[?] Spades hacking? - Page 264

Hi guys, tonycheese here. I'm currently a ~950 Masters (on my smurf I suppose), fairly inactive player on the NA server. There has recently been a touch of controversy about the Spades situation. Although it does seem painfully obvious to me that Spades is maphacking in the replays provided when compared to his normal replays, many still call for "conclusive" proof. And there has been fear that conclusive proof cannot and will not exist.

Well, I *believe* I have found a way to conclusively detect if camera locks (and therefore maphacks) have been used.

The way my detection method works is as follows:

When a maphacker's camera is locked, he is looking around in the fog of war, but the replay shows him looking somewhere else. So, the maphacker claims that his actions are done on the minimap, when in reality they are done directly on the normal map in the fog of war.

*But*, the minimap is limited to a grid, with the smallest unit of measurement being the pixel. When you click on the normal map you can achieve a much greater level of precision than on the minimap. And as far as I can tell, the minimap's grid is consistent and reproducable.

What does this mean? If we can find a point in the replay where the camera is not focused on the spot the maphacking player is clicking, and the player clicks on a spot that does not exist on the minimap grid, that player had the camera locked.

In my first attempt, I found a spot that I could not click on the minimap, when Spades' camera was focused on his back third in daybreak, and he sent an scv to build the forward third "on the minimap".

Here are the sc2gears logs of my attempts at minimap clicking the spot next to the gas geyser. I will upload replays of these two attempts momentarily.



If people cannot find holes in my detection method, I believe we do have a painful, but systematic way of detecting any maphacks where camera locks are used.

\*\*\*\*\*

Edit: Here are my replay files, not sure how long rapidshare will keep these up, though: https://www.rapidshare.com/#!download|433|104782746|Minimap_clicking.zip|156

EDIT: As batterseapower pointed out below, I have not tested minimap clicking in, uh, every available resolution. And I don't really have the means to do it right now, so if someone else is super bored that would be appreciated.

However, you can see that the jumps on the map happen in units of 1.4 and .8 when running at 1600x900 resolution, meaning the jumps will still be very large at a larger resolution, meaning this is hopefully still a viable method of detection.

EDIT AGAIN: Actually, we don't need to test this in other resolutions. If you look at the two clicks at 12:14, the x coordinates are 117.9 and 117.6. This is only .3 units apart and I believe this is impossible in any known resolution that normal people have on their computers. At 1600x900 my computer jumps 1.4 units for 1 pixel on the minimap.

**EDIT 4:30 PM: A/some TL player(s) have claimed that they can get better than 1 pixel granularity on the minimap and I believe one player claims to have hit the point that I tried to check. I do want to remind you that I have not yet tried to look at other Spades replays or other instances of clicking (this was simply my first try), so my method is still open to discussion I suppose.**

**Right now, the two ways I have proposed to check are*

**1) Seeing if clicks in the fog of war snap onto the minimap grid and are reproduceable at some resolution or**

**2) Seeing if close together clicks in the fog of war are impossibly close for a normal resolution on the minimap.**