There is no reward for doing well in SubProjects

[MarkRBright]

Great stuff thank you. I still wonder if we have been at cross purposes again though as that new screen is now beginning to look remarkably like the existing (but now updated nicely with regard to sequence) Userbycpidmilestones screen.

On the Userbycpidmilestone screen the only "fault" I can now see is that the Top100 an Top 1000 Ranks are the wrong way round, but I suspect that depends on when you look, as I suspect they may still share the same TypeNum at the moment of 3998 as previously mentioned. But it doesn't have the "No. required for a Maxwell" column which would be nice - for me :-)

On the new MMs Stats screen (https://stats6.free-dc.org/mms/fa1a6...617eb6b8903f8c) The only problems for me are that the RankT and Rank P column titles are the wrong way round (or the columns are, but I think P/T/C makes more sense than T/P/C), and secondly, that it doesn't have a row for Maxwells. Both of these I can happily live with but the first probably should be corrected.

What I meant re the # for a MiRB and TeraMiRB was a cheeky reference to my old request that it would be great to have two additional columns like the "#for Maxwell" column but "#for TeraMiRB" and "#for MiRB" This is information that would, like the Maxwell column, be the same for every BOINCer on any day, and would therefore only need to be calculated once per day, and perhaps held on the same table/array as the "#for Maxwell" data, and could even be calculated at the same time, and it would be a breeze to do so. The "# for Maxwell" is calculated as half of the number held by the person in rank 1 for each MM. The additional columns I am after are simply the number held by the person in rank 100 and rank 1000. They don't even need to be halved! I don't care about pretty pictures showing if you have it or not, just the numbers.

Glad you had a productive database upgrade supervising session.
Thanks again for all your work. Your site is important to me and many others.

[Bok]

Great stuff thank you. I still wonder if we have been at cross purposes again though as that new screen is now beginning to look remarkably like the existing (but now updated nicely with regard to sequence) Userbycpidmilestones screen.

On the Userbycpidmilestone screen the only "fault" I can now see is that the Top100 an Top 1000 Ranks are the wrong way round, but I suspect that depends on when you look, as I suspect they may still share the same TypeNum at the moment of 3998 as previously mentioned. But it doesn't have the "No. required for a Maxwell" column which would be nice - for me :-)

They shouldn't be the wrong way around. P100 has 3998 for me and p1000 has 3997

On the new MMs Stats screen (https://stats6.free-dc.org/mms/fa1a6...617eb6b8903f8c) The only problems for me are that the RankT and Rank P column titles are the wrong way round (or the columns are, but I think P/T/C makes more sense than T/P/C), and secondly, that it doesn't have a row for Maxwells. Both of these I can happily live with but the first probably should be corrected.

Fixed the titles. And change the SQL to use a LEFT JOIN which was causing the missing Maxwell row

What I meant re the # for a MiRB and TeraMiRB was a cheeky reference to my old request that it would be great to have two additional columns like the "#for Maxwell" column but "#for TeraMiRB" and "#for MiRB" This is information that would, like the Maxwell column, be the same for every BOINCer on any day, and would therefore only need to be calculated once per day, and perhaps held on the same table/array as the "#for Maxwell" data, and could even be calculated at the same time, and it would be a breeze to do so. The "# for Maxwell" is calculated as half of the number held by the person in rank 1 for each MM. The additional columns I am after are simply the number held by the person in rank 100 and rank 1000. They don't even need to be halved! I don't care about pretty pictures showing if you have it or not, just the numbers.

Still don't understand this. It doesn't use any number from the user in rank 100 ?? This is the main piece of sql

$sql = "replace into static.boinc_milestone_makers select a.cpid,'$mtype',a.nick,0,0,0,count(*),' ',' ',0,$ranknum from static.boinc_milestone_makers a join static.boinc_most_projects b on a.mtype = b.type where a.projrank < $min and b.subtype in ('I','O','S') group by a.cpid having count(*) > $number";

$min being sent in is 100, then 1000. Am I missing something ????

Glad you had a productive database upgrade supervising session.
Thanks again for all your work. Your site is important to me and many others.

It was uneventful, which is always good

[MarkRBright]

Still don't understand this. It doesn't use any number from the user in rank 100 ?? Am I missing something ????

All I meant was that 2 new additional columns on any/all of the screens showing MM stats, to sit next to the "Number required for a Maxwell" column - "Number required for MiRB" and "Number required for TeraMiRB" - would be as easy to calculate as the current "Number required for a Maxwell",
The Maxwell one that you do currently display shows half of the number currently held by the person ranked 1st, for each MM.
The new columns would do pretty much the same but would show the number currently held by the person ranked 100th for each MM for the TeraMiRB column, and the number currently held by the person ranked 1000th for the MiRB column.
They would be simpler to calculate because you don't even need to half them!
Thanks again.

[Bok]

But why would you want those numbers, I guess that's what I'm struggling to understand as they aren't used in any of my calculations??

So from This page you'd want the number 43 ? Which is how many the user at Rank 100 has ?

[Bok]

But why would you want those numbers, I guess that's what I'm struggling to understand as they aren't used in any of my calculations??

So from This page you'd want the number 43 ? Which is how many the user at Rank 100 has ?

[MarkRBright]

Sorry, I am clearly failing to be clear here. Maybe this will help. What I am after is the last two columns shown in the picture I have attached, obviously no colours are required.



Funnily enough the example that you gave is actually only one of two that isn't worth giving as it would represent the number of MiRB's you would need to get a MiRB for MiRBs which you can't get, nor can you get a TeraMiRB for TeraMiRBs in the same way you can't get a Maxwell for Maxwells.

Hopefully what I said now makes sense.
Cheers
Mark

[MarkRBright]

Sorry, you asked for the reason why.

I am ranked at around 2000th for stones earned, but I get a kick out of doing much better with the MMs, so I am ranked 38th for Maxwells, 48th for TeraMiRB's and a massive 16th for MiRBs. It's just one of these things that help me keep plugging away at BOINC and getting a little bit more fun out of it as well as hopefully doing some good. So if I can easily see that, for example I have 24 Sub750ks and I would get another TeraMiRB if I could get another 2, then that is what I will probably do the next time I find myself mulling over what to do. Then I might say Oooh! If I can take 7 of those 750's up to Sub1M's then I would get another TeraMiRB, not to mention the corresponding Squarepants, Badges and possible Teraboks for some of them.

Like I have said before, whatever reason people have for doing this is OK, as long as they do it, and any way to make it more interesting and fun is cool with me. I recently got a buzz out of bringing all of my MM Ranks down to below 500 - though I now need another Sub10M to get that back at the moment - and then another 100M to keep it a while longer.

Whatever floats your boat! Free-DC makes it much more fun than just collecting Cobblestones and badges, though there is nothing wrong with that.

[Bok]

Ok, I get it now. I'll have to create a new table to put this data into in order to join to it.

Something like this? Please check the numbers as I'm just doing an insert ignore using projrank < 101 to get the first available score for a rank < 101. Sometimes there is no exact 100 if multiple scores are the same as I group them together

[Bok]

Ok, I get it now. I'll have to create a new table to put this data into in order to join to it.

Something like this? Please check the numbers as I'm just doing an insert ignore using projrank < 101 to get the first available score for a rank < 101. Sometimes there is no exact 100 if multiple scores are the same as I group them together