Samuel H

[quote name='jgharding' timestamp='1347897281' post='18359'] Here's an interesting quote from Canon Rumors forum: [i]"According to Canon USA the image format of 6D is 35.8mm x 23.9mm, 5D2 is 36mm x 24mm.[/i] [i]I guess Canon just trim down the 5D2's sensor a bit. Maybe the AF is also modified from the 9 point AF in the 5D2. In this way, Canon is able to use these 4 years old parts to create a new body, reduce the overstocked 5D2 sensors and AF module and cut the cost a lot. Three birds with one stone."[/i] [i]"As it stands right now it's not really a FF cam - it's a crop camera with a 1.006x crop factor..."[/i] I suppose it won't be long til we find out. That would be seriously cheap, even for Canon. If we can find out the dot pitch of the sensor, then we should know... [b]EDIT: "With a pixel pitch of 6.55 microns, that could be a boon in the sensitivity department, as the 5D II was 6.4 and the 5D III was 6.25 microns." I'm assuming it's different then?[/b] [/quote] 5D2 is 36x24mm, 5616 x 3744 6D is 5472 x 3648 For those to be the same photosites on a firmware-cropped sensor, the active area on the 6D would have to be 35.1x23.4mm (1.0263x crop)

Yes, the specs are close to those of the 5D3... but they're even closer to those of the 650D (T4i). I want to hear about the aliasing/moire (or lack of it), and confirmation that the LCD screen resolution posted everywhere is a typo (102400 is the highest ISO for stills, and would mean 226x150 or so)

oh, the joy of shooting RAW!!! with the BMC, a better demosaic algorithm can get rid of that moire (better in the sense that it's geared towards reduced moire, instead of towards maximum detail) and if said better algorithm is released six months from now, you'll be able to apply it to the RAW footage you shoot today

(nearly) definitive solution to "will my glass work on that camera" troubles: buy vintage Leitz glass for the Leica-R mount It's not the cheapest vintage glass you can buy, but it's awesome glass indeed, and much cheaper than, say, Canon L glass. And it works on basically anything, including Sony A, Canon EF, Nikon F, and Pentax K: [url="http://www.similaar.com/foto/lensmount/lensmount.html"]http://www.similaar.com/foto/lensmount/lensmount.html[/url] The only downside is that you'll have a hole in the wide angle of things... the widest I've gone is 35mm, the 28mm, 24mm, etc, are not as awesome as the others, but are much more expensive But for a full frame camera, my set of Elmarit-R 35mm f/2.8, Summilux 50mm f/1.4, Elmarit-R 90mm f/2.8, Elmarit-R 135mm f/2.8 is just perfect: awesome image quality, sharp and with the best bokeh most people can buy, all for less than the price of a Canon 24-70 f/2.8L. Want to switch body brands? Spend $150 on a new set of adapters, and you're done.

I wouldn't be THAT optimistic. "We wanted to put real great PRO video into this camera" and "we added a continuous LED light on top of it" are somewhat difficult to mix together. I'd bet they've put more effort into AF than in using a great codec. As usual, wait and see...

If you can find a sensor that will deliver that, nothing prevents you from recording 20 stops of light on an 8-bit H.264 stream. You'll just have poor gradation, with something like 16 values per stop of light in the midtones. If you use a 10-bit codec, this problem dissappears, since you now have 64 shades per stop of light in the midtones. RAW is a completely different beast. It's not about dynamic range or color depth, but about preserving everything the sensor records, and making your choices in post. Of course, it usually also means 12-bit, 14-bit, or 16-bit, but it's usually on a linear color space, and I'm not sure I'd prefer 14-bit linear if I can get 10-bit log instead. Of course, you can also record 10-bit log RAW (the cinemaDNG standard supports things like this), and that would be my preferred option: the flexibility of RAW, with the gradation and bitrate savings or 10-bit log. And I want high bit depths even if my monitor can't display them, because I'll need them if I push the colors in post (otherwise, banding and codec issues will show up).

I think B is the GH2, and if that's the one you like, congrats, you can get the camera of your dreams in a tiny package and for a ridiculous price. I didn't like it, though: it looks much more "digital" than any other one. My take: My top 2 would be a tie between A and F: they both look great The next 3 would be tied too: C (which looks great except for the blown-out highlights), E (which has very slight aliasing) and H (which looked slightly too digital at times) After that would come G (somewhat soft, maybe less DR than others, some issues in the highlights) Then a tie between I (somewhat soft, less DR than others, lots of aliasing, some noise) and B (impressively clean -except for some slight aliasing-, but looking very digital) And finally D (horrible) Also, I think H is EPIC, G is FS100, I is 7D, B is GH2, D is iphone

all I was expecting from the new rebel is that it wouldn't have aliasing/moire, and it would have the new codecs (all-I and IPB) codec thing seems to be missing, and that doesn't bode well for the aliasing/moire thing I just bought myself a NEX-5N, if it's similar to the old APS-C canons but slightly improved, it will get me by while I wait for a blackmagic camera with a bigger sensor

great article and it confirms exactly what I was saying: for a true final 4K image, 4K Bayer RAW is not enough

it makes a lot of sense, but there are also a few silly notes in this rumor for example: [quote]It is 4K out of the box but from a 8K sensor. It downsamples for TRUE 4K unlike the Epic[/quote] if it records 4K RAW, it doesn't downsample to true 4K if it records 8K RAW, it downsamples to true 4K, but then it's not a 4K camera, it's an 8K camera! it doesn't add up

I thought this was a crazy idea when I first read about it, and still think it is after seeing your pics I have declicked and repaired lenses, repaired point-and-shoots, opened up laptops... but I'll stick to post sharpening, this is frightening  ??? ??? ??? plus: reading that the filter affects video but not stills makes me think it's not a standard passive filter: it's somehow controlled by electronics, and I look forward to having Magic Lantern solve the 5D3's softness issue in a much less aggressive way :)

[quote author=MattH link=topic=481.msg3119#msg3119 date=1332938713] You then say "This goes for contrast as well."  but I have to say in my experience I wouldn't agree with this.  I find you can easily add contrast to a dull image by raising the black level or applying curves, whilst it is difficult to do anything with a really contrasty image with crushed blacks and blown highlights.[/quote] You're absolutely right. What I mean is this: * if you only use chrominance values in the range [0,128] and push them back to [0,256] in post, you're using 7 bits instead of 8 for the color channels, and your final image will look worse * same way, if you only use luminance values [16-199] and push them back to [0,256] in post, you're again throwing away another big chunk of the codec's color space, and the effect on final IQ will be similar; and by doing this you're not expanding dynamic range at all (clipping points in highlights and shadows are the same in CineStyle and Flaat_10, but Flaat_10 uses all of the codec's color space, not just a small portion of it) (actually CineStyle uses luminance values [16-256], but reserves [200-256] for the brightest stop of light: 23% of the used color space is dedicated to 10% of the recorded DR, without affecting the point at which the highlights clip, which is basically the same in all picture styles I've tested, including "portrait with contrast=-1")

11 stops on the old cameras (Flaat_11 and Flaat_12) is noisy in the heavily-pushed-up shadows; some people say it's useable, some say it isn't this is the light response of Flaat_11: [img]http://www.similaar.com/foto/flaat-picture-styles/chart-waveform/flaat_3.jpg[/img] I count 32 unclipped wedges, which means 11 stops of DR (my synthetic chart works in 1/3 stops) the bottom 9 wedges are very noisy, not easy to distinguish from each other, but those are 1/3 stop marks, if you jump every two then they're clearly distinct; but that noise is nasty, so for many they're unuseable anyway; it's subjective, in any case the cleaner image of the 5D3 should lend itself better to these expanded-DR picture styles

thanks a lot for the recommendation!! I'll pay back saying the following here first: if the 5D3 gets Magic Lantern, this will probably unlock the extra DR that's available in stills mode but not in video mode so, without ML, all I can get on the 5D3 in video mode is 11.5 stops of DR, basically the same as with a T2i but with ML, all the DR captured by the sensor will (probably) be available in video mode too; that means 11.5 stops on the T2i, but clearly more with the 5D3 (the dxomark review is not out yet, but from my quick tests there's clearly more than the 11.5 I can get in video mode!!)

I've been working on Flaat v2 for the 5D3, here's what I've found so far: * the good news: picture styles for the old cameras work just as well on the 5D3; that includes CineStyle, and of course Flaat; cutting 5D3 and 5D2 footage shot with the same picture style should be relatively easy, at least in terms of light and color * the bad news: this means that, compared with my old APS-C canon, with the same picture style, dynamic range is basically the same; if the 5D3 has any extra DR to offer in video mode, it will require new picture styles * the even worse news: it's not easy to get more dynamic range with picture styles; the information is there, I can see it playing with the RAW stills in DPP, but no matter how hard I fight with the picture style editor, so far 11.5 stops seems to be the limit; unless Canon gives us a new way to control how the RAW information from the sensor is converted to video footage, the 5D3 has exactly the same DR as the 60D or the old rebel * the still unknown: even if DR is not any wider, useable DR may be better, given the 5D3's lower noise levels and hopefully improved codec; so if you have a 5D3, try Flaat_12p and let us know how it looks! you can download version 2.0 of my suite of Flaat picture styles here: [url=http://www.similaar.com/foto/flaat-picture-styles/index.html]http://www.similaar.com/foto/flaat-picture-styles/index.html[/url] you'll see there's a name change, now they are called by the number of stops of DR that they get: Flaat_1 is now Flaat_09 (slightly modified) Flaat_2 is now Flaat_10 (very slightly modified) Flaat_3 is now Flaat_11 (very slightly modified) Flaat_4 is now Flaat_12 (very slightly modified) (it doesn't get to 12 stops, but it's more than 11) also, there are two sets of picture styles: * one based on Portrait, for nice skin tones, but with some color shifts (e.g. blue goes a bit towards cyan) * one based on Neutral, without those color shifts I always use the ones based on Portrait, the others I only made them because some people wanted to avoid any color shifts. I'd rather have nice skin tones than correct blue, but it's your choice. Also, note that the recommendation for Tone has gone from =0 to =1.

you should not use unsharp mask on the 5D3 - you should use a convolution sharpening filter (in PP it's called just "sharpen") - here's why: [url=http://www.dvxuser.com/V6/showthread.php?278332-5D-Mark-III-Plenty-Sharp-and-Still-Anti-aliased-%28video-examples%29]http://www.dvxuser.com/V6/showthread.php?278332-5D-Mark-III-Plenty-Sharp-and-Still-Anti-aliased-%28video-examples%29[/url] also: the fact that CineStyle works on the 5D3 doesn't mean it will work well with it; it may, or it may not I'm making a version of my Flaat picture styles custom-tailored for the 5D3, it should be out this week if my picture styles need tweaking, CineStyle will need too; I don't know if it's the case yet; I'll post my findings around as I get results

I still have to create my Flaat picture styles for the 5D3, those could give this camera a clear win in the high-DR niche I would need a set of very specific (but easy to shoot) RAW stills, if you shoot those for me you'd be the first to use the resulting picture styles  ;) (I would also need to have a version of PSE and DPP that can work with those 5D3 files, and I'm not sure the updates at the canon site are up yet)

I'm extremely disappointed, for two reasons: * Interlaced? really? That's so 20th century... Thanks, but no, thanks. * See those horizontal lines in the notebook page? See how they randomly change from green to red? Guess what that is. Exactly. Any camera movement and you'd be watching full-rainbow psychedelic footage. Can't wait for the brick wall tests to appear. By the way, completely off topic: I tweeted you about a short film shot with my Flaat picture styles for Canon DSLRs. Did you get to watch it? I think it looks absolutely lovely, and most commenters seem to aggree :) [url=http://www.similaar.com/foto/flaat-picture-styles/real-world-tests-3.html]http://www.similaar.com/foto/flaat-picture-styles/real-world-tests-3.html[/url]

very clear explanation indeed my point was that, given a level of background noise in the sensor (as you explained, created both by heat and by inefficiencies in the ADC), a bigger sensor catches -overall- more photons, so the signal-to-noise ratio is boosted but also that, given the GH2 uses a much more modern sensor (the background noise is lower), the end result can go both ways (of course, the line-skipping thing noted above is a very big deal, as it nearly equalizes the number of photons hitting the used sensor area in a 5D2 and a GH2)

^ correct

re: sensors for low light first: with a lens at f/1.4 on both cameras, the 5D2 gets roughly 4 times as many photons hitting the sensor than the GH2, but that doesn't mean the final image will be any brighter (check my previous post); what it should mean is that the camera should be able to go to a higher ISO without displaying a lot of noise (that's why the 5D2 can get to a higher ISO than its APS-C brothers without more noise); but the 5D2 is quite old technology by now, and new cameras (GH2, 5N) can give you images with more detail and less noise even starting out with just a fraction of the photons also, on the Bayer pattern thing: even better than a random array would be to have clear photosites (white, instead of R, G or B) Kodak already has a few patents regarding this and I have this post proposing black (ND) photosites too (in order to protect highlights and increase DR) and a ratio of color-to-white photosites and an ordering that delivers exactly the amount of color information that 4:2:2 color subsampling will keep [url=http://www.similaar.com/foto/bayer/rgbwk.html]http://www.similaar.com/foto/bayer/rgbwk.html[/url]

one thing to take into account is that "ISO" doesn't mean the same on all cameras * GH2 at ISO 12800 can be brighter than ISO 20000 on the C300 * FS100 at 24db (ISO 6400) can be a lot brighter than any other camera at ISO 6400 (really, a lot), and even significantly brighter than the C300 at ISO 12800 * NEX-5N can be a lot brigter than any other camera at ISO 3200 (including FS100 and GH2) My ordering in terms of brightness would be: 5N > FS100 > GH2(a) > F3 > 5D2 >= 7D >= AF100 = GH2(b) > C300 where GH2(a) is above ISO 3200, and GH2(b) is below ISO 3200 (this one seems to be very non-linear, starts out dark at low-ish ISO and gets quite bright at high ISO, probably because it captures less latitude) [url=http://www.similaar.com/foto/iso/pbtest.html]http://www.similaar.com/foto/iso/pbtest.html[/url] of course, the main thing is how much noise there is in the images, not how bright they are, but still: it's something to consider when doing comparisons

this is a great indication that digic5 on its own is not enough but that doesn't mean we have to lose all hope: what we need is a faster sensor, the 1DX seemingly got it, and this one on the G1X is not the one that will be on the new APS-C canons yes, the one in the new APS-C canons could be just as bad as this one, but it could also be great (say: a smaller version of the one in the 1DX: same technology and pixel count, but smaller photosites) I know we're all losing patience with Canon, but, again: wait and see...

so there seems to be line-skipping on the G1X luckily, that doesn't mean there will be line-skipping on 1DX, 5D3, 650D, etc: line-skipping is there on current models mainly not because of digic-4, but because of the slow sensors they have; for example, the 18mpix sensor on the 7D, 60D, 600D, 550D has a read-out time of 60ms, which they can cut to around 20ms by reading out only one out of every 3 lines in the sensor (leaving it at 60ms would lead to incredibly horrible rolling shutter issues) if the 650D shares that same sensor, it will have to do line skipping too, no matter if it has digic4 or digic5 (I think) if it has a new sensor with faster read-out speeds (like the one on the 1DX, I'd hope), it could do without line-skipping (and only in this case, there's a new condition: ...only if digic5 is able to process all that information fast enough)

that's why Flaat_1 and Flaat_2 are there: they provide no more DR than Neutral with contrast=-4, but reassign the luminance response more smoothly (devoting less to the highlights and more to the skin tones) this is Flaat_1 (green) vs Neutral-4 (orange): [img]http://www.similaar.com/foto/flaat-picture-styles/others/compare_flaat1_neutral4.jpg[/img] and this is Flaat_2 (green) vs CineStyle (orange): [img]http://www.similaar.com/foto/flaat-picture-styles/others/compare_flaat2_technicolor4.jpg[/img] that slow slope line on CineStyle, as you said, leads to some issues - Flaat_1 and Flaat_2 should avoid them, because they're not flat in the sense that there's no contrast in the image, they're flat in the sense that contrast in the image is distributed evenly Flaat_3 and Flaat_4 deliver extended DR, but should not be considered a day-to-day tool, because that increased DR comes at a price - they're specialty tools, for those special times when you need them the workhorse option would be Flaat_2 also: * by being based on Portrait, Flaat delivers nicer skin tones (looking at the vectorscopes at the end of my page, it seems to do some sort of "skin squeeze" and "skin smooth", like what you'd find on Magic Bullet Mojo and Cosmo) * by not being overly desaturated, they should lead to smoother color transitions and nicer colors in the final image than other flat picture styles, specifically CineStyle * talking about final image: this smooth gamma response makes the footage really easy to grade: use a luma curve -or RGB curve and desaturaion- to redistribute contrast and get quickly to a reasonably good starting point