Hey guys, just in response to my post last night. I have an equation from which I can analytically determine the value of a constant of integration. To confirm this I use some numerical methods to get a list of values for the constant of integration. So I have a huge table of values that are almost exactly 5/9. Now I want to plot my constant C-5/9 and show that it is almost zero everywhere but it is very rough and jagged. I think it is because the numbers are so close together. Any suggestions?

@Rumplestillskin, PlotRange->Allsometimes helps with that. Or you can set it explicitly.

@KraZug Great. I will give it a try

@KraZug Any other suggestions? I tried using MaxRecursion, and PlotPoints but the results weren't smooth in any way. If anything it made the result look botched!

No, if you data is truly all almost exactly the same, MaxRecursion or PlotPoints are not going to help. Your issue is there is no signal, only noise, and the default plot will try and find some signal

MaxRecursion and PlotPoints just increase the effort Mathematica goes to in order to try and make the plot look smooth.

Hmm I tried this but it doesn't seem to work it gives me a graph like this still

Guys one last question... I have another exponential term where the argument is inversely proportional to a large number and mathematica just assumes its value is 1. Is there any way around this?

@halirutan I'm flattered and also very far from having a tag gold badge. Interesting policy, personally I'm unsure about the principle behind.

@Rumplestillskin, yes, it is showing that your residual is fluctuating around 0. Put PlotRange->{-0.001,0.001} and it'll be just a straight line.

@rhermans Might make more sense if you think about Stackoverflow. On Stackoverflow, one tag might be Python, another Scala, yet another C. Gold badges are easy to get. It's as if one gold badge on here was "wolfram language" – a lot of people would have one. If you give all those users the right to instantly close duplicates and not having to have four others users agree then the review queue will probably rather dramatically decrease in size.

On these websites where we only have one language, I wonder if it might not be better to give it as a privilege for a certain amount of reputation instead.

@rhermans Ha, I see now that this doesn't apply to us. I don't care for badges either but I checked now. It seems we don't have badges for tags. I wasn't aware of this.

@KraZug @KraZug That's very helpful thank you!

@Rumplestillskin, np.

@halirutan Do you mean you personally or this community? Some have gold badges for tags, including m_goldberg.

@C.E. You are right. I must have been blind.

@C.E. Yes, the most popular Gold Tag Badge is for plotting, awarded 10 times.

All people with gold tag badges have 40K+ reputation.

Are our review queues considered to be too long? I doubt this policy it's justified for Mma.SE. I'm sure all our 40K+ and Gold Tag Badge holder users are trustworthy and with noble intentions, but the value of consensus goes beyond personal trust, is a mechanism to minimize (apparent) subjectivity to average out outliers to favour a balanced attitude. I would have preferred better apparent fairness even at the cost of performance.

@rhermans - that's why I tend not to make closed votes on plotting questions any more, since I know it will be closed right away without consensus. Although earlier today I did close one as a dupe because it was obvious.

@rhermans yes, I agree with this.

@JasonB. The incentives are all wrong if we have sensible high reputation people like you self censoring and medium reputation left with a less meaningful vote and most users potentially feeling that their question may have been closed on a single person whim.

Probably the self censoring part could be solved if golden tag users have the choice to vote with standard or extraordinary weight.

Can moderators vote with standard weight if they choose to?

@rhermans No, I believe I've seen both Mr. Wizard and J.M. say at one time or another that they have the same concern as Jason B.

@JasonB. have I asked before if you're going to support common chemical IO formats in your paclet? I currently hook into OpenBabel for things like export to InChI or SMI, but I do this in a roundabout way via RunProcess which is a little bit slow...

It'd be nice to have direct export.

I've got InChI, SMILES, and a couple of file formats like MOL and SDF

The SDF importer will be very much faster - I don't know if you've ever tried to import one of these >1GB SDF files they sometimes put with datasets, but Mathematica doesn't do it very well right now

But for a lot of formats, like Gaussian output files, NWChem or molpro output files (you can tell I'm mainly a quantum chemistry person) I'd like there to be a community solution

Have looked at other formats, ones that aren't 30 years old, like CML or chemical JSON, but don't see a ton of adoption elsewhere

@JasonB. these are the big ones I was interested in. All the Gaussian stuff can be pretty easily handled by the community I think, although a C-level .cube file reader might be nice for efficiency.

I've never use NWChem but if I do so I'll find myself writing importers and maybe some exporters for that too, I figure.

@JasonB. @C.E. @rhermans Since very recently I don't have any choice when voting. Every vote is final. My behavior of reviewing and closing has changed completely since then.

Basically, I now don't have the possibility to say "I think this should be closed, what do others think?". Instead, I have to decide carefully what to do and I'm always relying on all of you. So what do is I go through the review-queue and I rarely close something that has only 1 close vote and no comments. Instead, I'm watching carefully for your comments, why this should be closed, suggested duplicates or if the OP said it is solved and was a simple mistake.