Nice camera but I feel one flaw that was only mentioned on the "noise" page (not on the conclusion page), is actually magnified by another. Like most of the X series cameras, ISO's are overstated up to 2/3 EV (1 stop difference compared to many other cameras). But in RAW, the camera also tops out at an indicated ISO 6400. Which means that if you want to shoot RAW, compared to many other cameras, the top sensitivity is actually comparable to ISO 3200. Not exactly high in the current age, where a lot of crop cameras are already used at even higher ISO's.
And while I acknowledge that users can push RAW files further themselves, choice is removed and it does affect the LV experience.
HappyVan: What a strange conclusion?
Sensor size is 58% of FF. Resolution (19mp) is 53% of D800. So, small improvement over D800 is expected. NO big deal.
The real deal breaker is the price and weight (5 lb).
I don't think anyone in here implied that the amount of light collected does not matter. In fact, I'd be the first to say sensor size matters, hence my first reply in this thread.
DXO scores do not quantify visible impact of type of noise, patterns in particular. That being said, patterns aren't tied to pixeldensities or format size either, but rather technology used. The other important factor.
In practise (key) there is little to no correlation between pixel density and for example shot noise at a given sensor size.
That's for example why an A7R performs as well or better in low light than the A7 (both shot noise and read noise actually).
On the other hand, ceteris paribus there does seem to be a small correlation between pixel density and colour sensitivity at low ISO (with the higher density positively affecting colour sensitivity).
Why would you expect a small improvement based on a marginally lower pixel pitch on a sensor that is almost half a the size?
Must... shoot... potholes now. Then again, trying to find one here is like trying to find a needle in a haystack.
Combatmedic870: This really does look like a great phone. All of the latest and greatest.
Gotta wonder if the screen is IPS or not though.
IPS according to specs.
munro harrap: Very uneven resolution, at the borders scarcely ever lifting above 1500lpm- This is LESS than an 8MP sensor. OK, at f5.6 in the middle its OK, but as the sensor resolves 4000lpm, and this is an f2.8 lens it is a disappointing one. Compare to Samyang 35/f1.4...We buy wides to get more in at closer distances and to grasp situations. Now, I don't know about you, but if I had to display prints from an A7R I would be very unhappy to with this lens, since only the central area will be sharp until at least F8. Not only that, but the step down into mush at wider apertures outside the central area renders it useless as a wide-angle prime. It is under-designed- it should be equally sharp allover at this stage in optical history- it is a 78rpm in the age of Blu-ray, and there are many sharper consumer f3.5-4.5-5.6 zooms at a fraction of the price.And check the Sigma 35mm f1.4
Correction, the corners are only slightly behind the Samyang at f/2.8 and f/5.6, centers a tad better and the corner gap narrows as you close down further. Point stands, if you'd be unhappy with these 35mm results (especially considering the lens is >6 times smaller and >5 times lighter), its hard to imagine you being happy with marginally better corner performance at some apertures. Unless you're that easily swayed from happy to unhappy of course. ;)
Cool story, but according to DXOmark, the FE 35mm outresolves the Samyang fitted on a 36MP sensor from f/2.8 to f/22. The corners are usually better, never worse, the center is better. Most tests shows its similar to or even better than the Sigma throughout it's apertures, which is the benchmark right now.
Which means that if you'd be unhappy with the results this lens would give you, what's your reference really? Because there would be little to no improvement to be found elsewhere.
TN Args: Let's say Sigma released a 35mm f/2.8 for $150 and its optics were the same. Everyone would say "well la di da, I expect no less, 35mm and f/2.8 is hardly a challenge."
And they would be right.
The next CaNikon will have lots of character like the Nikkor 58mm. And worthy of the price tag well over 1K, regardless of size and corner sharpness. Because that's how these arguments work.
radissimo: f1.4 in the 4/3 terms ,sharp from wide open and small! wow
In fact you can't, since you fail to mention a single µ4/3 lens within those relevant specs, which you did mention in your initial claim.
Instead you come with Fujinon lenses that are irrelevant (different equiv. FL) except one: the 23mm f/1.4. Which would translate to about f/2.1 equiv. aperture and it's 2.4 times the volume (2.5 times the weight too) of the FE 35mm, being both wider in diameter and quite a bit longer. Not smaller by any stretch of the word.
InTheMist: The first test I compared it to was the D800 (non-e) and Sigma 35 f/1.4 Art.
I'll take the Sigma.
I seriously doubt he would use IS running tripod stabilized resolution tests...The effect you mentioned is not noted in the said article, probably brought up elsewhere for different shooting purposes.
James First 007: Not a great lens. The DxoMark review got it right…When compare to the 55mm, it is very disappointing…
In fact, the reason why I am staying away from the A7R is because there is no 35mm capable of performing well on this 36 mg camera !
However, the 55MM is an outstanding lens…worthy of its price tag !
Typical non sequitur.
Frank_BR: Most likely, any reasonably capable Japanese optical company can make today a virtually perfect 35mm lens, especially if the aperture is a modest F2.8. Indeed, except for the high price, this "Zeiss" lens does not seem to be an authentic Zeiss, but a Japan-made lens with a Zeiss tag.
The question is, why Sony charges a reasonable price for the cameras, but so much for these faked Zeiss lenses?
In the near future, the prices of FF entry-level models should fall below $1000. Then, it will be clear to everyone that today's prices of most prime lenses are too high.
Since when was increased quality ever reflected in a linear increase in price..... in any product category for that matter?
That's only if you care about average, which conceals the weakest link: corners.
σ would have been more telling than the average you quote.
But if you care so much about the average, the FE scores 24% higher than the Sigma average across the entire frame (and as said before, 21% higher in the corners too). Quick, pick your poison and swallow it. ;)
RichRMA: Nothing to write home about for that price. They took the "Sigma" path, making the lens sharp in the centre but not so much the edges. Also, why does the body have to be as large as it is when the elements are so small (35mm f/2.8 only needs elements about 15mm across)?
According to tests done by Roger Cicala, the Sigma is quite a bit softer at the edges at f/2.8. According to DXOmark, they are similar.
The Sigma is also about 6.6 times the volume, 5.5 times the weight and not weather sealed.
Still waiting for a valid example.
Can you point me to a 35mm lens that is clearly/significantly sharper at the edges on FF?
I looked long and hard for an AF ~17.5mm f/1.4 lens for µ4/3 that is smaller and couldn't find one. But I'm sure you can point to one to support that claim.
Gary Dean Mercer Clark: Makes the Sigma 35mm F1.4 DG HSM | A lens look like a bargain comparing the optical tests. Zeiss lenses are overrated and overpriced in my humble opinion.
Horses for courses. The Sigma looks like a truck next to a classic Mini, side by side with the FE.
FEb 25th 2013Sigma at f/2.8, D800: 1171 center, 575 corner average.
As for the Canon, you conveniently left out the significant difference in measured corner performance. 560 average for the Canon on the A7R at f/4, where the FE scored 750. 34% higher on the FE.
JackM: f/2.0 would have made it a home run.
But Yabokkie can be trusted instead.