We are excited to announce the launch of our new forums! You can access it forums.aavso.org. For questions, please see our blog post. The forums at aavso.org/forum have become read-only.
American Association of Variable Star Observers (AAVSO)
Thu, 10/04/2018 - 00:23
I see Binary Maker referenced in many JAAVSO articles and AAVSO forum posts so I know it is widely used. I am curious if folks who use it feel it is worth the price.
Re other options, AFAIK, the W-D code is available for free, but is written in FORTRAN, has a card-image interface and no graphical output. But it has differential corrections.
PHOEBE is, I think the current standard, but doesn't have a GUI frontend, and to see the results I think you graph them using mathplotlib in python. Very powerful, but... Has differential corrections.
BM3 does have a GUI for model parameter entry, and directly produces graphical output. The numbers are available as output too, for graphing yourself, or other analysis. The GUI isn't all you could wish it to be (e.g., no "undo", no way to compare results from current run to previous run), but it is a heck of a lot easier than W-D. Might be easier than PHOEBE too, but because you can drive the latter from python, there is a lot of potential there if you are comfortable doign a bit of coding. (I used W-D a little bit a long time ago, but have never used PHOEBE.)
BM3 does not have "differential corrections", i.e. the mathematics to guide how you should update the parameters to improve the fit. Thus you have to develop an intuition (understanding) of how a parameter change will affect the output.
I am currently using BM3 to prepare a paper for the upcoming AAVSO meeting. I keep telling myself I should try PHOEBE, but BM3 is so straightforward. It also comes with a big manual (on disk). A lot of it is pretty straightforward walk throughs of menus, but there is a lot of meat there too.
This very cool tool updates the model and the system light curves as you move the sliders. VERY useful for getting a sense of what parameters affect the lightcurve, and how. Requires Adobe Flash to be installed and enabled. Free, and worth a whole lot more. Not sufficient for producing a publishable result, but good for developing understanding.
BM3 was pretty easy to install. I did have a bit of a glitch installing it on a linux box (years after I had bought it). Dr. Bradstreet sent me a solution within a day or two. He also very kindly let me share my licensed copy with a high school student I was mentoring through a project in another country.
I have actually never used binary maker, so I can't comment on it specifically. However, my science interests rely heavily on binary modeling. In fact, I am on the development team for the PHOEBE code and I want to clarify a couple things that Gary said. PHOEBE 1 does have a gui. It is in fact a gui overlay of the WD code. There are a lot of options and it will take some getting used to, but it is reasonably approachable. However, I believe Gary was referring to PHOEBE 2.0 which is a complete from scratch rewrite including a lot of new physics. It doesn't "yet" have a gui, at least that is at a releasable stage (and honestly it is still probably a year or two away). If however, you are familiar with python and willing to code a bit it is very powerful. In fact, it is probably the most comprehensive binary simulation program to date, but that flexibility and additional physics comes at a computational cost. If you wanted to do any real fitting you would likely need a cluster. Apologies for the PHOEBE centric post as I am a bit biased. Though since it is free, open source software I have no problems trying to sell it a little bit. I am more than happy to discuss this further if you are interested in trying out PHOEBE. For the new version I can even point you to some tutorials and we are good at responding if you have issues.
It's been quite a while since I've been involved in EB modeling so I was thinking that it might be useful to have an area in the EB section web pages to summarize what software packages are available to do the modeling and their features.
15+ years ago I was compiling the WD code in Fortran and plotting the output with gnuplot. I'm sure things have advanced and improved since then!
Thanks folks for the feedback and I like Tim's suggestion.
Gary, I look forward to your paper.
Bert, PHOEBE sounds interesting, but it looks like a bit of planning is required for setup and use. I don't find the python frontend any problem. I see version 2.1 is in the works. Can you say yet whether that version will be fully tested on python 3.x?
You're not wrong. Learning PHOEBE will definitely will definitely take some time. The question then becomes what exactly are your goals in using it as to whether or not that effort is worth it? As far as I know the 2.1 release will not be python 3.x compliant. It might work but has not been thoroughly tested.
Well, hard to say.
Re other options, AFAIK, the W-D code is available for free, but is written in FORTRAN, has a card-image interface and no graphical output. But it has differential corrections.
PHOEBE is, I think the current standard, but doesn't have a GUI frontend, and to see the results I think you graph them using mathplotlib in python. Very powerful, but... Has differential corrections.
BM3 does have a GUI for model parameter entry, and directly produces graphical output. The numbers are available as output too, for graphing yourself, or other analysis. The GUI isn't all you could wish it to be (e.g., no "undo", no way to compare results from current run to previous run), but it is a heck of a lot easier than W-D. Might be easier than PHOEBE too, but because you can drive the latter from python, there is a lot of potential there if you are comfortable doign a bit of coding. (I used W-D a little bit a long time ago, but have never used PHOEBE.)
BM3 does not have "differential corrections", i.e. the mathematics to guide how you should update the parameters to improve the fit. Thus you have to develop an intuition (understanding) of how a parameter change will affect the output.
I am currently using BM3 to prepare a paper for the upcoming AAVSO meeting. I keep telling myself I should try PHOEBE, but BM3 is so straightforward. It also comes with a big manual (on disk). A lot of it is pretty straightforward walk throughs of menus, but there is a lot of meat there too.
For purely "what if" playing with model parameters, albeit a limited parameter set and only low-res graphical output, I highly recommend: http://astro.unl.edu/naap/ebs/animations/ebs.html
This very cool tool updates the model and the system light curves as you move the sliders. VERY useful for getting a sense of what parameters affect the lightcurve, and how. Requires Adobe Flash to be installed and enabled. Free, and worth a whole lot more. Not sufficient for producing a publishable result, but good for developing understanding.
BM3 was pretty easy to install. I did have a bit of a glitch installing it on a linux box (years after I had bought it). Dr. Bradstreet sent me a solution within a day or two. He also very kindly let me share my licensed copy with a high school student I was mentoring through a project in another country.
I hope all this helps.
Gary Billings
Hey All,
I have actually never used binary maker, so I can't comment on it specifically. However, my science interests rely heavily on binary modeling. In fact, I am on the development team for the PHOEBE code and I want to clarify a couple things that Gary said. PHOEBE 1 does have a gui. It is in fact a gui overlay of the WD code. There are a lot of options and it will take some getting used to, but it is reasonably approachable. However, I believe Gary was referring to PHOEBE 2.0 which is a complete from scratch rewrite including a lot of new physics. It doesn't "yet" have a gui, at least that is at a releasable stage (and honestly it is still probably a year or two away). If however, you are familiar with python and willing to code a bit it is very powerful. In fact, it is probably the most comprehensive binary simulation program to date, but that flexibility and additional physics comes at a computational cost. If you wanted to do any real fitting you would likely need a cluster. Apologies for the PHOEBE centric post as I am a bit biased. Though since it is free, open source software I have no problems trying to sell it a little bit. I am more than happy to discuss this further if you are interested in trying out PHOEBE. For the new version I can even point you to some tutorials and we are good at responding if you have issues.
Thanks,
Bert Pablo
Staff Astronomer, AAVSO
how can ] use phoebe for plot the Roche lobe
Bert, thank you for the authoritative corrections re PHOEBE...
It's been quite a while since I've been involved in EB modeling so I was thinking that it might be useful to have an area in the EB section web pages to summarize what software packages are available to do the modeling and their features.
15+ years ago I was compiling the WD code in Fortran and plotting the output with gnuplot. I'm sure things have advanced and improved since then!
.....Tim
Thanks folks for the feedback and I like Tim's suggestion.
Gary, I look forward to your paper.
Bert, PHOEBE sounds interesting, but it looks like a bit of planning is required for setup and use. I don't find the python frontend any problem. I see version 2.1 is in the works. Can you say yet whether that version will be fully tested on python 3.x?
Clear skies,
Cliff
Hey Cliff,
You're not wrong. Learning PHOEBE will definitely will definitely take some time. The question then becomes what exactly are your goals in using it as to whether or not that effort is worth it? As far as I know the 2.1 release will not be python 3.x compliant. It might work but has not been thoroughly tested.
Thanks,
Bert Pablo