Re: [Plplot-devel] Two new tests (was Re: time)

2009-02-08 Thread trc
Hi, Alan W. Irwin wrote: > On 2009-02-08 21:46+0100 Werner Smekal wrote: > >> Hi Alan, >>> I confirm that. I found identical qsastime_test results and identical x29 >>> PostScript driver results for revision 9458 compared to previous on my >>> Linux >>> (Debian testing) platform. >>> >> I would

Re: [Plplot-devel] time

2009-02-08 Thread Jerry
On Feb 5, 2009, at 1:50 PM, Alan W. Irwin wrote: > Terrence's report of success on his particular Windows platform for > what was > effectively revision 9458 is most encouraging. Hopefully, the > requested > further testing on all the different Windows and Mac OS X platforms > accessible to o

Re: [Plplot-devel] time

2009-02-08 Thread trc
Hi Alan, > On 2009-02-04 22:27-0800 Alan W. Irwin wrote: > In sum, it is looking extremely encouraging as of revision 9472, but there > is still more testing to do, and I also have some additional development > plans to add some features to libqsastime (see README.qsastime_API) before I > fin

Re: [Plplot-devel] Two new tests (was Re: time)

2009-02-08 Thread Alan W. Irwin
On 2009-02-08 21:46+0100 Werner Smekal wrote: > Hi Alan, >> I confirm that. I found identical qsastime_test results and identical x29 >> PostScript driver results for revision 9458 compared to previous on my >> Linux >> (Debian testing) platform. >> > I would propose a new test: if a change is d

[Plplot-devel] Two new tests (was Re: time)

2009-02-08 Thread Werner Smekal
Hi Alan, > I confirm that. I found identical qsastime_test results and identical x29 > PostScript driver results for revision 9458 compared to previous on my Linux > (Debian testing) platform. > I would propose a new test: if a change is done in the PLplot core and output of examples changes as

Re: [Plplot-devel] time

2009-02-08 Thread Werner Smekal
Hi Alan, > Speaking of cross-platform, it would be good if OS X and Windows developers > did an ordinary PLplot build to make sure that libqsastime gets built > automatically and without build errors as part of that process. If no build > errors occur for those platforms, further testing of examp

[Plplot-devel] New PLplot features and nifty examples (was Re: using plplot for interactive financial graphs?)

2009-02-08 Thread Werner Smekal
Hi all, > Since we've now had two requests for this, shall I go ahead and commit > this example to svn? At least we show how plplot can be used for > financial plotting. If someone comes up with a new API routine for > handling this then we can always update the example to use it. > I'm actuall

Re: [Plplot-devel] using plplot for interactive financial graphs?

2009-02-08 Thread Werner Smekal
Hi Vadim, > Thanks for your answer and advice! I've started looking into actually > doing something like this and realized that there is actually a third > potential problem: is it possible to provide the plot data on demand with > PLplot? E.g. with these financial plots you may have data for som

Re: [Plplot-devel] patch for building svn plplot with svn wxWidgets

2009-02-08 Thread Werner Smekal
Hi Vadim, > I've just started looking in using plplot with wxWidgets backend and so > it's perfectly possible that I'm doing something wrong but it didn't build > for me out of the box because of a couple of missing headers. To be more > precise, here is what I did: > > 0. Checked out latest (r94

Re: [Plplot-devel] using plplot for interactive financial graphs?

2009-02-08 Thread Werner Smekal
Hi Vadim, nice to have one of the main wxWidgets developers around here ;). I have been away this last week, so I only could answer today. > I'm looking into the possibility of using plplot for a project I'm working > on and would like to ask for your collective wisdom about whether it's > goin

Re: [Plplot-devel] time

2009-02-08 Thread Alan W. Irwin
On 2009-02-07 16:47-0800 Alan W. Irwin wrote: > [...]I plan > to run one [test of setfromMJD versus timegm] with much finer spacing. I did such a test with 10 million calls and it turns out setfromMJD is roughly 4 times faster than timegm. That is, on my 2.4GHz box, the time per call is 200 ns f