Lumix 28–200mm f/4–7.1 Sample Images

[pdk42]

With all due respect, that is not what I would consider "tack sharp." The feather detail, while OK, is not great. I have seen more detail (at 200mm) from the 70-300, 100-400, and 150-600, and I'm sure any of the 70-200s that are available on L-mount would as well. Now, to be fair, ISO 8000 will never allow you to see the full potential of the lens, even when using DxO. The noise level is remarkably low, however.

I'd love to learn that some copies of the 28-200 are sharp & contrasty at 200mm, but IMO this image doesn't show that.

But it is a nice shot of a beautiful bird.

I agree George. Even at the highest resolution available on the post, it's clear that the feather detail is somewhat compromised.

But - as Carter Bresson said - "sharpness is a bourgeois concept" ! A good image isn't necessarily an uber sharp image.

 

[GeorgeHudetz]

As Roger Cicala (Lensrentals.com) pointed out, the final resolution of a shot is a result of both the lens's resolving power and that of the sensor. He says a way to think about it in simple terms is to imagine a "resolution factor" for both lens and sensor. A perfect example of either would have a factor of 1. Real lenses and sensors would have values less than 1. He then says that to get a feeling for the combination, you should multiply the factors together.

So:

- A 0.7 lens on a 0.7 sensor would give you a final score of 0.49.
- A 0.9 lens on a 0.5 sensor would give you a final score of 0.45.
- A 0.5 lens on a 0.9 sensor would give you a final score of 0.45.

The message therefore is that a weak lens on a higher res sensor will deliver a better overall image resolution than the same lens on a lower res sensor.

Although, interestingly, I think the observation that HR mode helps overcome less-than-stellar optics to be something else. After all, the HR shot only requires the lens to resolve (in this case) a series of 24 MP images. It's the sensor shift used to generate that series of images & the combination algorithm that delivers the higher resolution. I've always scoffed at using anything but my best lenses for HR (HR mode is actually a great way to expose optical flaws like soft corners) but I could see it also as a way to overcome those soft corners. Never thought about it that way, but it makes sense.

Interesting. As always, work with what you've got.

 

[GeorgeHudetz]

Just did a quick test, and indeed a HR shot from the 28-200 @ 200mm (F10), which is then downsampled back to 6000x4000, recovers that missing clarity - at least in the center of the frame - and also offers more clarity than a standard 200mm shot from the 70-300 @ F8. Nice trick, but it requires a tripod (at least on my older bodies).

 

[Prosecutor]

With all due respect, that is not what I would consider "tack sharp." The feather detail, while OK, is not great. I have seen more detail (at 200mm) from the 70-300, 100-400, and 150-600, and I'm sure any of the 70-200s that are available on L-mount would as well. Now, to be fair, ISO 8000 will never allow you to see the full potential of the lens, even when using DxO. The noise level is remarkably low, however.

I'd love to learn that some copies of the 28-200 are sharp & contrasty at 200mm, but IMO this image doesn't show that.

Yep, was the worst of the series. Look at this selection:

Lumix 28-200

www.picdrop.com

 

[Prosecutor]

For those who fear external links

• Panasonic - DC-S5
• LUMIX S 28-200/F4-7.1
• 200.0 mm
• ƒ/7.1
• 1/125 sec
• Pattern
• Auto exposure
• ISO 12800

Yes, it´s high ISO again. But the S5 doesn´t care about ISO 12.800.

 

[GeorgeHudetz]

These are all nice shots, but none of them convince me that your copy of the 28-200 is "tack sharp." Not that you need to care about what I think, but if you really want to back up your claims, post a raw file.

 

[Prosecutor]

I don‘t want to „convince“ anybody of this lens. Just showing the results that I am happy with. It’s better than I expected and there is no alternative if you need a light and small superzoom. I also own the Sigma 70-200/2.8 and know what perfect (!) sharpness means. But it‘s senseless to compare these lenses.

 

[Markuswelder]

I don‘t want to „convince“ anybody of this lens. Just showing the results that I am happy with. It’s better than I expected and there is no alternative if you need a light and small superzoom. I also own the Sigma 70-200/2.8 and know what perfect (!) sharpness means. But it‘s senseless to compare these lenses.

Indeed. Tack sharp does NOT mean the sharpest lens ever made, or available. There is a difference. I think my tiny little manual focus 26mm f8 bodycap lens is tack sharp. Are there sharper lenses out there? Without a doubt.

 

[GeorgeHudetz]

I don‘t want to „convince“ anybody of this lens. Just showing the results that I am happy with. It’s better than I expected and there is no alternative if you need a light and small superzoom. I also own the Sigma 70-200/2.8 and know what perfect (!) sharpness means. But it‘s senseless to compare these lenses.

Oh I agree - comparing the 70-200 to the 28-200 does not make sense. After all one is very sharp - one might even say "sharp as a tack." The other is a little soft and fuzzy, at least at 200mm. So, maybe "butter-knife sharp?" LOL, OK, I'm just having fun now.

Anyway, your images are fantastic, and if any of those are actually 100% crops from a 24 MP image, then those images are testimony as to what DxO can do - super impressive. Like you, I am quite happy with the images produced by my 28-200, and it fulfills an important niche that had been missing from L-mount. And I also agree, it exceeded my expectations as well. I'll be traveling with mine soon, and leaving all the bulky lenses at home.

I just honestly we don't try to redefine "tack sharp" with the output of this lens at 200mm.

 

[CharlesH]

I'll be traveling with mine soon, and leaving all the bulky lenses at home.

Same here, I'm on a trip starting Saturday and will take the 28-200. I've grown to like it as a carry-around lens. I do have DxO PrueRAW as a Lightroom plugin should I really need it.

 

[Nevyn72]

Same here, I'm on a trip starting Saturday and will take the 28-200. I've grown to like it as a carry-around lens. I do have DxO PrueRAW as a Lightroom plugin should I really need it.

I love my 28-200, but in a crowded city I definitely need a wider lens as well. For me, this and the Sigma 16-28 F2.8 make a dream couple, especially as I also have a low-light lens with me.

 

[GeorgeHudetz]

Also wanted to point out that your (@Prosecutor) skill at wielding the tools - both camera and PP - is excellent. And that, in the end, is what is most important.

 

[inspiredtimothy]

Same here, I'm on a trip starting Saturday and will take the 28-200. I've grown to like it as a carry-around lens. I do have DxO PrueRAW as a Lightroom plugin should I really need it.

DXO Photolab 8 is out now and supposedly includes better corrections for raw profiles that reduces sharpening artifacts. I can confirm from examples I've seen that DeepPrime XD 2 or whatever the new denoise algorithm is also reduces artifacts a ton and makes the advanced NR introduced after DeepPrime useable.

 

[inspiredtimothy]

I love my 28-200, but in a crowded city I definitely need a wider lens as well. For me, this and the Sigma 16-28 F2.8 make a dream couple, especially as I also have a low-light lens with me.

I was kind of disappointed with the 16-28. It always seemed like no matter the aperture there was a huge mid zone dip and I was fighting with raw processing to get an image I was happy with. Tried several copies of the lens, too.

 

[Nevyn72]

I was kind of disappointed with the 16-28. It always seemed like no matter the aperture there was a huge mid zone dip and I was fighting with raw processing to get an image I was happy with. Tried several copies of the lens, too.

I use it mostly for video, but I'm very happy with my copy...

 

[pdk42]

Oh I agree - comparing the 70-200 to the 28-200 does not make sense. After all one is very sharp - one might even say "sharp as a tack." The other is a little soft and fuzzy, at least at 200mm. So, maybe "butter-knife sharp?" LOL, OK, I'm just having fun now.

Anyway, your images are fantastic, and if any of those are actually 100% crops from a 24 MP image, then those images are testimony as to what DxO can do - super impressive. Like you, I am quite happy with the images produced by my 28-200, and it fulfills an important niche that had been missing from L-mount. And I also agree, it exceeded my expectations as well. I'll be traveling with mine soon, and leaving all the bulky lenses at home.

I just honestly we don't try to redefine "tack sharp" with the output of this lens at 200mm.

There’s a way to measure sharpness objectively on an optical bench - it’s called the MTF and acuity values are measured in line pairs per millimetre. Not sure how big a standard tack is, but if we really want to be objective in measuring lens performance, we need an optical bench and a machine to measure the MTF.

 

[CharlesH]

DXO Photolab 8 is out now and supposedly includes better corrections for raw profiles that reduces sharpening artifacts

Great to hear. Hope this works its way to DxO PureRAW.

 

[GeorgeHudetz]

I was kind of disappointed with the 16-28. It always seemed like no matter the aperture there was a huge mid zone dip and I was fighting with raw processing to get an image I was happy with. Tried several copies of the lens, too.

Not sure if you shoot landscape, but I consider my Sigma 14-24 DG DN Art to be my finest landscape lens (which means: how it performs at F8). There is a bit of distortion on the wide end, but unless you are willing to buy a big & expensive 14mm prime, I'm not sure you can do much better. It's fantastic at 24mm (better than the Sigma 24-70 Art), and even beats my Sigma 20 F1.4 DG DN Art at F8 in the corners, due to the correction distortion the prime requires. Of course, the 20 is super sharp in the center at F4, but at F8 it's mostly the same as the zoom, but with less correction, the zoom gives better corners.

 

[GeorgeHudetz]

There’s a way to measure sharpness objectively on an optical bench - it’s called the MTF and acuity values are measured in line pairs per millimetre. Not sure how big a standard tack is, but if we really want to be objective in measuring lens performance, we need an optical bench and a machine to measure the MTF.

Yes indeed, but I've sort of given up finding objective test numbers for LUMIX L-mount lenses. If you know of a good site I'd love to know.

 

[SoJuJo]

I'm just trying to sell my FZ1000 II as the 28-200 took its place. Not completely, but complete enough. In front of the fp L I can crop easily what I'd get from the "400 mm" equivalent (146mm) out of the fp L files and still have plenty of reserve. On my last trip through UK I used it for 68% of all images, super-versatile, easy to carry.