Rework the SAP system

[ilikexd]

The SAP system is a clear necessity in the FFR submission process to assure quality control. But I think there is a problem in how points are collected and in the submission cap tiers.

The idea behind submission caps, as I understand it, is as follows: Everyone should have the opportunity to submit simfiles to FFR, but the judgement process is a long and laborious one, so in order to save time and effort for simfile judges and batch organizers, simfile artists are restricted to a certain number of simfiles they can send per period. Naturally, this compels them to only send simfiles that they think are their best and have a chance at being accepted, thus raising the average quality of all submissions. When a simfile artist gets many simfiles accepted, his or her submission cap is raised based on the idea that the simfiles being produced are of a higher quality and less likely to be rejected. The problem is that the whole concept of a higher cap for the purpose of having a higher ratio of acceptance-worthy to rejection-prone simfiles in a batch isn't being executed optimally.

I do not think there should be an arbitrary limit of five at 40+ points. I can more sympathize with the limit of five back when batches would become full and close, but now that the batch is permanently open, I don't. If the simfile artist has truly proven him or herself by acquiring so many points, there should be even further cap tiers, as long as the points are a reliable measure.

That brings me to another point: I don't think they are that reliable. While having a lot of SAP is indeed a measure of simfiling capability, it becomes pretty inaccurate for a few reasons. If one in every three simfiles a simfile artist submits is accepted, he or she will be getting SAP over time, and his or her submission cap will raise indefinitely. But the intended purpose of the SAP system assuring quality control is not fulfilled; the opposite happens (this is an extreme theoretical example to demonstrate a point and doesn't happen to this extent in practice). Instead of SAP being granted for every file acceptance, even when those acceptances are accompanied by the same number or more rejections in a batch, SAP should be granted based on overall performance inside a single batch. One acceptance accompanied by more than one rejection is not a positive performance and should not yield +SAP.

There is also no real way to lose SAP. Technically SAP is lost when receiving 1.5/10 or lower average on a simfile. But as far as I know this has never happened to any simfile artist who has any charts in game, so this measure is entirely useless. But it seems that based on its existence, somewhere along the line there was some intention to implement a system that punishes negative performance. Now to be sincere the actual prospect of having SAP reduced from a negative performance is pretty harsh, but as I wrote earlier I think SAP should be granted based on the overall/net performance in a batch, so that would mean considering both positive and negative performance within the same batch, which would not be a punishment/deduction in itself but would be so indirectly and would be an even better motivator for sending one's best simfiles and only one's best simfiles, and limit any frivolous submissions.