Philip Bloom's first test

yes philip please the only thing I'm interested is dynamic range. DSLR's have just about everything else down pat, gh13 has sharpness, canons have DoF. The only thing that truly sticks out as pure crap in the DSLR world compared to something like Red One or film is the bad DR. Can you test some shots outdoors in bright sunlight on people's skin see how it holds highlights etc...?
 
Homunculus said:
The only thing that truly sticks out as pure crap in the DSLR world compared to something like Red One or film is the bad DR.
Click to expand...

WHAT???

Most video cameras have a DR of 6-8 stops. DSLRs have been measured 10-12. That isn't a huge gap from the Red One's 13 stop range.
 
NextWaveG said:
WHAT???

Most video cameras have a DR of 6-8 stops. DSLRs have been measured 10-12. That isn't a huge gap from the Red One's 13 stop range.
Click to expand...

10-12? Try again. That's in stills mode. DSLR's all have DR of about 8.3, that goes for gh1, 5dmkii, 7d etc. This has been tested and proven in many different tests. This is not even close to Red MX DR, or even original Red DR of about 11.

You think non MX Red One has LESS DR than a DSLR? Are you kidding me? DSLRS's have about 8.x stops, and AF100 was already tested at 10.5 that's why I'm so eager to see if it lives up to that 10.5, now THAT is close to Red.
 
Philip, on the sensitivity vs. noise thing -- there's two mindsets on that. Some folks think you should equalize around noise, so that at any given noise level, when noise is equal, then you're on even footing to go comparing sensitivity.

Others think that the sensitivity is what it is, and the noise level is what it is.

I have traditionally fallen in the second camp, based on my experience with film. 800 ISO is more sensitive, and grainier, than 200 ISO film. That's just the way it is. You accept the higher noise level, for the better sensitivity. But I've been argued with quite a bit on that, and I do see their point -- if the question was "in any particular given light level, which camera delivers the cleaner picture?" then that seems to be an entirely appropriate way of measuring which one is more sensitive. Clearly the more sensitive camera should be the one with the less noise at the same brightness, as it can then be pushed to higher brightness while retaining similar noise...

So, anyway, two ways to look at it. And that might be how you'd get a case of "more sensitive but noisier", whereas other folks would say "well, then, that's not really more sensitive, because I could always gain up the other camera and add noise and brightness."

Nice matchlight test, by the way. And that's not with the fastest lens... gotta try it again with that Noktor .95, because that'd give you another .4 f-stops of light! That's crazy...
 
Jan_Crittenden said:
the factory is just saying that it is a new chip, very non-committal on which chip it is, but it just isn't the GH1 chip.
Click to expand...

I had read about this a while back regarding the GH2 sensor and video spec and can be found at http://www.eoshd.com/entry.php?29-Panasonic-engineer-reveals-workings-of-GH2-dual-exposure-sensor:

With Panasonic's recent camcorder update paving the way for 1920x1080 at 60p (Progressive) - the GH2's video spec certainly has to be impressive to stay state of the art, even by the company's own standards. Sony are ready to become a competitor once again too. (More on Sony's plans at the bottom of this article).

Panasonic engineer T. Azuma's talk at the International Solid State Circuits Conference (ISSCC) in San Francisco took place last week. (This builds detail on top of an earlier expose on sensor design made in advance schedule publications by ISSCC a few months ago).

His presentation dealt with Panasonic's innovations in sensor design for forthcoming digital cameras such as the GH2.

Here are the key points:

* Panasonic have developed a 4K x 2K image sensor (approx. 12 megapixel) based on 'dual exposure technique'
* The sensor is described as having new architecture that tries to improve sensitivity by a factor of 4 or more
* The architecture change is a new design which allows green pixels to be read independently of the blue and red ones on the sensor.
* The green pixels are exposed 4 times longer than the red and blue, thus achieving the sensitivity gain
* Motion artifacts are introduced as a result, but clever signal processing of the red and blue channels corrects the motion artifacts introduced in the green channel by the long exposure
* Red and blue channel gain achieved by pixel binning in 4x4 mode


The technique is similar to the Intelligent Resolution feature of the new Panasonic TZ10 sensor, but on a larger scale.

A factor of 4 increase in sensitivity will allow the Micro 4/3rds camera to surpass the full frame Canon 5DMkII. GH2 is currently slated for a September release date - partly due to delays last year in Panasonic commissioning their new Japanese sensor fabrication plant in Tonami and strategical decisions related to the worldwide economy and consumer electronics market performance.

I am told that since the fall of 2009, manufacture of the older GH1 sensor was transferred to the new Tonami plant to get things up and running there. Unforeseen circumstances resulted in a worldwide shortages of the GH1 at the same time period of the cross-over. These are now resolved and the plant is ready to produce the new GH2 sensor and other new Lumix G series image sensors. As a side note, I have noticed that newer stock of GH1s produce a slightly better image with less codec break up and banding than in the GH1's whose image sensors were produced at the older fab.
 
NextWaveG said:
Art Adams test claiming 10 stops of DR: http://provideocoalition.com/index....d_how_much_dynamic_range_does_it_have_really/

...and Shane Hurlbut has claimed on several occasions that he has squeezed 12 stops with the right lenses and settings.
Click to expand...

This has been discussed ad nauseam. There are about 12 other tests that all prove that DSLR's have 8 stops. And Shane Hurlbut has already been denounced as saying many strange things, such as that different lenses give him 2 extra stops of DR, that the compression in 5dmkii gives him very nice "skin tones" and that he loves the compression etc.

Seriously, the Red One (non MX BODY ONLY) costs 17,500$ you really think a DSLR has better DR than that? Red One is proven at about 11 stops or a littel more. The MX sensor added a couple stops to 13 or 13.5.

Go to DP review and check their tests, the DSLR's are doing those DR's in STILLS mode not video mode. You actually think compressed Video mode si giving you same DR as a RAW still? Come on buddy.
 
I've never run my own tests, but all I can say is that my 5D has given me more dynamic range than any other camera I've used. As long as the AF100 does as good or better, I'll be happy.
 
I think this test shows a lot quick and dirty...there is noise and banding..cleaner then the 7d at 3200 iso that's for sure.
On his face its pretty clean ....
Im impressed.

I was hoping Philip would do a re enacment of the T4 with out the swearing ofcourse :)
0.jpg
 
Dear Philip,

Thank you for putting together this quick "preview" of the AF-100. I look forward to seeing more of your tests.

As Barry mentioned before, it looks like the camera's noise falls mostly in the luma category and stays out of the chroma badness. This is most pleasing. It does look a little chunky, but that could just be compression - who knows.

I also have to wonder about cause of the purple halos. Lenses or sensor? Hmm...

It's a pity all we can see now is web-compressed, but I'm sure there will be a time later on for lots of raw footage.

Thanks again, Philip.
 
Last edited:
I do think the ISO performance is great! I like the quality of the "grain". Looking forward to seeing more, thanks for posting the footage Philip!
 
A good first test Philip and thanks, Panasonic is to be congratulated for getting to the market with a working camera that will change the shape of video.
Does anyone know what the SDI output will be, 8 bit or 10 bit, this camera with a nano flash or the soon to be released Atomos ninja a pro res unit could be awsome.
 
Barry_Green said:
Nice matchlight test, by the way. And that's not with the fastest lens... gotta try it again with that Noktor .95, because that'd give you another .4 f-stops of light! That's crazy...
Click to expand...

I have the Noktor but just find it a bit too soft wide open compared to the Voigtlander so that's why i went with the slightly slower lens...

Going to do tests with Canon, Nikon, Zeiss and Panasonic lenses...test out moire, aliasing etc.

Lots to do. Liking it very much so far!
 
What in the bloody green heck is DVCAm and how is it possibly more important than the most seminal camera in the history of civilization the mighty AF100?
 
Am I the only one here old enough to remember Barry Lyndon?

Am I the only one here old enough to remember Barry Lyndon?

Dear Mr. Bloom,

While others nit pick that bit of video that you posted, let me tell you that I am beyond amazed. I was immediately taken back to the drawing room scene in Barry Lyndon, the scene which was lit only by candle light, and photographed, by Mr. Kubrick's imploring, with special Zeiss lenses that were f0.7! The scene was lit to just 3 candlepower.

Very exciting— this test has tremendous implications for my work in the documentary field.

Mahalo!

Jeff DePonte
Honolulu, Hawaii
 
You must log in or register to reply here.
Back
Top