GH2 Ready for Prime Time? Finding the Right Hack

So we decided to go all in and see if we can make a GH2 work for a Network series. We did a series of tests with the camera hacked from 20mbs to 176mb/s hack by "Driftwood". This is almost eight times the data rate of a Canon 5D, five times that of an EX3, almost double Panasonics AVC Intra and so on. But not all is in the numbers. Can the camera handle this much data? The chip is more than capable of delivering that level of data but can the camera's DSP process, chew and digest the footage. Well, yes and no. Yes, if your takes are below four minutes. No, if you overload the codec with codec breaking shots. For example, we did several high speed shots weaving in and out of traffic from a skateboard. Every now and then when I got too close to parked cars, the amount of information from the moving cars, sidewalk, trees, and leaves, would make the camera say "no mas!".


Other caveats are cold weather. Our fist week of production was all exterior in 20-30 degree weather. The camera went bananas. Shutting down after a minute, two minutes, or whenever it felt like it. Menu items seemed to randomly disable themselves. Now I'm thinking surely, the developers and testers of the hack must have run into this. But then again, the hacks were developed over the Summertime and perhaps no one had a chance to stress test it under extreme conditions. We we're in a codec crisis. Shooting with children who often give their original interpretations with the lines with every take, we'd routinely run one to two cameras and keep them running until they got the scene right. When a camera would shut down after a minute, this was completely unacceptable. Being the head of the camera department, we were deep shit.


So back to the drawing board, we found a less intensive hack that only boosted the bit rate to 88 mb/s. And to our relief, no one in post could tell the difference. It seems after the 90 mb/s the margin of returns is negligible. But the stress on the processing is way less and our cameras returned to their predictable little selves. They performed in the cold and did takes as long as we'd like without a hitch. There was much rejoicing.  


Here's the link to the hack starter packs: 


https://www.dropbox.com/s/6wz0e4dlrl6xow1

You'll need this

http://winebottler.kronenberg.org/

And this (GH2 original firmware)

http://www.mediafire.com/?cwok7orwdegp2pg

This is new version PTool 3.6.3

http://www.gh1-hack.info/ptool3d.zip

I can't stress how much that I am not a computer guy. I can edit on FCP, set up email, start a blog, and do photo shop (on a good day). With that in mind, should you cheat the gods and attempt to hack your GH2, if you're not much a computer guy, find your local hackster and when he can figure it out, film him doing it. Having them explain it to you will do you no good unless you're fluent in Geek Speak. 

Once we had the hack working, I filmed someone on my I phone doing it, telling him not to explain anything, you're just complicating matters, just do it as simply as possible. One more peep from you I'll staple your lips shut.

Anyway, I digress. But this was by far the most challenging and frustrating part of incorporating these cameras into a production MO. For some, it's a walk in the park. I had many a contentious word with a superior who felt it was my responsibility as the DP to know how to do this. All I could say is, "would you let a computer programmer light your night exterior or lens up a beauty close up on a starlet?". He basically harrumphed and told me to figure it out. 

So there you have it. To hack or not to hack? 

the toy becomes a beast...

Next post: So we figured out the hack, how the hell to we make this toy camera into a production ready work horse?