Chipping broken, bugged or changed?

[forgo]

So I am baffled, perhaps someone can help me out.

I have a couple empty ESI's.

I decided to see how much skill an ESI would take out of my top skills, and found some differences I cant explain.


10.04 ped ESI

BLP weaponry Tech 10884 -> 12 pts to fill
Rifle 10544 -> 8 pts to fill
RDA 8440 -> 12 pts to fill

20 ped esi was same ratio, doubled ofc.

As you can see there are 2 skills with a 20% difference in total skills causing the same point extraction on an ESI.

Feel free to test it out if you have one on you, implant-er is 15 ped can sell back full TT.

Let me know what you think please.

I chipped out 225k skills a few years back and it was a linear extraction skills to ESI regardless of skill then.

Perhaps something changed in my break?

Perhaps mission rewards need to be re-considered with this in mind as well.

Thanks in advance.

 

[Manique]

afaik it should be the same
Maybe it affects the decimal cases you cant see, but in this case, point wise? (for example as in guns decays past the 2nd decimal case)

 

[Fiona Simmons]

Sometimes you have a spike in the skill curve where you need more TT skillgains to go up a point and after you broke trough its going up faster again. Maybe you in the middle of such a spike on RDA ?

 

[girtsn]

imho its this, and it becomes way more noticable when you have over 9k skills (take care you might waste a lot of esi value as well if using small esis)

Sometimes you have a spike in the skill curve where you need more TT skillgains to go up a point and after you broke trough its going up faster again. Maybe you in the middle of such a spike on RDA ?

 

[Doer]

This actually may be normal behavior although a bit odd to see and shows some imperfections in the top end of the skill calculators.

Here's what my chipping tool http://www.entropiawiki.com/SkillCalc.aspx calculates:
Skill Skill tt of tt of
start end implant implant with +1 skill extracted
10884 10872 12.28 13.28
10544 10536 9.96 11.21
8440 8428 9.17 9.92

It's not really calibrated for levels above ~9k (wasn't much data at that level when we were figuring out the skill curve), but it illustrates what's happening, especially because it has an accurate period and approximate relative magnitude for the sine wave embedded in the exponential growth of the skill level to tt. So look at the size of the interval between the number of levels you could chip out and the next level.

So to extract one more skill it's:
(10884->10871) 1 ped difference
(10544->10535) 1.25 ped difference
(8440->8427) 0.75 ped difference

Which is reasonable because of the sine mentioned before, which leads to this curve:

So you see that this extraction would have coarse granularity (small chip size compared to size of level), which means potentially large remainders of ESI (which would be returned to you as PECs at amounts less than 10 ped). Of course, this we know because there is only ~10 levels going on a 10 ped chip, which means you are losing up to ~1 ped ESI due to remainder.

So it's a combination of where in that sine wave you are (which is how you can have larger intervals between levels at lower levels) and that the remainder is large in some of the specific cases you checked and smaller in others.

PS Note that the graph is the tt of skill per level (in other words, how much effort it takes to go from any particular level to the next level of skill -- this is the infamous "slowdown" one experiences periodically in each skill). This is not the tt of skill vs. level, which would show how much total tt of skill you have at any particular skill level. The mathematically inclined will note that the latter would be the integral of the first.

 

[jetsina]

10.04 ped ESI

BLP weaponry Tech 10884 -> 12 pts to fill
Rifle 10544 -> 8 pts to fill
RDA 8440 -> 12 pts to fill

20 ped esi was same ratio, doubled ofc.

If it's the fluctuations as suggested above, then maybe try with a quite a lot larger esi to get a better idea of the longer gradient? I've heard that the gradient doesn't really get worse at some point. Maybe the area around 10500 is a much harder patch (slow progress) and thus only 8 pts lost for 10.04 ESI, though.

edit: if you accept the suggestion of max chipout for ESI you have, you never get any pecs back, surely? Is it still theorised that parts of points are still being lost, though? I would doubt that that happens, personally.