Science & Industry

Hello again people,


This will mostly be an update of the numbers. As nothing has changed to Project Discovery it would be rather pointless to discuss faults and possible fixes again. I still have hopes that Project Discovery will be receive some love in the future, even if it is at least tradeable suits.


The numbers

Month | ISK | High(99%) | Low(30%) | Average
---|---|----|----|----
March | 340.341.808.085 | 3437796 | 11344727 | 7391261
April | 298.249.556.815 | 3012622 | 9941652 | 6477137
May | 176.913.198.620 | 1787002 | 5897107 | 3842054
June | 106.751.414.870 | 1078297 | 3558380 | 2318339
July | 78.072.857.880 | 788615 | 2602429 | 1695522
Augustus | 75.846.604.055 | 766127 | 2528220 | 1647174
September | 91.013.635.495 | 919330 | 3033788 | 1976559



So we have four months extra now, this translates in the following graphs:

ISK per month
Classifications per month


Much to my personal suprise, the curve is actually up since last month. Though taking the traditional summer dip into account this may not be that suprising. We may also be seeing what can be considered the lowpoint of Project Discovery, settling at around 76B~. This translates to 770K~ images classified at peak accuracy. This is actually not bad for a baseline! 25.000 images per day is very nice and will surely add to science. It is however still a far cry from the 150K~ samples per day we used to see.


Going forward

With the 11th of November coming closer, and thus the release of Alpha clones dubbed Free-to-Play EVE, I do think it is imperative that Project Discovery get an update. The expected influx of players is quite massive by all accounts. This could be a real boost to participants for Project Discovery.


However as we all know Project Discovery still got a few flaws and I fear that develop time will be put in different areas in preparation for the Alpha clones. Therefor I would like to ask you to raise this issue with CCP, HPA and the CSM. Project Discovery can already be hugely improved by some minor changes and will potentially have great real life value. To even retain a small percentage of the incoming new players is of notable importance.

Reddit discussion

The max accuracy (which is tied to the payout in ISK rounded up) is 99. The lowest % accuracy you can get without not getting any ISK at all is 30%. This means that if everybody would have a 99% accuracy less samples would have been analysed and if everybody would have 30% accuracy more would have been analysed. Of course the actual situation is different but since we cannot directly get those numbers unless Mmosch releases them we have to estimate. I think most of the current players are well above 80% accuracy.

Meanwhile
monthly report for december
171.2 billions of isk payed as rewards for Project Discovery participation.
Obviously, swithching to free2play model revitalized project.
Yet, It seems project backend is no longer supported:
most offered non-presolved samples are darker and blurrier in green channel (fit for *not identifiable*) than they were couple of months ago. With some expirience you can 90% say if sample is presolved.
My guess is that only bad samples stay due to lack of players' consensus, while samples with good consensus are no longer offered.

My Alpha clone is using Project Discovery to fund ship replacement.

I find it earns my Alpha more than ratting.

I'm glad CCP made it available to Alpha clones!