timballic wrote:
They will include Nikkor Ai 24/2.8, and an MD comparison between ø49mm and the ø55mm, (before the ø55mm goes back after the w/e.)
I had the Nikkor 24/2.8 Ais and was impressed by both it's color and contrast on the A7r. Very nice rendering overall. It was not as sharp at the borders and corners compared to the late MD 24/2.8 I had but it also did not have the mid frame dip I saw from that MD. I don't think either compares to the Olympus though.
The new line up:
OM 24/2.8; MD 24/2.8 ø49mm; MD 24/2.8 ø55mm; X-Fujinon EBC 24/2.8; FDn 24/2.8; Nikkor Ai 24/2.8
In use, I quickly came to love the MD ø55mm the best, for it's easy to find aperture ring on that raised ridge, and it was the easiest to focus positively. Conversely, I disliked the FDn, for its small, "tight" aperture ring and the image jumps on reversing focus direction. The Nikkor Ai has a very long focus throw, making it hard to be sure where the best focus is, as it doesn't snap in and out, (Ais better for this), but the silvered grip ring is brilliant, makes changing so easy, only the OM has anything similar to grip that doesn't turn. The Fuji is the least sharp wide open but nice to handle, similar in feel to the MD ø49mm. I found the OM aperture ring too close to the focus ring and I kept wondering if I'd moved focus unintentionally.
The differences between the MDs.
Obviously different coatings and size (but little difference in weight, 200:215g) I just wish they'd kept the raised ridge for the aperture ring in the ø49mm's design, it makes a big difference. Apart from that the ø49mm was very nice to use. Also note the raised aperture lock for AE (?) on the ø49mm.
For Nikon F and X-Fujinon, I'm using Fotodiox-Pro. I couldn't recommend them! Very short and the Nikon one is uneven (~ 4 thou" variation, but I reduced it to 2.5 thou" with careful shimming) ...yes, these are the "pro" series too. I shim them to get to right length
The Cheaper VA Pixco series used on the MD , MC and ZS are superior to the Fotodiox Pro, IMO, and less than half the price. They are v. slightly short, but variation is within 1-2 thou". I shim to exact length too.
The OM is on Kindai (Rayqual). The best adapter I have. (Bought in a moment of madness, as I already have it in VA Pixco) Contains baffles to help reduce internal reflections. The exact length and variation under 1 thou", that's pretty impressive, but expensive too, almost Novoflex price.
All the adapters fit snugly on the A7 camera mount and hold the lenses firmly, though again the Fotodiox-Pro are a bit "coarse".
I've added flocking inside each adapter as discussed elsewhere on Alt. Gear. However, as I noted there, the flocking is like a fine dense velvet material, which sheds tiny fibres from cut edges.
After two days of changing adapters, over and over again, my A7's sensor looks like a fibre-dust storm's hit it! Hopefully it will all just blow off!
I'm keeping the order as in Pt 1, but with both MDs replacing the MC VFC, and the Nik AI replacing the ZS.
So: 1= OM; 2= X-Fujinon; 3= FDn; 4= MD ø55; 5= MD ø49; 6= Nik Ai.
I realised that my earlier test shots all lacked a good distance, so I went out to my local fells, overlooking Pendle Fell, (of 17c witch notoriety). First, at Stonyhurst RC College for boys, (where Sir Arthur Conan Doyle of Sherlock Holmes fame, was educated.) It dates back to 1372, but became a Jesuit college in 1794...(and I bet the students call it "Hogwarts!)
All I know of it from personal experience, is that they always thrashed us at rugby, when we played them.
Back to the comparison, which throws up many interesting points.
I focussed on the seed heads in the foreground.
All results sharpened in Lr5 with my "standard" 60: 1.0 :40 . All F11 again. (Sensor dirt showing!)
First the OM, still the "reference" lens. Good close, not quite full dof. OM 24/2.8 F11 by timballic2, on Flickr
Next the Fuji. Great for close distance, but really struggling with the distance here.
EDIT: After seeing this result, I have re-shimmer the adapter and reset ∞ on the lens, and it's looking better immediately...so I'd discount the result below! I'll reshoot when I get time to revisit and the weather allows. X-Fujinon EBC 24/2.8 F11 by timballic2, on Flickr
Canon FDn. The most yellow in colouring, and I don't like using it! But good even sharpness. FDn 24/2.8 by timballic2, on Flickr
MD ø55mm. NB. Allow for this copy's soft left edge. Everything else is the strong Mid Zone dips. I'm beginning to wonder why all the hype that surrounds this lens? This is my 3rd copy, and apart from the soft left edge, all performed like this. The Elmarit-R 24/2.8 that was based on it (why??), shows a strong mid zone dip in the MTF chart.. However it is the sharpest centrally wide open, making focussing it a delight. Ergonomically, it is my favourite lens of the set. MD ø55mm 24/2.8 F11 by timballic2, on Flickr
MD ø49mm. Warmer than the ø55mm. More noticeable curved field. The closest performance to the OM, though better distance than close in comparison with OM. I'm SO glad I got one to try. MD ø49mm 24/2.8 F11 by timballic2, on Flickr
Finally the Nikkor Ai. Classic Nikkor. The sharpest of all centrally, but lagging in the far corners. Nikkor Ai 24/2.8 F11 by timballic2, on Flickr
I've shot many more sets, which I haven't studied yet.
Hey Tim, you should give Photo Ninja a try. I notice some stair stepping in the grass and limbs near the corners of the frame. The CA removal tool in Photo Ninja will most likely get rid of that. There may be some loss in sharpness in the process, and some lenses react better than others, but then you can sharpen a little harder to compensate.
Just saying it might be interesting for you to try...
NB. Re the Fujinon results above. I have re-shimmed the adapter to correct 25.5mm length, (it was at 25.4mm after I'd shimmed it before, because I couldn't get ∞ at the correct length), and it only varies by 1 thou". I've also been inside the "nose" of the lens under the name disc, 2 layers down, to get to the 4x screw focus set.
It's now all reset and immediately looks better! I'd assumed that the X-Fujinon probably wasn't a floating element design and that 25.4mm would do, as long as it was even, but perhaps I was wrong. (No information beside the basic spec on web.)
I need to reshoot the test shots with it now. Sadly Stonyhurst isn't just round the corner, and clear skies are rare here!
I don't know, but I've often wondered what it was...until I started using PN. Only recently I realized it was most likely related to CA. This might be worth reading, look for the "zipper" comment.
tsdevine wrote:
Does the zippering go away when you check the box? (Lightroom never grew in me...)
timballic wroteI've not got the "remove CA" box ticked/checked in Lr though?
It certainly "smooths" it, from those definite squared steps, but it doesn't completely eliminate it.
I noticed it most with results from the MD ø55mm for some reason, but I haven't seen any with the OM.
It's visible (without CA removal) from where sharpening has any effect at all, so it isn't an "over" sharpening artifact, more a Lr5 artifact, but only with certain files. I'll repost the first set of branches against sky (pg 2)at full size where it is very clear on the MC VFC file. (though it's probably still visible at half size, it's so clear.)
I meant to mention it then, but forgot, so thanks for reminding me Tim.
CA is a shift in color, and if not corrected prior to demosaicing it will introduce artifacts. That's what I think we're seeing. If corrected prior to demosaicing, then the error isn't exposed.
Most likely the Olympus has less CA to begin with....just a guess.
In any case I didn't mean to hijack the thread to another topic.
tsdevine wrote:
CA is a shift in color, and if not corrected prior to demosaicing it will introduce artifacts. That's what I think we're seeing. If corrected prior to demosaicing, then the error isn't exposed.
Most likely the Olympus has less CA to begin with....just a guess.
In any case I didn't mean to hijack the thread to another topic.
-Tim
Not hijacked Tim, as I said, I meant to ask the question but forgot.
Is that part of the difference between Lr and Photo Ninja, Lr doesn't correct before demosaicing and PhN does?
I honestly have no idea to be truthful. But I've seen this behavior in just about every RAW converter I've used. With PN, it pretty much goes away when the CA removal box is checked. The tradeoff is that you lose some sharpeness/detail. But some of that is probably faux sharpness and detail anyway, due to the artifacts.
In your shots though, it makes it a little hard to judge sharpness, at least for me. I would imagine there are others here who might have ideas/input. I have an old Lightroom license, so I don't think it will open my a7R files. I've noticed this, maybe to a lesser extent, with DxO too.
If there's some way of avoiding it, let's hope someone more knowledgeable chips in before I download a lot more to Lr, these full size files take so long to load!
I had thought of cutting out all sharpening, but then everything looks mushy and comparisons are even more difficult.
If you have what you consider a "junker" RAW file, and you can put it somewhere I can download it, I could give it a shot in PN. I understand if you wouldn't want to do that. I'm pretty protective of my RAWs unless it's really something that is absolutely boring to the point that no one would want it.
This is the "stepping" I'm seeing in Lr5:
Almost 200% of a portion of the MC VFC shot on pg2, which shows it the most clearly. This is also from the mid-zone-dip area of the MC VFC. The sharper the edge, the less it appears to "step".
Ticking the "Remove CA" box certainly helps smooth it at the expense of some "clarity", but it's present from the first amount of sharpening (here 25:1.0:20) I don't want to remove CA anyway, as for a comparison I want to see how much each lens has of it.
Is this typical of Lr5 processor, or have I got something turned on, or up, that shouldn't be...in camera or processor?
This is how I'm seeing it in Lr5, so it's nothing to do with sharpening or artefacts added by Flickr, on its way to here..
timballic wrote:
/MD ø55mm. NB. Allow for this copy's soft left edge. Everything else is the strong Mid Zone dips. I'm beginning to wonder why all the hype that surrounds this lens? This is my 3rd copy, and apart from the soft left edge, all performed like this. The Elmarit-R 24/2.8 that was based on it (why??), shows a strong mid zone dip in the MTF chart.. However it is the sharpest centrally wide open, making focussing it a delight. Ergonomically, it is my favourite lens of the set.
I'm glad I stopped after one copy of this lens. I hated the mid zone dip and sent the lens back to the seller (it was slightly decentered too)...