Superproducer Ted Hope on...

June 15, 2010

With hacked firmware, is the Panasonic GH1 better than the Canon T2i?

Short answer: no. Long answer: the GH1 is now a viable tool for filmmaking, whereas before I didn't view it as such. So, what is this hacked firmware? As I mention on the new Choosing a DSLR page of the DSLR Guide, the Panasonic GH1 recently got its own Magic Lantern-eque hacked firmware. As spotted at EOS HD, this firmware enables two very important features for filmmakers:

The first is a true 24p frame rate; originally the GH1 recorded 24p as 60i and the resulting footage needed to be deinterlaced required pulldown removal. The second feature the firmware enables is a much higher MJPEG bitrate -- the original bitrate of 17Mbit has been bumped up to 50Mbit somewhat unreliably, and 32Mbit more reliably. It's not a simple numbers game -- the 50Mbit MJPEG codec on the GH1 does not mean it's "better" than the T2i's comparably more efficient h.264 codec -- but this firmware is a game-changer for using the GH1 on serious video projects. While there are reportedly issues with the increased datarates, if Tester13 can come up with a stable version, the GH1 becomes a unique filmmaking tool, as it's the only DSLR on the block to offer somewhat viable autofocus -- not to mention its articulating LCD screen (seen below), which allows for a variety of shooting angles nigh impossible with other DSLRs. Beginners GH1 Custom Firmware Guide ยป

[viaPhilip Bloom]

Your Comment

17 Comments

Perfect timing! Today is the day my GH-1 arrives from B&H, to join the 5dMkII and T2i in my toy box. Woo hoo!

June 16, 2010

-1
Reply

Blah, fact check what you write. The GH1 always recorded a true 24p, nothing needed to be de-interlaced. Your had to perform a pull down removal which is not the same as de-interlacing.

June 19, 2010

0
Reply
Amy

Interesting thread at DVXUser comparing the GH1 to the 5D -- to my eye, the GH1 is oversharpened:

http://www.dvxuser.com/V6/showthread.php?t=214113

But then again the GH1 is $1200, and if you like more of a "video" look it's a heck of a deal with this new firmware.

June 27, 2010

0
Reply
avatar
Ryan Koo
Founder
Writer/Director

Hello Ryan,
the 'comparison' you linked above i find ridiculous, every frame grab from the 5d is far superior colorwise and there i believe there is something wrong with the focus. there are a few videos from the gh13 now to download from vimeo and i have the impression that the image is cleaner than the 5dmkii's. to my eye the gh1's footage is closer to raw (red), the saturation is too high though, i wonder if the gh13 can be set 'flatter'. if so, and if the hack is stable and if rolling shutter is not worse than on the 5d - i think the image quality is better. we will probably be able to judge all that in the next weeks.

July 2, 2010

0
Reply
Alfred

3 2 pull down is not de-interlacing. 3 2 is used to make 23.97 clips work in a 59.97 or 29.98 environment. De -interlacing is the reverse of the above information.

July 4, 2010

-1
Reply
Ralph Elliott

The GH1 stores 23.97 clips as 59.94 footage using 2:3, which does store interlaced frames, correct? Then the resulting 59.94 footage requires a 3:2 pulldown to extract the original frames. Whether we're talking about the 2:3 or 3:2 part of the process is inconsequential in my mind, as the pre-hacked GH1 was storing 24p footage as 60i. That's not a good thing, and that was my point...

July 5, 2010

0
Reply
avatar
Ryan Koo
Founder
Writer/Director

Ryan, you are a tad confused... The GH1 uses 3:2 pulldown to lay the 24 progressive images it shoots inside of a 59.94 wrapper... I assume they did this so it would play on native 60p consumer sets that don't support 60p. When you bring the footage into you editing bay, you do a pulldown REMOVAL, not a pulldown.

July 6, 2010

0
Reply

Mind you this is on the standard camera. When applying the tester13 hack which is now fairly stable at the 'c' settings for most people with cameras made in japan (newer models, not older original models made in china) you can get 24p native at 50 mbps and it is very stable. I have both the 7D and the GH1 and to my eye it clearly blows the 7D out of the water, and is pretty much equal to or a tad superior to the 5D MkIIs image, with a bunch of features the mark II does not have. This is not to start a war... I have no loyalty to canon or panasonic just for the sake of it.... you can do great things with either.

July 6, 2010

-1
Reply

Clearly my 3:2 knowledge is from the antiquated DVX era when it was 2:3:3:2 in 24pA mode. The point still stands -- that the GH1's pre-hacked 24p footage was not ideal -- but I've updated the post. Thanks for the corrections, guys -- can't wait to see some new footage from the hacked GH1.

July 6, 2010

0
Reply
avatar
Ryan Koo
Founder
Writer/Director

We got to have the Panasonic GH-1 in our inventory. Oh yes and a hacked firmware!

July 28, 2010

0
Reply

What a losd of uninformed cr*p.

The codec variant Canon use is no more efficient than that in the GH1...they all lack B frames.

There is virtually none of the Canon's line-skip aliasing artefacts on the hacked GH1.

"C" settings are totally stable...and the resultant image quality blows the Canon 5D2, 7D, or T2i away in terms of resolution and color purity (unless you love Canon's magenta disease). Have a look at some of the comparisons on Vimeo.

I was a Canon fanboy for a year...until I did a side-by-side of my 5D2 with the hacked GH1.

Hi there Blackout...have the cats kissed and made up?

August 26, 2010

2
Reply
Skeptikal

When I wrote this, it was the first release of the firmware, which wasn't even stable -- so I wouldn't stop by and call it "uninformed crap" two months later. It just so happens that a GH1 arrived at my doorstep today, so I'm looking forward to testing the two side-by-side...

August 27, 2010

-1
Reply
avatar
Ryan Koo
Founder
Writer/Director

The Canon output is under resolved. Line skipping is a resolution killer...it manages only about 650 lines.

Don't confuse resolution and sharpness. The hacked GH1 resolves way more than the Canon 5d2...I've compared both before I got rid of all my Canon gear.

August 26, 2010

0
Reply
Skeptikal

My issue was with your opening assertion of "NO".

The GH13 totally soundly outperforms the Canon VidDSLRs as far as video quality is concerned. No "if, buts or maybes".

Side-by-side, controlled tests are the way to go...you should have had a look at some already done months ago on Vimeo and Youtube.

September 11, 2010

1
Reply
Skeptikal

This post was written June 15th. The GH1 wasn't even STABLE with the hacked firmware then. So at that point, there were a lot of "ifs, buts, and maybes," holmes.

September 13, 2010

0
Reply
avatar
Ryan Koo
Founder
Writer/Director

Curious where the GH1 stands today. Tried to click on the beginners hacking data in this article and it was a dead end. Hoping to get current data on the GH1 or 3 camera and whether hacking firmware is still necessary. Thanks for any input. Robyne

August 27, 2012

1
Reply
Robyne

Robyne (or anybody else reading this in the future....), have a look here:
http://web.archive.org/web/20110404194616/http://wiki.eoshd.com/index.ph...

As of at the moment, GH3 (or the G6, which is an even better deal than the GH3!) have not yet been hacked. But both the GH3 and G6 are already very very good cameras in their stock state, so the non-existence of a hack shouldn't be stopping you from buying one! Additionally official firmware updates have been released by Panasonic for G6 and GH3, which is a crucial ingredient needed for VK to work on the hack (waiting for the first official firmware update is usually the biggest holdup for a new Panasonic camera to be hacked) so hopefully the first hack for the GH3 and G6 shouldn't be too far away now.

October 20, 2013

-1
Reply