Lens Rentals reviews the 35-100 / 2.8

Started Nov 2, 2012 | Discussions
Anders W Forum Pro • Posts: 21,468
Re: This lens is great wide open
1

mpgxsvcd wrote:

I posted RAW samples that show it really is about the same throughout the entire zoom range. It also is better than the 14-140mm throughout that whole range and it achieves that quality right from F2.8.

As already pointed out by micksh6, 2 MP images like those you posted are hardly ideal for judging sharpness. Even so, these shots tell me the lens does better at the short end than at the long one, which is fully in line with Roger C's results.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: This lens is great wide open
1

Anders W wrote:

mpgxsvcd wrote:

I posted RAW samples that show it really is about the same throughout the entire zoom range. It also is better than the 14-140mm throughout that whole range and it achieves that quality right from F2.8.

As already pointed out by micksh6, 2 MP images like those you posted are hardly ideal for judging sharpness. Even so, these shots tell me the lens does better at the short end than at the long one, which is fully in line with Roger C's results.

That is why I posted the Original RAW files and unprocessed JPGs files in a zip folder at the link below. I guess everyone missed my other post.

RAW image samples

I also have 100% crops in the gallery below.



 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: Lens Rentals reviews the 35-100 / 2.8
1

thinkfat wrote:

Anders W wrote:

MTF-50 is chosen precisely because it has been found to correlate best with human perception. You can look at the entire MTF curve if you like and some test sites, e.g., DxOMark, let you do that. If you do, you will find that lenses that do well at MTF-50 also tend to do well at other contrast requirements than 50 percent.

I don't recall exactly, where, but I read a pretty nice article about pinhole lens design (and about how sometimes a bad engineer kicks out a good engineer who happens to not understand the problem). It went to say that it is most important for a lens to have good contrast (i.e. a high MTF value) in low and mid frequencies and even a steep drop-off towards high frequencies will still give the perception of sharp images. Whereas a lens with the MTF value declining steadily across frequency will appear soft even though performing better at high frequencies.

I'm not ready to accept that MTF-50 really models human perception in the best way. To me it's just as likely that it just gives nice curves that are easy to read and compare.

I am not sure myself whether MTF-50 is the best criterion (due to the simple fact that I don't have the real-life examples I would need in order to judge the matter based on my own perceptions). I would also guess that the conclusion depends on things like display size, resolution of the output medium, and viewing distance.

One may of course additionally debate whether a single number (per FL, f-stop, and area of the frame) is sufficient. Seeing the entire MTF-curve (for each of four different distances from the center of the frame), as DxOMark lets you do, is of course more informative.

However, if you start looking at those curves, you will find, as I pointed out in the post to which you replied, that lenses that do well at a contrast requirement of 50 percent as a rule do well even at higher contrast requirements (corresponding to lower frequencies). That's why I am not so concerned about the widespread use of MTF-50. I am pretty sure that using a higher contrast criterion would not yield significantly results in most cases.

Once you get to very low contrast requirements, however, the picture gets more diffuse. If I have understood things correctly, the current MTF-50 practice came about as a better alternative to the old practice of simply counting how many lines you could distinguish no matter how low the contrast, which in practice meant a very low contrast requirement.

For their scoring of lenses, and for the lp/mm values they report unless you start digging, DxOMark, unlike most other sites, uses a contrast criterion as low as 20 percent. Since I am pretty sure that's too low for my perceptual standards, I prefer to look at their MTF curves, where I can choose whatever contrast requirement (or frequency) I wish.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
Jeremy_T Senior Member • Posts: 1,224
Re: This lens is great wide open

mpgxsvcd wrote:

I don't have the 75mm so I can't comment on that. However, it can't possibly be as flexible as the 35-100mm and I really don't see where people are coming up with the fact that this lens isn't sharp.

I think people are confusing "not perfect" with "not sharp." Whatever minor "problems" the MTF charts show at 100mm are essentially meaningless for actual human perception.

It's tempting to hold this to a "higher standard" due to the high price, but it appears to meet any possible quality requirements already, and it wins by default since it has no competition in this aperture/zoom range.

 Jeremy_T's gear list:Jeremy_T's gear list
Olympus OM-D E-M5 Panasonic Leica Summilux DG 25mm F1.4 Olympus M.Zuiko Digital 45mm F1.8 Panasonic 20mm F1.7 II
mpgxsvcd Veteran Member • Posts: 8,094
Re: This lens is great wide open
1

Jeremy_T wrote:

mpgxsvcd wrote:

I don't have the 75mm so I can't comment on that. However, it can't possibly be as flexible as the 35-100mm and I really don't see where people are coming up with the fact that this lens isn't sharp.

I think people are confusing "not perfect" with "not sharp." Whatever minor "problems" the MTF charts show at 100mm are essentially meaningless for actual human perception.

It's tempting to hold this to a "higher standard" due to the high price, but it appears to meet any possible quality requirements already, and it wins by default since it has no competition in this aperture/zoom range.

I am quoting this post because it is 100% spot on.

Did lensrentals ever post the images that they determined were below their expectations?

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: This lens is great wide open
1

mpgxsvcd wrote:

Anders W wrote:

mpgxsvcd wrote:

I posted RAW samples that show it really is about the same throughout the entire zoom range. It also is better than the 14-140mm throughout that whole range and it achieves that quality right from F2.8.

As already pointed out by micksh6, 2 MP images like those you posted are hardly ideal for judging sharpness. Even so, these shots tell me the lens does better at the short end than at the long one, which is fully in line with Roger C's results.

That is why I posted the Original RAW files and unprocessed JPGs files in a zip folder at the link below. I guess everyone missed my other post.

RAW image samples

I also have 100% crops in the gallery below.

Well, I am afraid that other post was rather easily missed. But thanks for the additions. I looked at the crops in your gallery and found two series: one at 100 mm, one at 99 mm but none at 35 mm. What I could see by looking at the beginning (widest apertures) of the 100 mm series is that your samples are well in line with Roger C's measurements. The lens sharpens up noticeably when you stop down from f/2.8 to f/4.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: This lens is great wide open
1

Well, I am afraid that other post was rather easily missed. But thanks for the additions. I looked at the crops in your gallery and found two series: one at 100 mm, one at 99 mm but none at 35 mm. What I could see by looking at the beginning (widest apertures) of the 100 mm series is that your samples are well in line with Roger C's measurements. The lens sharpens up noticeably when you stop down from f/2.8 to f/4.

That is weird because when I looked at those images I really couldn't see much of a difference at all.

I didn't post the 35mm crops because the distortion happens mostly arround the edges. I thought that the crops gave it an unfair representation of its overall performance. Instead I gave everyone the RAW files to make their own determinations. That is the only fair way to do it.

I have some more full size pictures. However, I am running out of upload space to post them. Can we buy more space here on dpreview?

I wonder how long it will be before dpreview reviews this lens? It seems like there is a lot of interest in seeing the review even if most of those people are not willing to pay the price to actually purchase it.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: This lens is great wide open

mpgxsvcd wrote:


Well, I am afraid that other post was rather easily missed. But thanks for the additions. I looked at the crops in your gallery and found two series: one at 100 mm, one at 99 mm but none at 35 mm. What I could see by looking at the beginning (widest apertures) of the 100 mm series is that your samples are well in line with Roger C's measurements. The lens sharpens up noticeably when you stop down from f/2.8 to f/4.

That is weird because when I looked at those images I really couldn't see much of a difference at all.

I didn't post the 35mm crops because the distortion happens mostly arround the edges. I thought that the crops gave it an unfair representation of its overall performance. Instead I gave everyone the RAW files to make their own determinations. That is the only fair way to do it.

I have some more full size pictures. However, I am running out of upload space to post them. Can we buy more space here on dpreview?

I wonder how long it will be before dpreview reviews this lens? It seems like there is a lot of interest in seeing the review even if most of those people are not willing to pay the price to actually purchase it.

It's pretty easy to see the difference between the 2.8 and the 4.0 crop if you know what to look for. Just put these two crops next to each other so that you can see them at the same time or switch quickly back and forth between them. Then look at the contrast level, more global as well as in finer details. How black are the blacks and how white are the whites?

As to the 35 mm crops, its up to you whether to post them of course. But I didn't quite see the point of the 99 mm series on top of the 100 mm series so the former can perhaps be deleted to make room for the 35 mm series.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: This lens is great wide open
1

I posted the 100mm vs. 99mm crops to show how the 14-140mm compared to the 35-100mm. The 14-140mm is fine at 100mm and so is the 35-100mm.

I think the problem is that lensrentals expectations were that sharpness is everything. They basically assumed that in order to justify its price the lens had to have an equally high level of sharpness.

Well at some point increasing sharpness becomes unnecessary. What is important and what actually makes this lens expensive are the fact that it is an extremely small and lightweight zoom lens with excellent results throughout its entire focal range and it has one of the best IS systems I have seen for an m4/3s lens OR camera.

I haven't seen any of the IS problems and double images that plague some of the other less expensive lenses.

I deleted a bunch of my old pictures so I can post the full sized images here. What exactly does everyone want to see? Is there a resolution chart you want to see or is there a specific real world image you are more interested in?

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Horshack Veteran Member • Posts: 6,725
Re: Results indicate this is an f/4 lens

dotborg wrote:

Good job pulling charts from two different sources. What I meant by "direct comparisons" is actual pictures that show the Panasonic to truly have "mediocre wide-open corner performance relative to lenses of the same class on other systems".

The lens just came out, it'll take a while. In the meantime if you can find any review of the Canon 70-200mm II @ 200mm that shows its corner performance being 30% worse at f/2.8 than f/4 then you'll have proven your point and proven me wrong.

mpgxsvcd Veteran Member • Posts: 8,094
Re: This lens is great wide open
1

I think there is a much bigger difference between the F2.8 image and the F8.0 image than there is between the F2.8 image and the F4.0 image. However, that never stopped the “Pro” Panasonic photographer from shooting at F8.0 with this lens during the Olympics.



35-100mm F2.8 @ F2.8

35-100mm F2.8 @ F4.0

35-100mm F2.8 @ F8.0

It would be a far bigger mistake to stop this lens too far down than it would be to shoot with it wide open. I still don’t trust any reviews that post opinions without any images to back up their claims.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: This lens is great wide open
1

mpgxsvcd wrote:

I posted the 100mm vs. 99mm crops to show how the 14-140mm compared to the 35-100mm. The 14-140mm is fine at 100mm and so is the 35-100mm.

I think the problem is that lensrentals expectations were that sharpness is everything. They basically assumed that in order to justify its price the lens had to have an equally high level of sharpness.

Well at some point increasing sharpness becomes unnecessary. What is important and what actually makes this lens expensive are the fact that it is an extremely small and lightweight zoom lens with excellent results throughout its entire focal range and it has one of the best IS systems I have seen for an m4/3s lens OR camera.

I haven't seen any of the IS problems and double images that plague some of the other less expensive lenses.

I deleted a bunch of my old pictures so I can post the full sized images here. What exactly does everyone want to see? Is there a resolution chart you want to see or is there a specific real world image you are more interested in?

I don't think that LensRentals (Roger Cicala) thinks that "sharpness" is everything. Roger often points out that there are other considerations too. But MTF is what he decided to test and report. Fine with me. There are other sources where we can learn about the things he didn't cover.

Further, I don't think this lens is at a point (if there is one) where increased "sharpness" (I'd rather say MTF-values which is a combination of resolution and contrast criteria) becomes unnecessary (if there is such a point). The performance of a lens with regard to MTF is a gradual rather than discrete phenomenon and the effect is visible (in contrast to what many people appear to think) even at rather moderate display sizes (as an increased crispness and clarity if the lens has a good MTF curve).

As Roger pointed out in his post earlier in this thread

http://forums.dpreview.com/forums/post/50191873

he is rather satisfied with the lens but is nevertheless a bit disappointed in one specific regard, the reason for which is also spelled out in that post.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: Results indicate this is an f/4 lens
1

Horshack wrote:

The lens just came out, it'll take a while. In the meantime if you can find any review of the Canon 70-200mm II @ 200mm that shows its corner performance being 30% worse at f/2.8 than f/4 then you'll have proven your point and proven me wrong.

I don't believe that the 70-200mm would show a 30% degradation going from F2.8 to F4.0. I also haven't seen that in the 35-100mm F2.8 either.

Quite Frankly I don't believe the lensrentals chart at all. I haven't seen/shown anywhere near that difference.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: Results indicate this is an f/4 lens
1

If there was a problem with the 35-100mm and sharpness I would expect it to look more like this.

Please don't ever use F22.0 with an m4/3s camera unless you are actually photographing the sun.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: This lens is great wide open

mpgxsvcd wrote:

I think there is a much bigger difference between the F2.8 image and the F8.0 image than there is between the F2.8 image and the F4.0 image. However, that never stopped the “Pro” Panasonic photographer from shooting at F8.0 with this lens during the Olympics.

It would be a far bigger mistake to stop this lens too far down than it would be to shoot with it wide open. I still don’t trust any reviews that post opinions without any images to back up their claims.

There appears to be something wrong with your test shot at f/8.0. The next one in the series, at f/9.0, is noticeably sharper, which shouldn't be the case if the result at f/8.0 would reflect the increasing negative effect of diffraction. The one at f/9.0 looks roughly on a par with the one at f/2.8 to my eyes.

It's otherwise no surprise that f/8 isn't an ideal aperture for an MFT lens. Top-level MFT lenses often peak around f/4.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: Really an f/2.0 to f/2.8 lens
1

RCicala wrote:

That is exactly what I had in mind. Somehow people seem to have missed that I said "i'll buy one, it's the best zoom available." I had truly hoped that it would be as sharp at the long end as the near end, but it's still sharper at 100 than anything else I have available, and at f/4 it's perfectly sharp.

I've been shooting with it all weekend and enjoying it a lot. Very nice lens.

The problem is that a lot of people didn't get that from your review. It seemed like a lot of people ran with the fact that you said "where resolution dropped significantly". However, just before that you said "There was a SLIGHTLY different story at 100mm".

So which is it? "Slightly different" or "dropped significantly". Please show some examples that illustrate whichever stance you finally decide on.

How significant a difference is "100mm mtf 50 is 86% of what it is at 35mm". Those numbers sound alarming. However, we haven't seen any examples that illustrate that much of a difference.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: Really an f/2.0 to f/2.8 lens
1

What really disturbs me about the lensrentals conclusions and charts is the fact that 7 lenses were tested and their results at 100mm varied by about 15% and yet they still said this “The difference is worth showing in chart form since it’s quite consistent with all 7 copies I tested.”

The difference between the 35mm and 100mm results was 12.4% and yet the samples at each focal length varied by as much as 15%. That is not a good scientific test.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: Really an f/2.0 to f/2.8 lens
3

mpgxsvcd wrote:

What really disturbs me about the lensrentals conclusions and charts is the fact that 7 lenses were tested and their results at 100mm varied by about 15% and yet they still said this “The difference is worth showing in chart form since it’s quite consistent with all 7 copies I tested.”

The difference between the 35mm and 100mm results was 12.4% and yet the samples at each focal length varied by as much as 15%. That is not a good scientific test.

Do you know of any other lens testing site that takes the trouble to test seven different copies of the same lens and report the entire distribution? I don't.

Do I consider the conclusions reasonable in view of the evidence presented? Absolutely.

Am I happy that Roger does what he does even if, on occasion, he might find that a lens I own have this or that shortcoming. Absolutely.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
mpgxsvcd Veteran Member • Posts: 8,094
Re: Really an f/2.8 lens
1

Flawed results and conclusions are not better than no results or conclusions.

 mpgxsvcd's gear list:mpgxsvcd's gear list
Panasonic LX100 Panasonic Lumix DMC-GF1 Olympus PEN E-PM2 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GH4 +5 more
Anders W Forum Pro • Posts: 21,468
Re: Really an f/2.8 lens
4

mpgxsvcd wrote:

Flawed results and conclusions are not better than no results or conclusions.

My point was that the test, the results, and the conclusions are not flawed. Your understanding of them is.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
Keyboard shortcuts:
FForum MMy threads