GH2 Are all these new hacks and patches creating a placebo effect?

D

dtjohns1

Well-known member
While I will say that the inital hack has definitely brought more quality out of the GH2 and I give many thanks to all those that sacrifice/d their time to create these hacks and patches does it seem to anyone else that as the bitrate gets higher and higher nothing is visually changing that much but people are going crazy and saying they are seeing amazing new things. In my opinion Mugso shot on the 42mb hack still looks like the best GH2 footage I've seen quality wise. As more and more test of flowers and locked down shots of houses, barns, and wildlife (all in amazing 166mb and higher as the days go by) get posted to vimeo and youtube I get fearful that the main purpose of shooting meaningful short films and features is being lost. Then we wonder why no one will ever take our amazing GH2s serious. Ok guys we can shoot stuff that will fill our 64gb sdxc cards in 15 minutes but we've got to get back to the bigger picture: PUTTING OUT MEANINGFUL CONTENT
 
Actually, if you haven't installed Quantum V2, you should. Or V5.

It's definitely not a Placebo effect. Of course, there's only so much you can gain.

All of my testing's been for the purpose of knowing what I can and cannot do, should or should not do, what to watch out for and what to let go of, in preparation of a feature. Nothing's being lost, those who shoot will still shoot. Testing is good, that's how we got here in the first place.
 
If you were to test the various patches in a double blind test, I'm pretty sure the results would be random. I would even go so far as to say that you could mix in GH1 footage as well, unless the lighting and exposure produced a "tell", a tiny band or pattern that identifies a camera. However, I'm pretty sure no one will do this: not to be cynical, but it doesn't make anyone look good.
If you look at Philip Bloom's latest shootout, you will see that people can't even tell Canon from Panasonic, and so on, much less a subtle difference in the hack of a single cam. Just to be clear, we are talking about failing to even identify vastly different cams with different sensors by different makers.
But so what? The Hack is fun, and it is fun to try to get different looks with it. I enjoyed hacking my GH1, until I lost some video because of file spanning, and then I enjoyed even more developing a Hack that did file spanning reliably. And that work could not have been possible without the work of the Hack pioneers, those brave souls who sometimes brick their cams in search of knowledge.
 
yoclay said:
Would you mind pointing to it with a link?
It's just such a bloody mess over there, that it really is difficult to find anything.
I have been trying to sift through the mass of threads and posts for ages now and it's quite difficult to deal with.
Click to expand...

http://www.personal-view.com/talks/discussion/comment/31084#Comment_31084

here ya go. That's GOP3 Quantum V5

I have not tried this patch out at all, I'm on GOP1 Quantum V2 right now. I noticed an immediate difference between Driftwood seaQuake 2 and Quantum V2. I probably won't patch any further until there's something super significant going on. The main thing is to pick up some 95mb/s cards (For me) so that I can use the 720/60 mode for slow mo.

Although Sage says 1080i is the best deal for GH2 Hack and Slow Mo... Not sure what he meant by that.
 
Re the question of the thread. Take a look at Ralph's HDMI vs AVCHD tests we were doing on PV a few months ago. That put to bed very clearly the idea that we were wishfully seeing results. No, the lesser rates DID NOT hold up as well, and only when driftwood's old 132 GOP 3 was added to the mix of hacks being tested did people begin to see what a huge difference in the dark shadows and with movement the lower gop settings could have. Of course we are now long past the 132 GOP3... the latest Driftwood hacks are amazing to me. This coming from a Red owner, for what it's worth.

The new and better settings don't make art, but they allow for better art to be made given time, resources, and enthusiasm of the shooter.
 
I saved for over three years to buy a RED Scarlet. Then November 3rd happened. So I wanted a new but less expensive camera and came back to Panasonic. I tried out an AF100 and grew to like it. Then I saw the hacked GH2 footage. Having experience with footage from both I humbly propose it's not a placebo effect. And this includes having an external digital recorder (Sound Devices PIX 240) plugged into the HD-SDI out of the AF100. Was the GH2 footage better? Nope. But it was more than acceptable. And I don't know that an average audience member could really tell the difference with good lighting, composition and post.

And without the hack, I wouldn't have shelved my AF100 plans and gone with my GH2. I would have written it off as just more DSLR hype. But the footage I saw got me curious and here I am. Granted, it's a totally different way of shooting but I'm growing into it. I'll still eventually buy a dedicated motion camera. But now I don't have to start my fund from zero and I can still get great images. But yes, I'm guessing we will eventually hit the ceiling on what can be squeezed out of the GH2.

I totally see where the OP is coming from. We do get a little bit enthusiastic about new benchmarks. Well, I do anyway. Speaking only for myself, I think it's just sheer amazement about what can be done with this little consumer device. That's the key I think. As more people move over (a few of us from Scarlet User even...) we'll get more and more great footage. But without the testing and pixel peeping we might not convince some of them. I know I wouldn't have given in without it.
 
It's fair to say a piece of footage must contain imagery that will reveal the extra performance of a good hack.
Assuming that - The GOP-1 patches with the highest bit rate you can work with are the best way this GH2 device will capture what comes in through the front of the lens. Period.
So the tool has had all it's shortcomings removed - you are free to create with confidence, anything you want to put before your lenses.
 
The testing phase is a natural evolution of the hacking, as time settle and everything is tested, we will go more in the production phase. I have bee using quantum 2 and don't see myself hacking the camera until something that really warrants it.
 
Cosimo Bullo said:
Re the question of the thread. Take a look at Ralph's HDMI vs AVCHD tests we were doing on PV a few months ago. That put to bed very clearly the idea that we were wishfully seeing results. No, the lesser rates DID NOT hold up as well, and only when driftwood's old 132 GOP 3 was added to the mix of hacks being tested did people begin to see what a huge difference in the dark shadows and with movement the lower gop settings could have.
Click to expand...

If that is true, and I have no reason to doubt that it is, then it should be easy to prove in a double blind. I doubt that that will happen, even though I would be delighted if it would. No one gains in such a test, except the cold truth. But if your average viewer cannot pick out their own cam, that they own, against different makes, in even a single blind, which has been shown to be true, then it is a tough sell. Just as an example, after years of debate in the audio world they finally double blind tested some of this stuff, including gear, bit rates, resolution, all kinds of things. And the results were random. And even then, some people did not believe the results. So they won't be likely to do that again anytime soon, it is a buzz killer, doesn't sell gear.

I think the joy in using the gear is the feeling, the emotional investment, not the tiny changes in technical results. Or maybe one gets great technical results as a result of some inner journey from thinking hard about all aspects of the creative process, including grain, hacks, shadows, and so on. Maybe Vermeer agonized over how much ochre to put in on brush stroke of one pane of one window, even though it was a subtlety only seen by his eye. Obviously, most could pick out the winner in a low light shootout against a camera from five years ago, but subtle changes in bitrate, in the same model from today, I'm not so sure. Would be happy to believe it if I see it. On a medium to big screen, sitting at least two meters away, dunno.

But then again, maybe it is there, and some see it and others don't. After seeing the results on Bloom's site, I would simply assume that different people see very different things.
 
Ah, "The eye of the beholder" post.
Well, I can agree with the OP on his point that more focus should be on content rather then tech.
However, when im producing "content" and my image starts to degrade the more I push it. I need/see a huge difference with the high bit rates.
Even simple grading using MB - "Looks", I can really see a difference with lower bitrate footage and the higher stuff.
Granted, im using a medium patch. I dont have expensive cards, nor huge ones. (A few of Transcend Class 10s - 32G and 16G)
I usually shoot weddings or fashion sessions, and so I need to be able to run the entire day without dumping. So, its always a balance between power and storage.

I can really tell the difference with these patches... at least in the medium range.
As for the higher ranges, I take what others tell me.
I know its there if I need that quality, but you can only push these cameras so far... Where the medium range patchs, the diffrence is night and day.
Im using driftwood - SpanMyBitchUp patch
http://www.personal-view.com/talks/...ult-most-popular-patches-in-one-place#Item_22
 
DrDave said:
If that is true, and I have no reason to doubt that it is, then it should be easy to prove in a double blind. I doubt that that will happen, even though I would be delighted if it would. No one gains in such a test, except the cold truth. But if your average viewer cannot pick out their own cam, that they own, against different makes, in even a single blind, which has been shown to be true, then it is a tough sell. Just as an example, after years of debate in the audio world they finally double blind tested some of this stuff, including gear, bit rates, resolution, all kinds of things. And the results were random. And even then, some people did not believe the results. So they won't be likely to do that again anytime soon, it is a buzz killer, doesn't sell gear.

I think the joy in using the gear is the feeling, the emotional investment, not the tiny changes in technical results. Or maybe one gets great technical results as a result of some inner journey from thinking hard about all aspects of the creative process, including grain, hacks, shadows, and so on. Maybe Vermeer agonized over how much ochre to put in on brush stroke of one pane of one window, even though it was a subtlety only seen by his eye. Obviously, most could pick out the winner in a low light shootout against a camera from five years ago, but subtle changes in bitrate, in the same model from today, I'm not so sure. Would be happy to believe it if I see it. On a medium to big screen, sitting at least two meters away, dunno.

But then again, maybe it is there, and some see it and others don't. After seeing the results on Bloom's site, I would simply assume that different people see very different things.
Click to expand...

If you look at the embeded 1080p images on his site the difference are startling. In terms of sharpness I mean, the C300, F3 and gh2 are very very close to each other, the rest are soft soft soft compared to these three.
 
Great discussion everybody. I'm not saying that some of the more recent hacks aren't providing anything to the overall image but I just fear that a ceiling will be hit and just because a new patch arrives with a higher bitrate people will eventually start to "find" things that aren't even there. I just fear that an endless cycle of "this hack is amazing and I just finished testing it from top to bottom lets shoot something... oh wait quantum v20 just released let me test that first..." will happen and then the GH3 will arrive and the same cycle will repeat itself 1. find a way to hack it 2. create a patch that doubles it's bitrate 3. people test 10 sec clips or outdoor montages until a new patch is released 2 days later 4. repeat step 3 until GH4 arrives.
 
God forbid Panasonic should ever produce an S35 consumer camera. If Vitaly hacked it, it would be over for RED, Canon, the whole field of players.
 
The bitrate ceiling of the Gh2 has already been maxed out. The camera/cards can't put down any faster. The highest labelled hack was Driftwood's '220', though no one uses that now. Driftwood has been working to develop the most efficient intra, because extremely well distributed bitrate is necessary to allow intra to look its best. Essentially, I don't see any way to achieve a better hack than Quantum. All settings have been finely tuned, and no available bitrate has been left to waste. Technically, Quantum would be labelled '160' under the old naming convention (remember 176)... but that number is misleading. This is the best looking, most efficient setting. The image is visibly stunning.

To recount my own experience... I developed a 3gop 88 setting that was perfect and practical (could do 80%). I reasoned that to step up to intra would be hugely inefficient - the I frames could only peak in the 700s (600s at the time I think), while my own could peak at 1K. I thought it would be a step back. Fast forward a few months, and I heard that Driftwood had come up with a limited intra setting called aquamotion 2 that could record 80% continuously (slow motion is important to me). So I became curious and tried it out. But - Driftwood had just released Seaquake, and I installed this by accident. Hey, the names were similar! Water, you know... I thought I was testing Aquamotion 2, and shot some material. 80% wasn't working... but when I took a look at the footage on my computer, I had a holy sh-t moment.


I knew I could not go back. 80% be damned.
 
dtjohns1 said:
Great discussion everybody. I'm not saying that some of the more recent hacks aren't providing anything to the overall image but I just fear that a ceiling will be hit and just because a new patch arrives with a higher bitrate people will eventually start to "find" things that aren't even there. I just fear that an endless cycle of "this hack is amazing and I just finished testing it from top to bottom lets shoot something... oh wait quantum v20 just released let me test that first..." will happen and then the GH3 will arrive and the same cycle will repeat itself 1. find a way to hack it 2. create a patch that doubles it's bitrate 3. people test 10 sec clips or outdoor montages until a new patch is released 2 days later 4. repeat step 3 until GH4 arrives.
Click to expand...

I have a fear that people will keep coming to these technical camera specific sub-forums and chastising shooters to stop shooting tests and "get out there and shoot something real!"

</sarcasm>

Fear not. :)
 
You must log in or register to reply here.
Back
Top