• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

OVERTAKE?!?!

K

Kido1986

[H]ard|Gawd
Joined
Mar 10, 2003
Messages
1,180
It says we can be overtaken in 5 days!!!

What the hell is Austrailia doing? Ramp it up people. Ive heard a few people saying they were converting their boxen over to UD now that we're #1. I dont think that is gonna work as we are in the danger zone currently,
 
i hope it is just the morning thing

they always turn in more points in the AM then we come back in the PM

hopes....

u passed me yesterday :(


 
Well if things are as usuall they probably will pass us again pretty soon due to the seasons. The thing is, the longer we stay in the top spot the less time they have to build a lead once they do regain #1 (if they do). I wouldn't be surprised if our re-taking #1 has cause a few of their bigger producers to start folding again after a long summer. Just a guess though.
 
i like my idea better then yours pageian

of course ur is prob right... i know they are cranking up those b0rgs


 
Well these guys like to tout the fact that they have some heavy hitters on the way back, but I think the seasons are starting to turn in their favor... I've been trying to take advantage of this latest cool spell however and have been cranking the OC's.

However, I'm all for a large number of producers with smaller borgs, than one big producer trying to maintain the pace. So we just gotta keep up the recruitment efforts and borging everything in sight!

 
Kido,
Need to know where you are getting this information. What website? Saying something like "It says we can be overtaken in 5 days!!!" doesn't mean anything without back up.

The website I use for my data (EOC) says that we are outproducing, on a 24 hour average, OCAU by 20k point per day. Other systems will say something else.

Link to my data: http://folding.extremeoverclocking.com/team_list.php?s=
 
Kido was going on the most recent update at hardfolding.com. Currently it shows our gap as being about 31k.

 
http://www.hardfolding.com/index.php?go=39

That says 5 days. No offense intended, King_N, but I'll get my stats from Stanford if I'm really interested in accuracy. The new [H] graphs and things are really cool, though. For individual stats Hardfolding is getting better and better.

Edit: Perhaps you could make the overtake date be based on an average of the 7 day production, 24 hour production, and update production? That would weight it a little toward the recent stuff, but not as drastically.
 
Hardfolding is the one that had it as 5 days... I checked earlier.... I think that it based off of the production that day or since the last update, rather than a 7-day 24hr average... But, I've been known to be wrong in the past :D


Keep on Folding!!

 
Thanks guys. I assumed it was Hardfolding.com but was making the point for the reason that unrefferenced data stated that way causes panick that isn't understood.

I'm sure you all know this.

Maybe King_N can tell us how the stats are calculated so we at least undestand the data. I'm sure the discrepancy is due to calculating average production differently.

EOC runs a 7-day running average to come up with their number. I'm not certain about Hardfolding.

edit for grammer
 
I'll probably post this as a separate thread eventually - but have you seen the number of sluggers coming in?
party2go9820 - 43 machines 21K points last week
Kido1986 - 40 machines 20K points last week (thanks!)
and
testocules 14 machines and 19K points which is his total production. 19k points for a n00b!

These guys are smokin!

I've been busting my butt to get a 15K point week.
party2go9820 posted about swingin his office
And Kiddo - your doing great.
But I couldn't find testocules.

If we can continue to get folders with a decent farm like this maybe we can hold the summer.
 
That would seem most likely Hito, it would also explain why when we were second, that the estimated overtake was acting a little spastic over those last 4 or 5 days.

I second that it would be nice if King could give some insight into what sort of range these estimates are being calculated off of.

 
About the calculations - what you need to do is run a standard deviation on the daily average points produced by each team you are interested in, then factor in the variation due to vacations and major power outages, add a 10% factor if it is cold.
Print this out, burn the printout, carefully scatter the ashes in a circle (a frisbee will do) blow gently, then count the pieces left that show any of the teams names, and the team with the most pieces is #1.
There's another way to do it with goat entrails, but it's messy. And unkind to goats.
 
But I thought you only did the calculations that way on the second Tuesday of every month while you're trying to patch your Window's boxen?!? :eek: I knew I was doing my math the wrong way...

 
Dammit.. now they know I'm on a non-work-related site.... dammit....

gotta find a new work keyboard too...


Though, I did think there was something about bending over backwards and licking your heels while praying for a moose hug/reach around/reach through! LOL


Keep on Folding!!

 
I should point out that;

OCAU as of the 6am pacific, 8am central update they were 11,317 ahead of [H] in daily production if the other team's stats site doesn't consider that a threat then I don't know what to tell you.

hardfolding's approx takeovers are based on 24 hour(last 8 updates), the reasoning is a *lot* can happen over a week's time so choosing to go with a 7 day average is not the way to go for this, it's in fact less accurate. IE: 50 new members join the team, they produce for 2 days then 30 of them stop folding, with a 7 day average you would have to wait a week to get the *real* data everytime that happens.

As far as the accuracy is concerned hardfolding is accurate, about as accurate as you can possibly get when dealing with approximate dates based on ever-changing data, if we're being outproduced in the last 24 hours hardfolding will let you know that if the current situation continues the overtake is imminent and give you the date/time.

I'll let someone else that's noticed the inaccuracies of the other team's site explain them.

1 [H]ard Team, 2 Great Projects, 1 Common Goal..... The Cure



Webmaster: www.hardfolding.com
Lurker Extraordinaire
 
Thanks for clearing that up King. Your reasoning is sound, and IMO it's probably better that the data set isn't any larger.

 
Here's a larger dataset for you -

1356 - Team OCAU active members
1353 - Team [H] active members

That's why we're now down in the daily points race. We added new members, but they added a bunch more.
 
[H]ard|OCP
Active CPUs within 50 days 6730

OCAU
Active CPUs within 50 days 6878

.....

- Fold [H]ard - - Or go [H]ome - ​
 
Don't forget however, running with -advmethods and big WU's can and have been to our advantage. They're definitely the reason that we've been doing more with less for quite some time.

 
You can make data tell you whatever you want it too.

Now if you take hardfolding's 24hour average and take EOC's you will see that OCAU is trending towards outproducing us and that the line has crossed (i.e. they are outproducing us).

What I would also gather is that they are on a trend to increase their lead in production. We will know this is not true with their Hardfolding 24hour average is lower than the EOC 24hour production rate.

Know your statistics and how they were gathered and you can always infer the pertinent information.


 
Hito Bahadur said:
You can make data tell you whatever you want it too.

Now if you take hardfolding's 24hour average and take EOC's you will see that OCAU is trending towards outproducing us and that the line has crossed (i.e. they are outproducing us).

What I would also gather is that they are on a trend to increase their lead in production. We will know this is not true with their Hardfolding 24hour average is lower than the EOC 24hour production rate.

Know your statistics and how they were gathered and you can always infer the pertinent information.


Click to expand...

What was that saying of Relic's bout statistics.................I think it applies here :D
 
Spectre said:
What was that saying of Relic's bout statistics.................I think it applies here :D
Click to expand...


See Sig. :D


Keep on Folding!!

 
Spectre said:
What was that saying of Relic's bout statistics.................I think it applies here :D
Click to expand...

Agreed. We can guess at the statistics all we want but we don't even know IF there will be an overtake, much less when it might happen. For all we know MF might have kicked his machines back on and is getting ready to dump thousands of points on us. Statistics can be dangerous if you try to use them to predict the future. ;)
 
gnewbury said:
About the calculations - what you need to do is run a standard deviation on the daily average points produced by each team you are interested in, then factor in the variation due to vacations and major power outages, add a 10% factor if it is cold.
Print this out, burn the printout, carefully scatter the ashes in a circle (a frisbee will do) blow gently, then count the pieces left that show any of the teams names, and the team with the most pieces is #1.
There's another way to do it with goat entrails, but it's messy. And unkind to goats.
Click to expand...

I did this as a quote, because I did not want to get King_N upset. I was attempting to point out that in the few years I've been folding virtually no prediction was accurate. They often come close, but that's about it. I have not followed the stats at hardfolding.com. But every other stat site is like weather reporters - That's it - they report, but forecasting is not worth much.
I've been lucky to have been able to borg/sneakernet recently. This is about to crash, But:
You got to know when to hold em'
You got to know when to fold em'
You got to know when to get up and just walk away.

So let's try to keep our numbers up this summer!
 
Hito Bahadur said:
You can make data tell you whatever you want it too.

Now if you take hardfolding's 24hour average and take EOC's you will see that OCAU is trending towards outproducing us and that the line has crossed (i.e. they are outproducing us).

What I would also gather is that they are on a trend to increase their lead in production. We will know this is not true with their Hardfolding 24hour average is lower than the EOC 24hour production rate.

Know your statistics and how they were gathered and you can always infer the pertinent information.


Click to expand...

From what I see there is no advanced statistical analysis going on here. This is simple calculation done as Stanford reports the numbers.

I see no attempt to make the stats say anything but what the numbers actually are at the time.

I’m no statistician by any stretch of anyone’s imagination, but the variables here are many and totally unknown. Any team can and will add or subtract any number of boxen for any reason. Pretty hard to “predict” anything on an ongoing basis.



 
bill, this is off topic, but a)what kind of server(s?) are running your website and b)how many ppD do you get? (from those servers)


edited for clarification
 
BTW, I noticed the other day the OCAU has 3 more active members than us. Thousands of members between us and something like 2700 active total, and they have 3 more than us. I'd say that's a pretty close competition. Just like the points stats though it doesn't really tell us anything about what's going to happen tomorrow or the next day.

 
As of today... we actually have had 91 new people to the F@H project (according to hf.com) - from the past week
 
KodiakStar said:
As of today... we actually have had 91 new people to the F@H project (according to hf.com) - from the past week
Click to expand...

Proof positive that all the mentions on the front page, and recruiting efforts haven't gone to waste. Hopefully this puts us over the active members for the other teams, now all we have to do is get the CPU count up there.

 
Not as good as OCAU's 300 new users but a significant feat none-the-less.
 
Carnival Forces said:
bill, this is off topic, but a)what kind of server(s?) are running your website and b)how many ppD do you get? (from those servers)


edited for clarification
Click to expand...

Heh, my server is Comcast, so no folding going on there. My home farm, the only one machines I have is partially shut down due to power costs. At the moment I am running 5 machines total, all AMD, all over 2 gig.
 
BillR said:
Heh, my server is Comcast, so no folding going on there. My home farm, the only one machines I have is partially shut down due to power costs. At the moment I am running 5 machines total, all AMD, all over 2 gig.
Click to expand...
This is an excellent example of the variability which is VERY difficult to predict.
 
"Bye Bye Boys! Have fun borgin' the castle!!"
*waves*
 
gnewbury said:
This is an excellent example of the variability which is VERY difficult to predict.
Click to expand...


I think you just made my original point, thanks ;)

 
You must log in or register to reply here.
Back
Top