Store item selection algo?

Random I would say is random...

Just focus on the end result rather than the journey there... or your end up quitting...

On a related note I have always wanted to run this simulation but never had the time or math.

Suppose you have (let's pick a weird number) 17 different 10-sided dice that you can distinguish by color or pattern or whatever. You roll all of them at the same time and the ones that come up 10 (or '0' for you D&D players), you put a tally in that die's record, numbered A through Q.

You keep doing this until you get 10 tallies in a record, at which point you remove that die from the pool.

If you graph the tallies over time, what does it look like? How often do you get weird spikes where (say) die H reaches 10 before anything else reaches 4? How often do you get nice even distributions where dice A through Q all reach 10 within (say) 3 tosses of each other?

The above is basically how the ME3 RNG worked, to my understanding.

I was going to say "Bioware will never reveal the drop weighting" but they have been surprisingly forthcoming about that stuff. (Ex. Bastiaan talking about the 10% chance for a variant weapon to drop, or the previous discussions around the pity timer.)

I will add a question about this to my list.

Can you also ask for him to elaborate on the "10% chance for a variant" thing?  It isn't very clear, nor does it match my experience (I started tracking when my first Uncommon weapon, Mattock, hit 10).  And that's where the algo comes in again...

Mattock at 10, others all below 10. 
a) If it selects by weapon type, I now have a choice of Mattock variant (which can either be 3 items or 1 item, then split into 3) and 9 other UC types.  One way it is 1/10 chance of a Mattock variant, the other it is 3/(9+3) = 3/12.  So it is either already 10%, or it is 25%.

If it selects by cards remaining. Well, then I had 30 Mattock variants vs. remaining 53 UC's for others (since it hasn't unlocked the 9*3*10 variant cards for those).  In this case, I have a whopping 36% chance of a Mattock variant!

What I got was 3 Mattocks out of 14 UC drops, or 21%.  Yeah, I know, tiny sample space.  I now have a total of 14 variants out of 46 UC drops, 30%.  But it is more clouded as I now have amongst four L10+ UC's to drop from.

Bottom line: While it isn't clear what they did, the "10% chance" will mislead you into thinking it is a small chance.  Clearly, it isn't.  You will get plenty of variant drops before maxing out the others (I've gotten 14, while my six under 10's stand at ranks 4,5,6,7,8,9 respectively).  If they lower-weighted it, they did so minimally.

We might ask them to recheck their assumptions if they intended to make the variants anywhere near unlikely.

Since my Indra reached rank X (the only of my rares I've maxed), the only rares that have since dropped have been it's variants (albeit from a small sample size of ~3 packs).  Also, I thought they'd taken mods out of the packs, but last night I bought my first 100K pack, and got a mod.