Re: AF100 successor?
 

Wish list? OK, that's pretty easy:

- Real 1080 resolution (Panny seemingly very reluctant to issue a cheap, real HD camera - they did the same with the HVX).

- Real HD resolution when over-cranked.

- Expand focus (crazy that that feature has not been added in some sort of firmware update - even the Sony VG20 has it).

- S35 sensor. No use for MFT sensor. I shot for far too long on either 16mm film, or 1/3" chips to go backwards on the sensor size scale now.

- In-camera 4:2:2

Keep everything else.

Re: AF100 successor?
 

If you are a designer tasked with "build us a new camera" there are certain fundamentals that are common no matter who the company is, and certain basic principles have to be decided upon right from the start.

Maybe the first and most basic is what cost can the camera be built to, and that is coupled with what the marketing department thinks they can sell it for! Close behind that is whether existing technology should be used, at least in part. The advantages are obviously cost, but also speed of bringing product to market, and arguably reliability - if an existing chip etc, the bugs should already have been got out of it.

Given all that, there's then the job of matching the engineering together with such as styling, features etc. It's a complicated job! I agree with what Gabe said earlier about aspects of the design - select the best from the AF100 and the FS100.
There is no reason why a broadcast codec need cost any more than such as AVC-HD. Any company that says otherwise is doing so for reasons of marketing, to keep up the price of their higher end range. As illustration look at Canons XF100 - which manages to keep full broadcast spec XDCAM 422 50Mbs in spite of it's low price, AND reliably record it to relatively inexpensive media.

I don't expect Panasonic to go with XDCAM for obvious reasons, but even if AVC-Intra 100 meant P2, then AVC-Intra 50 should be easily possible to SDXC cards.
There is an even more basic point here, and it relates to what was said earlier about "do you go for existing technology, or develop from scratch?" In the case of the AF100, Panasonic basically used existing - the camera was initially designed around the GH1 chip, then adapted to launch with the GH2 chip. (Albeit without the improved electronics of the GH2.)

Advantage was cost and speed of development - disadvantage is that the chip is primarily designed for stills. They basically put a DSLrs innards in a video camera body! Here, I'll slightly disagree with what you said earlier, Gabe. More pixels does mean more resolution - but only up to a point. Above the "right" number and they become a problem. Go up to the numbers found necessary for still sensors and it becomes impossible to read them all at video framerates. Which is a fundametal problem for DSLRs used for video, and a problem the AF100 (with a designed-for-stills sensor) shares.

I believe the AF100 works in a very similar way to the Canon C300, basically directly reading out blocks of 2x2 pixels, and directly reading out R,G,B values for each block. The reason the C300 has better performance is that it has the EXACTLY the right number of photosites for 1080 video, EXACTLY 4x1920x1080. As simply as possible, the AF100 has too many - so has to ignore a proportion, and the resolution, sensitivity issues follow from that.

Basically, purpose designing a chip for video will always give far better results than using one designed for stills. So from the chip point of view, the first strategic decision Panasonic have to make is:- use the latest still sensor, or develop one optimised for video?
And this is the second fundamental issue. Yes, s35 is fundamentally better, 4/3 has an attraction in terms of legacy and pleasing past customers. The danger with going with 4/3 and purpose designing the chip is that a lot of money gets spent in development - but the result is still seen as "second best" from day one.

Sorry, but I don't think the idea of s35 with 4/3 crop is remotely feasible. It's a nice idea in concept, but couldn't really work in practice. If you take the Canon way of "designing for video" it relies on the chip having 4x1920x1080 photosites, to (easily) give full 1080 video. If that is then centre cropped for 4/3 the photosite count is halved, and (far worse) it is no longer possible to easily do the direct 2x2 read.

There are other variations, but as far as I can see, they will all have similar problems - either the s35 or the 4/3 mode will end up compromised relative to the other.

(The only possibility may be if the s35 output was true 1080p, then by reading the cropped sensor in the same 2x2 manner, the 4/3 window could be made to directly give 720p in exactly the same way. Would that be acceptable from a users perspective? Use s35 glass and get "true 1080", use 4/3 glass and get 720p, which could be upscaled, albeit without improving the resolution?)

Re: AF100 successor?
 

On paper, we were all set to buy the FS100. Our latest contract to produce 10 films for the BBC Academy was a good excuse for a tech refresh. We wanted integrated audio and were looking for a small but not insignificant shift up in image quality from our in-house GH2s.

So we went out and tested both the FS100 with the kit lens and an AF101 with our set of MFT glass - new Leica Summilux 25 and 45, 14-140 and some old but good quality Nikon glass I've had for 20 odd years. Both cameras were used on paying jobs.

We bought the AF101 for several practical reasons:

* We can continue to use the lenses we know and trust and, in my practical experience, lenses make a far bigger difference to the overall viewing experience than sensor size and codec. Of course, we could use the Nikkors on the FS100, but not the nice new MFT stuff.

* For me, the built-in ND is tangibly better than fumbling around with variable ND when we're under time pressure. And there is none of the baked-in colour cast that I've experienced with a couple of expensive brand variable ND filters.

* Great build quality - the Sony felt weak and plasticky to me. Our stuff gets used by several different freelancers and they're not always as caring as I'd like them to be.

* It produces tangibly better video than the GH2. Meaning that I'm confident I can tell the difference immediately between the two - and not just when A/B tested. We used both cams together this week for a couple of interviews. They cut together fine, but I always want to cut back to the 101 pics over the GH2, which always has noticeable artefacts around subjects' eyes and mouths. Hacking? I've tried it, tested several of them extensively and I'm not a believer. However, we're not getting rid of the GH2 any time soon - it's still a very useful tool in our armoury.

I had the same questions about both the AF101 and FS100 and thoughts about them both being towards the end of their notional product cycle. For us, it came down to:

Do we want to sell all of our expensive MFT glass and have to do loads more testing with new lenses and crop factors? (no)

Can we make more profit by buying one of these over renting? (yes)

Is lack of easy to use high quality ND a significant choice factor for us? (yes)

Is the camera well established with bugs wrinkled out? (yes)

That's where we're at right now. It will all change again sometime soon. And, yes, a game-changing successor from Panasonic which uses MFT glass would be exciting. Might let the early adopters wrinkle out the bugs, though!

Cheers,

Ben.

Re: AF100 successor?
 

I don't know that it has to cost F3 money for what I wanted. If you
look at the FS700 it has all those features and it doesn't cost
F3 money. OK, so it may not give you 10 bit 4:2:2 YET but it should
be able to do that once it gets the '4k' upgrade. (This is to
an external recorder of course). And of course they'd need to change
the LCD position or make it hinged in the back. But otherwise, it
pretty much gets everything for about 8K. Do away with the 4K chip,
stay with the FS100 chip and how much would that camera cost? I'd
like to think it should be cheaper than the FS700, which would be
less than 8K for the features I would like to see.

As for the feature list of why to chose the AF100 over the FS100
that Ben posted, I could post an equally long one about why to
chose the FS100 over the AF100, but as Troy says why not instead
focus on getting SOMEONE (and I don't care if it's Sony, Panasonic,
Canon or even if someone conjures it up from Black Magic :-),
to combine the best of both the AF100 and FS100. I think a lot
of us can agree on some things we'd all like. I think maybe the
biggest issue is the sensor. See, those like Ben that have
invested in the M4/3 format would like to keep that format.
Others like me, who have not invested in the M4/3 would rather
have them go to a S35 chip that is purposely designed for
motion, not stills. And they will have a hard time doing
both. Perhaps, Panasonic will stay with M 4/3 and Sony
and Canon will continue to go with S35 which I guess will
give both of us a choice as where to buy.

David, I think I understand about the practical limits of pixels
on a sensor. I guess I should have said 'Until you get to that
point, either the larger chip will give you more resolution
for the equal dynamic range and low light ability, or it will
give you more dynamic range and low light ability for equal
resolution.'
For the same NUMBER of pixels, each one can be bigger, or you can
have the same SIZE pixels, and more of them, because the chip is bigger.
So if you assume both manufacturers want to use the optimum number
of pixels for motion and not go over that number, then you see
how there is ALWAYS an advantage to a bigger sensor, in terms
of one or all of the three....resolution, dynamic range, and
low light ability. That is something the M 4/3 fans just cannot
get around, because it is simple physics.

As for recording format, of course I'd like it to be in camera.
I'd be willing to 'put up' with it on a external recorder, only
because it would still be possible to get. If the choice was
between hitting a 4-6k price point and F3/C300 price territory,
I'd rather stay at the lower price point and deal with external
recorders. I know you are saying it shouldn't be more expensive
to do in camera, but for some reason no one is doing it yet.
I was under the impression that it was more expensive
to make a camera with less compressed codecs, but I don't
know that much about it. Perhaps they will do this in camera
in future large sensor cameras at the AF100/FS100 price point,
one can only hope!

Still, lest anyone think I am complaining, I think it's pretty awesome
to have two cameras at this price point that can do what they can do.
They are better cameras than I am a shooter, that's for sure. Just human
nature and all to want something even better. But that won't stop me
from using what I have in the meantime, no doubt about that!

Re: AF100 successor?
 

I'm now strongly inclined to believe that the FS100 and the 700 share the same chip. If you're interested in why, then look at http://www.dvinfo.net/forum/sony-nxc...ml#post1737180 . The assumption for a long time was that the FS100 had the F3 chip (well, they came out at the same time, they are both s35.....) but there were anomalies in the measured results that couldn't be explained. Believing the FS100 to have the same chip as the 700 explains a lot......
Exactly. To Ben, 4/3 was a positive point for the AF101 - to others it's a negative.

One thing I would say is that it's as equally possible to make a "designed-for-video" chip for 4/3 as for s35. Simplest way for 1080p being to use a quad-HD count (3840x2160) and read as the C300 does. That means an 8 megapixel sensor - either less or more will give issues. (The F3 has far less - but then needs a far more complex processing system, hence the far higher power consumption, cost etc.)

But an 8 megapixel sensor is not these days considered enough for a true stills camera - it's only likely to sell if far more. That's why, come an AF101 successor, the most key point is whether it uses a chip designed primarily for still cameras.
Correct. Of course, all else equal, the bigger the sensor, the bigger the camera is likely to be, the bigger the lenses, etc, so at some point it becomes a law of diminishing returns. Generally, I think it's fair to say that s35 or full-frame is probably around the sweet spot for this type of camera unless you have legacy issues. (Just as 1/2"-2/3" is for cameras where dof is less important than flexibility - long zoom range servo zooms etc.)

Well, maybe not in large format - but Canon are with the XF100. If they can do it (50Mbs XDCAM 422 - fully broadcast compliant) in a camera with the size, power consumption, price etc of that, why is it not possible with cameras like the FS100, AF100 etc? Simple answer is that it is possible - but it doesn't suit the marketing strategies of the likes of Sony, Panasonic - and "not technically possible" is an easier stock answer than "we don't want to".
That may be true under more extreme circumstances - for example, to record such as HDCAM SR the data rate is far higher, and that alone can have a cost attached, not least in the cost of more and much faster memory! But not when you talk about the difference between such as AVC-HD and XDCAM 422. They can both be reliably recorded to pretty cheap cards, and the only factor is you'll need about twice as many GB for the latter as the former. Hardly an issue given the price of SD cards.

Re: AF100 successor?
 

The original question was as to whether there is a successor in the wind for the AF100. When I bought the camera in the spring it came with a $500 rebate (in the form of a fixed value AMEX 'credit' card - just had another great lunch on Panasonic yesterday). When a manufacturer does this it is a sign that he is trying to clear inventory and one reason they will want to do that is to clear all of an old model out so they can introduce a new one. OTOH while I really like the concept of this camera (especially being able to use every Nikon lens that I've ever owned with it) I don't sense the buzz that I have with some of the other HD cameras. Perhaps that was just the excitement of the early introduction of that technology.

How would they improve the AF100?. Higher resolution could be had at the cost of more aliasing by raising the corner frequency on the antialiasing filter. Higher resolution without more aliasing would require more pixels (more over sampling) followed by digital antialiasing (more processor demand, a new sensor, more power consumption and definitely more $). Certainly pixel for pixel mode in the EVF is something sorely needed. Slower shutter speeds would be nice etc. I don't think the noise performance of this camera is that great. Improving that would require a different (larger) sensor and dramatic redesign - again many more $. Maybe a redesigned DSLR with XLR's on it isn't the shining path it at one time appeared to be.

Re: AF100 successor?
 

Is this only for the US or other countries as well? Ive never heard of it. Seems the ROW gets left out.

Re: AF100 successor?
 

That I don't know. I just checked to see if they are still giving this rebate in the US and they are through the end of September. I've looked at a couple of web sites for sellers of the AG-AF101 and do not see it mentioned. So I guess the answer is that it is Panasonic USA that is offering this.

Re: AF100 successor?
 

I doubt it - look at charts for the AF100 (and I suggest Adam Wilts - ProVideo Coalition.com: Camera Log by Adam Wilt | Founder | Pro Cameras, HDV Camera, HD Camera, Sony, Panasonic, JVC, RED, Video Camera Reviews ) and it's pretty clear that fine detail is already getting to the chip well beyond the level where "real" resolution has turned to aliasing.

Look at the horizontal and vertical trumpets - the lines that converge towards the centre are real resolution, the ones that diverge are aliases. The point where they go from convergence to divergence is the real resolution limit, and it's about 650 lpph. But the aliases are strong at around 1200 lpph, and you're even getting second order aliases faintly visible even at around 1750 lpph.
The problem with the AF100 is not too few pixels, but the wrong number! The F3 has just over 3 megapixels, the C300 about 8 megapixels - both far fewer than the AF100 at around 16 megapixel - yet both the F3 and Canons C300 produce full res 1080 video. It may be thought that the more pixels the better - but it becomes impossible to read and process so many at video framerates. Hence the need to simplify the read out system - only read a fraction of the total. This also explains why the raw sensitivity is so far below what may be expected for this size of sensor. (All that likewise applies to DSLRs.)

Hence what I said earlier - the lower performance is the price paid when a chip designed for stills use is used for video. The easy way to (much) better resolution would be to REDUCE the photosite count, yet be able to use them all. If you are designing the chip specifically for video use, the approach Canon have taken is the obvious way to go, 8 megapixels and direct read.

Re: AF100 successor?
 

The resolution of a sensor with n pixels in the horizontal direction and m in the vertical is n/2 lp horizontally and m/2 lp vertically. Period. If you present such a sensor a strictly band limited image (an impossibility) such that no spatial frequency above n/2 or m/2 are present then you can reproduce that image perfectly (ignoring aperture error). In the real world, of course, images are not strictly band limited by any means. If the image has spatial frequency above m/2 and/or n/2, as the test charts do intentionally, then the sampling theorem is violated and aliasing occurs. To combat aliasing in the real world requires an antialiasing filter i.e. something which blurs the image. MTF limitations in the lens does this to some extent but not sufficiently if good quality glass is used, not stopped down too far etc. So some manufacturers put a blurring filter on the sensor itself. Hardware antialiasing filters are not perfect as the trumpet photos you reference show. Bear in mind when looking at these that they have probably been sampled and resampled many times such that some of that aliasing will be caused by the resampling necessary to make the picture fit your display. Images can be introduced in upsampling regimes as well as down sampling. The fact that you see response at spatial frequencies above the ultimate resolution of the sensor says that energy with spatial frequencies above the fs/2 is reaching the sensor which means that the antialiasing filter is not removing those spatial frequencies. As I noted above hardware solutions are not that effective. The response of a hardware device cannot reproduce the 'boxcar' response the design engineer would love to have.

This said, if a hardware (optical) filter with a higher corner frequency were used, the resolution of the camera would go up. But the aliasing would be worse because the attentuation of frequencies above fs/2 would be less. In the early days of CCD photography the option of removing the antialiasing filter was given by some manufacturers. You decided whether you wanted a sharper picture or more moire. I guess you could argue that more resolution is not an improvement of it introduces more moire but then whether its better or not depends on the spatial frequency distribution of whatever it is you are photographing.

So if an optical antialiasing filter doesn't work, then how do you get a sharper picture without aliasing? The answer is to recognize that an optical filter can only knock the response down so many db/octave and over sample so that the folding frequency occurs at a high enough frequency that the response of the antialiasing filter is sufficiently down at the folding frequency. Oversampling is acheived by increasing the number of sensors on the chip. So now you have more data than you can use. What to do about that? The answer is a digital lowpass filter. These can be made to have much much better performance than optical filters. You can get the sharpness without the moire. But of course there is a cost. All that extra data has to be shuffled off the chip in a frame time and all that data has to be processed. The sharper the filter the more 'taps' it must have and as we are working with a two dimensional image doubling the number of taps requires a filter kernel four times larger. Memory, processor speed, power consumption all go up. All this is, of course, complicated by the necessity to derive R an B channels if it's a Bayer sensor and so on.

Thus increasing the number of pixels on the chip and using digital antialiasing filtering can increase sharpness without increasing moire up to the limits of n/2 by m/2. But at cost and that's why I said it is unlikely that Panasonic would do it.

Reducing the number of photosites and using them all is a nice idea but it's like trying to go faster than the speed of light. Physics won't allow it. Perhaps you can do better than a particular manufacturer is doing and that's what this game is all about but the limits are still there

Re: AF100 successor?
 

Not so simple, (assuming we're talking about colour, and a Bayer sensor) and the easiest way to illustrate why not is with the example of the C300.

In this case, a Bayer block of 2x2 photosites is treated as a single resolution unit, with values for R,G, and B. To get resolution matching the 1080 system it follows there must be 1920x1080 blocks - so 4x1920x1080 photosites in total. (8,294,400 - we'll call it "8 megapixel") The beauty of the Canon approach is that not only will it give full 1080 resolution for R,G, and B, (so 4:4:4 off the front end) but it will get it with very simple processing - no deBayering etc. Each block corresponds to a single output pixel of the image.

What therefore would be the point of moving up to a sensor with a higher number of photosites? If every photosite is still read, then not only does that impose difficulties in the reading (at videoframerates), but it will give a base raster of GREATER than 1080 resolution - which then needs downconversion down.

In practice, cameras such as DSLRs and the AF100 with high count sensors have to cheat. Their reason for using such sensors is that they are neccessary for high quality stills (for DSLRs), and in the case of the AF100 that the GH2 chip was readily available. In every case, they get round the problems by ignoring a percentage of the photosites. Early DSLRs did it in quite a clumsy fashion which gave considerable colour aliasing, such as the AF100 and GH2 are better in that they employ virtually the same technique as the C300 - reading out blocks of 2x2 as a resolution "unit".

The reason that the performance of the AF100 is so much poorer than the C300 is that it only reads out one block in two horizontally and vertically - so only one in four in total. Consequently, a resolution of a quarter of the total photosite count on both axes. Using the measured figure of 650 from Adams chart, that predicts a total chip count of about 2600 vertically and hence about 4620 horizontally. Multiply them together and you get 12,012,000, say 12 megapixel - as would be expected for such a sensor windowed to 16:9. (Look earlier in the thread and you'll see more exact figures and an explanation of why even the GH2 outperforms the AF100 for resolution.)

If you go the route of deBayering, downconverting etc, then it's possible to get full 1080 res with much less than the 8 megapixels of the C300 - at the expense of complexity, power, hence cost etc. But in that case the minimum necessary is about 3.5 megapixels - exactly as with the F3.

Yes, in this case oversampling may be beneficial for the reasons you mention to do with aliasing (as the Alexa does) - but there is absolutely no point in going even up to 8 megapixel, let alone 12 or above!
The "laws of physics" lower limits are about 3.5 megapixel if you fully deBayer etc, and 8 megapixel for direct read. That's what Sony are doing with the F3 in the former case, and Canon in the latter - and no, I couldn't do any better than those manufacturers. But using substantially more won't bring significant benefits for a 1080 output, and is likely to have the reverse effect. As the results from DSLRs and the AF100 prove.

The block skipping means that whilst about 3 million 2x2 blocks are AVAILABLE, only about a quarter (0.75 million) are USED. Far better to do what Canon do and have about 2.2 million and be able to use them all!! But then you can't make use of an already existing still camera chip......

Re: AF100 successor?
 

Rumor has it that the AF1000 or what ever they call it will be announced soon.
NAB 2015 starts April 13th.
New better 4/3 Sensor, Usable 12,800 ISO?

Twice as high as GH4 would trigger my buy button as it is maxed out for indoor poor light at closer to 1600 for me and I need to go higher.
I got rid of my HC-W850 and HC-X1000 because they were not as good as the GH4 in low light.
But I really miss the handling and controls of the X1000.
And I really hate the handling and workflow of my A7s.

Re: AF100 successor?
 

This is the only web rumor I've seen re a 101 replacement. Lifted verbatim, make of it what you will.

Chris Young
CYV Productions
Sydney

--------------------------

Panasonic AF101 Replacement Camcorder Coming on April 13, 2015

Published by dcnadmin on March 10, 2015

inShare
Panasonic will have a press conference on April 13 at the NAB 2015 show in Las Vegas. Panasonic AF101 replacement camcorder expected to be announced the same day.

According to latest rumors, the upcoming MFT camcorder – which is said to be an upgrade of the Panasonic AF101 – will be able to capture 4K videos and feature a newly designed 18-megapixel image sensor.

The camcorder with Micro Four Thirds image sensor to have a high price tag between $3,000 and $4,000, similar to the AG-AF100.

New Panasonic AG-AF 4K MFT camcorder coming on April 13

panasonic-af101-successor-nab-2015

Here are the previously rumored specs of the upcoming AF101 replacement camcorder.

4K video
18MP photo
16+ stops of dynamic range
Improved lowlight capabilities
Improved skintones
Improved highlight rendition
Useable ISO (almost no grain) up to 12,800
The new camera will basically have one of the best stills and videos images/color rendition/dynamic range compared to the competition
The big event for the new camcorders is the National Association of Broadcasters Show 2015, which takes places between April 11th – 16th.

Re: AF100 successor?
 

If it's 4/3 and 18MP sensor, the photosites must inevitably be quite small, certainly compared with a normal 4K (approx. 8-9 MP) s35 sensor.

In such case I find talk of 16+ DR and "almost no grain" up to ISO12,800 on the highly optimistic side.

It seems quite likely there may be some substance to the basic rumour, but it's most likely to be the basic innards of the GH4 in a more video friendly form factor. The AF100 was, after all, just that using the GH1 innards and GH2 sensor. I think such will be welcomed by many, and sell plenty well enough, but talk of "almost no grain up to ISO12,800" almost certainly an exaggeration in the rumour mill.

Re: AF100 successor?
 

I too doubt "usable" ISO 12,800.

I am all for a 4/3 interchangeable lens VideoCam but think that 12 MP or less could be optimized for lowlight performance much easier and with better results.

Re: AF100 successor?
 

Like most of the wild rumors that float around prior to any big broadcast show... I believe very little especially specs like these.

But there again who knows. Sometimes revolutions take place. I remember the first time I used a DXC-3000. A CCD camera that never lost it alignment on the tubes. Wow! Those CCDs created a whole new world in cameras. I will never forget sitting on a pavement in London prior to an interview trying to auto align a Sony BVP-3 that had it tubes go out of alignment after a knock.

Maybe there is something special at this year's NAB. We can always hope this year is the next big revolution. I'm a a bit skeptical though I must admit.

Chris Young
CYV Productions
Sydney

Re: AF100 successor?
 

Revolutions cannot undo the laws of physics. These specs are ludicrous. Wishful thinking by someone who doesn't know what they're talking about.

Re: AF100 successor?
 

As Gary says: Revolutions cannot undo the laws of physics.

And I'm afraid it seems to be an April Fool, or based on one and spread in good faith, anyway. I've been sent this link about an "about to be released GH5" Panasonic GH5: Feature Wishlist - the author apparently didn't originally say it was an April spoof - and a lot of the detail seems very, very similar to the "AF200" rumour above. Eg:
This is not to say nothing may be coming from Panasonic at NAB - if an AF200 is predicted every year, then eventually you may get lucky! :-) - but I'll put a pretty heavy bet that especially if 4/3 it will NOT be "Useable ISO (almost no grain) up to 12,800"!

Re: AF100 successor?
 

Well... we got one, but it ain't got a removable lens, and it's not part of the AF series.

Re: AF100 successor?
 

I went to NAB expecting to see the AF200 in the Panasonic Booth.
Instead, they were exhibiting the AG-DVX200 4K which does not fit my needs.
Across the isle was the JVC booth and the Camcorder I expected the LS300.