Hacked GH2 posterizes RED Epic (link)

Well, not quite actually. GH2 lightly hacked to 35mb/sec. I agree with some of the comments, that a better comparison would use the 175mb/sec intra frame hack.
http://vimeo.com/30751603

<iframe src="http://player.vimeo.com/video/30751603?title=0&byline=0&portrait=0" width="960" height="540" frameborder="0" webkitAllowFullScreen allowFullScreen></iframe>
 
Very interesting. There were some shots where the RED won out (though not by much) and some shots where the GH2 won out (though again, not by much). And there were some shots that were pretty much even.

And then there were the two high ISO shots, where the GH2 literally kicked the RED's ass. Usable footage at 2500 ISO??? That was where the 5D MkII was really winning out in the DSLR wars, and after seeing this GH2 footage, I'm thinking Canon has really lost its edge in that department.

Guess I'm definitely adding a GH2 to my wishlist (I've already got a GH1, so at least all my lenses and adapters will carry over!).
 
This seems like more of an issue of lenses and compression. Online video players have their own form of compression which makes video with lots of detail (like a red) lose a lot of it. Also, I think that the Red lens had a softer look than the olympus which contributed to what could be perceived as its sharpness. The GH2 is an awesome camera, especially if your target audience is online, but for big screen releases, the Red Epic comes out way ahead.
 
One thing I've noticed is that cameras that start out with more compression and lower resolution video hold up a LOT better when played online. I noticed it first when Film Riot did their test between the RED and the 5D MkII, and this GH2/RED test further reinforces that point.

I'm curious to see how Like Crazy holds up on the big screen, since that was shot entirely on a 7D. If the quality doesn't suffer as much being upsampled as the RED does being downsampled (I know those aren't the right terms, but I think everyone will get what I mean), then that gives indie filmmakers something else to consider when figuring out which camera they want to invest in (especially since you could buy a pretty awesome DSLR setup for less than renting a decent RED setup), particularly for those of us who aren't doing a whole lot of compositing or other effects where having higher res and RAW footage is more valuable.
 
This seems like more of an issue of lenses and compression. Online video players have their own form of compression which makes video with lots of detail (like a red) lose a lot of it. Also, I think that the Red lens had a softer look than the olympus which contributed to what could be perceived as its sharpness. The GH2 is an awesome camera, especially if your target audience is online, but for big screen releases, the Red Epic comes out way ahead.

I'd agree with you about the lenses/compression being an issue except when it comes to the high ISO footage. Lenses and compression aren't going to cause the kind of noise the RED was producing compared to the GH2.
 
As a personal contribution to the Deathworld project, I'd be willing to part with one of my GH2's in exchange for Nate's Epic. I've been tough on Nate and this is my way of offering an olive branch. I can't give both my GH2's, I'd need to keep one as A camera and then use the Epic for the B cam.
 
As a personal contribution to the Deathworld project, I'd be willing to part with one of my GH2's in exchange for Nate's Epic. I've been tough on Nate and this is my way of offering an olive branch. I can't give both my GH2's, I'd need to keep one as A camera and then use the Epic for the B cam.

If you had both cameras as your disposal, one would be foolish to use the GH2 as A camera and the Epic as a B cam.
 
Very interesting. There were some shots where the RED won out (though not by much) and some shots where the GH2 won out (though again, not by much). And there were some shots that were pretty much even.

To my eyes, the GH2 never came close, and never stood a chance to begin with. The faux sharpening and details, crunchy colors and posterized sky.

There's an illusion of "resolved detail" because the Epic sensor is much larger than the GH2's as well.

And then there were the two high ISO shots, where the GH2 literally kicked the RED's ass. Usable footage at 2500 ISO??? That was where the 5D MkII was really winning out in the DSLR wars, and after seeing this GH2 footage, I'm thinking Canon has really lost its edge in that department.

The way I see that RED footage is someone who doesn't know how to post RED doing camera tests. It looked like he ran the footage through Compressor, and if he didn't, he should probably study up a bit on how to use REDCineX.

The MX footage in lowlight versus the GH2 footage in lowlight when processed properly is no contest. I'd shoot MX @ 6400 ISO (And have done it) without a single worry in the world.

I would not shoot GH2 @3200 ISO with the same fearlessness. Nor a Canon camera.
Guess I'm definitely adding a GH2 to my wishlist (I've already got a GH1, so at least all my lenses and adapters will carry over!).

On the other hand, it's definitely worth its weight in gold.

One thing I've noticed is that cameras that start out with more compression and lower resolution video hold up a LOT better when played online. I noticed it first when Film Riot did their test between the RED and the 5D MkII, and this GH2/RED test further reinforces that point.

Naturally. Garbage in, garbage out.

I'm curious to see how Like Crazy holds up on the big screen, since that was shot entirely on a 7D. If the quality doesn't suffer as much being upsampled as the RED does being downsampled (I know those aren't the right terms, but I think everyone will get what I mean),


RED suffers being downsampled by people who don't know what they're doing. AS blunt as that sounds, it's the truth.

... then that gives indie filmmakers something else to consider when figuring out which camera they want to invest in (especially since you could buy a pretty awesome DSLR setup for less than renting a decent RED setup), particularly for those of us who aren't doing a whole lot of compositing or other effects where having higher res and RAW footage is more valuable.

RAW is valuable to the entire process. Every single step of the way, from concept to delivery, whether or not VFX come into play. If one can afford to have it, that is.

If not, get what you can afford.


I'd agree with you about the lenses/compression being an issue except when it comes to the high ISO footage. Lenses and compression aren't going to cause the kind of noise the RED was producing compared to the GH2.

But, exporting through Compressor would.

Now all Hot Rod needs to do is create a kit for the GH2 and we'll be all set!

They've had one for a while.
gh2.jpg


That's a GH2 with a cooke 18-100/T3.1 and an RPP 18mm/T1.8 prehack + Hot Rod GH2 setup, the camera I used for pickups on the feature.

I've got footage from a prehack GH2 + Hot Rod Kit + Arri Masterprimes on my Vimeo. It's a pretty great combo. The hack makes it a little golden camera.

Why? You can mount PL lenses with a simple adapter on the GH2. No need for any modification.

There's one floating around ebay that's pretty cheap and is the closest you'll get to Hot Rod quality, but I've heard two people snapped their camera mounts using it.

If you had both cameras as your disposal, one would be foolish to use the GH2 as A camera and the Epic as a B cam.

Indeed. There's no real contest. But, the GH2 is pretty good on its own for what it is.
 
Everything Kholi said +1

That RED footage consistently looked better.. But the GH2 footage looked good too. It's like comparing a Honda and a Lambo, yeah the Lambo is better but the Honda gets you A-B for a lot less change.
 
Back
Top