Where best to read about the hacks?

[Don Litten]

You know Ben....there's an awful lot of truth in what you're saying..

The fact is, there is a difference but it's mostly for the Cinema crowd. Some hacks have more detail (The new CM soft skin tones and night are prime examples) and they color grade better.

That's great except some of us that make our living with a camera don't especially care that much about tiny details. In my case, I need low light and repeatable settings or more often, reliable function in auto for fast, unexpected shots or long recording for interviews.

I don't see much advantage in lpowells 100mb hack at least in 720-60 which is mostly where I'm shooting, but there is some in Ralph_B's Sanity.

And yes...the hype runnith over on Personal View but no worse than most of the Canon Sites or even the lowly GoPro.

Of course, experimentation has advantages. Yesterday I bought an old Brownie Bullseye, cleaned it up and tried it out for nostalgia's sake. I had forgotten how much trouble film developing was.

[Ben Giles]

Mmmm... I thought my last post may have been a little curmudgeonly - but I do have near-fatal man flu at the moment and that can't help. So...

I did a little test just now. I downloaded the Driftwood patches from OSG Films - How to hack the Panasonic GH2 *Updated* - OSGFilms - and I loaded up "Span My Bitch Up".

I had an issue with a couple of takes - one froze the camera and the other then provoked the "Camera Stopped Recording..." message.

But...

Actually, the material I then got was quite considerably cleaner - none of the aliasing and digital artifacts I accept as given with the 24 Meg native codec - and, yes, I had taken my expectation bias pills. So, I'm going to do some more controlled tests this weekend (my wife's overjoyed) and do some A/B comparisons - one body with the hack and t'other without. I'll use my 2 stock 14 - 140 lenses and try varying light setups.

That will be my punishment for being such a curmudgeonly old fart. Will report back as soon as I have anything vaguely interesting to say.

Ben.

PS Hi Colin! Sun's coming out! Woo Hoo!

[Don Litten]

Just to get you started on the pixel peeking Ben, bump the images up to 4 or 500 percent and compare the macroblocking.

Then remove all but one color channel and compare the noise, then do another channel, etc.

BTW..Spanmybitch up is pretty mild. There are better low bitrate patches.

[Ben Giles]

Thanks Don.

What else would you suggest?

Ben.

[Pat Reddy]

Ben, I don't think your penance is complete until you have run a full Alan Roberts style assessment of two or 3 of the latest hacks.

Send your wife on holiday.... you're sick anyway.

Pat

[Don Litten]

Originally Posted by Ben Giles

Thanks Don.

What else would you suggest?

Ben.

Boy...that's a big laundry list Ben <G>

Going back to older to newer:

No Adverse Affects (Pretty mild but really sharp)

Cluster V1 (not quite as tame as the next two but very good low light and 720)

100 mb Flow Motion (very reliable but not geared to AVCHD really)

Sanity 5 (My vote for the best all around patch) Pretty low bitrate and the image quality is better, works well in 1080 and 720 NTSC & PAL. Almost bullet proof and very good low light. Handles well in post or if you wish, as good or better than stock firmware with the camera settings bumped up.

There are dozens of others. These are just my reliable favorites.

[Nick Gordon]

I played around with hacks on a GH1 (just got a GH2 and haven't done much). Anything by lpowell has been solid and reliable. He seems to find a good balance between worthwhile improvements in bitrate and reliability in use.

"nitrate"?!? What am I on?

[Don Litten]

As Nick said LPowell is very stable. The 100mb Flow Motion I listed is his latest/greatest/

[Paul Mailath]

back to the original question - if you haven't got time or can't be bothered reading up on the hack and it's variations - don't hack the camera!

it's a great little camera on it's own and while the hack can make a dramatic improvement in certain areas - it's a hack and that involves risk.

it's like saying you want to learn skydiving but can't be bothered doing the course - you just want a few tips and a plane - good luck with that!

[Ben Giles]

Paul, what you're saying is correct - there are apparent risks in hacking GH2s (although I've yet to read about anyone who has actually damaged their camera while installing a hack, beyond the usual caveat of leaving the camera alone while it installs any firmware update.)

The spirit of the original question is asking for a summary, as the poster (Aaron Holmes) seems to be experiencing what many of us have felt - the information out there is disorganised, fractured, rapidly evolving and lacking in evidence and any obvious "controls." It's quite overwhelming for those who have busy lives and are trying to make a living in the film/video business. The hacking work is all voluntary and gratefully received, but it is, nevertheless, lacking in "curation."

So, I'm going to just add to the noise with my point of view, having run some very unscientific tests for a couple of hours this afternoon. I've done many "proper" technology tests for the BBC and continue to consult for them as a post production adviser - but this was a relatively quick and dirty test in an uncontrolled external environment, so please take it with a pinch of salt.


The subject was an interview situation - but on a hammock, so the background was moving slightly.

2 GH2 bodies with the latest 1.11 firmware:

* One with the "Sanity 5" hack - which averages around 70 Mbps, 25 FPS
* One with the native Panasonic 24 Mbps (averaged around 15 Mbps in these tests) HBR codec, 25 FPS.

The lenses:

* Panasonic Leica 25mm/1.4
* Contax 135/2.8 (circa late 70s vintage)
* Nikon Series E 50mm/1.8 (1988)
* Nikon 50mm/1.4 (1983)

I'm not going to post the visuals as I'm convinced all subsequent compression will only add confusion - so I'm afraid you'll have to take my word for it!

So, I've just spent an hour watching through the results. Here's what I think:

* On full screen playback on a 27" Apple Display, the differences between the 2 codecs are so subtle as to be indiscernible. The kind of situation where you think you can see the difference, but subsequent watching then has you doubting. It wasn't a double blind test, so I'm not saying this is scientifically valid or anything - I just can't tell the difference. Going forward, will I now risk the potential for lost shots and lower recording capacity/larger file sizes? Nope - I'll stick with the native codec for now.

* As many of you will already know, by far the most dramatic differences were in the lenses - the Leica was the sharpest and had the greatest contrast and I'd probably use it for a corporate head and shoulders shot, as it's crisp and accurate. But for beauty shots, I'd go for the softer and more forgiving presentation of the older lenses.

* Workflow-wise, I was following my usual transcode to ProRes422 using Apple Compressor. However, I did spot some noticeable additional artifacts which don't show up when playing the original AVCHD files - especially around the subject's mouth and teeth - which surprised me. It's got me thinking I might pursue CS6 (Adobe treated me to a course on CS6 last week and I've been wavering about when I take the plunge and give it a proper try) as Premiere Pro will work with the AVCHD files natively. BUT, smooth playback will depend very much on my processor (currently a 2.8 i7 Mac), so time will tell.

There you go - that's my perspective so far. I'll keep an open mind and perhaps try some moving stuff when I get a moment - I've got to do a beach shoot over the next week or so, so I'll try and get a couple of additional tests in then.

So, in the spirit of the original question, yes, Aaron - I recommend the GH2 but suggest you try hacking for yourself if you want to, once you've properly got a feel for what the unhacked camera can do. Be prepared to dedicate a reasonable amount of time to working out how to do it over on "Personal View."

There's a useful "how to" video from this guy who has got some nice results:

DSLR Video - Advanced - Hacking the Panasonic GH2 - YouTube

However, what I haven't been able to find is much evidence to suggest that the great results people are getting are due specifically to hacking, rather than the innate abilities of the vanilla camera - just a lot of conclusions along the lines of, "I've hacked the camera, I've got nice results - it therefore follows that I've got nice results because of the hack."

I hope this isn't too contentious - just my 2p worth and the gift of my time to the greater good :-) If you think there's anything wildly obvious I've overlooked, do please say.

Ben.

[Don Litten]

I hope this isn't too contentious - just my 2p worth and the gift of my time to the greater good :-) If you think there's anything wildly obvious I've overlooked, do please say.


That was a well written and intelligent post Ben!

There is one thing I don't think any of us have posted though and that's the difference between the hack itself and the patches/settings being written for it.

The hack will load the stock firmware but you can change certain things like recording limits, etc...again with the regular firmware.

[Ben Giles]

Agreed, Don - the ability to remove the time limit on recordings adds considerable value to these cameras. I've been running both of mine on the stock codec, but with the time limit removed - and they've proved themselves perfect for event filming. On a recent health service seminar we shot, the GH2 images were far superior to the 2 other EX3s we were using and my 2 little cameras paid for themselves on one job.

Ben.