AF-100 Wishlist/Suggestion Discussion Thread

Status
Not open for further replies.
grimepoch said:
Most people answered already, but for me the purpose of a windowed mode would be for smaller lenses and/or deeper focus with more light on a cropped view of any lens.

That said, to do this, it would require a selectable OLPF and is that value good or bad?
Click to expand...

If you window a sensor, the OLPF can stay the same because the pixel/photodiode density of what you are sampling stays the same, you're just reading out a smaller area at its full native resolution.

If you sample the whole sensor but reduce the resolution through line skipping or binning, then ideally you would want to change the OLPF.
 
Jan_Crittenden said:
I am assuming you mean like a Fujinon or Canon Zoom or something like that and right now, that looks like it might be a little tough right out of the box, as those lenses draw power from the camera for the zoom motor. There isn't enough power available on the camera to do that. I have engaged in a conversation with a 3rd Party manufacturer that is interesting in coming up with a solution. That will be easier to say sometime in October/November.
Click to expand...

I'm beginning to think that an ENG lens for a large-format camera like this one would likely have a maximum f-stop of f/4. Also, I think that it wouldn't be that bad.

First, why do I think it's f/4? Size. There simply seems no way around the physics of lenses (unless someone finds a major new design). Most people, by ENG, mean a lens with like a 13x zoom (Although I myself could probably get by with 5x, and I don't really need anything more than 11x. For reference, 11x could bring you, in full-frame focal lengths, 28-300mm).

Now I am not a lens expert, but I have shot film and video for about 15 years, and I have been paying more attention to why lenses are bigger or smaller than others. One of the biggest correlations to size is, as many of you know, maximum f-stop. And if you do the math, you see why. To go just one f-stop faster, your aperture's diameter grows about 40%. And when you compute the area of a circle, the hole gets nearly twice as big for each f-stop.

100mm lens
f/2.8 = 36mm aperture = 1,017 sq. mm
f/2 = 50mm aperture = 1,963 sq. mm

This is partly why those Hollywood zooms are huge. They often open up to f-2! Meanwhile, our nice, portable SLR zooms often max out at f-4. F-stop also apparently adds even more to cost. Last year, Cooke came out with a new line of lenses called Panchros (not to be confused with a really old line of their lenses also called Panchros). Now these new Panchros have all the optical capabilities of the "regular" Cookes --- no breathing, great mechanics, excellent sharpness (if you have read anything about Cooke lenses you know that they are great). The only difference is that the new Panchros max out at f/2.8 instead of f/2. Yet Cooke said this saves 8 times the cost. 8 times! For one f-stop.

Anyway, back to size, every lens I've seen with even a 5x zoom for this format that also does f/2 or f/2.8 is at least 10 pounds.

Another reason to think f/4 is if you look at real 2/3" ENG lenses and scale up. Most 2/3" ENG lenses max out at about f/2. To make an equivalent 4/3 lens, you basically double the focal lengths. So a 7-126mm (18x) lens becomes a 14-252 4/3 lens. If you don't enlarge the iris, then, it is an f/4 lens. It still might be a bit larger, since the focal lengths are longer --- although at this point lenses get funny. You would think that a 100mm lens would be twice as long as a 50mm one. But lens designers must know something we don't. Looking even at Hollywood prime lenses, where they take no shortcuts, they are all about the same size and weight no matter the focal length (for example, again, the venerable Cookes: http://www.cookeoptics.com/cooke.nsf/products/s4i_specs.html). Or look at two of RED's zoom lenses. The 18-50mm is exactly the same weight as the 50-150mm, and the 50-150 is just 40% longer, not 100%. Or look at all the 2/3 ENG zooms. It doesn't seem to matter whether you get a 13x or a 20x, they're all about the same size and weight.

All that to say, you could probably make an ENG lens for 4/3 the same size as ENG lenses for 2/3, if you keep the max aperture at f/4. (Which, mind you, is still too big. Most ENG lenses are 4 lb. For the AF100's handheld style, you really want something more like 2-3 lb. This is where we look back at our needs and say we don't need the normal 13x or 18x zooms and could have just a 7x zoom or something).

Now, why isn't it that bad, to max out at f/4. Most would object that f/4 is (1) too much depth of field, and (2) too dark. Let's look at the objections one at a time.

Too deep. What's the point of 4/3 if you're right back to the depth of field in 1/3" cameras? Well, there are still advantages, of course. When you want shallow focus, you could always screw on a prime for that shot. And do ENG shooters want that shallow of a focus? Well, as it turns out, they're gonna get it, even at f-4.

For the same angle of view, a 4/3 lens at f-4 is just as shallow as a 2/3 lens at f-2. In the 1/3 format, you would have to use an f-1.0 lens.

Furthermore, I've heard from more than one place that Hollywood cinematographers don't like to go below f-4 anyway. That super-shallow look at f-1.4 or whatever is not exactly realistic, and pretty hard to pull focus at. Point being, if you are just seeking the Hollywood look, shoot around f-4 or f-5.6.

Too dark. Especially for ENG shooters, who often shoot in natural light, you need all the help from the camera you can get. But I say that, for exposure, f-4 is also going to be the new f-2. Until this century, most film and video cameras normally used ISOs of 100, 200, 400. The latest CCD and CMOS technology can do better than that. For example, RED is saying its new M-X sensor can baseline at 800. Now I haven't used a GH1, but how does it look at ISO 800? Well, f-4 at ISO 800 is the same as f-2 at ISO 200.

Also, the AF100 should be just the first generation. I really hope this format takes off. If so, they should stop re-using DSLR sensors and use a chip made for video, with no more pixels than it needs. Such a sensor should easily achieve ISO 1600.
 
Last edited:
Ironic that I'd start with a DVX100, graduate to a Red, and come full circle back to panasonic. This sounds like it could be the camera for me. Right price. Right features. Right form factor. I'm very interested.
 
Redundant Recording to both flash cards

Redundant Recording to both flash cards

If the AF-100 will record redundantly to both card slots that should eliminate any card failure issues.
Click to expand...

I think this is an important feature, and something we should make sure Panasonic knows we want. If it can redundantly record to both cards simultaneously then you can buy two 64gb SDXC cards and be pretty close to the reliability of a 64GB P2 card. A P2 card, as I understand it, has two solidstate hard drives in it in RAID-0, so if one dies you have a backup.

The likelyhood of two SDXC cards dying within minutes of each other is very very low unless the camcorder itself frys them. P2 cards are also more reliable with regards to temperature/moisture etc. but for most users two SDXC cards redundantly recording will be enough reliability. This is one of the big selling points of the new Sony HXR-NX5U.

Of course this doesn't change the AVCHD vs DVCPROHD/AVCINTRA codec debate but for many users redundant 64gb SDXC cards with high bitrate AVCHD is a very attractive option + pricepoint.

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

I know that the camera is already physically designed, but as for the firmware, can I please request redundant simultaneous recording to both flashcards ???
 
Redundancy would be an excellent feature that would certainly put people at ease. It should be east enough to implement.



Nityananda_SCSMath said:
I think this is an important feature, and something we should make sure Panasonic knows we want. If it can redundantly record to both cards simultaneously then you can buy two 64gb SDXC cards and be pretty close to the reliability of a 64GB P2 card. A P2 card, as I understand it, has two solidstate hard drives in it in RAID-0, so if one dies you have a backup.

The likelyhood of two SDXC cards dying within minutes of each other is very very low unless the camcorder itself frys them. P2 cards are also more reliable with regards to temperature/moisture etc. but for most users two SDXC cards redundantly recording will be enough reliability. This is one of the big selling points of the new Sony HXR-NX5U.

Of course this doesn't change the AVCHD vs DVCPROHD/AVCINTRA codec debate but for many users redundant 64gb SDXC cards with high bitrate AVCHD is a very attractive option + pricepoint.

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

I know that the camera is already physically designed, but as for the firmware, can I please request redundant simultaneous recording to both flashcards ???
Click to expand...
 
Redundancy

Redundancy

bgundu said:
Redundancy would be an excellent feature that would certainly put people at ease. It should be east enough to implement.
Click to expand...

Believe we mentioned the redundancy issue in the official wishlist suggestion thread and it is covered there. Hopefully they were listening!
 
is it just me or do the buttons in that 'leaked' foto all look like they're recessed and not very easy to push in?
wish the lcd was bigger...it had better be higher resolution!
 
OK - after reading through all of the many many pages on this thread (thank you "Show Printable Version" and "view 40 threads on the same page" buttons), here is my 2c:

1. SENSOR
I hope that this is not just the GH1 sensor with a different anti-aliasing filter. Whatever pixel binning technique the GH1 uses is good - but doesn't really resolve 1080p. I'd expect this camera to resolve 1080p properly. Hopefully some kind of new (GH2?) sensor combined with a good OLPF would do the trick?

2. 60i BUT NOT 60p WTF??
If the camera can do 1080 60i, then the sensor is running at 1080 60p and they are throwing away vertical lines, right? Can they not throw away the vertical lines and give us 60p instead?

3. FOR 3D RIGS (for those shots with interocular distances that the AG-3DA1 is not suited for)
- old-school way: do genlock and offer lens control (LANC or whatever) for electronic 4/3rds lenses (we don't all want to buy a C-motion system for focus although I'd imagine redrock etc are working on stereo manual lens control motors - still it's heavy and a pain)
- OR just have a "camera sync cable" which syncs everything between two cameras (or more than 2 for those making bullet time rigs :) - ideally with a "gain trim" function to compensate for one camera getting slightly less light than the other in the beamsplitter rig

These features may add to the cost of the camera - but hey Panasonic, you'll sell more 3D screens if there is more content.

Bruce Allen
www.boacinema.com
 
bruceallen said:
1. SENSOR
I hope that this is not just the GH1 sensor with a different anti-aliasing filter. Whatever pixel binning technique the GH1 uses is good - but doesn't really resolve 1080p.
Click to expand...
True -- but that's the GH1, a $600 consumer camcorder. Who knows what the limited processing power it has can handle, as compared to what a pro $6,000 camcorder has in it? The sensor might be capable of much better results, but the GH1 might not have the oomph to deliver it.

I'd expect this camera to resolve 1080p properly. Hopefully some kind of new (GH2?) sensor combined with a good OLPF would do the trick?
Click to expand...
I think we all expect proper crisp sharp 1080p. I don't really care how they do it, as long as it delivers.

2. 60i BUT NOT 60p WTF??
Click to expand...
1080/60p is not a broadcast standard. No broadcast camcorder on the market does 1080/60p. There isn't a professional recording format on the market that supports 1080/60p. If they give us 60p (and I hope they do) it would be nonstandard.

If the camera can do 1080 60i, then the sensor is running at 1080 60p and they are throwing away vertical lines, right? Can they not throw away the vertical lines and give us 60p instead?
Click to expand...
No, not necessarily! Look at the Canon XLH1/XHA1 series. They scan their chips in interlaced fashion. They only scan half the chip in one field, and the other half the other field; it's actually scanning the chip in interlaced mode. EX1 does the same thing when in 1080/60i mode. Scanning a chip in interlaced fashion has been done for decades, and actually has some benefits (lower noise, better sensitivity, plus some chips just can't be scanned that fast progressively anyway.
 
Points well taken, Barry!

Thank you for taking the time to read through my post and explain everything with such patience and care. I need to post here more often - no way I'd have gotten that response on some other forums :)

Ah well, can't wait for Dec 2010. Depending on 3D / high speed features, it's between this cam and Scarlet for me. But then I will be shooting 3D as well.

Oh, forgot to mention: hope they're putting in a good viewfinder for the AF-100!

Bruce Allen
www.boacinema.com
 
LuckyStudio 13 said:
True 10 bit Sdi output and not padded 10 bit out.
Click to expand...

The still mode on a dSLRs barely produces 8bits of luma and 7bits of chroma over ISO100. 8 bit YCbCr is ok.

Check the DXO results on the GH1:

http://www.dxomark.com/index.php/en/Camera-Sensor/All-tested-sensors/Panasonic/Lumix-DMC-GH1

8bit tonal range in best case (lowest ISO), 21bit color sensitivity (lets say 3x7bits) in best case. On higher ISO, it's less than that.

A 10bit container leaves more room through post though, so it has advantages. But nothing produces 10bit chroma really. Not even medium format backs. The high bit precision is on RAW/linear domain. After adapting to video color space and applying white balance, a full 8bit RGB can only be achieved by one or two medium format digital backs and an 8bit YCbCr covers just about any camera on the market that has a normal gamma curve.

The advantage of the uncompressed output is not the 10bit support, it's the lack of compression artifacts which preserves the image quality down to the last bit. 8bit YCbCr on 10bit YCbCr container will be wonderful:)
 
Barry_Green said:
1080/60p is not a broadcast standard. No broadcast camcorder on the market does 1080/60p. There isn't a professional recording format on the market that supports 1080/60p. If they give us 60p (and I hope they do) it would be nonstandard.
Click to expand...

Since it's being sold as a digital cinema camera it makes sense that 1080/60p should be included regardless of its place in broadcast (is 1080/24p a broadcast standard?) so people can overcrank it and get slow motion. And yes I know you know that and is why you hope they include it.
 
We all hope for it, I've kind of demanded it, and I've explained to them in no uncertain terms that hey, if a TM700 can do it, then by gum the AF100 should do it too.

But, with that said, there are certainly reasons why it may not have it.

I certainly hope it does.
 
People mentioned the Lumix G Vario HD 14-140mm F4-5.8 OIS as a possible kit lens. What about the M Zuiko Digital 14-150mm F4-5.6? Doesn't have optical stabilization, but it sounds interesting. Here's dpreview's review.
 
Status
Not open for further replies.
Back
Top