Ron Pfister wrote:
Thanks - yes I saw these. I hope this perceived edge is indeed real. But as you say, making a valid comparison regarding sensor resolution is a difficult task, even if the lenses are the same. Where did you find the RAW files?
In the Imaging-Resource camera reviews, you will find a page with the test images which contains a link for the raws. Here it is for the D800E:
Phillip Reeve wrote:
Do you think field curvature is somewhat affected by the sensor?
It is pretty obvious in these pictures, but i don't know how the Biogon would behave on a leica M9/M240
Sorry for being a bit off topic with regard to rangefinder wides in showing the D800e vs A7r.
I think this is a great question and I'm curious to hear what folks think. I do wonder if it's possible that the micro-lenses themselves at the borders/ edges of the sensor could be resulting in a focus difference with lenses that have extreme ray angles (such as this Biogon) in that area IF they are not optimized perfectly for the lens. I suspect the answer is yes though I'm not 100%.
Tariq Gibran wrote:
I do wonder if it's possible that the micro-lenses themselves at the borders/ edges of the sensor could be resulting in a focus difference with lenses that have extreme ray angles (such as this Biogon) in that area IF they are not optimized perfectly for the lens.
My bet is still the filter stack, not the micro lenses. They should only affect IQ on a pixel-to-pixel level.
If you remember, I was mentioning something very similar in regards to the Contax G 21 and the NEX 5N, in which I was focusing on the borders and it’d be (more or less) ok, whereas if you focus somewhere else, they’d be out of focus to the point of thinking it was smear. And this is the most problematic of all Biogons, one that was discounted by many as hopeless.
To me, it’s pretty clear that with biogons you can get the whole frame in focus, but it needs to be a certain type of scene and you need to be very careful with your aperture.
Makten wrote:
My bet is still the filter stack, not the micro lenses. They should only affect IQ on a pixel-to-pixel level.
Fully agreed. Also I would like to mention that all Zeiss wides I have used, ZM 18, 25 and 35 have a very flat field, especially at infinity. The ZM 18 may benefit from stopping down slightly but not the 25 and 35. That is on the M9. The ZM 21 and 28 seem to have more field curvature that affects the corners but nothing like what we see in the test images on the 7.
Phillip Reeve wrote:
Do you think field curvature is somewhat affected by the sensor?
It is pretty obvious in these pictures, but i don't know how the Biogon would behave on a leica M9/M240
If you're talking about the ZM, it has Leica style field curvature, but the corners should be fine. I have not tried this lens, but the shown results seem odd to me. I believe the sensor toppings are exaggerating the field curvature.
Makten wrote:
My bet is still the filter stack, not the micro lenses. They should only affect IQ on a pixel-to-pixel level.
Does it really matter from an end user's perspective if it is the filter stack or the microlens that can't handle the angle of incidence for proper refraction to the sensor? It would seem to me that either way the angle of incidence from the optical projection is too steep for the components to properly pass the projected image onto the sensor ... i.e. if it is too steep, it's too steep.
RustyBug wrote:
Does it really matter from an end user's perspective if it is the filter stack or the microlens that can't handle the angle of incidence for proper refraction to the sensor? It would seem to me that either way the angle of incidence from the optical projection is too steep for the components to properly pass the projected image onto the sensor ... i.e. if it is too steep, it's too steep.
It does matter in my opinion because the sensor toppings can be removed or modified. Anyhow it can't be the microlenses fault here.
p.106 #10 · A7/A7r - performance with WA RF lenses
edwardkaraa wrote:
It does matter in my opinion because the sensor toppings can be removed or modified. Anyhow it can't be the microlenses fault here.
Isn't it the unusual, tall oblong shape of the Leica M240 Microlenses that supposedly result in better corner/ border performance? As has been mentioned elsewhere, the filter stack can result in astigmatism with wides outside of the center, which I guess shows up as general smearing, but for a difference in focus point that can be corrected by focus (I don't think that's the case with general smearing?), perhaps the microlenses do play a role. The difference may be academic to many.
p.106 #11 · A7/A7r - performance with WA RF lenses
RustyBug wrote:
Does it really matter from an end user's perspective if it is the filter stack or the microlens that can't handle the angle of incidence for proper refraction to the sensor?
Yes, since many seem to believe the a7R is "better" just because it has got these offset microlenses. Nothing we've seen so far would indicate that it helps smearing even just a bit.
Tariq Gibran wrote:
Isn't it the unusual, tall oblong shape of the Leica M240 Microlenses that supposedly result in better corner/ border performance?
Regarding vignetting, yes. But it's most likely the thin filter stack that makes the M sensors work well with M lenses. Remember the M8, where they chose not to use an UV/IR filter at all. It must have been because they saw the issues with a thick filter, that we now can see with the Sonys.
Edit: I might be wrong, but I cannot understand how microlenses could change focus of the projected image.
p.106 #12 · A7/A7r - performance with WA RF lenses
Interesting illustration ... thanks.
As with all things optical ... trade-offs/compromises exist when changes are made. What is the compromise/trade-off of the taller/oblong microlens design (i.e. choose your poisons)?
p.106 #13 · A7/A7r - performance with WA RF lenses
Tariq Gibran wrote:
In the Imaging-Resource camera reviews, you will find a page with the test images which contains a link for the raws. Here it is for the D800E:
Thanks. Since the crops you had posted were quite a distance from the center of the original images, and because I thought the difference between the A7R and the D800E samples may well be due to lens aberrations, I took a closer look myself. Below center crops of the same A7, A7R and D800E sample files made available on imaging-resource.com. The resolution advantage of the A7R over A7 is very obvious. The A7R sample also clearly shows more detail than that of the D800E, but - as discussed - that may well be due to the different lenses used. That said, I prefer the result from the A7R over the others.
NB: all files were first converted to DNG and then processed in LR4 with default settings except for appropriate sharpening with identical settings for all images. I'm aware this is somewhat off-topic, but I hope still relevant to everyone's interest.
p.106 #14 · A7/A7r - performance with WA RF lenses
Tariq Gibran wrote:
Isn't it the unusual, tall oblong shape of the Leica M240 Microlenses that supposedly result in better corner/ border performance? As has been mentioned elsewhere, the filter stack can result in astigmatism with wides outside of the center, which I guess shows up as general smearing, but for a difference in focus point that can be corrected by focus (I don't think that's the case with general smearing?), perhaps the microlenses do play a role. The difference may be academic to many.
There is an article in LFI magazine issue of this month about the oblong microlenses of the M240. It says clearly that the purpose of this design is to minimize vignetting and Italian flag without the recourse to offset microlenses. The corner image degradation is caused by the sensor glass and Leica does use the thinnest stack possible even at the price of slight IR contamination.
p.106 #15 · A7/A7r - performance with WA RF lenses
We need an expert! Where is Joakim/ theSuede?
Just to re-state the original question, with a lens that has known field curvature (Biogon ZM28), is it possible for the sensor design to exaggerate (worsen) that field curvature? I think it's a great question.
p.106 #16 · A7/A7r - performance with WA RF lenses
Tariq Gibran wrote:
We need an expert! Where is Joakim/ theSuede?
Just to re-state the original question, with a lens that has known field curvature (Biogon ZM28), is it possible for the sensor design to exaggerate (worsen) that field curvature? I think it's a great question.
I know you don't trust me (and I'm certainly far from being an expert) but the answer is yes. In fact even the filter that you put in front of your lens affects field curvature.
p.106 #17 · A7/A7r - performance with WA RF lenses
edwardkaraa wrote:
I know you don't trust me (and I'm certainly far from being an expert) but the answer is yes. In fact even the filter that you put in front of your lens affects field curvature.
p.106 #19 · A7/A7r - performance with WA RF lenses
Nice to see how politely we disagree...
The whole affair might end up being a clever strategy by Sony. Weeks before releasing the product, they start the marketing buzz about 'gapless design, offset microlenses' etc. and at once they get the drooling interest of the 'Alternative RF brigade'. Actually, Sony fits the cameras with very thick toppings, so the wide RF glass won't work as it could. Of course, hundreds of users after receiving their cameras start to rant and cry but, WTF, the A7(r)s are so good in everything else they do...Instead of sending it back, most users will think it twice and, instead they'll sell those useless Biogons and get the FE 35, FE 55, eventually a future FE 24 ..
p.106 #20 · A7/A7r - performance with WA RF lenses
Many a true word spoken in jest!
Actually, I'm finding the NEX7 has grown more attractive over the last few days. I'll be interested to hear Phillip Reeve's views on these two cameras as he now has both.
artur5 wrote:
The whole affair might end up being a clever strategy by Sony. Weeks before releasing the product, they start the marketing buzz about 'gapless design, offset microlenses' etc. and at once they get the drooling interest of the 'Alternative RF brigade'. Actually, Sony fits the cameras with very thick toppings, so the wide RF glass won't work as it could. Of course, hundreds of users after receiving their cameras start to rant and cry but, WTF, the A7(r)s are so good in everything else they do...Instead of sending it back, most users will think it twice and, instead they'll sell those useless Biogons and get the FE 35, FE 55, eventually a future FE 24 ..
