October 21, 2014, 12:35:43 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Aglet

Pages: 1 ... 26 27 [28] 29 30 ... 70
EOS Bodies / Re: Black-Cap event
« on: August 31, 2013, 01:37:59 AM »
Only 4 stops? Are you sure that's enough?  Or is that just the limit of the exposure slider?
yup, it's enough, and it's the limit of my slider.  So it's pretty consistent.

..maybe shooting in bizarre, oligochromatic light would be a good way to test auto WB?

Maybe.  If you'd like to develop a technique, fill yer boots.
Don't forget to be mindful of the filters peak admittance and asymmetrical shoulder responses and the effects that has on QE & debayering algorithms and how they differ from mfr to mfr and model to model .. and temperature.

Well, the PhD in neuroscience certainly suggests I'm a 'real' scientist.  So do the publications listed on my CV.  As for time, science often isn't a 9-5 job, which is great for me because I'd really dislike something so mundane.  But thanks for the snide remark.  Looking forward to more lens cap shots....but be careful - if you take too many, you may find yourself forgetting to take the cap off for an actual picture, and that would be a great loss.

It wasn't intended as snide.
I don't pay too much attention to you so don't know if you're for real or just a wannabe neuroscientist with a nifty avatar.
Just seems kinda odd you spend probably as much time on this forum as you do being a scientist... None of the science PhDs I work with divert so much of their energy in alternative directions.  But they're not in neurology; is that where the money is?

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 11:12:31 PM »
New thread with 70D lens cap tests and text interpretation.  sorry no time for visuals tonite.


Happy long weekend everybody.
Go shoot something!   w i t h   a   C A M E R A.

EOS Bodies / Re: Black-Cap event
« on: August 30, 2013, 11:09:23 PM »
70D in stock at my local store.
Goin' in for some lens cap shots.

Case in point - this is how Aglet evaluates a new camera: 'pictures' with the lens cap on.  He'll then boost the black images 4-5 stops, and tell us all how crappy the 'pictures' are, and how Canon still hasn't made their sensors cameras any better.


raw files are pushed 4 stops

Do you look for bio-luminescent markers with the room lights on?

(BTW, are you a real scientist?...  cuz you sure spend a LOT more time on this forum during the work day than I'd like to see any employee of mine doing)

EOS Bodies / 70D black-cap tests - Now with samples
« on: August 30, 2013, 11:02:53 PM »
Lens cap shots.  You bet! :)

Well, I had to process them using DNG-converter 8.2rc but that's likely adequate for now.

What I can see is that the noise structure of the 70D's sensor is still typical Canon; lots of noise in the red and blue channels and still a fair bit in green but the noise structures at base ISO are mostly fine-grained and should respond well to NR software.

There is still some larger structured noise banding tho it's mostly broad horizontal banding and unlikely to severely impact most images.  Dark-sky photographers pushing to look for dim details my be about the only ones who would run into problems.

edit:  BUT - I accidentally took the first 100 ISO dark shot with the info display activated, so rear LCD was ON. The resulting noise levels were much higher (~ between 400 & 800 iso dark shots) than the next 100 iso shot when I turned the display OFF.  The 100 iso dark shot with LCD display ON had much more red channel noise than any of the other shots.

200 and 400 ISO are similar but the noise structures are (more balanced) with both H & V elements.  Again, these are quite fine and will probably respond very well to typical NR software.

Noise starts to pick up at 800 ISO, noise structure similar to iso 400

1600 and 3200 iso show increasing overall noise levels as expected.  Noise structures become a little more obvious but, again, would respond well to NR software and only pushed large areas of continuous shades or tones may bely some of this non-uniformity.

6400 and 12800 are pretty noisy and continue to show some banding structures in both H & V axis but


FPN exists but is at a workably low level and may even be comparable to Toshiba's new sensor as far as noise structures.  However, overall TOTAL NOISE is still much higher than the Sony or Toshiba sensors and this will reduce effective maximum DR somewhat but that can be partly negated with NR software.

I would not steer anyone away from using this camera, it's comparable to the 6D for noise quality and certainly bests my 60D and blows away the 7D I used to have.

If you need the ultimate in clean black levels in your raw files then this is still not the camera for you.

BUT, using some decent noise-reduction software you can comfortably use the 70D in ways that the 7D and 60D before it would disappoint.  The 70D is "good enough" for most uses unless you're as picky as I am.  I know I could still find some situations where this camera might not perform as well as I'd like.

I could also not find any significant variation in noise structure I could attribute to the areas that were normal pixels VS. the AF pixels so they've done a really good job on that.

Polite applause for Canon.  They've certainly made some progress in this model.

EOS Bodies / Black-Cap event
« on: August 30, 2013, 06:24:17 PM »
70D in stock at my local store.
Goin' in for some lens cap shots.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 05:44:45 PM »
I've done well to stick to the topic of 70D and DxOmark, I think.
I've merely presented DxO's data in a way which tellingly demonstrates just what they measured on the 70D's sensor in comparison to what they measured to a 10 year old model. 
A per-pixel level comparison is important because this is what defines the quality of the electronics in the system.

Can we imagine how fabulous the 70D would be if it also had 2 full stops better SNR or DR?..  I'd be drooling!  The capabilities of such a CAMERA would be truly class-leading.
As it is, there's just a lot of compromising.

@Neuro; sorry, your big corp style capitalist justifications are not relevant when talking about sensor tech unless they're invested more of those profits into R&D or patent licensing to improve the foundation of the actual product, it's sensor technology.  And coming up with RESULTS.  It seems dividends were more important than technology investment in the last 10 years.
Can't argue with success tho, Canon's doing very well in that regard.

I will repeat, what they've done in the 70D with the AF ability is very impressive.  I just wish they'd also bring more than minuscule improvements to the base stills performance of ALL their sensors.

FWIW, 2012-2013 saw me liquidate a lot of newer Canon gear and give that money to Ricoh-Pentax.
I'm now only looking for a high performance crop body for long glass, dim light and high fps rates.
Will it be a 7d MkII, a D400, a K-3(?), or possibly even a new mirrorless system?
If the old 7D had less FPN, it'd have stayed in my inventory.  If the 7d2 has a vastly improved sensor, it'll top my list.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 03:27:36 PM »
The 70D...has not improved and is slightly worse than the 9 year old 20D. 

Oh, I think it's a little better.  I hope your analysis really was quick, because I'd hate to think you wasted even more time.  Per-pixel SNR?  Funny, I haven't seen that phrase on the display placards at Best Buy or my local camera shop.  I wonder why?  I know...because notwithstanding a minuscule number of DR-obsessed Canon-bashing forum jockeys, no one who buys cameras cares.  The 70D is a massive improvement over the 20D in 99.9999% of ways that matter to people. Canon will sell loads of 70D bodies, quite likely more than the D7100 by a wide margin.

DxOMark measures sensors, but people buy cameras, not bare silicon sensors.  You can rehash DxOMark data until hell freezes over, it doesn't change the fact that Canon has been outselling Nikon for years, nor the fact that the 5DIII outsells the D800.  The obvious conclusion is that 'better' sensors (where 'better' is defined as low ISO DR) have not helped Nikon or Sony sell more cameras.

So. . you're waving a flag and cheering because the sales/SNR charts for the last 10 years look good?..  ;D

perhaps if Canon's sensors improved as much as their oft' touted sales figures these recurring discussions wouldn't exist.


I think the time I spent putting together the animated graphs and essay demonstrates that perfectly. ;)
Don't forget, I didn't just compare the 70D in a previous graph with the d7100, I deliberately chose the sub-$600 consumer-class D5200. ;)

But yes, I'd prefer a 70D to a 20D for the wealth of other improvements and the fact that, at least at base ISO, the performance is comparable with higher resolution and the few higher ISO levels I use are also still adequate.

but that doesn't change the facts about the sensor system tech |  NO SIGNIFICANT IMPROVEMENT IN 10 YEARS FOR STILL IMAGES

But maybe Canon is finally approaching a corner of sorts and, when they finally change their course, they'll be able to provide customers with actual sensor performance improvements along with all the other bells, whistles and flashing lights they're so good at adding to the mix.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 03:16:13 PM »
This makes ZERO sense.  If there is no real world difference the where is the inferiority - or the superiority?  The sensor camera combo is only inferior/superior if there is a REAL tangible REAL WORLD Difference.  If there is no real world difference then logically - one is not greater than the other.  Again, this is photography, people buy images, people hire you because you craft good images - they don't hire/buy because the sensor is better.  And you can even take that to the consumer level - ohhh...thanks for taking some pictures of my sons first birthday, but, I saw that you used a canon so I don't even want to look at the pics because nikon has better sensors?????does anyone in the real world do that?????

The real world also includes hobbyists, who do not buy or sell photos. They are curious about the challenges pros face and about the way the run their business but do not really relate to that.

It is like being a car enthusiast and discussing taxis which professional taxi drivers drive. Every taxi driver would tell you than the clients could not care less about handling, acceleration but they care about space and a smooth ride. The drivers themselves want reliability, trunk space, fuel economy. This automatically excluded the hottest car brands.

+1, well stated analogy.

I have a few "taxis" in my garage which get regular use; a couple 4x4s (d800/e), Austin Mini (Pentax Q), and a few other assorted oddballs including a high performance daily commuter (Pentax k52s). I'm still waiting for an affordable sport coupe (d400 or preferably improved 7d2) to work with long lenses and dim light with high fps.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 03:34:56 AM »
OK, put the popcorn away for a minute. :)
Let's do some quick analysis using DxOmark's data.

Putting aside the 70D's sensor-based AF achievement for a moment, let's look at the Signal to Noise Ratio (SNR) of today's 70d, compared with the 20d from 9 years ago, and the 10d from 10 years ago.
As SNR goes, so does DR to a large extent but SNR is a little more telling and when you compare them overlaid you can see how they differ and where the changes in sensor performance are.

The upper blue line is the ISO 100 SNR plot for all 3 cameras.
On all 3 cameras you'll see that the 100% gray scale (white) is at about the same 42dB level.  So all 3 have the same SNR at white.

Where the line meets the bottom axis is where signal = noise and the image information gets lost in the noise and vice-versa. Well, unless there's a pattern that's discernible but DxOmark is still not publishing noise pattern data as far as I know. (I've requested that they do so we can estimate FPN severity of a sensor.)

Anyway, the signal=noise level of 0db is the cutoff level for all ISO measurements.  That is at gray scale 0.052% for the 70d at 100(claimed) ISO.  That's about 10.9 stops of highlite-to-dark range where dark = average noise. (It's late at night, somebody please correct me if I make a math mistake here, auto-correct may also mess up some spelling)

Without getting the specific data points for the other 2 cameras we can see that they also have about the same end points for 0db and white.

Net result.  At a per-pixel level, the 70d's base ISO performance has not improved in 10 years.  But, because it's pixels are about half the total area of the 10d's pixels (likely slightly more because of better fill-factor in modern sensors where more surface is actually actively used but lets go with half as that's in the 70d's favour) the 70d has made a technical achievement of about 1 full effective stop for one full effective pixel.
This may be even better by one more stop because of the way the split-pixel AF system functions but since I have no technical details on how this is really done at the sensor level I'll leave this to other tech types to expound. I'm also not considering each camera's real-effective ISO performance, just the rated level.

If we look at some other ISO levels:

- the 20D has better SNR results at white for all ISO levels.  This is not something that's easy to see in prints since a little noise on a big signal pretty much disappears.  This is also due in part to the physics of larger pixels, which is why full-frame sensors do even better. The 10D is similar to the 20D except for its highest ISO which falls down a bit

- the 20D's curves are ALL higher in every part of the graph, denoting that its SNR is better than the 70D's at all intensities from white right down into very dark shades at matching ISOs.  The 10D's curves are considerably lower; it has worse SNR than the 20D pretty much at every point below base ISO.


At a per-pixel level, the 20D was a significant improvement over the 10D which came 1 year before it.  The 70D, at the same per-pixel level, has not improved and is slightly worse than the 9 year old 20D.  Taking into account the smaller pixels of the 70D's sensor, the practical net improvement is less than 1 stop in 9 years. (possibly close to 2 stops, depending on how the dual-sensel-AF-pixels really work)

Compared the 20D to the 60D/7D and Rebel xx0 using 18MP sensors we have an honest improvement of less than 1 stop in the period between the 20D and the 7D's release.

So, if Canon's 70D sensor is read out in such a way as to provide the same level of read noise as a regular single sensel pixel then they may be able to achieve a measurable improvement (nearly 1 stop) in base ISO dark noise which should translate into a similar improvement in total DR of a similar sensor NOT using this new AF system.

So, will the 7D Mark II have no such split AF pixels and offer us slightly improved SNR and DR because of this method or will they do something more akin to the Exmor's superior noise elimination?
Or will they provide these new AF pixels and compromise the stills performance for the sake of video and improved live-view AF?

Please Canon, give us the best possible STILLS camera in the 7D Mark II.  I might then buy one.

Landscape / Re: Beautiful sunsets
« on: August 30, 2013, 01:15:17 AM »
out in the back yard watching the sunset and a Great Blue Heron flew past....
I do like when a bird positions itself nicely in the frame for me while I'm shooting a sunset or sky shot. That's a nice shot.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 01:11:48 AM »
Well, as innovative as the 70D's new dual-sensel pixels for continuous AF are, the overall signal to noise ratios, as reported by DxOmark, have changed very little.  Hopefully there'll be less banding in dark shadow for those who need to push it but I thought I'd put together some animated gifs to compare the difference between the 70D and the 60D and then the 70D compared to the Nikon D5200.
I hope DxOmark will allow this editorial use of their material here.  If not, it can be removed easily enough.
I find the complete SNR graphs are more useful to see where the low ISO deep shadow SNR limits are and what the highlite end shows for difference, which combined can help indicate DR and more.  Switching between them is a useful way to quickly see the differences.
One take away from all of this though is that somehow they managed to double the number of pixels and add a significantly useful feature (the Dual Pixel AF) without breaking anything.  Of course it doesn't look like they made any significant improvement in the low end noise either -- oh well.  However, I think the jury is still out until we get a look at the back of a lens cap (or something similar) to see what the pattern noise looks like.  I am very curious if any improvement has been made in that area.

+1, the AF development is quite remarkable and accomplished w-o breaking still IQ performance, actually improving on it ever so slightly in some areas.
I'll get some lens cap shots as soon as my local dealer has some stock.  I'm curious about not just what FPN there may be, but whether I can also see where the dual sensels are bordered by the regular ones.

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 01:08:29 AM »
On my 1DX I just hit a couple of button and bracket, never failed me this far. Generally prefer the results I get from blending exposures over what I get from using sliders on 1 exposure too  :).

Exposure blending's a great work-around, but not always practical. (moving scene elements)

EOS Bodies / Re: 70D and Dxomark....
« on: August 30, 2013, 01:06:08 AM »
.. learn how to expose properly, and take some freaking pictures for God sake or sell your gear and jump to sony for all I care... Just stop this nonsense. 

how do YOU "expose properly" for a scene that exceeds your Canon's DR?
Are you content to clip highlites and shadows and live with the out-of-camera tone curve for every shot?
If so, your advice may not register with the more artistic photographers.

Aglet Aglet Aglet ::Shakes head::  I expose the way I expose.  I've been shooting professionally for the last 10 years and been shooting even longer...  I look at a scene, look at what the meter tells me, and I either go with it or call it's bluff and compensate as i see fit.  It's not rocket science.  I get a shot, if it looks the way I want it to, great, if not, I compensate more.  When it's exposed the way I want it, It's good.  If you dont know how to expose, go, get off this forum, and start shooting. 

As far as going into a scene with too much DR... what absolute non-sense...  I've shot back in the days of 4x5 film, shot transparency, medium format, the early canon DSLRs which had what, 5 stops of DR?  If a scene is too dark, brighten it, if you cant brigten it without over exposing something else, use flash, or even better off camera flash, or reflector or some other way to manipulate the light.  Dear god son, this is photo 101, well maybe 102.  This isn't hard.  A real photographer doesn't blame his gear for not getting the shot.  A real photographer knows what needs to be done, and makes the photo even better. As i've said before, a client NEVER has looked at my photos and said "it's a shame there wasn't more DR"... Hell, the average client doesn't even know what noise is.  This is pure pixel peeping madness and it's disgusting.

Well, that was kind of a rhetorical question.  I know the answers. :)

What you seem to be saying is "compromise your exposure" or "manipulate the light" to fit your gear.
Perfectly good advice, that's what has to be done whether it's to fit the limitations of the camera or to change the appearance of the scene.  However, such methods are not always desirable or even possible.  So that advice, good as it is, may also be a bit trite.
I just prefer to have equipment that's less limiting than Canon's, especially since there have been better options available for a few years now and Canon's only improved marginally (base ISO) since their original CMOS sensor hit the scene years before that. They've in fact gone backwards and have only recently returned to where they used to be a few system generations ago when it comes to FPN.

I'm probably not the only one that's not surprised, and also disappointed, that the 70D is not significantly improved for still photography over its predecessor except for an improved AF system.
This simply means we're STILL waiting for Canon to catch up to the competition in the area of basic sensor system metrics.
That said, what they've accomplished with this new sensor-based AF is impressive!  Adding the 7d's PDAF system is a big plus too.

I've still made lots of great images with my Canons, (especially the Digic 3 and older ones) but I've also experienced plenty of situations in landscape shooting where FPN (Digic 4 systems) has shown up, not just in shadows, but in midtones.  Some of this FPN is not just read noise in shadows but also seems to be intra-sensor inconsistency.  Translation, their sensor production lines are not as precise as they need to be.  But I digress…

70d's to be lauded for its AF breakthrough, but not much else as far as I can see.

EOS Bodies / Re: 70D and Dxomark....
« on: August 29, 2013, 01:11:53 PM »
As for jumping ship, I think people tend to exaggerate how hard it is.  In the past few years I've jumped ship from Nikon APS-C to Pentax APS-C to Canon FF, bought a second ship (Olympus) and toyed with a third (Nikon FF) before deciding against it after renting a couple.  Each time I switched I sold all the previous equipment I had bought.  Depending on whether I had bought it new or used I received less/more/the same as I had paid for it in the first place.  I may have overall "lost" but I don't look at it that way - I think if it as the (not very high) price of using that equipment during the time I owned it and an extremely useful learning experience.


Pentax, by the way, provided a rather good example of why DR isn't enough.  I owned a K-5, with a K-x as back-up.  At the time there seemed to be fairly wide agreement that the K-5 had the best sensor of any APS-C camera (the same Sony sensor as the D7000 but run by slightly better software), and it was a good camera in other ways too (esp. ergonomics).  The dynamic range was simply astonishing - when I first bought it I would amuse myself by fooling around with deep shadows in DxO and LR, amazed by what it could reveal (not that the results were worth keeping...); and yes, there were a few times, mainly involving sharply contrasting light in the alleys of Lugano, when it proved useful.  But the relative shortage of first rate lenses with fast, accurate focusing soon became old....

I've added Pentax gear to my kit over the last year.  I really like the k52s, the thing will AF in near dark w-o assist light.  And it likely still does have the best overall low ISO raw performance of any crop body... per-pixel anyway. and the high iso end is also very good.
But I agree, Pentax lenses are a different collection compared to the competition but I've managed to find some that work extremely well for me, tho only my body-driven primes focus super fast.  Their SDM AF is kinda slow on my 16-50/2.8 and that's cost me a few shots.
Still, I use it because I love the images I get with it and the ergonomics and highly customizable interface.  It's a very good photographic tool and has replaced my 60D + 15-85mm as my go-to rig.  If they would bring out a fast lens with more range, like 15-85mm f/2.8-4.5, that'd be even better.

EOS Bodies / Re: 70D and Dxomark....
« on: August 29, 2013, 01:01:15 PM »
... awkward when handholding a d800 ..to change the ISO settings (top left of the camera), I could not find how to assign it to any of the other buttons.

?? In Av mode I've got my front control wheel for aperture, rear wheel for ISO

The same to change autofocus points, hard to reach that switch...

short thumb?

The af-on switch needs too much pressure..

OK, maybe your hands aren't the best fit for a d800
I have mid-size paws and d800 feels perfect in my mitts with a 70-200 hanging off it.

4fps? really?

5d2 was pretty slow too, who's complaining about that?

Nikon bodies are quite different, you need to spend some time with them to adjust yourself to how to get the best handling with them.  No different than any other mfr.  I have to admit, my long term Canon use gave me a similar bias but no more.
D800 isn't a spray'n'pray kind of camera, it requires more deliberate use.. but has more forgiving (under)exposure latitude

.. learn how to expose properly, and take some freaking pictures for God sake or sell your gear and jump to sony for all I care... Just stop this nonsense. 

how do YOU "expose properly" for a scene that exceeds your Canon's DR?
Are you content to clip highlites and shadows and live with the out-of-camera tone curve for every shot?
If so, your advice may not register with the more artistic photographers.

Pages: 1 ... 26 27 [28] 29 30 ... 70