Glad to be here too Geek!!  Thank-you for starting up a home thread for us where this stuff is king...
 

 
Press those buttons team-mates and let's unleash flaming hell! 

I'll be down the track a little, throwing spike strips, tacks, bent nails and sharpened forks all over the place.  Hope you guys have insurance before I start throwing cinder blocks.

Yay, our thread is alive!  Now I'm sad because I lost the sig image that 415krunch made for me last year :-(
Point estimation is a slippery beast.  I guessed 200K because that is what my rig does on 8101's.  Who knew that for the first time in several months I would get a string of 8103's instead?

 

Afterburner

We just need to be tactful. The calculation tool they are using is sweet! So it looks like I will put my excess resources towards a fallen member instead of our contest. Unless someone can show me my error in thinking as I have produced 127K and am only giving us 43K in contest points...

I had a horrible time figuring out what my production would actually be, so I was safe and said 15k. Looks like I should have over guessed instead. Very sorry about that fellas 

 
An unfortunate consequence that comes with addition of our crunching section is that most of the crunchers could blow their estimates out of the water by switching over additional hardware.  While it is certainly exciting to watch teams heat up their production to overtake one another, it is impossible to create teams based on total potential output - the only number we have is the numbers that are reported.
 
At the moment, you are producing more than 2.5 times your reported PPD.  The amount you are allowed to overproduce before partial credit for additional points kicks in is based on how much everyone else is over their reported number.  Its designed to be a flexible and self correcting system that encourages people to fold as much as possible without completely destabilizing the contest.  Essentially, only the extreme outliers will feel a noticeable pinch.  I am certainly open to allowing a bit more leeway if it would improve things.  I will also be adding more stats that will shed additional light on it as I have the time. 
 
PS: When I saw your 15k, I assumed you were entering a single GPU.

You are correct that points never go down.  As you produce more, you will always get more points, but you may not receive full credit for them.  If your production slows, you will reclaim some of the partial credit as well, essentially it is dynamic in the sense that there is no disadvantage to large points drop on one day vs steady points spread over time.
 
While some type of correction is necessary, the amount of leeway allowed is certainly an open debate.  I don't know that there is an optimal level, but we can certainly arrive at a consensus.  The exact formula is detailed here (granted a more technical than intuitive explanation). 

texinga

Yes, it is difficult to forecast, especially with Bigadvs.  It is "feast or famine" at times.  If I lock-on to repeat 8103s, my PPD can be almost double what it is with the 8101 WUs.  I didn't know exactly what to do about that and then there was my 1Mil Quest that became a reality yesterday.  I hope that I'm not hurting the team's standings and am willing to take advice to help out.  My other motivation is to push as hard as I can for the 1Billion month which suggests keeping the foot on the "points-pedal". 

 
You are not hurting your team by producing more.  The strategy I would recommend, especially with variable PPD by WU, is to aim toward the lower end of what you are capable of so that you do not hurt your team based on getting bad WUs.  The tradeoff, however, is that if things go really well you may not get full credit for all the points you produce - but it never hurts the team.  Either way, at the end of the day it is science that wins!

troy8d

 

Afterburner

We just need to be tactful. The calculation tool they are using is sweet! So it looks like I will put my excess resources towards a fallen member instead of our contest. Unless someone can show me my error in thinking as I have produced 127K and am only giving us 43K in contest points...

I had a horrible time figuring out what my production would actually be, so I was safe and said 15k. Looks like I should have over guessed instead. Very sorry about that fellas 

An unfortunate consequence that comes with addition of our crunching section is that most of the crunchers could blow their estimates out of the water by switching over additional hardware.  While it is certainly exciting to watch teams heat up their production to overtake one another, it is impossible to create teams based on total potential output - the only number we have is the numbers that are reported.

At the moment, you are producing more than 2.5 times your reported PPD.  The amount you are allowed to overproduce before partial credit for additional points kicks in is based on how much everyone else is over their reported number.  Its designed to be a flexible and self correcting system that encourages people to fold as much as possible without completely destabilizing the contest.  Essentially, only the extreme outliers will feel a noticeable pinch.  I am certainly open to allowing a bit more leeway if it would improve things.  I will also be adding more stats that will shed additional light on it as I have the time. 

PS: When I saw your 15k, I assumed you were entering a single GPU.

This has nothing to do with my crunching personally. I have not stopped any crunching do participate. So your reference is false in this situation. I chose to sacrifice my "Stress reliever" called gaming and put the 680's to work. When I tested them it was showing 15K ppd. Yet now it bounces from 80Kppd predictions down to 30k... Just like many of the contests.. I try to be as honest as I should be, and instead should overestimate sense there is no penalty for under producing.
 
I have nothing negative to suggest about the calculation. I only said I was incorrect in my estimates and consequently hurts the points of the contest. And rather than stop producing, I will convert those resources to the ultimate cause, called producing for others. 
 
This crap of crunching vs. folding has reached my last nerve. We are one team. Period. I am choosing to help crunching because PG is so far behind. I would rather offer my 24/7 components to those projects supporting my components the best and projects I wish to help. Period.
 
I should have tested more than a few days so I had a better avg. THis is not your calculations fault, it is mine. Pure and simple.
 
Oh and this did not help at all... http://forums.evga.com/tm.aspx?m=1871866

I hope I can keep my projected numbers, these 12hr wu's killing me. Specially the heat they giving off and not to mention my last drop didnt count for yesterday! upset about that.

Idk usually my drops before 10pm pst always count for same day, sometimes before 11pm. Some reason it counted for the next day.
 
12.5k yesterday from my 28k projected, and that was a 14k drop. Would have been 26k finish. 
Just dropped another 14k just now so I'm even for the day for my projected.
 
Would over producing on a different day to make up a previous day balances out correctly does it? Or not really?

AzN-SoLjA
Idk usually my drops before 10pm pst always count for same day, sometimes before 11pm. Some reason it counted for the next day.

Over the past several days, I've noticed the Servers are sometimes not collecting data right at the "top of the hour" as they normally would do.  I don't know what's going on with that, but have watched it happen several times.   Maybe that is part of what is going on with your credits not showing when you expected them to.

AzN-SoLjA

Yea I got credit for today and dropped another which makes today easy lols. But yea I coulda sworn that dropped around 930pm I think, EOC site shows it on their 3am update.

It is because of this contest. We are overloading them MUHAHahahahahahahaha

Checkin in.
I have my run-flats on, spike strips don't scare me.

Afterburner

This has nothing to do with my crunching personally. I have not stopped any crunching do participate. So your reference is false in this situation. I chose to sacrifice my "Stress reliever" called gaming and put the 680's to work. When I tested them it was showing 15K ppd. Yet now it bounces from 80Kppd predictions down to 30k... Just like many of the contests.. I try to be as honest as I should be, and instead should overestimate sense there is no penalty for under producing.

I have nothing negative to suggest about the calculation. I only said I was incorrect in my estimates and consequently hurts the points of the contest. And rather than stop producing, I will convert those resources to the ultimate cause, called producing for others. 

This crap of crunching vs. folding has reached my last nerve. We are one team. Period. I am choosing to help crunching because PG is so far behind. I would rather offer my 24/7 components to those projects supporting my components the best and projects I wish to help. Period.

I should have tested more than a few days so I had a better avg. THis is not your calculations fault, it is mine. Pure and simple.

Oh and this did not help at all... http://forums.evga.com/tm.aspx?m=1871866

 
It wasn't directed at you personally, but it was a situation that occurred among a handful individuals last year.  Its nothing against crunching either, both crunching and folding are noble causes and everyone is encouraged to contribute what suits them best.  I did not say anything negative about crunching, simply pointed out that crunchers typically have additional hardware at their fingertips that isn't devoted to folding.  I believe you may be overreacting a little. 
 
You may be more disciplined than some, but you cannot deny there is the temptation of those individuals with machines devoted to crunching to switch them over for a short period.  While I don't anticipate this happening, as I mentioned history has proven this to be a possibility as the situation occurred last year that caused some hard feelings.  I felt was necessary to address in order for the contest to have any measure of balance.
 
Believe me, I understand its not perfect and there is a tradeoff that occurs because of this as well.  It certainly open to any improvements we can make to it.  An interesting idea that we could implement, perhaps, is allow captains to allocate additional points among their folders.  For instance, each team can raise reported points of one or multiple folders totaling an equal amount (i.e. divvy up 100,000 reported points among their folders that are doing better than expected).
 
Given the system we are using, I personally believe it is better to be slightly on the underestimating side than overestimating side - but ideally you want to be relatively close to accurate.

Thanks for the new sigs krunch! I'll sport em both when I remember to switch every now n then :D

Nah, I'd rather do this.  *drags laser-tipped claws over all the cars on the Fast and Furious Folders' team*