I compare the shots between my 18-70mm and my 17-35mm and I see the difference. The 18-70 fringes and isn't anywhere near sharp wide open or closed down as the 17-35mm is (although at f5.6-f16 there isn't a noticeable difference)
[closed]
Lens Discussions - DxO Lab test etc ...
(70 posts) (19 voices)-
Posted 2 years ago #
-
SquamishPhoto said:
Talk about worrying about the wrong thing. Get over the lab results. You are obsessed and its getting old. The only lens you ever mention is the 17-35 and you seem just baffled that it even exists. Go rent one and make some comparisons for us cause Im getting pretty tired of reading you whine about this stuff. You've had some very exhaustive explanations given to clarify all this to you and yet you are still here bemoaning pro glass. We get it, you love your 18-135 and you feel like you've outsmarted everyone whos spent money a pro wide angle lens. We know how you feel about this, so let it die.15 people answered the topic thinking it is worthwhile and shared their experiences in detail before you showed your face with your usual charm & wisdom ( Thanks for holding that long ) . I had actually concluded the subject but felt like ( after apologizing ) I had to answer what testing123 said .
Check again , there is also the 17-55 mm f2.8 test ( it failed as well :-)) ) . The only reason I did not add more is that I did not want to bore everyone with similar results .
SquamishPhoto said:
If I want a sharp photograph I do not use a wide angle lens. I start with the 50mm and if it doesn't do it for me the 85mm does. And if that doesn't do it for me the 300mm crushes both of them. And I didn't use any lab info to figure that out. I just took pictures. You clearly need to spend more time with your camera and less time reading bullshit about tests on the internet.I understand I must use 50mm , 85mm and 300mm for sharpness in my landscapes ( ooops , you missed the part I mentioned that ! ) ?? It will be hard on a D300 but will try my best and keep you posted once I have the results :-)
Posted 2 years ago # -
"My conclusion - as many of you have implied - is that I am probably expecting too much from a 12 Mpixel APS-C sensor and I won't get any better sharpness ( let's say visible ) on my D300 no matter what I use . The "maximum" I should be buying without considering it "throwing away money " is maybe a used 12-24mm for $600 or so."
No, I dont think that is a good conclusion. Glass will last much longer than a body. so in a few years time if you got very good glass that out resolves your current sensor (Like I did(6 MP camera) then when you upgrade your camera you wont need to upgrade your glass to take advantage of the higher resolutions that are bound to arrive (which is as it should be) ( Unless you like to upgrade your lenses often :-) ! (which isn't a bad thing ! and can be part of the fun! )
Still as you are a land scape photographer I understand that sharpness is one of your higher priority parameters. Bokeh, and wide open performance is usually of no importance for landscape, neither is the shape of the of focus plane. Simply discounting these 3 parameters gives you a much wider choice of lenses that will give you equal performance for your subject. However, that is not to say that those 3 parameters would not be highly important to some other target subject.
Posted 2 years ago # -
True when you consider I will eventually end up buying some expensive glass as better sensors come out - and no harm doing it a bit ahead of time. Still , I'd opt to replace the D300 first with the upcoming model - which should cost me only half as much as a pro lens - and continue from there.
Can't agree more with the last paragraph ...
Posted 2 years ago # -
Paperman said:
@studio . . . I do run my own lens tests . . . I started doing them after I found it hard to believe some primes could not be as sharp as my simple kit 18-135 zoom. I'm sure it must have felt funny for you to see the $100 18-55mm kit being sharper than most.Ha! "Funny" wasn't quite the word I had in mind at the time, although, it did start with the letter 'F.' I haven't kept up with this thread for a while, but I did want to mention a couple of other points.
1. I shouldn't be surprised that a plastic kit lens outperformed my metal-barreled 18mm f/2.8 Nikkor. Fast glass is harder to design. Kit lenses, with their ultra-slow f/5.6 maximum apertures can keep flare, dispersion, aberration, etc. to a minimum because there's far less glass in the lens to mess things up.
2. While I'm no expert in optics, it does appear that soft wide-angle lenses are more noticeable, and therefore, more objectionable, than soft normal or telephoto lenses. Edge-to-edge sharpness in a wide-angle lens (in addition to being more difficult to attain in the lens' design) is likely to be under increased scrutiny, since in many wide-angle images, there happens to be a lot of detail spread throughout the frame (e.g., landscapes).
Whereas, in a normal-to-telephoto shot, often a subject is framed with a background, purposely, either slightly or severely, out-of-focus, and edge-to-edge sharpness becomes less of an issue, and therefore, is less noticeable (at least, for the types of wide-angle and normal-to-telephoto images I, personally, tend to shoot).
3. The main problem with DxO data isn't their methodology, it's their sample size. An externally valid sample would be far more than one or two copies. A lens with a given DxO score is no guarantee that the copy you buy will be anywhere near that mark.
4. The advisories given here to, "Stop worrying about numbers, just go out and shoot . . . " don't consider the fact that there's often a lot of time and expense involved in making a photograph. If your tools are sub-standard for personal photos, it may be of little or no consequence (other than sheer disappointment). However, if it's for a commercial client, it may require a re-shoot (at guess who's expense?). Many cinematographers personally perform acutance tests on lenses before they shoot a feature. Why? There's a lot at stake.
Posted 2 years ago # -
studio460 said:
2. While I'm no expert in optics, it does appear that soft wide-angle lenses are more noticeable, and therefore, more objectionable, than soft normal or telephoto lenses. Edge-to-edge sharpness in a wide-angle lens (in addition to being more difficult to attain in the lens' design) is likely to be under increased scrutiny, since in many wide-angle images, there happens to be a lot of detail spread throughout the frame (e.g., landscapes).Whereas, in a normal-to-telephoto shot, often a subject is framed with a background, purposely, either slightly or severely, out-of-focus, and edge-to-edge sharpness becomes less of an issue, and therefore, is less noticeable (at least, for the types of wide-angle and normal-to-telephoto images I, personally, tend to shoot).
True indeed ... We are worried about the whole landscape being sharp where as the tele or portrait shooter just cares about his object being perfect ( That's maybe why SquamishPhoto above thinks he's getting his sharpest results with a 300 mm . With all the blur in the background , everything looks sooo sharp ) .
Pro photographers shooting for commercial clients naturally can not take risks using " so so " equipment ; they have every reason to go for the best . Too much at stake / too much to lose moneywise/timewise in case of even a minor fuck up ...
Posted 2 years ago # -
Lots to read. Ansel Adams, landscapes. 35mm no. large format yes. Bigger sensor, bigger pixels. Just get the bigger sensor, learn the sweet spots of the lenses you have or get 1 new one per year if needs be. I have engineers in my family that love to roll around in the numbers, but all the portraits are in the dead center and the landscapes, well, they are few and unimaginative. Strive for balance, read some numbers, but trust that the pros who make and use these lenses know why the large aperture primes are sought. Photography is about balance. Science (an engineering) balanced with art. Go too far in either direction and it quickly becomes apparent to those around you that you are ... out of balance.
Posted 2 years ago # -
pbull221 said:
Lots to read. Ansel Adams, landscapes. 35mm no. large format yes. Bigger sensor, bigger pixels. Just get the bigger sensor, learn the sweet spots of the lenses you have or get 1 new one per year if needs be. I have engineers in my family that love to roll around in the numbers, but all the portraits are in the dead center and the landscapes, well, they are few and unimaginative. Strive for balance, read some numbers, but trust that the pros who make and use these lenses know why the large aperture primes are sought. Photography is about balance. Science (an engineering) balanced with art. Go too far in either direction and it quickly becomes apparent to those around you that you are ... out of balance.Balance is often good advice on any topic, but no one's arguing the fact that a 4x5 image can outperform an image from an APS-C sized sensor. My advice was hard-won, and I just want others to beware that, "caveat emptor," applies here, too, even to brand new, Nikon lenses:
1. Perform some kind of acutance test on any lens you buy, new or used, dealer or Ebay, prior to the expiration of your return period. There is for a fact, manufacturing variance, even in brand new, current-production Nikkor lenses.
2. Use DxO, or any other published quantitative data, to assist in making informed pre-purchase decisions, but know that the data may not be statistically significant.Posted 2 years ago # -
studio460, good advice. My point, and I am still working on being clear when I post, is that the bigger sensor is what will lead in the direction of satisfying this user most in the end. Perseverance about the nuts and bolts is fun, understanding the details of the physics at work is good for anyone doing this. But I personally learned more about making good pictures from Feininger's(sp?) Photographic Seeing than I ever did from all the tech studies I did combined. I landed on Nikon because of a gift when I was 10. I stayed with Nikon even though I could not afford it because they didn't go button crazy like Canon, and had for my eye, superior optics. And it was not superiority in resolution, it was a quality that I can only occasionally repeat, even today. Obsess on optimizing the technical quality of every move you make and you get to the point where you can't move at all. Example, Bokeh has a name, but listen to the definition: "The quality of blur" (my words). It is recognized that even this highly valued element has it's roots in a subjective interpretation created (in part) by engineers using their knowledge to create a thing that barely exists. Let's face it, when was the last time you studied the bokeh in your own eyesight, with or without glasses on. studio460, I especially like item #2. Use the numbers to "ASSIST, ... know the data is not necessarily significant."
Posted 2 years ago # -
pbull221 said:
studio460, good advice. My point, and I am still working on being clear when I post, is that the bigger sensor is what will lead in the direction of satisfying this user most in the end.Thanks, and I'm sure we're in general agreement, but not to be argumentative, this is a Nikon DSLR forum. We're primarily discussing how to wring every bit of technical quality out of DX and FX sensors that we can. Again, I think we all agree that shooting on a meduim-format Phase One back or on an 8 x 10 view camera will lead to superior results. A discussion about DxO data is both a relevant and legitimate topic. Topics discussed here are generally narrow in scope by design.
pbull221 said:
Obsess on optimizing the technical quality of every move you make and you get to the point where you can't move at all.While I get your point, I still maintain argument #4 in the previous post. The time and money spent producing a photograph warrants the best quality tools within our budgets. Also, that doesn't mean that obsessing about both technical quality and artistic merit are necessarily mutually exclusive pursuits. Ansel Adams was among the most technical photographers on record.
pbull221 said:
It is recognized that even this highly valued element has it's roots in a subjective interpretation created (in part) by engineers using their knowledge to create a thing that barely exists . . . studio460, I especially like item #2. Use the numbers to "ASSIST . . . know the data is not necessarily [statistically] significant."Thanks. Again, I get your point, but it's not necessarily relevant to the discussion here. Art is subjective. Optical performance is based on empirical data (although, interpretations of those data may be considered subjective). I think we can all agree on that.
I know what you're getting at, and it's been a constant theme throughout this thread. Some accuse others of obsessing about numerical data, with the included subtext, that it's the art of shooting that counts. Well, both things count. Photography is a technical art. The better the tools, and the better the understanding of those tools, the better baseline you have to start with.
Posted 2 years ago # -
Better for me personally to not go too far afield in the technical (theoretical) aspects of this craft. I spent many years reading and reading about the points in macro photography that were required to make that photo work. What I turned into for years was, as my martial arts teacher called, " a technique hound." So preoccupied with the hows and the whys that it actually drew me away from making photographs with what I was learning. I used Ansel's name in vain to over emphasize the point that his ultimate end product was the print on the wall. Well, I'd like to be more like that too. I usually spend more time in the discussion of the cool toys (I admit, I am a tech head too,) than gazing at the photo. In Astronomy, as in Micro-photography we are rendering things beyond the human eye. My absolute favorite lens is my 200mm Micro Nikkor, a macro lens. The scientists, engineers, and managers at Nikon are rendering tools that allow me, personally, to approach the recording of my child's eye view of the wonder of the beauty that surrounds all of us to one degree or another. I have a D300s, high end photo printers can not reproduce what I can see on my computer screen, and it is not even a 10bit monitor yet. I seldom print beyond 16 X 20 " and can thus far not justify starving myself and my dogs to get the D3X. Is the original posting about landscape photos that will be used to cover a wall? If so, I submit she/he will want the bigger sensor, the high end primes, and a portable weather station that can tell him where the cloud deck is at sunrise so the reflected light from the clouds will cover his subject in a light of the quality that will make people want to put all their savings into establishments where this perfect photo will be located. My problem with DxO is that as Bruce Lee said, "... boards don't hit back." When you study quantities of flat targets you get a tiny glimmer of what the lens would do with a flat subject. Life is not flat. Is the thread about what it looks like when you take out your loupe and compare just how the system recorded the detail in that security feature of that Yen note? Maybe I am too stupid to understand what has been going on here, or maybe the coffee is finally kicking in. I'd like to see some images from the original poster's collection, even with his most limited lens, I bet we could find some unique elements that could be improved on but have not got a chance of being repeated. Let's hear it for those amazing Japanese inventors who have pushed the limits of this art so far, as to be so unique, that a trained eye could likely see an image and tell which camera took it. Let that be their signature, let that be their standard, for 500 years.
Have a great debate, all this reading has made me want to go take a few hundred photographs with the camera system that satisfies my inner eye - Nikon - .
Posted 2 years ago # -
pbull221 said:
I used Ansel's name in vain to over emphasize the point that his ultimate end product was the print on the wall.It's funny (and, not the "ha-ha" kind) how few of our treasured images do make it to prints on the wall. I'm planning to rectify that situation this year with some large digital Type-C prints, and some silver, black-and white-prints, printed from actual, drum-scanned, Tri-X film.
pbull221 said:
I seldom print beyond 16 X 20 " and can thus far not justify starving myself and my dogs to get the D3X. Is the original posting about landscape photos that will be used to cover a wall? If so, I submit she/he will want the bigger sensor, the high end primes, and a portable weather station that can tell him where the cloud deck is at sunrise so the reflected light from the clouds will cover his subject in a light of the quality that will make people want to put all their savings into establishments where this perfect photo will be located.Ha! I really enjoyed that post! I just fear we'll soon be scolded for going off-topic. I could talk about morning cloud cover and flat targets all night long! It's true, many of us don't have the wherewithall, stamina, means, or dedication it takes to make outstanding landscape photographs.
Many of us take the occasional 10-14mm wide landscape or two, maybe even a sunset (hopefully, not dead-center in the frame). But, it's true, few of us will aspire to the really great landscape images of the masters (the ones who rent the cottage, sit for days or weeks, with their view camera in tow). But, I digress . . .
Posted 2 years ago # -
I don't see that discussion as off-topic, studio. I think the original question was basically about how much weight should tests like these be given in lens buying decisions, and you and pbull have been discussing exactly that.
Posted 2 years ago # -
Paperman said:
While looking for a new lens as always for my D300 , I checked DxO Labs page to compare my current $150-250 lenses ( 50mm 1.8, 28mm 2.8, 18-135mm ) with what is supposed to be a dream lens - the $2000 17-35mm f2.8D IF-ED .My take on everything . . . seems like you have a stacked an argument by making some false claims to make things appear more in your favor, and it actually troubles me that everyone seems to be fighting more about these lenses compared to each other and scratching their heads, so I will spell out my take on things.
The Dream Lens (making the competition seem better than it really is)
I think you are a bit biased (and mentally high) that your lens scored so well, but you are not making a true comparison. First you make the 17-35mm seem better than it really is in your first paragraph, calling it a "dream lens".
I would never dream of a 17-35mm the way you do, since you called it a dream lens. I never had the desire to get it, it is not a perfect lens. It was designed for speed at the time it was released, in a time where digital cameras were first starting to come out, and they were only DX format anyway, so a compromise was made to make it a suitable wide angle range for DX (26mm - 52mm).The High Cost (making the competition seem more expensive than it really is)
You state the costs of the lenses, comparing your $150-$250 lenses to a $2000 lens. Well your first mistake is that you are comparing some used costs of yours, to a new cost of a discontinued lens. If you look on eBay, the 17-35mm ranges from $650 to $1000. If you want to buy it new for $2000, you are an idiot, but I will get to that in a bit. Let's just say for argument sake that you inflated the cost of the lens to make it seem higher than it really is.The Model T vs The DeLorean
The 17-35mm was released in 1999, the 18-135mm was released in 2006. In that 7 years between lenses coming out, much has changed. Lenses are designed 100% by computers now. We have AF-S, and NANO coatings. Lenses that were impossible before are absolutely possible now. Where was the 14-24mm until the last few years? It didn't exist because it was impossible to design without computers.
So now you want to compare lenses that were designed "generations" ago. It has been 10 years since the 17-35mm was released, and I think a fair comparison would be with a 16-35mm instead. Comparing the 18-135mm and the 16-35mm would be a massacre and it has only been about 4.5 years between these two lenses. Also the cost of about $1000 for the 16-35mm. How about some other modern lenses for comparison. Compare the 18-135mm with a 14-24mm, or a 24-70mm. Same difference, there is a huge difference about them, whether you admit it or not.This brings up my idiot point again. You would be an idiot to buy a new lens, and for it to be a 17-35mm.
Posted 2 years ago # -
Segura said:
The Dream Lens (making the competition seem better than it really is)
I think you are a bit biased (and mentally high) that your lens scored so well, but you are not making a true comparison. First you make the 17-35mm seem better than it really is in your first paragraph, calling it a "dream lens".
I would never dream of a 17-35mm the way you do, since you called it a dream lens. I never had the desire to get it, it is not a perfect lens. It was designed for speed at the time it was released, in a time where digital cameras were first starting to come out, and they were only DX format anyway, so a compromise was made to make it a suitable wide angle range for DX (26mm - 52mm).You're confusing the 17-35, which predates Nikon digital SLRs, with the 17-55.
The High Cost (making the competition seem more expensive than it really is)
You state the costs of the lenses, comparing your $150-$250 lenses to a $2000 lens. Well your first mistake is that you are comparing some used costs of yours, to a new cost of a discontinued lens. If you look on eBay, the 17-35mm ranges from $650 to $1000. If you want to buy it new for $2000, you are an idiotSo I get edited for saying ****, but a personal attack is cool.
We'll ignore the fact that it has already been settled that he's hitting the resolution cap of his camera and just leave your words to stand for themselves.
Posted 2 years ago # -
The Real Comparison
So I went to DxOMark and compared the 18-135mm, the 17-35mm, and the 24-70mm. The following is their score and lp/mm resolution.
Nikon D90:
18-135 Score 12 / 43 lp/mm
17-35 Score 7 / 38 lp/mm
24-70 Score 15 / 46 lp/mmNikon D300:
18-135 Score 10 / 45 lp/mm
17-35 Score 7 / 40 lp/mm
24-70 Score 14 / 48 lp/mmNow unfortunately your 18-135mm cant compete in the full frame tests, where we see the following results:
Nikon D700:
17-35 Score 14 / 45 lp/mm
24-70 Score 23 / 48 lp/mmNikon D3S:
17-35 Score 18 / 48 lp/mm
24-70 Score 28 / 51 lp/mmNikon D3X:
17-35 Score 15 / 57 lp/mm
24-70 Score 28 / 61 lp/mmSo what we see from the results is the best resolution the 18-135mm can do is 45 lp/mm. The other lenses were designed for full frame, not DX, so their results show that. We see that on FX, the "crappy" 17-35mm starts from 45 lp/mm and goes up to 57 lp/mm. Wow.
Now we get the results from a modern pro zoom lens. The 27-70mm. It ranges from 48 lp/mm - 61 lp/mm. Even better than the old lens design, in fact it is entirely possible Nikon optimized it for DX as well as FX.
So the conclusion? The 24-70mm bested the 18-135mm easily. The 17-35mm on an FX body is actually pretty good, resolving up to 57 lp/mm, where the best the 18-135mm could do was 45 lp/mm on DX.
But honestly, compare modern lenses to modern lenses. You will really see the difference.
Posted 2 years ago # -
Segura said:
So what we see from the results is the best resolution the 18-135mm can do is 45 lp/mm. The other lenses were designed for full frame, not DX, so their results show that. We see that on FX, the "crappy" 17-35mm starts from 45 lp/mm and goes up to 57 lp/mm. Wow.
When the 17-35 performed poorly I thought the problem was that the 17-35 was a Model T and the 18-135 was a Delorean?
When you have numbers showing it performing well, now the problem is the D300 is DX and the 17-35 is a FX lens?
Pick one.
Better yet, understand what the numbers mean.Now we get the results from a modern pro zoom lens. The 27-70mm. It ranges from 48 lp/mm - 61 lp/mm. Even better than the old lens design, in fact it is entirely possible Nikon optimized it for DX as well as FX.
So the conclusion? The 24-70mm bested the 18-135mm easily. The 17-35mm on an FX body is actually pretty good, resolving up to 57 lp/mm, where the best the 18-135mm could do was 45 lp/mm on DX.
Again, the fact he is at the D300 resolution limit has long been established.
Posted 2 years ago # -
Testing123 said:
You're confusing the 17-35, which predates Nikon digital SLRs, with the 17-55.
So I get edited for saying fuck, but a personal attack is cool.
We'll ignore the fact that it has already been settled that he's hitting the resolution cap of his camera and just leave your words to stand for themselves.1) I am not confusing the 17-35mm wit the 17-55mm
2) The 17-35mm does not pre-date Nikon digital SLR's. In fact it was released in 1999 at the same time as the D1.Here is a quote as to why this focal length was made:
"To offset the effect of apparent focal length lengthening, Nikon have designed a special 17 - 35mm f2.8 zoom lens which will serve as an excellent 'standard' lens for the D1. This ultra-wide zoom has lens angles approximately equivalent to 25.5 - 52.5mm on a 35 film camera. No price has been given for this lens. Although it's good to have a zoom, it will need a well-designed rectangular lenshood which takes into account the reduced format size."So what personal attack are you pointing out, and you dropped the f-bomb again . . . stop taking it so personal, and get your info straight (correct).
My point was that the 17-35mm is not a good comparison because it is an old design, that was probably made in a hurry to get out when the D1 was released. It is not a good modern lens.
Posted 2 years ago # -
Testing123 said:
When the 17-35 performed poorly I thought the problem was that the 17-35 was a Model T and the 18-135 was a Delorean?
When you have numbers showing it performing well, now the problem is the D300 is DX and the 17-35 is a FX lens?
Pick one.
Better yet, understand what the numbers mean.
Again, the fact he is at the D300 resolution limit has long been established.Jesus Testing123, why do you take it so personal? I guess you have this lens too . . .
1) I never said the 17-35mm performed poorly, those were your words. I think it did okay, but it is an old design.
2) I can pick more than one reason why a lens does not perform as good as another if there are more than one reason. Why pick one?
3) Do you understand what the numbers mean?Take a time out man and relax. Your lens is okay for your camera, but try a 16-35mm instead.
Posted 2 years ago # -
OK guys enough of the back and forth. Consider this a warning to both of you. Personal attacks and "extremely harsh" language are NOT OK. Anyone who uses the F-word again will be dealt with according to the forum rules. PM me if you want to debate what an adult forum means.
This is thread has boiled over so I'm locking it.
Posted 2 years ago #
Topic Closed
This topic has been closed to new replies.
