00:56:54"       3102  i
"11/05:00:56:54"       3102  i
"11/05:00:57:04"       3102  i
"11/05:00:57:04"       3282  i
"11/05:00:57:04"       3282  i
"11/05:00:57:14"       3282  i
"11/05:00:57:14"       1459  i
"11/05:00:57:14"       1459  i
"11/05:00:57:24"       1459  i
"11/05:00:57:24"       2697  i
"11/05:00:57:24"       2697  i


# Curve 2 of 3, 21 points
# Curve title: "ARRAY01/B/3"
# x y type
"11/05:00:56:30"          0  u
"11/05:00:56:30"          0  u
"11/05:00:56:30"       2094  i
"11/05:00:56:34"       2094  i
"11/05:00:56:34"       2094  i
"11/05:00:56:34"       1181  i
"11/05:00:56:44"       1181  i
"11/05:00:56:44"       1181  i
"11/05:00:56:44"       1428  i
"11/05:00:56:54"       1428  i
"11/05:00:56:54"       1428  i
"11/05:00:56:54"       2053  i
"11/05:00:57:04"       2053  i
"11/05:00:57:04"       2053  i
"11/05:00:57:04"       2638  i
"11/05:00:57:14"       2638  i
"11/05:00:57:14"       2638  i
"11/05:00:57:14"       3554  i
"11/05:00:57:24"       3554  i
"11/05:00:57:24"       3554  i
"11/05:00:57:24"       1641  i




On Tue, Nov 23, 2010 at 5:35 AM, Thomas Sefzick <t.sefzick@...> wrote:
>
> try this:
>
> y1=y2=y3=0
> plot './eva.dat1' using 1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC")
> ?(y1=$2+$3,y1) : y1) with linespoints title "ARRAY01/B/1",''using
> 1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(y2=$2+$3,y2) : y2) with
> linespoints title "ARRAY01/B/2",'' using
> 1:(strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(y3=$2+$3,y3) : y3) with
> linespoints title "ARRAY01/B/3"
>
> explanation:
>
> stringcolumn(.) == "string"
> doesn't work, but
> strstr(stringcolumn(.),"string")
> does.
>
> using 'NaN' sets the datapoint to 'undefined' and this interrupts
> the plot - this way you don't get lines between the points.
> so it's better to just plot the last valid point again.
>
> (and omit the 'multiplot', you don't need it for a single plot.)
>
>
> mbm329 wrote:
>>
>> Hello,
>>
>> I'm trying to plot read and write requests from a storage array.  I
>> have an application that dumps data with each array controller's read
>> and write requests on a separate line.  I'd rather not mangle the data
>> from the original application's dump if possible.  Within gnuplot, I'm
>> trying to conditionally say, "If the 4th column is <predetermined
>> name>, then plot the sum of columns 2 and 3."  It's not working out
>> for me, so far.  Instead of three lines, I get a single line with
>> multiple points on the same y axis.  Below is my config and data file.
>>  Is this even possible with gnuplot?  Any help is certainly
>> appreciated.
>>
>> Thanks,
>>
>> Mike
>>
>> ############################
>> A sample of the gnuplot config that is not working properly:
>>
>> set title "Array I/O"
>> set terminal png size 1024,768
>> set output "/var/www/html/eva.png"
>> set key below reverse height 1 width -1 box
>> set grid x y
>> set datafile separator ","
>> set multiplot
>>    set origin 0.0,0.0
>>    set size 1.0,1.0
>>    set xlabel "Time (MM/DD:HH:MM:SS)"
>>       set xdata time
>>       set xtics rotate by -70
>>       set timefmt "%d/%b/%Y %H:%M:%S"
>>       set format x "%m/%d:%H:%M:%S"
>>       set xrange [ "5/Nov/2010 00:56:29" : "5/Nov/2010 00:57:25" ]
>>    set ylabel "Req/sec"
>>       set format y "%10.0f"
>>       set yrange [ 0 : 5000 ]
>>    plot\
>>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDC" ?
>> ($2+$3) : NaN) with linespoints title "ARRAY01/B/1",\
>>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDD" ?
>> ($2+$3) : NaN) with linespoints title "ARRAY01/B/2",\
>>       './eva.dat1' using 1:(stringcolumn(4) == "5000-1FE1-5000-BDDE" ?
>> ($2+$3) : NaN) with linespoints title "ARRAY01/B/3"
>> set nomultiplot
>>
>>
>> ############################
>> A sample of the data file:
>>
>> 5/Nov/2010 00:56:30,3043,71,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:56:30,1043,71,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:56:30,2023,71,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:56:34,3371,61,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:56:34,3173,60,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:56:34,1139,42,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:56:44,2905,110,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:56:44,3074,88,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:56:44,1352,76,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:56:54,3355,58,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:56:54,2932,170,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:56:54,2015,38,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:57:04,3115,120,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:57:04,3197,85,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:57:04,2546,92,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:57:14,1163,121,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:57:14,1408,51,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:57:14,3521,33,5000-1FE1-5000-BDDE
>> 5/Nov/2010 00:57:24,1412,58,5000-1FE1-5000-BDDC
>> 5/Nov/2010 00:57:24,2612,85,5000-1FE1-5000-BDDD
>> 5/Nov/2010 00:57:24,1589,52,5000-1FE1-5000-BDDE
>>
>> ------------------------------------------------------------------------------
>> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
>> Tap into the largest installed PC base & get more eyes on your game by
>> optimizing for Intel(R) Graphics Technology. Get started today with the
>> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
>> http://p.sf.net/sfu/intelisp-dev2dev
>> _______________________________________________
>> gnuplot-info mailing list
>> gnuplot-info@...
>> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>>
>>
>
> --
> View this message in context: http://old.nabble.com/Plotting-row-oriented-data-tp30282816p30283789.html
> Sent from the Gnuplot - User mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> gnuplot-info mailing list
> gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

mbm329 <mbm329 <at> gmail.com> writes:

> Unfortunately, it still did not plot properly.  However, it is much
> closer than before; at least I get three lines now :)
> 
> With the change you suggested, I have 2 issues.
> The first item, "5000-1FE1-5000-BDDC", plots correctly all the way
> through, (1) but the second and third lines (strings ending with D and
> E) plot starting at 0.
> ...

if forgot to store the x-value too.
this plot command should work:

x1=x2=x3=NaN
y1=y2=y3=NaN
plot './eva.dat1' using \
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC") ?(x1=timecolumn(1),x1) : x1):\
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC") ?(y1=$2+$3,y1) : y1) \
 with linespoints title "ARRAY01/B/1", \
'' using \
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(x2=timecolumn(1),x2) : x2):\
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(y2=$2+$3,y2) : y2) \
 with linespoints title "ARRAY01/B/2", \
'' using \
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(x3=timecolumn(1),x3) : x3):\
 (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(y3=$2+$3,y3) : y3) \
 with linespoints title "ARRAY01/B/3"

Thanks much.  That solved the issue for me.

Mike


On Tue, Nov 23, 2010 at 5:11 PM, Thomas Sefzick <t.sefzick@...> wrote:
> mbm329 <mbm329 <at> gmail.com> writes:
>
>> Unfortunately, it still did not plot properly.  However, it is much
>> closer than before; at least I get three lines now :)
>>
>> With the change you suggested, I have 2 issues.
>> The first item, "5000-1FE1-5000-BDDC", plots correctly all the way
>> through, (1) but the second and third lines (strings ending with D and
>> E) plot starting at 0.
>> ...
>
> if forgot to store the x-value too.
> this plot command should work:
>
> x1=x2=x3=NaN
> y1=y2=y3=NaN
> plot './eva.dat1' using \
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC") ?(x1=timecolumn(1),x1) : x1):\
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDC") ?(y1=$2+$3,y1) : y1) \
>  with linespoints title "ARRAY01/B/1", \
> '' using \
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(x2=timecolumn(1),x2) : x2):\
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDD") ?(y2=$2+$3,y2) : y2) \
>  with linespoints title "ARRAY01/B/2", \
> '' using \
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(x3=timecolumn(1),x3) : x3):\
>  (strstrt(stringcolumn(4),"5000-1FE1-5000-BDDE") ?(y3=$2+$3,y3) : y3) \
>  with linespoints title "ARRAY01/B/3"
>
>
>
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> gnuplot-info mailing list
> gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

removing the unwanted points or replacing them by the average of the
y-values of the adjacent points is the same, at least for discrete gaps.

you could store the last valid point and plot it instead of the unwanted
one.

xlast=ylast=NaN
p 'aa' u (($2>5)?(xlast=$1,xlast):xlast):(($2>5)?(ylast=$2,ylast):ylast) w l


bsmile wrote:
> 
> Dear members,
> 
> I am trying to skip data which is beyond the valid range. The usual way
> like
> 
> 
> p 'aa' u 1:(($2>5)?$2:1/0) w l
> 
> leaves in the figure broken lines, which does not look very good.
> 
> I am wondering whether I can do the following two ways
> 
> (1) remove these points on the x axis?
> (2) reset the value of these points as average of the nearby two
> points(one
> before it and one after it)?
> 
> Thanks,
> 
> Sincerely,
> Jon
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> gnuplot-info mailing list
> gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> 
> 

-- 
View this message in context: http://old.nabble.com/ignore-invalid-data-from-an-input-dataset-tp30336749p30337764.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Clever trick! Many thanks!

Jon


On Tue, Nov 30, 2010 at 2:30 AM, Thomas Sefzick <t.sefzick@...:

>
> removing the unwanted points or replacing them by the average of the
> y-values of the adjacent points is the same, at least for discrete gaps.
>
> you could store the last valid point and plot it instead of the unwanted
> one.
>
> xlast=ylast=NaN
> p 'aa' u (($2>5)?(xlast=$1,xlast):xlast):(($2>5)?(ylast=$2,ylast):ylast) w
> l
>
>
> bsmile wrote:
> >
> > Dear members,
> >
> > I am trying to skip data which is beyond the valid range. The usual way
> > like
> >
> >
> > p 'aa' u 1:(($2>5)?$2:1/0) w l
> >
> > leaves in the figure broken lines, which does not look very good.
> >
> > I am wondering whether I can do the following two ways
> >
> > (1) remove these points on the x axis?
> > (2) reset the value of these points as average of the nearby two
> > points(one
> > before it and one after it)?
> >
> > Thanks,
> >
> > Sincerely,
> > Jon
> >
> ------------------------------------------------------------------------------
> > Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> > Tap into the largest installed PC base & get more eyes on your game by
> > optimizing for Intel(R) Graphics Technology. Get started today with the
> > Intel(R) Software Partner Program. Five $500 cash prizes are up for
> grabs.
> > http://p.sf.net/sfu/intelisp-dev2dev
> > _______________________________________________
> > gnuplot-info mailing list
> > gnuplot-info@...
> > https://lists.sourceforge.net/lists/listinfo/gnuplot-info
> >
> >
>
> --
> View this message in context:
> http://old.nabble.com/ignore-invalid-data-from-an-input-dataset-tp30336749p30337764.html
> Sent from the Gnuplot - User mailing list archive at Nabble.com.
>
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> gnuplot-info mailing list
> gnuplot-info@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

On 23.11.2010 20:09, Dr. Ing. Dieter Jurzitza wrote:

> this works without any problem with linux, however I am facing a strange issue
> on the windows platform. The placement of the arrow - nice, as expected, but:
> the arrow tip is missing. In contrast, the tip is cut so that there is a
> straight end of the arrow the width of which is about the size of the line
> thickness.

Platforms rather certainly don't have any relevance to this.  This is 
pretty much guaranteed to be due to either a version difference or a 
configuration difference.

Configuration difference can consist of commands in your ~/.gnuplot or 
its MS Windows equivalent, gnuplot.ini.  See "help startup".   Or maybe 
you modified your PostScript template files on only one of the two 
machines (--> "help postscript prologue"), and don't remember doing it.

Dear Mr. Broeker,
dear listmemebers,
knowing that this problem persists across different gnuplot releases on M$ and 
knowing that there is no gnuplot.ini - file neither on M$ nor on linux I tried 
to replace the directory PostScript and it's contents on the M$-tree by the 
same directory from the linux tree.

This fixes the issue. So I am sure that there is a buglett in the postscript 
configuration on M$ that causes arrows to be printed without a real peak on M$ 
platforms (see the example I provided).

Lacking the capability to interpret postscript code I cannot fix this, maybe 
someone with more insight can do this in my place. This is not the fault of a 
misconfiguration but an erratic initial configuration of gnuplot.

I never ever touched this directory before. The opensuse source rpm does not 
contain any patch touching the files going into the PostScript directory as 
far as I can see.

Thanks for pointing me into the right direction, I found my solution, however, 
this solution is not generic.

Take care




Dieter

************
> you modified your PostScript template files on only one of the two
> machines (--> "help postscript prologue"), and don't remember doing it.
************

-- 
-----------------------------------------------------------

                               |
                                \
                 /\_/\           |
                | ~x~ |/-----\   /
                 \   /-       \_/
  ^^__   _        /  _  ____   /
 <°°__ \- \_/     |  |/    |  |
  ||  ||         _| _|    _| _|

if you really want to see the pictures above - use some font
with constant spacing like courier! :-)
-----------------------------------------------------------

On 24.11.2010 19:53, Dr. Ing. Dieter Jurzitza wrote:

> This fixes the issue. So I am sure that there is a buglett in the postscript
> configuration on M$ that causes arrows to be printed without a real peak on M$
> platforms (see the example I provided).

If anyone's to have a chance of investigating that, we'll need details: 
precise versions of gnuplot packages used on that Linux and that M$ box, 
primarily.  If those are binary packages you got from somewhere other 
than our project download site on SourceForge, we'll need the actual URLs.

> Lacking the capability to interpret postscript code I cannot fix this,

If you send us (or post to the bug tracker on SourceForge) sample output 
or, failing that, the original PostScript prologues on those machines, 
someone will be able to do that.