|
|||||||||
|
Thread Tools | Search this Thread |
December 19th, 2010, 11:25 AM | #1 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
Does the AF-100 Line skip?
Need to know more before considering a purchase.
|
December 19th, 2010, 12:02 PM | #2 |
Major Player
Join Date: Mar 2006
Location: Chicago
Posts: 706
|
Why does that matter if the output is good?
|
December 19th, 2010, 01:32 PM | #3 |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
It does not appear to. But neither do the GH1 or GH2.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
December 19th, 2010, 01:56 PM | #4 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
|
December 19th, 2010, 02:01 PM | #5 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
|
December 19th, 2010, 02:59 PM | #6 |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
We see none of the aliasing and moire usually associated with line skipping in the AF100 footage. Looks pretty darn good to me.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
December 20th, 2010, 06:55 PM | #7 |
Panasonic Broadcast
Join Date: Sep 2002
Location: Secaucus, NJ 07094
Posts: 271
|
there is no line skipping being used. It uses an optical low-pass filter to resolve the aliasing and moire that is typical from high count imagers.
Best, Jan
__________________
Jan Crittenden Livingston Panasonic Solutions Company, Product Manager for 3D and Handheld Cameras Last edited by Jan Crittenden Livingston; December 21st, 2010 at 07:11 AM. |
December 21st, 2010, 04:29 AM | #8 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
Thanks Jan. Could you also confirm the size of the sensor IE a 16:9 slice taken from a 17.3 x 13mm sensor?
Cheers Mark |
December 21st, 2010, 07:10 AM | #9 | |
Panasonic Broadcast
Join Date: Sep 2002
Location: Secaucus, NJ 07094
Posts: 271
|
Quote:
Best, Jan
__________________
Jan Crittenden Livingston Panasonic Solutions Company, Product Manager for 3D and Handheld Cameras |
|
December 21st, 2010, 09:11 AM | #10 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
Thanks Jan
All questions answerd Sounds really good now! Mark |
December 22nd, 2010, 12:33 AM | #11 |
Old Boot
Join Date: Aug 2002
Location: London UK
Posts: 3,633
|
Perfect Jan.
Thank you Grazie |
December 22nd, 2010, 04:38 AM | #12 | |
Inner Circle
Join Date: Jan 2006
Posts: 2,699
|
Quote:
There is no question that the OLPF will largely resolve the issues of moire/aliasing (it will), but this is quite independent of whether the chip is read in total or via line or pixel skipping. Saying that it uses an OLPF, and has little evident moire/aliasing, says little about how the chip is read. The AF101 is reckoned to have about the same native sensitivity as a 1/2" 3 chip camera, with comparable technology. If every photosite was being read, every frame, the expectation would be that the sensitivity would be (should be) much higher. Technically, the AF101 results are in line with not every photosite being read or used in every frame readout - line skipping or pixel skipping. If what is basically a still camera sensor is used, the alternative to line or pixel skipping is to read out the entire sensor (all 12.4 million pixels) every frame, deBayer, then downresolve with software filtering. The technical challenge of doing such in terms of complexity, power requirements, heat dissipation etc shouldn't be underestimated, and it is extremely difficult to see how they are overcome in a camera of this price, consuming relatively little power? From a designers point of view, far easier, sensible and more cost effective to pixel skip and deal with the aliasing with an OLPF. If "line skipping" isn't used, what about other forms of pixel skipping, such as in blocks of four? It would be interesting to point the GH2 at a zone plate in 40fps burst mode, and see what the patterns reveal. That mode restricts resolution to 4 megapixels, and is likely to use the same principles that the AF101 does - but since the GH2 doesn't have the OLPF of the AF101, the alias patterns should be unaffected and tell us more about exactly what's going on. |
|
December 22nd, 2010, 08:54 AM | #13 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
So it's clear now it does line skip.... Difficult to see how a camera like this wouldn't.... I have to wonder now about other decisions made in its construction and design and possibly if this was why it has an 8 instead of 10 bit out and how image quality is affected if a Ki Pro is used, and then undergoes a lot of post colour correction for example.
|
December 22nd, 2010, 09:43 AM | #14 |
Major Player
Join Date: Dec 2007
Location: Hertfordshire UK
Posts: 414
|
|
December 22nd, 2010, 11:12 AM | #15 |
Regular Crew
Join Date: Feb 2007
Location: London UK
Posts: 71
|
Wide angle for GH2 in Video Mode?
Sorry, I'm a bit lost! Which micro 4:3 (m4/3) focal length should one have to achieve a wide angle view on a GH2, in video mode? By 'wide angle' I mean an angle of acceptance similar to that of a 35mm or 28mm lens on full frame stills 35mm.
In fact I'm so confused that maybe this post should have gone in the GH2 thread; except I want to use the GH2 I just ordered, alongside the AF 100/1 that I have just ordered. (Rush of blood to head.) And is my old Nikkor 50mm only good for C.U.s of the Sea of Tranquillity? (On the Moon.) |
| ||||||
|
|