AC130: Panasonic ag ac130

[Mike Harvey]

There are certainly some features that make these new cameras very appealing to me; the extra zoom, the dual SD card slot for backup recording, the widescreen lcd. But the cmos is what kills the deal for me. I hear a lot of conversations about how hmc150 users could never go dslr only or use the af-100 for event shooting due to the cmos sensors. So why the lack of concern over the change here? And can we assume that Panasonic is done with CCD in this pricerange? If they had added the things I mentioned above to the HMC150 and called it the HMC150B, I'd have my money down already. But as it stands, I'm taking a wait and see approach.

Personally, CMOS was never an issue with me that much. DSLR's are designed for photography first, foremost, and above everything else. As such, you're going to get a lot of skew (amongst other things) because they aren't really designed for video as a primary use. The only issue I would have using my new AF100 for event work has nothing to do with the CMOS chip. The newer CMOS chips seem to handle the issues with rolling shutter a lot better. Unless you're doing crazy pans and such, you probably aren't going to see it that much. And personally, weddings/events aren't where I do most of my work, so it's really a non-issue with me.

 

[Vaughan Wood]

As I mentioned in another thread, I've done hundreds of weddings with EX1's over 5 years, and even though photographer's flashes drove me crazy, (especially when slowing the footage down over the bride getting ready), I honestly did not have ONE bride ever mention it!
You can also see flash bands in every second news interview on TV (especially on politician's news bites), and no-one ever recoils in horror when watching the news, so I have to say, despite doing a lot of swearing over the last five years, flash banding is a non issue with the general public.
....and also, if you get a really bad one, it takes 5 seconds to put a white colour matte over those few frames with a quick fade to hide it, so it's not an issue at all really.

The ONLY time we ever got some citicism for flash banding was during a school play in a very large and dark hall, when the headmaster did not tell parents there was to be no photographs, and we got poor feedback on the quality of our video, especially the start, when the kids first appeared on stage.
We went back and counted frames and counted 157 flashes in the first 5 minutes of footage, and thereafter the school has not allowed flash photography in their school plays, and we still have the gig!!

Cheers,

Vaughan

 

[hendosan]

I never correct flash banding unless someone goes out of their way to complain about it - most people don't even notice!

 

[VideoMaker]

Jan, I did some video editing this weekend with some footage shot with a Panasonic Lumix. These were fairly static handheld shots and there was no apparent jello effect UNTIL I applied image stabilization. Will the AC130 footage hold up well to image stabilization in post?

You will not have the jello effect unless you totally shoot something you wouldn't be paid for, like rapidly swishing the camera back and forth 10-15 times. The clock speed on the AC130 is much fast than what you will find on any of the DSLRs or Flip phone. Jello is not a concern. If you were using it in a highly vibrational scenario like a helicopter, put it on a stabilized mount and it will be beautiful.

Best,

Jan

 

[Mike Warren]

Jan, I did some video editing this weekend with some footage shot with a Panasonic Lumix. These were fairly static handheld shots and there was no apparent jello effect UNTIL I applied image stabilization. Will the AC130 footage hold up well to image stabilization in post?

The 130/160 will have significantly less jello than a still camera. You only need to look at the example videos that have been posted shot from a moving vehicle. The bending of verticals is visible, but only we will notice. The customers won't.

Apart from that, the reason you see the jello more after stabilisation on the Lumix shots is that the overall movement was helping to disguise the rolling shutter distortion.

 

[Mike Warren]

I never correct flash banding unless someone goes out of their way to complain about it - most people don't even notice!

How do you fix it? I've never tried, but I guess it could be done by duplicating the frames either side of the 2 with the flash, but surely that would cause a slight pause.

 

[Jan_Crittenden]

Jan, I did some video editing this weekend with some footage shot with a Panasonic Lumix. These were fairly static handheld shots and there was no apparent jello effect UNTIL I applied image stabilization. Will the AC130 footage hold up well to image stabilization in post?

Hi, Saying that because it is a Lumix and thus Panasonic made, that it has to behave like a professional camera just is a little naive. I don't say that to be in your face, but please realize that these cameras are built to completely different specifications and with completely different engineering teams.

Now you applied image stabilization in Post? or in the camera. That isn't very clear.

Best regards,

Jan

 

[MichaelUK]

The picture would be similar but not exactly the same as the Digital Signal Processing is different. But you would think they were similar. The HDSDI is full uncompressed 8 bit 4:2:2.

Best,

Jan


Best,

jan

Many thanks to Jan and Phil for their replies. I'm now the very happy owner of a 160, thanks also to the (as always) super service from H Preston's here in the UK.

 

[Mike Warren]

I'm now the very happy owner of a 160,

Congratulations.

 

[Dstar]

Michael post some footage if you can PLEASE. My Ac160 is being shipped, the wait is killing me!

 

[VideoMaker]

Jan, I applied the stabilization in post using Sony Vegas pro 10 and the Mercalli stabilizer plugin. I've seen this with a couple different consumer grade cameras - not just the Lumix. I am just trying to understand what might be different in a prosumer cmos camera that would eliminate the "jello" effect in footage stabilized in post.

Hi, Saying that because it is a Lumix and thus Panasonic made, that it has to behave like a professional camera just is a little naive. I don't say that to be in your face, but please realize that these cameras are built to completely different specifications and with completely different engineering teams.

Now you applied image stabilization in Post? or in the camera. That isn't very clear.

Best regards,

Jan

 

[Jan_Crittenden]

Jan, I applied the stabilization in post using Sony Vegas pro 10 and the Mercalli stabilizer plugin. I've seen this with a couple different consumer grade cameras - not just the Lumix. I am just trying to understand what might be different in a prosumer cmos camera that would eliminate the "jello" effect in footage stabilized in post.

Hmm, I wouldn't use Stabilization in post, I would use the camera to stabilize. there are many variables in software that could give you jello in post. What does the camera do with its own stabilization. Your software may not be up to the task. So in a more direct way of answering your question, I don't know if there is anything that is better about the recorded image that would allow your software to do a better job. On camera, absolutely, the clocking out of the chipset is faster and thus the jello effect is minimized on camera, in post that is entirely up to the software and how good or bad it is.

Hope that helps?

Jan

 

[ullanta]

Not meaning to put words in Jan's mouth, but jello results from the process of reading data off the imager. In a global-shutter system (like CCD), the reading of data happens all at once, in between exposures, so all image data is from the same moment. In a rolling shutter camera (like most CMOS), the data is read off the imager sequentially during the exposure. Thus, the pixels towards the bottom of the image may be from a slightly later moment than those at the top, causing images with motion to warp (jello) and sometimes differ greatly if circumstances change very quickly (flashband).

So, better imagers can clearly influence this process. Faster read times off the sensor, different patterns of reading the sensor, added parallelisms, and likely some low-level compensation techniques (such as stabilization that is aware of and interactive with the sensor read process) can greatly reduce rolling shutter effects.

 

[VideoMaker]

Ullanta, this is the kind of information I was looking for. I'm still a bit unclear why CMOS footage if more prone to show the jello effect when doing stabilization in post. I know from experience you can take footage that doesn't exhibit the jello effect and stabilize it, and it will may the effect. Many others report this as well with CMOS cams. Even though I do use my HMC150's internal stabilizer, I often find I have to supplement it with stabilization in post - for example when shooting hand held and full zoom. Of course, I never experience jello effect with the HMC150 footage, only with CMOS cams.

I see from another post in this forum that the 160 did produce a small amount of jello on a fast pan. I'll wait and see how this plays out as the cameras hit the field.

Not meaning to put words in Jan's mouth, but jello results from the process of reading data off the imager. In a global-shutter system (like CCD), the reading of data happens all at once, in between exposures, so all image data is from the same moment. In a rolling shutter camera (like most CMOS), the data is read off the imager sequentially during the exposure. Thus, the pixels towards the bottom of the image may be from a slightly later moment than those at the top, causing images with motion to warp (jello) and sometimes differ greatly if circumstances change very quickly (flashband).

So, better imagers can clearly influence this process. Faster read times off the sensor, different patterns of reading the sensor, added parallelisms, and likely some low-level compensation techniques (such as stabilization that is aware of and interactive with the sensor read process) can greatly reduce rolling shutter effects.

 

[hendosan]

CMOS chips are going to be more prone to jello because of the rolling shutter. Even on newer camcorders with faster clock speeds and built-in stabilization, it's going to be there even if it's minimal. Depending on how your stabilization plugin works, this may exacerbate any rolling shutter issues.

 

[Jim Simon]

flash banding is a non issue with the general public.

That doesn't make it acceptable. It just makes the general public ignorant. But then, that's always been the case. How many people still unwrap their brand new HDTV, hook it up without any calibration to their noisy, analog, standard-definition cable and are perfectly satisfied?

We're professionals. We do know better. We want better. And it's very frustrating when manufacturers don't build "better".

 

[hendosan]

They do build better. You're just not willing to pay for it.

As I mentioned in another thread, I've done hundreds of weddings with EX1's over 5 years, and even though photographer's flashes drove me crazy, (especially when slowing the footage down over the bride getting ready), I honestly did not have ONE bride ever mention it!

I understand this, but we did an interesting poll. We asked some of the "top" wedding and event shooters and held a focus group with them. They all shot with CMOS cameras and EX1-EX3s. All of their gear had been purchased within the last 3 years, so they could have bought a camera with the feature. Not a single person knew what Flash Band Compensation was. And not a single one did anything to resolve the Flash Bands in the final videos they delivered. It was a rather stunning revelation. That we had a feature that addressed an issue they had and they didn't even know about it, nor did they find the problem to be that big of an issue.

Not only do audiences not care, pros don't seem to care either! You can keep on crusading for your full raster 1/3" CCD with huge zoom range HD camcorder for under $6k, but nobody has ever manufactured anything like this and because of demand and the cost of producing such a camera, nobody will.

 

[Justinec101]

Sorry if this is a stupid question, but if the move to cmos is inevitable, why not come out with a 4/3 fixed lens camera? It seems like if we're accepting the rolling shutter and flash banding, then the only thing that makes a camera like the AF-100 impractical for event shooting is the poor/lack of autofocusing. It seems like a camera like this would be the best of both worlds.

 

[hendosan]

It costs a lot to manufacture a servo zoom lens for a sensor that large. 2/3" servo zooms are already pretty expensive.

 

[Justinec101]

It costs a lot to manufacture a servo zoom lens for a sensor that large. 2/3" servo zooms are already pretty expensive.

Yes, I forgot about the zoom. That's another downside to the AF-100. (*For my situation)

But would proper autofocusing and a servo zoom really be enough to push a camera like I'm describing into the $10k+ range? There's obviously nothing like that on the market and I'm not sure if it's because it's completely impractical or non-cost-effective, or if it's simply a matter of micro 4/3 being so new that nobody's done it?