NikonGear'23
Gear Talk => Lens Talk => Topic started by: Tristin on December 17, 2015, 04:13:14
-
The f/1.8 is sharper and has more contrast, but I noticed that the copy of the f/1.8 I picked up is soft on the left hand side, so I didn't bother including a comparison of the left hand side as it seems to be a defect in this individual lens. All images shot in raw at the same exposures and converted with no adjustments or sharpening
Center
(http://i.imgur.com/fXurB3M.jpg)
Right Edge
(http://i.imgur.com/o3LSe58.jpg)
The f/1.8 has much more noticeable ghosts than the f/2.5, but also handles flare better. The f/1.8 produces more rays that are also more tightly contained than the f/2.5 does. The f/2.5 can produce more focused rays but it requires stopping down to f/22 or f/32 (which is apparently only on pre-Ai converted with the factory kit to Ai).
(http://i.imgur.com/tzlTOcs.jpg)
I was concerned about how short the f/1.8's hood is, but it worked perfectly at blocking ghosts I produced by aiming near the sun. So that was nice. With hoods out the f/1.8 is shorter, while with hoods off the f/2.5 is shorter. The HS-4 for the f/2.5 sure is irritating, though I wouldn't hold that against the lens itself. The aperature ring is a bit easier to access with the f/2.5.
Assuming I can get a copy of the f/1.8 that is not soft on the left side, I can't see where the f/2.5 would outperform it except in ghosting. The bokeh on both were so similar that I did not bother including a comparison. The f/1.8 is virtually vignette free at f/2.8, where-as the f/2.5 reaches vignette free at f/4.
-
As to edge performance, keep in mind that de-centering sometimes brings the sharpness towards the "non-affected" side.
Could the difference in star bursts be due to the more rounded diaphragm on the 105/2.5 AI, vs the 105/1.8 AIS that has straight blades? (May that matters less at f.11?)
Here is a sample from the 105/2.5 AIS stopped down to f/6.3 only; it is a crop to <1/2 of the frames length. It does not show the diffuse tips of the star bursts as in your example from the AI version:
(http://otoien.zenfolio.com/img/s5/v124/p1780735576.jpg)
NIKON D5100, f/6.3 @ 105 mm, 4s, ISO 100
-
The f/1.8 has much more noticeable ghosts than the f/2.5, but also handles flare better.
Comparing the two night shots at f/11, I'm not sure that is flare, the overall exposure with the 105/2.5 seems a shade brighter? It would also pay to ensure that all filters are removed and the lens surfaces are spotless, and check for internal haze or dust, which could easily affect the results here.
The f/1.8 produces more rays that are also more tightly contained than the f/2.5 does. The f/2.5 can produce more focused rays but it requires stopping down to f/22 or f/32 (which is apparently only on pre-Ai converted with the factory kit to Ai).
The f/1.8 has straight edge aperture blades, which will give nice sharp rays. The AI and earlier f/2.5 has curved aperture blades so the rays will be slightly spread out and diffuse (a perfectly round opening won't produce star rays at all). As you stop down the curvature becomes less relative to the size of the opening so the rays are more defined - also diffraction is greater at small apertures so the rays are larger. If you want a 105/2.5 to produce nice tight star rays, the AIS version is better since it has straight edge aperture blades.
And yes you are correct, an Ai converted 105/2.5 (gauss/black barrel version) stops down to f/32. For some strange reason the AI conversion ring retains the f/32 stop from the pre-AI lens, while the native AI lens only goes to f/22.
-
No filters were used and the lens were cleaned before testing. I thought the same on the flaring at night but multiple tests showed the f/2.5 to exbibit this behavior. Looking at the two at full resolution the f/1.8 images show the same details, but it looks darker because the contrast is much higher. The f/1.8 does have a slighter higher transmission, but the contrast drop on the f/2.5 was much too drastic to be attributa le to that.
I should add that the f/2.5 had consistent edge to edge performance right from f2.5 where-as the f/1.8 took until f/4 to achieve the same consistency edge to edge. At f/2.8, the f/1.8 was noticeably sharper and contrastier in most of the frame, at f/4 it was better everywhere. I find the lack of corner performance at f/2.8 a non-issue as I would normally use that aperature, or lower, for portraiture. So the corners don't matter at that point for me. If I want all over sharpness, I generally want more DoF, so stopping down to f/4 would be routine even if the lens had sharp corners at wider aperatures. The f/2.5s consistency wide open is certainly phenomenal, but I wouldn't shoot that wide an aperature on a 105mm for landscaping, etc.
You are definitely correct on the curved/straight blades. Unless the Ai-s version doesn't have contrast issues when presented with light sources at night though, I wouldn't find more refined rays to be much help.
-
The f/1.8 has a bit of veiling flare at the widest apertures as you already have discovered. Already at f/2.8 this is nearly gone. In direct comparison with landscapes I could not detect any significant difference between the f/1.8 and f/2.5 (AI) by f/5.6 and smaller.
The star burst appearance depends on the actual model of the f/2.5, as Roland already has elaborated. My overall preference is for the AI (or AI-modified mid '70s P.C/K versions), but optically speaking little differentiates the models after the change from the 'Sonnar' to the 'Gauss' in the early '70s. The better bokeh and handling are my main reasons for preferring these versions, but no doubt the latest AIS has newer coatings to give slightly more contrast. Each to their own. There are plenty of the various 105 models floating around and price is no stumbling block to get alternative samples.
The best hood for the f/2.5 is an HN-8. For the f/1.8 I'm using a no-name 62 mm screw-in type of the same build, but shorter. The slide-out hood is annoying.
-
"....and check for internal haze or dust, which could easily affect the results here."
My copy of my 105/2.5 AIS used for the image above is what I would almost call "horribly dusty" inside. I was able to compare it to a much less dusty sample of the same vintage with controlled torture tests of an LED flashlight shining into the lens at some distance etc., and if anything my copy caused less flare. So just do not look into these lenses, and everything is fine. ;D
(Haze/very fine dust particles would be an entirely different matter though).
-
Some internal dust inside contributes surprisingly little in terms of degrading the image and contrast. However, the grey film deposited on element surfaces as happens in a smoker's environment certainly makes image quality take a nosedive.
-
However, the grey film deposited on element surfaces as happens in a smoker's environment certainly makes image quality take a nosedive.
When does that occur? Does it imply leaving the lens unprotected in a smoker's room for several hours, days or months?
-
All lenses I have examined that originate from smoker's home have had this grey film sheen over the inner glass surfaces (and presumably, front and rear surfaces but here cleaned off). I am not going into smoking* practice to study how long it takes for the problem to develop, but assume this problem gets worse over time.
* quit smoking 35 years ago, a decision for which I am very grateful today :D
-
All lenses I have examined that originate from smoker's home have had this grey film sheen over the inner glass surfaces
I haven't noticed that, but that's not really surprising at all. There ARE holes and slits in the lenses through which tiny particles of the smoke can invade and settle on the internal coating.
-
Obviously a lens isn't hermetically sealed (unless made for submerged use). I had a nice demonstration of this fact when I recently operated one of my 50 mm f/0.75 lenses to remove a hatched small bug on its inside. The speck of the dead bug was big enough to be visible in the photos. I just considered it being some ordinary dust until looking closer and found it was indeed a bug with feet and antennae and all ... including dried excrements around it.... No idea how it got there or what it fed off inside the lens :D must have been there for years though.
-
It does imply careless treatment of your gear. I think it's normal to keep all the photo-equipment in a smoke-free room and in a dust-free cabinet.
-
I keep my photo equipment in a smoke-free and reasonably dust-free house :D
Compared to fungus attacks, inside bugs do far less damage.... The bug probably accompanied the lens when I purchased it (from the Netherlands) and hatched later
-
I keep my photo equipment in a smoke-free and reasonably dust-free house :D
good to know that Peugeot is building clean rooms
-
Gasfree? :)
-
Thomas made a joke about my 'lens matrix storage' Peugeot ... The fact is of course most lenses stay most of the time in the 'lens house' which is totally smoke free (anyone foolish to try to light up there would be evicted by the back of their pants before they understood what happened to them) and reasonably free of dust.
-
Obviously a lens isn't hermetically sealed (unless made for submerged use). I had a nice demonstration of this fact when I recently operated one of my 50 mm f/0.75 lenses to remove a hatched small bug on its inside. The speck of the dead bug was big enough to be visible in the photos. I just considered it being some ordinary dust until looking closer and found it was indeed a bug with feet and antennae and all ... including dried excrements around it.... No idea how it got there or what it fed off inside the lens :D must have been there for years though.
That's rather surprising. Rayxar should not have any movable part like aperture ring, aperture lever or focusing ring which all has some slits or holes. So, the inside of the optics should be pretty much tightly sealed...
-
Years ago I tested five 105mm lenses: 105/2.5 Nikkor-PC, 105/2.5 AI, 105/2.5 AIS, 105/2.8 AIS Micro and a 105/1.8 AIS Nikkor. I used a trusted Nikon F2 with a 6x, DW-2 finder. For focusing I used a 1000 watt quarts flood. I turned off the floodlight after focusing, opened the shutter on B with a locking cable, let the vibrations settle for one minute and exposed my film in total darkness with a flash from a Vivitar 285. The film was Kodak Tech Pan developed as a continuous tone film in Kodak Technidol LC. I viewed with a enlarger at 8x, Fujinon-EP 50mm f3.5 enlarging lens and a heavy 10x Omega critical grain focuser. This was the best I could figure out at the time. My test distance was 2 meters a common distance for portraits.
What I found was the best lens was the 105/2.5 AIS and almost tied was the 105/2.8 AIS Micro, then very close the 105/2.5 AI, then 105/2.5 Nikkor PC quite close. I figured the difference in the f/2.5 lenses was sample variation. I today doubt that the difference in the f/2.5~f/2.8 lenses would be seen in common sized prints. Bringing up the tail was the 105/1.8 AIS. All the others beat it from f/2.5~f/2.8 through f/5.6. All were equal by f/8.0. I had special ordered the 105/1.8 AIS so I sold it after a week at a loss as I could not return it. I also tested an 85/1.4 AIS and an 85/2.0 AIS. The f/1.4 was best with the f/2.0 lagging. The 85/1.4 was soft at f/1.4 but beat 85/2.0 from f/2.0 to f/5.6 and then they were equal by f/8.0. I didn't find the 105/1.8 useful anyway as it was quite soft unless stopped down to f/2.8. The 85/1.4 on the other hand was quite sharp by f/2.0.
I tested more lenses and would have liked a 105/2.5 to be the sharpest but it wasn't to be. A 50/1.8 AI and 55/2.8 AIS were tied for sharpest. I never did lens testing like this again.
---
At Gayson's Camera, Glendale California where I had an open account I saw a friend cleaning the window of a film refrigerator. The cleaning cloth had a dark brown residue on it when finished. I told my friend who smoked, “That's Tar.”
Dave
-
Purely out of interest where in the quality stakes does the 105/2.8 AF ( 1st generation AF ) sit ?, it was bought to go with the F4 and now sits fitted to my old F4 as a reminder of days past alongside my desk.
-
The early 105/2.8 Micro-Nikkor in AF and AF-D had a reputation for poor bokeh as I recall. I remember samples photos posted and the lens didn't interest me. For Portraits and close-up where a tripod can't be use I'd like an AF-S VR Micro-NIKKOR 105mm f/2.8G IF-ED but for macro and close-up from a tripod I'd use my 105/4.0 AI or 105/2.8 AIS Micro-Nikkors. I understand that the AF-S G breaths rather heavily and requires a lot of tripod re-positioning for close-up and macro. It doesn't sound fun.
I find VR very helpful for hand held close-ups near or just after sunset when using a friend's AF-S 16-85/3.5-5.6G ED VR DX Nikkor.
Dave
My 105s are a 105/2.8 AIS, 105/2.8 AIS and 105/4.0 AI.
-
Purely out of interest where in the quality stakes does the 105/2.8 AF ( 1st generation AF ) sit ?, it was bought to go with the F4 and now sits fitted to my old F4 as a reminder of days past alongside my desk.
It is not a bad performer. Not at all. In terms of sharpness not far from the renowned 105/2.8 AIS Micro-Nikkor. However, it could render the background a tad harshly if you were careless with background detail.
Its AF was glacial and more a joke than offering any practical value.
Do note it had the same tendency to break-up of the A/M switch ring as several other Nikkors of the same era.
-
My experience with the AF-D 105/2.8 micro echos other here. It is respectably sharp at medium-far distances, but bokeh is poor. Bokeh is fine at close range. I mostly used it mounted on a PN-11 for macro shots beyond 1:1 up to about 1.6x life size. Working distance was about right for this magnification and performance was very good (can't comment about wide open performance as I always used it stopped down for more DOF). The bokeh was pleasing, only the straight-edge 7-blade aperture can sometimes show up in out of focus blurs.
Manual focus feel was good for an AF lens, it lacks the buttery smooth feel of manual lenses, and has a plastic friction feel, but is otherwise reasonably smooth. Focus throw is very short (just 180° from infinity to 1:1) which makes accurate focusing tricky at far distances, but fine at close range.
-
Still doing some testing out of curiosity when time permits. This time around, a test of their close range performance with a poster at a bit over a meter away. All shot in raw, focused with live view, mirror lockup and converted with no sharpening or correction. Both lenses are clean as a whistle. Only thing modified was the WB to more accurately judge their color, the f/2.5 consistently AWB'd a hair more magenta than the f/1.8, +2-3 in ACR 9, which I matched to the f/1.8.
Full shot of the poster for reference
(http://i.imgur.com/ZChT9He.jpg)
(http://i.imgur.com/0miNW5v.jpg)
(http://i.imgur.com/1ITtqoH.jpg)
(http://i.imgur.com/H88eIt9.jpg)
(http://i.imgur.com/N2Hdpai.jpg)
At near range, the f/1.8 is superior at all aperatures and at all parts of the image. Most notable is the lack of CA in the f/1.8 while the same cannot be said for the f/2.5.
I am going to redo my distant performance tests as well. Looking through the first shots again there is quite a bit of variance, so I am going to do it again more strictly.
-
I have a late pre-AI 105/2.5, from sometime in 1976 judging by the serial number. I like the rounded aperture blades. Sharpness is simply phenomenal even wide open on the D800 but the lens has a tendency in combination with the camera's ground glass and electronic rangefinder to entice you into slight front focus at times. Bought this well-loved sample over a used but mint 105 2.5 AIS while in grad school to save 30 bucks and get a metal aperture ring, as I was feeling sour about plastic in any shape or form at the time. Finding out later about the deal with the bokeh and the aperture blades was a bonus.
-
At near range, the f/1.8 is superior at all aperatures and at all parts of the image. Most notable is the lack of CA in the f/1.8 while the same cannot be said for the f/2.5.
I wonder if that is because the 1.8 has a flatter field than the 2.5? A lot of lenses test poorly on flat subjects in the corners, but are fine in 3D.
-
Daniel, this f/2.5 is also a pre-Ai that was converted with the factory kit. ;)
Roland, that was exactly my thoughts. I need to find a more suitable scene to test their distant rendition, I am guessing the f/2.5 will fare much better there. I've also noticed that the f/1.8 seems to see plenty of use in reproduction, which I'm sure stems from it's fantastically flat curvature (or practical lack of?) and negligible distortion.
-
Try shooting the same target with the camera aimed up at a slight angle. You will no longer get a sharp image corner to corner since the top and bottom of the target will be outside the focus plane, but you will get a zone of sharpness across the middle of the picture. If the lens has a flat field, the zone of sharpness will be a straight line across the middle. If the lens has some curvature the zone will curve up or down at the sides, depending on the direction of the curvature.
This is a more useful test than shooting directly onto a flat target. It gives you information about field curvature, how the focus plane rolls off into the near foreground and background, LoCAs will show up, and you can still judge sharpness from side to side by looking along the line where the focus plane intersects the target.
-
Will do so and report back 😉
-
Conducted a comparison test at infinity focus with a challenging scene. The sun is ~45° to the left of the framed area and ~45° from the horizon, creating quite the glaring scene. The f/1.8 was tested with it's built in hood extended and the f/2.5 with the HS-4 hood on. This may have put the f/1.8 at a disadvantage given how short it's hood is, but I felt it better to test it with the hood I would use with either.
Full shot for reference
(http://i.imgur.com/GloaXB3.jpg)
(http://i.imgur.com/ohACjAv.jpg)
(http://i.imgur.com/Zgwqtog.jpg)
(http://i.imgur.com/doL4XhM.jpg)
At f/2.8, the f/2.5 manages the flare better than the f/1.8. Both lens sharpness is pretty much the same, though the f/2.5 doesn't hold up on the edges as well as the f/1.8. I was surprised to see that the f/2.5's field curvature was strong enough to knock the lower right corner oof, which had to be 100-150 meters away. The f/2.5's field curvature seems to only effect the outer corners, but it plunges pretty strongly.
(http://i.imgur.com/DpuXCJS.jpg)
(http://i.imgur.com/Zgwqtog.jpg)
(http://i.imgur.com/fLyrSX6.jpg)
The f/1.8 displays the same contrast drop at f/4, slightly improves at f/5.6 and levels out with the f/2.5 at f/8. At f/8, sharpness in the center area remains the same between both lenses, though the f/2.5 still has some CA visible. On the edges the f/2.5 still doesn't quite catch up to the f/1.8, though it is a pretty small difference.
At f/11 and f/16, both lenses are virtually equal. At f/22 the f/2.5 pulls ahead slightly in all areas of the image.
-
After getting a f/1.8 in nicer condition I re-did my tests which proved out my earlier findings. The f/1.8 beats out the f/2.5 in all areas of the frame and all focal distances, though the differences at a distance are small and differences at near focus are quite significant. The f/1.8 is virtually CA free at near distances and negligible at far distances, while the f/2.5 has quite noticeable CA at near distances and near negligible at far distances.
I decided to compare their bokeh and how they handle flare with a light source in front of the lens and came up with something I am not understanding. Perhaps someone can inform me how this is panning out. Here are both lenses at f/2.8.
105mm f/1.8 Ai-s
(http://i.imgur.com/C72cZ22.jpg)
105mm f/2.5 Ai
(http://i.imgur.com/oQiCnYx.jpg)
The f/1.8 handles the flare much better but exhibits some ghosting, while the f/2.5 fares worse in flaring but has no ghosts. This is the same thing I noticed with the night shot I did the first time around. What I'm confused by is I have learned that the f/1.8 handles flare from light sources outside of the FOV worse than the f/2.5. How does it handle it better when it is in the FOV, yet worse when out of the FOV? The difference in the handling of flare in the scenario pictured above is quite drastic. The f/2.5 must be stopped down to f/5.6 to match the contrast of the f/1.8 at f/1.8.
On bokeh, the f/2.5 exhibit smoother bokeh balls at f/2.8. Stopped down any further and the f/1.8 appeared smooth, due to more blades. Combined with the fact that the f/1.8 has very smooth bokeh at f/1.8 and f/2, I can't see any advantages here for the f/2.5 unless shooting at f/2.8 specifically.
I am done comparing these two. Below are the optical strengths of either compared to the other, available aperature differences aside as those are obvious.
f/1.8
- Better edge to edge performance up to f/16
- Better handling of CA
- Significantly better performance at close range
- Better handling of flare from light sources in FOV
- *Subjective* More rays from starbursts
f/2.5
- Better edge to edge performance from f/16-f/22
- Better handling of flare from light sources outside of FOV
- Much less ghosting
- Smoother bokeh at f/2.8
On a very minor note, the f/1.8 has a tighter FOV than the f/2.5. Though to a negligible degree.
-
Very interesting test Tristin, a pleasure to read. Thanks for posting.
What are your conclusions by the way between the old version and the obviously 'new' F/1.8?
-
Between the two copies of the f/1.8? Optically, no differences despite the second copy having not a speck of dust to be found. The second copy is just better condition mechanically. As there was only one version of this lens, you shouldn't find any significant differences between copies unless one has a de-centering issue.
Glad you enjoyed my testing. I hope it will help others decided which is best for them so they don't have to buy multiple copies and test them. This is obviously my personal opinion, but after testing them I believe the f/2.5 to have a much more popular legacy than the f/1.8 due to it's physical properties and cost rather than it's optical properties.
-
Tristin, When you say "light sources outside the FOV", do you differentiate between light sources 'just outside' the FOV but maybe within the image circle, or 'way outside' the FOV, like 170º off to the side?
Thanks for publishing your comparison...it has been interesting.
-
By outside the FOV, I mean out of the image circle. With these lenses I noticed flare with a glaring sun at roughly 45º to the side. Not too far outside the lense's FOV, but the worst possible offender for flare you could find. I couldn't imagine any light source at 50º+ to the sides causing flare with the lenses. Let alone 170º, which would be nearly behind ya ;)
-
The Nikon 14-24 zoom is infamous for colorful flare effects originating from light sources way off to the side. The tele lenses are much easier to shield from side lights by employing ordinary deep lens hoods.