nigelht

Joined on Sep 2, 2012

Comments

Total: 66, showing: 61 – 66
« First‹ Previous1234Next ›Last »
On article What is equivalence and why should I care? (2467 comments in total)
In reply to:

tokumeino: Very good article. More accessible than the "original" http://www.josephjamesphotography.com/equivalence/ . It will save a lot of posts in the forums, I hope.

The next step will be to make people admit that image noise heaviliy relies on lens size and not that much on sensor size. Indeed, whatever the format, you need a large absolute aperture diameter (in mm) to catch more light and reduce noise. And a large aperture diameter (in mm) needs a large lens diameter (in mm).

As a result, more than ever, before getting invloved into a format, one must choose what kind of lens he needs, and balance with lens size. Then, one has to choose a body that is comfortable with the chosen lenses.

FF are not good because of the large sensosr, but because of the large lenses with large aperture diameters. And such large lenses are only comfortable with large bodies.

There's also the point of low ISO, but this will come in a third step, I think.

Your assertion that it depends on lens size and not on sensor (or specifically sensel size) is easily shown to be incorrect when the same lens is used on cameras with different pixel sizes results in different outcomes.

The reason that the EOS 1-DX does so well against same generation FF cameras is because the pixel size is larger.

Compare the EOS-1D X vs the 5D Mk III:

EOS 1-DX : 41.75 um^2
5D MkIII: 38.69 um^2

The 1-DX sensel area is 23% larger than on the 5D MkIII. There are other (significant) differences by the 1-DX has a huge inherent advantage out of the gate for high ISO performance while using the exact same glass.

When comparing against the D4 at 53 um^2 you see a small 11% size advantage in the Nikon. Which matches the real life testing and reviews where the D4 seems to have a slight advantage over the 1 DX.

Link | Posted on Jul 8, 2014 at 23:04 UTC
On article What is equivalence and why should I care? (2467 comments in total)
In reply to:

nigelht: Total light is meaningless to noise. Given the same FOV then the total light will be the same. Given different FOVs with the same intensities (as in the example provided) then for the same sensel size then the same number of photons hit so shot noise is identical since the photon shot noise is the square root of the number of photons on a per sensel basis.

Read noise is more dependent on the sensor design and generation but there is a minor advantage for larger sensor since the base SNR for FF sensors is typically slightly better than m43 or 1" of the same generation. Mostly though the noise is dominated by shot noise.

All you've shown on your example charts is the effects of larger sensel size.

The sensel (pixel) sizes for the cameras listed are:

EOS 1-DX : 41.45 um^2
X-A1: 22.66 um^2
GH4: 13.99 um^2
V3: 6.3 um^2

The area of each EOS 1-DX sensel is 658% larger than a sensel on the Nikon 1 V3.

The 1-DX sensel area is 111% larger than on the X-A1.

@great bustard

Your "experiment" is faulty as when you crop the 50mm to 100mm FOV you lose half the resolution when you display at the same size.

This has nothing to do with "total light". it also is not equivalent to the scenario at hand because the pixel densities are much higher for most of the crop sensors, especially the 1" V3 vs FF.

You aren't losing half the resolution when using crop sensors.

Link | Posted on Jul 8, 2014 at 20:46 UTC
On article What is equivalence and why should I care? (2467 comments in total)
In reply to:

nigelht: Total light is meaningless to noise. Given the same FOV then the total light will be the same. Given different FOVs with the same intensities (as in the example provided) then for the same sensel size then the same number of photons hit so shot noise is identical since the photon shot noise is the square root of the number of photons on a per sensel basis.

Read noise is more dependent on the sensor design and generation but there is a minor advantage for larger sensor since the base SNR for FF sensors is typically slightly better than m43 or 1" of the same generation. Mostly though the noise is dominated by shot noise.

All you've shown on your example charts is the effects of larger sensel size.

The sensel (pixel) sizes for the cameras listed are:

EOS 1-DX : 41.45 um^2
X-A1: 22.66 um^2
GH4: 13.99 um^2
V3: 6.3 um^2

The area of each EOS 1-DX sensel is 658% larger than a sensel on the Nikon 1 V3.

The 1-DX sensel area is 111% larger than on the X-A1.

(cont due to character limit)

A FF sensor with the same sensel area (and general design) as the V3 would show a correspondingly high sensitivity to shot noise. Crop to the same FOV and you'll get largely the same results.

The tremendously larger sensel sizes in FF sensors and the resulting larger Full Well Cpacity (FWC) is key to the differences between sensors when dealing with photon shot noise.

The other primary noise advantage of FF over m43 and 1" is that they typically run at lower temps since the latter two typically are running in live mode. Higher temps = more noise.

Link | Posted on Jul 8, 2014 at 18:48 UTC
On article What is equivalence and why should I care? (2467 comments in total)

Total light is meaningless to noise. Given the same FOV then the total light will be the same. Given different FOVs with the same intensities (as in the example provided) then for the same sensel size then the same number of photons hit so shot noise is identical since the photon shot noise is the square root of the number of photons on a per sensel basis.

Read noise is more dependent on the sensor design and generation but there is a minor advantage for larger sensor since the base SNR for FF sensors is typically slightly better than m43 or 1" of the same generation. Mostly though the noise is dominated by shot noise.

All you've shown on your example charts is the effects of larger sensel size.

The sensel (pixel) sizes for the cameras listed are:

EOS 1-DX : 41.45 um^2
X-A1: 22.66 um^2
GH4: 13.99 um^2
V3: 6.3 um^2

The area of each EOS 1-DX sensel is 658% larger than a sensel on the Nikon 1 V3.

The 1-DX sensel area is 111% larger than on the X-A1.

Link | Posted on Jul 8, 2014 at 18:46 UTC as 332nd comment | 7 replies
On article Apple to cease development of Aperture (425 comments in total)
In reply to:

photogeek: People are missing the bigger picture. In another 2 years or so, tablets will have CPUs comparable to those currently found in laptops. For laptops the Moore's law is over (actually, not Moore's law, but Dennard scaling is over, but let's not get too technical). For tablets that's not the case. They are not held back by having to support software written 20 years ago. Their CPUs are much simpler, and they are being developed taking past (very expensive) lessons on how not to do it into account. Laptops will stay about the same, just as they did in the past five years. Tablets will take over for most people, with pros preferring laptops and desktops, 5% of the overall market. Put simply, there's not enough money in this market for Apple to be interested in it. Apple is about focus. They are betting on a different horse, nothing more, nothing less.

The top end tablets are about half as fast as a 2011 Core i5-2520M @ 2.50 GHz.

http://browser.primatelabs.com/ios-benchmarks

I don't think 2016 iPads will have the same processing power as a 2014 MBA.

Link | Posted on Jun 28, 2014 at 19:04 UTC
On article Apple to cease development of Aperture (425 comments in total)
In reply to:

MrTaikitso: iPhoto is one of the worst pieces of software ever developed, and I speak as a UX (User Experience) and software engineer. 1. A few years after it was launched, it managed to lose over 400 of my precious photos. And because they were being backed up, and the way the bug erased the images, I never got them back. Many others reported odd bugs too. 2. More recently, as part of a huge multi year archiving effort (& to escape it's bugs), I exported thousands of priceless photos & scans from iPhoto. Only later did I discover that EVERY SINGLE portrait oriented photo had been mysteriously cropped such that only about 5% of the photo was saved. I lost another 500 or so images & no idea which ones! 3. The UX is awful & inconsistent with Aperture, a superbly designed app, the very opposite of iPhoto. So why don't Apple do the reverse and make Aperture their default main photo/video app? Seems they are dumbing down to appeal to the idiots that buy iPhones purely to Snapchat & Instagram. Tragic!

As a software engineer and UI developer I have to say that anyone that can lose hundreds of photos "exporting" from iPhoto should turn in their software spurs because "exporting" from iPhoto consists of going into the iPhoto folder (using show package contents) and grabbing the masters folder and dragging it somewhere that their new library system can import.

Yes, the iPhoto metadata (faces, stars, etc) is lost in this method but losing the originals should be very very hard and should be step 2 of any migration process. Step 1 should be making a copy of the iPhoto folder.

Step 0 should be saving the DCIM from your card onto an archive disk and importing from the archive disk rather than the card. That way there's always a clean backup of the photos and until I see a good version in my library I don't erase the card. A good version in my library means I have 2 copies: the one in my library and the one in my archives.

Link | Posted on Jun 28, 2014 at 18:52 UTC
Total: 66, showing: 61 – 66
« First‹ Previous1234Next ›Last »