We may have our first cheater....

[psych1098]

Our lil time traveler just won't stay away from the daffodil

[psych1098]

Getting sick of the cheaters
Their doing it in packs now!
Darwyn and Spartan both just flew up past us all
And I've been on top (psycho) all along on the
Daffodil

[Adam Schmelzle]

Sorry I haven't gotten around to the auto detection for ultra growers yet. I'm actively trying NOT to work right now. Can we be sure that those guys weren't tag teaming the flower over night? I'm not saying they didn't cheat the clock, but I didn't see it happen and have to err on the side of caution.

One thing I wouldn't mind knowing: What's the most you've ever grown in a 1 hour period? I ask because I don't know if I've ever watered it as often as you guys while testing. If somebody could measure it and time the result to as close to 1 hour as possible, then post here it would really help

[stuberman]

I will run a 60 minute test and tell you what I get (I am guessing around 3.8 cm per hour)
I am averaging about 80 cm per 24 hour day. (I expect that the maximum is 91.2 cm per day)

[stuberman]

Confirmed - I just timed just a hair under 3.8 cm in an hour. (Will retest to verify)
Other measurements:
At 100% moisture, growth is 4.0 cm/hr.
At 95.0% moisture, growth is 3.61 cm/hr (at 12 minutes 2 seconds this is the first opportunity to water the flower).
Mathematical average is approx 3.8cm/hr and since the one hour cycle is very close to 5 rewaters per hour this average won't vary significantly.

[stuberman]

Rechecked one hour growth - this time got 3.802 cm in an hour.

[psych1098]

I think it can be ruled out that they tag teamed it overnight due to the fact that within a few minutes I watched them pop up under me by about 20cm and within about 1 minute they caught up to me and then passed me in the next couple of minutes.
What it did do to help me out was to make me realize just how obsessed I have become and definately not worth it if someone else that hasn't lost sleep over this can pop in and get on top.

[Adam Schmelzle]

Thanks for running that test! I'll implement the server side cheat detector as soon as I can, but don't know when it'll be yet.

[stuberman]

I assume you have logs of activity. I don't know what facilities are available but if you can get a unique identifier (per iPhone) that you track against server logins you could manage cheats.

When someone first downloads and 'registers' a flower you basically start the timer against a maximum growth rate. each time they login you compare the time if less than the maximum you reset the timer to their current reading. If greater than the maximum you ban that device based upon the unique identifier (IP addresses won't work since they are dynamic).

[Adam Schmelzle]

Well, as stated in my 'warning' screen, I don't actually save logs. However, I do plan on checking real time passed versus growth in the same way you mention. I didn't actually want to openly mention the cheat prevention mechanism, but I guess it's obvious to anyone who puts some thought into it.

This won't prevent cheating, just limit cheaters to a certain growth or risk detection.

[stuberman]

You already have the best cheat prevention mechanism in place: there is very little to be gained from cheating.

I deal with hackers and Internet attacks as part of my job. The more valuable your resource the more determined the attackers. In the case of your games it appears that it just looks like low hanging fruit to people who just want to show they can - sort of on the level of playing Sodoku. The more popular the games are the greater the attraction to attack. Since the game doesn't enforce any kind of validated user name this also discourages attacks since it is easy to spoof any name.