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

[DrDave]

I would enjoy working with a Red.
For anyone who is interested, there is a fun article about the JAES tests that were done with a sample of 550 trained listeners.
http://www.mixonline.com/recording/mixing/audio_emperors_new_sampling/
This is not the JAES article, but a sort of rehash of the important points.
As far as comparisons to video, those who tested measurably better in hearing scored worse on identifying higher quality. I imagine that is sort of like being able to see noise because you have sharper vision.

 

[GrahamH]

Motion picture is also a huge tell. Stills are a different story. Have someone setup a shootout like the one Zacuto hosted, with really well lit scenarios, HUMAN BEINGS IN FRAME, motion, etc. You do a double blind like that, balance the cameras like they did, the post image, then let us go at a double blind. The results in that test were frighteningly obvious, and I don't believe that it would've taken much to discern which one was which.

...... obviously something is triggering an entire lot of us to jump on the Hack Bandwagon, and it isn't placebo effect.

The first point is key - much of the benefit of the hack is in motion rendition. Bloom's test doesn't get at that.

Regarding the second point, and a number of other posts in this thread, please keep in mind that the OP was asking about a possible placebo effect between medium-bitrate and high-bitrate patches. He was NOT asking about the difference between hacked and unhacked, which should be strikingly obvious to most anyone in any halfway-sensible comparison.

 

[Kholi]

The first point is key - much of the benefit of the hack is in motion rendition. Bloom's test doesn't get at that.

Regarding the second point, and a number of other posts in this thread, please keep in mind that the OP was asking about a possible placebo effect between medium-bitrate and high-bitrate patches. He was NOT asking about the difference between hacked and unhacked, which should be strikingly obvious to most anyone in any halfway-sensible comparison.

Yeah, totally realize that he's talking about between the 66mb and up patches. It's more so in response to people saying that there's no difference between hacked and un-hacked... which straight-up blows my mind.

 

[Ian-T]

If I get close enough to my 1920x1080 monitor then I can tell more from my computer screen than I could from the largest big screen out there. It is all about matching the display to the source resolution of the footage and getting as close to the display as focusing will allow. You don’t need a big screen to examine the footage. Your monitor will do just fine as long as it is 1920x1080 for 1080p footage and 1280x720 for 720p footage.

Yes, I understand what you are saying. But if we are talking about shooting for a typical 2K screen (which most of us watch movies on in theaters) then shouldn't that be your point of reference? Also, (as far as your example is concerned) I don't think it should be just any 1920x1080 monitor...but it should probably be a calibrated monitor. Isn't that what professionals use when color correcting their films (I dunno....I'm just an amateur)? Anything outside of that would proly be innacurate and you would miss a lot of issues in your image.

EDIT: Ok...I seen DrDaves #78 post sort of talking about the same thing.

 

[omnidecay]

Regarding the second point, and a number of other posts in this thread, please keep in mind that the OP was asking about a possible placebo effect between medium-bitrate and high-bitrate patches. He was NOT asking about the difference between hacked and unhacked, which should be strikingly obvious to most anyone in any halfway-sensible comparison.

Without hacked vs unhacked there is no placebo (in the context of this thread). The bottom lines are experiences and reference points per individual. The more you know the more you can do with what you are given. It really is a linear progression but not always at the same rate...

 

[GrahamH]

Without hacked vs unhacked there is no placebo (in the context of this thread).

The OP was clear in his view of hacked vs. unhacked. He says "While I will say that the inital hack has definitely brought more quality out of the GH2 ..." He then goes on to ask his question about the possible placebo effect of the highest bitrate hacks.

Personally, I think we get very significant benefits from a hack up to say 44MB, and thereafter some further improvement but at a much diminishing rate as the datarate is pushed higher. Sort of like the "80:20 rule".

 

[Kholi]

The OP was clear in his view of hacked vs. unhacked. He says "While I will say that the inital hack has definitely brought more quality out of the GH2 ..." He then goes on to ask his question about the possible placebo effect of the highest bitrate hacks.

Personally, I think we get very significant benefits from a hack up to say 44MB, and thereafter some further improvement but at a much diminishing rate as the datarate is pushed higher. Sort of like the "80:20 rule".

Agreed. More than likely, beyond a certain bitrate it's all coming down to very tiny difference that you either want and can justify or you do not want. Heck, can't even see perhaps. Very simple conclusion. What people choose to do with that (shooting grass, laundry, etc) is fine, go out and shoot your own stuff and show us what you've got. Not you specifically, Graham, but anyone who has an issue with it.

Some of us are testing and here trying to figure out the camera so that we can go into battle with it. The only way I would have ever been comfortable enough to shoot with this thing is by shooting randomness in situations that I feel would be the worse possible scenarios to shoot in. I learned that the minute difference in grain structure between seaQuake and Quantum V2 were going to help me just VERY slightly in ETC mode, so I took it.

Also made me more satisfied with actually shooting 1600 ISO and doing a chroma denoise without ETC mode on. But the gains weren't a leap, just minuscule and barely noticeable.

 

[DPStewart]

I would enjoy working with a Red.
For anyone who is interested, there is a fun article about the JAES tests that were done with a sample of 550 trained listeners.
http://www.mixonline.com/recording/mixing/audio_emperors_new_sampling/
This is not the JAES article, but a sort of rehash of the important points.
As far as comparisons to video, those who tested measurably better in hearing scored worse on identifying higher quality. I imagine that is sort of like being able to see noise because you have sharper vision.

I know these test very well that you speak of. And I LOVE the silliness they illustrate.
HOWEVER- there is a HUGE point about them that is overlooked in this comparison.
People don't know how music is "Supposed" to sound.
Meaning: There are in infinite number of "right" ways.
An electric guitar or snare drum can sound any one of a million different ways and they are all "right" because it's just art.
But a recording of the voice of our child or partner? Well, we can tell you in an instant if something's not completely right. Or a recording of the sound of our own frontdoor bell....we CAN tell if there's any deviation.

The same holds true for photography.
You can make a human face in any one of a million crazy ways in a CARTOON and it is accepted.
But if you try to pass off a digital CGI creation of a face and you get the geometry a little off, or even if you make the face symmetrical - Everyone senses "something's not right".

We know what leaves look like - so it's pretty easy to spot Macro-Blocking.
We know what the sky looks like - so it's easy to spot banding.

The ears and eyes are crazy devices...

 

[DPStewart]

The OP was clear in his view of hacked vs. unhacked. He says "While I will say that the inital hack has definitely brought more quality out of the GH2 ..." He then goes on to ask his question about the possible placebo effect of the highest bitrate hacks.

Personally, I think we get very significant benefits from a hack up to say 44MB, and thereafter some further improvement but at a much diminishing rate as the datarate is pushed higher. Sort of like the "80:20 rule".

I think your point is very accurate.
BUT GOP is a whole 'nuther can 'o worms.
The motion difference and edit factors of GOP-1 are very significant...I wish GOP-1 could be done well at a lower bit-rate but I have not seen anyone pull it off.
Good point about the 80:20 rule.

 

[tired]

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.

care to share the results? - I'm having real dramas with the spanning problem.

 

[mpgxsvcd]

Yeah, totally realize that he's talking about between the 66mb and up patches. It's more so in response to people saying that there's no difference between hacked and un-hacked... which straight-up blows my mind.

I can't say that I have ever seen anyone say there is no difference between hacked and unhacked. Yes some question whether the benefits are worth it or not. However, I think everyone agrees that there is some difference.

Now with the 42 mb/sec vs. higher bit rates I have seen people(including myself) say that there is "no difference". That is incorrect. There is a difference just not the huge differences that some people have claimed. I have uploaded half of the test videos to my Vimeo account. The only ones I have so far are the V1.1 firmware and the 187 mb/sec GOP=1 videos.

You can download those videos as they came out of the camera and decide for yourself how big the difference is. I should be able to upload the rest of the videos tonight, complete the review video, and have it uploaded for tomorrow.

 

[Manu Delpech]

A question: what shutter speed are you guys using with the hack patches?! Which one is best?! I'm used to shooting 1/25, 1/30, but I realized it stutters a bit (Soldier Ryan style), and I've read that 1/50 was the closest to what we can see in movies, yet I find it way too smooth when looking at the LCD screen and moving the camera around. It's supposed to get smoother when you go up (when I mean up, to 1/60, 1/100, etc), but sharper as well no?!

 

[DBP]

A question: what shutter speed are you guys using with the hack patches?! Which one is best?! I'm used to shooting 1/25, 1/30, but I realized it stutters a bit (Soldier Ryan style), and I've read that 1/50 was the closest to what we can see in movies, yet I find it way too smooth when looking at the LCD screen and moving the camera around. It's supposed to get smoother when you go up (when I mean up, to 1/60, 1/100, etc), but sharper as well no?!

Are you shooting 24p? It sounds like you might be shooting either 50p or 60p. 1/25 or 1/30 should actually have less stutter than 1/50, at least at 24p. 1/50 shot at 50p will look very smooth.

 

[Kholi]

Are you shooting 24p? It sounds like you might be shooting either 50p or 60p. 1/25 or 1/30 should actually have less stutter than 1/50, at least at 24p. 1/50 shot at 50p will look very smooth.

Yeah, you shouldn't be getting stutter like Saving Private Ryan at 1/25 or 1/30, you should be smearing all over the place.

1/30 without a ton of motion in the scene is tolerable at night, 1/40 is what I primarily use, but would like a 1/45 or, of course, 1/48. I do not use 1/50. Will jump to 1/60 or higher for action material, anyway.

 

[Kholi]

I can't say that I have ever seen anyone say there is no difference between hacked and unhacked. Yes some question whether the benefits are worth it or not. However, I think everyone agrees that there is some difference.

Now with the 42 mb/sec vs. higher bit rates I have seen people(including myself) say that there is "no difference". That is incorrect. There is a difference just not the huge differences that some people have claimed. I have uploaded half of the test videos to my Vimeo account. The only ones I have so far are the V1.1 firmware and the 187 mb/sec GOP=1 videos.

You can download those videos as they came out of the camera and decide for yourself how big the difference is. I should be able to upload the rest of the videos tonight, complete the review video, and have it uploaded for tomorrow.

Unfortunately, I know I won't have the mental bandwidth or time to download them, but I am very sure that a lot of people will be thankful for what you're doing. Thanks for showing the community. AT the end of the day, of course some people will see a huge benefit, some won't. The differences between FS100 and Af100 resolved detail, some people think it's negligible, right? I think it's obvious and important to note.

But, that's because I think that any visible gain in image quality is a win, and should be noted. It is not the same for everyone else, though, so naturally some people will say it's not important enough to warrant new cards, others will shell out for the more expensive media to capture the best that they can.

Neither are wrong.

Again, thanks for the tests.

 

[PaPa]

kholi, i've read you several times mention not using 1/50th and going for 1/40th instead. What's your reason behind this? I mean, if a 180 shutter is around 1/47.955 of a second ( arguably 1/48th ) then would 1/50th match closer than 1/40th in trying to change the motion the least possible?

 

[Kholi]

kholi, i've read you several times mention not using 1/50th and going for 1/40th instead. What's your reason behind this? I mean, if a 180 shutter is around 1/47.955 of a second ( arguably 1/48th ) then would 1/50th match closer than 1/40th in trying to change the motion the least possible?

I'm hyper-sensitive to stuff like that .The 1/50 has never looked right to me. Even that tiniest value looks off, and I'm not the only one that sees it. DVXuser Geoff Reisner would always complain that Canon footage looked weird in motion, he's also hyper-sensitive to it like I am. But, whenever he watched my footage he'd ask me what I did differently.

The only change was 1/45 shutter instead of 1/50. Recently, a few DVXusers have PMd me or commented on videos and are switching to 1/45 and 1/40 because they felt the same way. Reem12 has two 5Ds I think, one of them had been set to 1/50 and the other to 1/45 and for the life of him he could not figure out why one camera's image he liked better. Not until I told him to shoot 1/45, and he noticed his camera were not set to the same shutters.

GH2 doesn't have a 1/3 step, so it's 1/40. Perhaps with the GOP1 hack the 1/40 is actually where it needs to be at. However, again, that's preference and I understand that a lot of people won't ever notice, won't ever care. That's cool too!

Note: I have tested my "doesn't look right" problem with RED, 1/48 and 1/50. No likey.

 

[dcloud]

100mb/s and up for me is just too much space for a small benefit. id prefer somewhere around 50mb/s hack

 

[DBP]

I'm hyper-sensitive to stuff like that .The 1/50 has never looked right to me. Even that tiniest value looks off, and I'm not the only one that sees it. DVXuser Geoff Reisner would always complain that Canon footage looked weird in motion, he's also hyper-sensitive to it like I am. But, whenever he watched my footage he'd ask me what I did differently.

The only change was 1/45 shutter instead of 1/50. Recently, a few DVXusers have PMd me or commented on videos and are switching to 1/45 and 1/40 because they felt the same way. Reem12 has two 5Ds I think, one of them had been set to 1/50 and the other to 1/45 and for the life of him he could not figure out why one camera's image he liked better. Not until I told him to shoot 1/45, and he noticed his camera were not set to the same shutters.

GH2 doesn't have a 1/3 step, so it's 1/40. Perhaps with the GOP1 hack the 1/40 is actually where it needs to be at. However, again, that's preference and I understand that a lot of people won't ever notice, won't ever care. That's cool too!

Hmmm. Glad I'm not the only one. I also switched to 1/40. I noticed 1/50 was off after having my HMC150 and GH2 plugged into my parents 50inch TV, doing a few pans. The GH2 stuttered way more. It was pretty alarming at first. I've since done tests with the two, and I still find that the HMC150 has a more pleasing, organic pan, but using the GH2 at 1/40 brings them much closer together. The difference is now negligible. I realize this could be just a by product of the GH2 resolving so much more detail.

Also, nice to eek that much more light out of 1/40 vs 1/50.

 

[maranfilms]

I also switched to 1/40. I noticed 1/50 was off after having my HMC150 and GH2 plugged into my parents 50inch TV

What you and Kholi say makes sense. I think the 180 degree shutter with film cams is right on, and may be because it's mechanical. Where there's an actual shutter. With the gh2, it could very well be the software that's not right on the money, where 1/40th digitally is actually closer to 1/48th mechanicaly. I may be way off in my thinking, but after just messing around, I like it better myself. And I like the added light gain as well.