PDA

View Full Version : GH1, meet the Chroma Du Monde and ToNi charts (dynamic range)



Barry_Green
10-16-2009, 04:47 PM
Next test involved seeing just what kind of dynamic range I can eke out of a GH1. I used two charts to do this, a Chroma Du Monde (Billups VF/X version) ...
http://dsclabs-us.intelex.ca/UserFiles/product/BillupsVFX.jpg

... and a ToNi (toe and knee)...
http://dsclabs-us.intelex.ca/UserFiles/product/toni3.jpg


Both charts have grayscales in them. I wanted to push the charts way past the ability of the camera to resolve all the chips, so I put both charts in the same scene, lit them both with a couple of fluorescents, and then hit the top chart (the Billups) with an HMI and a 2K fresnel (because I blew out my 575w HMI bulb during the last test... sigh.)

The idea was to get the top chart way brighter than the bottom chart, so I could get a wide tonal range. Setting it up this way, I used my spotmeter to verify, and the cavity black on the ToNi registered at f/0.7, whereas the whitest chip on the Billups chart registered f/45! That's a massive range of over 12 stops, plenty enough to see what the GH1 is capable of.

I put the GH1 in "smooth" mode, which claims to have flatter contrast, and then I also turned down the contrast to -2. I shot at all various ISO settings and racked the iris and shutter speed throughout their ranges to expose the chart at various light levels.

My conclusion is that ISO 200 may be the best for dynamic range; it looks like it might be showing about 1/2 stop more range than ISO 100. Curiously, other than that, all the ISOs seem to produce about the same dynamic range.

I tested with Intelligent Exposure on and off. Intelligent Exposure does appear to rescue about a half-stop of shadow detail that would otherwise be crushed to black.

Overall, it looks like the GH1 can handle a range of about 9.5 stops maximum; without Intelligent Exposure on, that drops to about 9 stops.

stephenvv
10-16-2009, 05:27 PM
Thanks - that's pretty good DR. Any chance of HMC40 or HV30 or HV40 comparisons?

commanderspike
10-16-2009, 05:51 PM
Good test, would there be a chance of doing the same for the 5D and 7D?

philiplipetz
10-16-2009, 06:25 PM
Here is what Barry found on the 7d and HMC40; http://www.dvxuser.com/V6/showthread.php?t=186746 HOwever, there appears to be slight difference in the way he did these two tests. This one appears to not include digging into the images to find hidden details, so it may be more a measure of what he calls usable range. If this is true then the GH1 has greater dynamic range than the 7d by approximately two stops. This is so startling that I have to suspect there is some difference in methodology. I hope these two tests are not directly comparable; it would not speak well for the 7d that already showing artifical detail galore in the resolution tests, and large sections of false color in B&W charts.

"On a Stouffer, the Red One resolves about 11.3 stops, but most users seem to put the "usable" range at about 9. On the 7D, I'd say it resolves 9 stops, with a "usable" range of about 7.5, maybe 8. Not too shabby, seeing as the 7D costs 1/10th the price.

"How does it compare against a comparably-priced video camera? I ran the same test with the HMC40, and found the HMC40 (with DRS on, equiv. to HTP, in CINE-D) to deliver a range from about f/32 on down to f/1.7, or a total range of about 9.5 stops, and when digging into the shadows with Premiere's Shadow/Highlight feature enabled, you could technically discern all the way down to the f/1 block but that's really pushing it. It also brought out some cleaner separation of the top steps, adding maybe 1/2 stop at the top and bottom, bringing the total potential dynamic range to 10.5 stops. However, the shadows are so noisy and grainy that I wouldn't say it's got 10.5 "usable" stops. In terms of visibly usable stops, with DRS off, I'd call it 8.5 without DRS, 9.5 with DRS."




Good test, would there be a chance of doing the same for the 5D and 7D?

Ian-T
10-16-2009, 07:56 PM
Can we see the actual results from the ToNi chart like we did on the 7D?

Barry_Green
10-16-2009, 08:04 PM
Thanks - that's pretty good DR. Any chance of HMC40 or HV30 or HV40 comparisons?
I already did the HMC40, in the Canon thread. Don't have access to an HV30 or HV40.

Barry_Green
10-16-2009, 08:11 PM
Good test, would there be a chance of doing the same for the 5D and 7D?
Already did the 7D, don't have access to a 5d.

Barry_Green
10-16-2009, 08:15 PM
HOwever, there appears to be slight difference in the way he did these two tests.
The first difference is that the lighting on the charts is different because my 575w HMI bulb popped, so I had to substitute in a 2K fresnel.

The second difference is that the GH1 shows about 9 stops without rooting around using the Premiere Shadows & Highlights filter. Putting that filter on dragged another half-stop out of the low tones, did nothing for the highlights.

I wouldn't say there's two stops difference between these cameras (7D and GH1). The GH1 didn't show objectionable noise like the 7D did when pulling up the shadows & highlights, but that may simply be because the codec mushed the noise together!!

I would say the GH1 appears to hold an advantage, perhaps a 1-stop advantage. But really I'd have to shoot them both side-by-side on the same charts to see. 9.5 stops appears to be about the maximum on the GH1, with about 8.5 to 9 usable. And note, this was with Intelligent Exposure on -- with it off you lose about 1/2 stop. The difference between the GH1 and the 7D, in dynamic range, isn't much.

But again, a proper side-by-side would be warranted to see. I might try to get to that tomorrow.

ydgmdlu
10-17-2009, 12:29 AM
Interesting how the GH1 beats the 7D in dynamic range, when a lot of people have assumed otherwise...

We need to reevaluate what makes the "superior image" that we claim to see.

Ozpeter
10-17-2009, 02:35 AM
Interesting how the GH1 beats the 7D in dynamic range, when a lot of people have assumed otherwise...Barry seems to be saying there's not much in it and possibly that there's room for error as it wasn't a side by side test.

My eye is caught by the use of fluorescent lighting, which seems to be a light source that the GH1 isn't good with (in reviews, as I recall it). Could that have any bearing on the results?

commanderspike
10-17-2009, 07:16 AM
Interesting how the GH1 beats the 7D in dynamic range, when a lot of people have assumed otherwise...

I am sorry but I simply don't believe it does, because these two tests were not done together and the settings weren't consistent enough on the camera. iExposure? Settings? 7D's colour mode? Contrast setting? Too many variables for my liking.

I'm more prepared to believe that the 7D is closer to the 5D, but if the GH1 beats it in dynamic range then that does not back up what I found comparing the GH1 to the 5D, in that it falls a long way short in this department.

Also in the stills orientated reviews I've read about the GH1, they say it has less dynamic range than the competition. I can see this in my own photos.

Exposure is very important for a chart test like this. I'd recommend against using Auto iExposure in the test as it's exclusive to the GH1 and a software trick which simply boosts exposure by upping the ISO for certain areas of the image. The shadow detail that iExposure or the shadow & highlights filter reveal is of very poor quality and of ugly, low contrast, noisy appearance.

Barry_Green
10-17-2009, 07:47 AM
I am sorry but I simply don't believe it does, because these two tests were not done together and the settings weren't consistent enough on the camera.
Settings were designed to extract the maximum from both. So iExposure and highlight tone priority, contrast set to the lowest, etc. 7D was set according to Stu Maschwitz' "flatten your 5D" article.


Too many variables for my liking.
I'll do 'em side by side, but somebody isn't gonna like the results. That's the way it always goes, someone always has to complain that something or other wasn't set right, so now's everyone's chance -- if you've got suggestions on how to set 'em up for maximum dynamic range retention, make the suggestions now please.


I'm more prepared to believe that the 7D is closer to the 5D, but if the GH1 beats it in dynamic range then that does not back up what I found comparing the GH1 to the 5D, in that it falls a long way short in this department.
7D should be close to the 5D, and the GH1 should be close to both of 'em, if you go off of pixel size. The GH1 has a lot fewer pixels, so even though it's a smaller sensor the fewer pixels allow the individual pixels to be bigger. The 5D should be the best, in that even though it has the most pixels, its sensor is massively bigger.


Exposure is very important for a chart test like this. I'd recommend against using Auto iExposure in the test as it's exclusive to the GH1 and a software trick which simply boosts exposure by upping the ISO for certain areas of the image. The shadow detail that iExposure or the shadow & highlights filter reveal is of very poor quality and of ugly, low contrast, noisy appearance.
Same as highlight tone priority. The idea is to find out how much total range is possible, and how much usable range. In the GH1, I found that the iExposure added detail was nowhere near as noisy and nasty as the recovered stop I could drag out of the 7D; using Premiere's Highlights & Shadows filter I could yank a stop out of the 7D but the noise was overwhelming. So I classified that under total dynamic range, but I disqualified it out of "usable" dynamic range. On the GH1, the noise introduced is nowhere near as bad, which doesn't make sense until you factor in the codec, I believe the codec is mushing the noise together, whereas the 7D's codec is retaining the grit. Doesn't make it "usable" either, but it didn't look as bad as the 7D's additional stop.

Barry_Green
10-17-2009, 09:21 AM
because these two tests were not done together
I was thinking about reshooting it today, but I'm not really all that thrilled with the setup I have going anyway (trying to use two charts, etc). I have ordered a Stouffer 14-stop wedge (41 steps in 1/3-step increments). When that arrives I'll put both cameras to that test, that's an absolutely repeatable scenario that doesn't require any tricky lighting setups, etc.

stephenvv
10-17-2009, 09:45 AM
I am sorry but I simply don't believe it does, because these two tests were not done together and the settings weren't consistent enough on the camera. iExposure? Settings? 7D's colour mode? Contrast setting? Too many variables for my liking.

I'm more prepared to believe that the 7D is closer to the 5D, but if the GH1 beats it in dynamic range then that does not back up what I found comparing the GH1 to the 5D, in that it falls a long way short in this department.

Also in the stills orientated reviews I've read about the GH1, they say it has less dynamic range than the competition. I can see this in my own photos.

I'm inclined to say Barry's results gibe with the still-oriented reviews. The 5D does well in RAW mode extracting more DR- in JPEG, it's squarely middle of the pack as is the GH1. No 7D results yet and being a new sensor, old results don't help but based on comparison, it looks mid-pack as well.

All three have more dynamic range than slide film or most film print stocks, so getting nice images is more skill than tech.

Imaging Resource chart (with GH1 & 5D) (http://www.imaging-resource.com/PRODS/DMCGH1/DMCGH1IMATEST.HTM)

DP review does not have comparison charts but looking at the individual reviews shows that Nikon and Sony DLSR lead the DR. Perhaps it is the color and contrast mode that you like the way the 5D renders an image better. But the actual results show both 5D and GH1 in the 8.5 to 9 stops of usuable range with minor variations depending on tester, testing method, ISO and camera settings.

Isaac_Brody
10-17-2009, 01:48 PM
Didn't Hunter do a dynamic range test a few months ago? I remember him saying 8 to 8.5 stops of dynamic range.

Is it typical to get varied results like this?

PappasArts
10-17-2009, 01:54 PM
Didn't Hunter do a dynamic range test a few months ago? I remember him saying 8 to 8.5 stops of dynamic range.

Is it typical to get varied results like this?


He did indeed..

The dynamic range test chart 1080P frame here.
http://www.dvxuser.com/V6/attachment.php?attachmentid=10788&d=1241039195

The thread!
http://www.dvxuser.com/V6/showpost.php?p=1623056&postcount=1



.

Barry_Green
10-17-2009, 01:55 PM
It all depends on how it was tested and what settings the camera was in, whether you're going for maximum dynamic range or "usable", etc. Red tested their Red One at 11.3 stops discernable, but most users put the usable range at about 9.

A more accurate test will be done once my Stouffer wedge arrives.