Weekly Nikon news flash #404


→ There are a lot of fake listings/scams on Amazon (in this case: used DL cameras) - make sure you are purchasing from Amazon directly or from a reliable reseller.


Capture One ($299) 10.0.2 released with support for the Nikon 1 J5 camera (release notes).


Lensrental published their MTF test results for the Sigma 85mm f/1.4 DG HSM Art lens ($1,199): "From an MTF standpoint, it’s better than any other 85 except the Otus, and it makes a very respectable showing against that fine lens" (they tested the Canon version).


£200 trade-in offer on the Nikon D810 in the UK.

→ Interesting: Laowa 105mm f/2 STF lens ($699) vs. Nikon 105mm f/1.4E ED lens ($2,196.95).


Nikon announced presence and speakers at the 2017 Photograqphy Show.

Another camera magazine find- this is an article on the Nikon SP 20, a mock-up of an interesting compact camera that never went past the design phase. It had a 24-80mm f2.8 AF lens which could also be manually focused with its rangefinder, an auto film advance, and was designed in the spirit of the earlier RF Nikons. The article said this was spotted at the Nikon booth at a camera industry fair and that the staff on site were not able to say much about it. (Online sources say it's from 1990). The writer noted that the film plane was far too close for either 35mm or APS film and concluded that it was merely a Nikon designer what-if side project to test industrial design ideas. Funny how there's probably a market in 2017 for something new like this. #NikonSP20

A photo posted by Tokyo Camera Style (@tokyocamerastyle) on

This entry was posted in Weekly Nikon News Flash. Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • Eric Calabros

    a 24-80 f/2.8 lens shouldn’t be much bigger than that?

    • Ritvars Pujats

      It’s f2.8 – f4 😉

      • Spy Black

        …and butt-ugly…

        • dunlap.megan@mail.ru

          I’ve earned 104 thousand dollars previous year by doing an on-line job a­n­d I did it by work­ing part time f­­o­­r 3+ h a day. I used an earning opportunity I found online and I am so thrilled that i earned so much extra income. It’s really newbie friendly a­n­d I am just so blessed that i discovered it. Check out what I did… STATICTAB.COM/5rq2qty

    • Thom Hogan

      IIRC, this was posited to use the larger current sensor size of the time (think what the Coolpix 990 came out with).

      • From 1990? The sensor was film. The QuickTake came out in 1994 according to Wikipedia (I thought it was a little earlier than that.)

        • Thom Hogan

          Nikon’s sensor work dates at least back to 1988, and I believe even before that. Sony’s dates back further. When I was dealing with Nikon engineers in 1993-1994 timeframe, they were already prototyping with far bigger sensors.

          The real issue early on was cost. The reason we went with TI sensors in the QuickCam was that we were able to get them dirt cheap. But larger sensors like those eventually used in the late 90’s were available, but just not at the price that would justify putting them in a consumer product.

  • peter w

    luxury shooters… 🙂

    • TwoStrayCats

      That would be over in the Leica blog.

      • peter w

        (I’ld consider a Df with a 58 F1,4 true candidates for a luxury shooter. Personaly I’ld prefer the D810 with 35 F1,4 kind of luxury. I simply love the luxury of lots of beautiful data. Bèta. )

  • Spy Black

    “Capture One ($299) 10.0.2 released with support for the Nikon 1 J5 camera…”

    That’s hilarious. They just got a D7200 in a studio I freelance in, and C1 doesn’t have the 18-140 lens they got it with in their database. Camera Raw, on the other hand, does.

    • dredlew

      I requested J5 support back in v9 and their reply was that they wouldn’t support it until there was more demand. According to their logic, people have to request an item first, hope that many others do too and then maybe they’ll support it sometime down the road. – Absolutely ridiculous. These guys are so far removed from reality, it’s not even funny.

      • Spy Black

        Yeah, although Adobe has it’s own set of problems (mainly rental software), they have vast support for cameras AND lenses. They even have my ~12-year-old Tamron 28-75 f/2.8, which I was rather surprised about.

        • Eric Calabros

          Maybe C1 software team is a lot smaller than Photoshop department in Adobe, so they have to prioritize. However they should know a huge number of people are doing product shoots with kit lenses.

          • peter w

            “However they should know a huge number of people are doing product shoots with kit lenses.”
            That being serious?
            A huge number of people do their own product shots directly as jpegs with compact camera’s. They don’t need COP. At what point do you need COP as a product photographer? When you are a pro? When you are a pro, do you choose a kitlens for products? Answer: Yes, because then I don’t have to think before the shoot…

            Perhaps, since they – COP devellopers – are accustomed to customers with digital backs for like Alpa camera’s and like Schneider lenses and other non-digital stuff, and now their own fancy Phase One camera systems on the other hand, they have a little problem with adapting to the needs of customers with kit-lenses. They should evaluate that… 😉
            (They do inform the potential buyers about supported camera’s and lenses).

            If I hire a professional photographer any time in future, I’ll make sure he/she uses Nikon, because then I can supply the proper lenses…

          • Spy Black

            I’ll assume you’re being cynical. 😉 The D7200 is essentially a test mule at a company who’s pretty tight with spending, and the studio manager is looking to prove to the top brass the effectiveness of the alternate platform. Although the intent was to invest in full frame replacement bodies down the line, the D7200 and it’s kit lens however are proving themselves to be quite capable in product shoots.

            • Thom Hogan

              Duh.

              The actual need for more pixels, bigger sensors, more dynamic range is fairly small. And this brings up Nikon’s real issue again (buzz, buzz): Canon, Fujifilm, and Sony are all supplying APS/DX in mirrorless form as DSLR sales continue to decline. Ultimately, this erodes the ability of Nikon to sell DX, and mostly because Nikon never defended DX.

            • Allan

              “The actual need for more pixels, bigger sensors, more dynamic range is fairly small.”

              Then….

              What a poor decision to try to convince DX shooters to either buy FX bodies or FX lenses (especially for the D500). Especially with third-party lens manufacturers making good to excellent DX lenses.

            • Thom Hogan

              Well, what this means is that Nikon basically considered DX->FX upgrading a marketing problem. Bigger has to be better, right?

              In essence, Nikon was making decisions that helped Nikon more than they helped Nikon users. FX being more expensive, it also meant that Nikon collected more money from those upgrading. FX was part of Nikon’s regaining parity with Canon in 2007 (actually, slightly better than parity), thus might help them retain parity with Canon. It’s also possible that Nikon caught wind of Sony deciding to move to full frame, too.

              Marketing, marketing, marketing.

              Meanwhile, Nikon left this big gaping hole in DX: lenses (and for a long time, no D300s update). Fujifilm is now right in that spot, and didn’t make the same mistake. It’s not surprising that Nikon’s big FX push resulted in:

              1. Some Nikon folk upgrading to FX.
              2. Many Nikon folk sticking with DX.
              3. Some Nikon folk leaking to other brands such as Fujifilm.

              Nikon must have felt that 1 produced better financial results than 3 removed. But this is wrong thinking. There’s no reason why you couldn’t optimize 1 while protecting 2.

            • Hysz

              Does FX “AF-P” lenses give a much better performance? And, you seem to know things, which of FX cameras can use AF-P lenses?

            • Thom Hogan

              Only DX AF-P lenses so far.

            • Hysz

              Oh OK. Thank you for having time to answer. I appreciate your reviews and such. Very informative.

            • I think Nikon could defend itself better against Mirrorless if they put PDAF on sensor so that live view was more usable. The ability of Nikon to pioneer a technology and then drop the ball never ceases to amaze me.

            • Thom Hogan

              You might want to try a D3400 with AF-P lenses. It’s not just PDAF that imparts speed, it’s how the lens handles camera input. As I noted in my D3400 review, something dramatically changed with that camera and the new AF-P lenses in Live View, and it isn’t PD ;~).

            • That’s interesting, but I’m not going to switch down to a D3400 and new (kit!) lenses to get improved live view. My Nikon V1 does a better job of AF using the FT-1 adapter than my DSLRs in live view. Why can’t Nikon take that technology and stick it in their DSLRs the way Canon has with the 80D?

            • Spy Black

              Well I never doubted the body, but I had reservations about the lens, which turned out to have even worse distortion than I had anticipated. Fortunately for the final resolutions the detail loss after correction will not be an issue. It’s an unfortunate shortcoming of an otherwise good lens.

      • Thom Hogan

        You can’t have it all ways. Handling everything means that you need to do a full on test scenario with now thousands of items. If you divert resources to deal with 10 of your users at the expense of hundreds of thousands of your users, you’ll be in trouble in no time.

        Or…you’ll have to charge like Adobe does: an ongoing fee that may have to eventually rise to deal with all the combinations and permutations that come up. And even Adobe uses a fair amount of user-supplied data.

        • dredlew

          Um, it’s the top end mirrorless camera from Nikon, not some obscure camera. All prior versions were supported. The camera was out for over a year when I submitted the request. Gimme a break.

    • That lens is so good though I never find myself needing to correct mine. 18-140mm FTW – I’ve banged and drenched mine in rain and snow for years now, never lets me down even in terrible conditions. And it was $200 with amazing VR performance. I don’t know of a better deal than that in general purpose glass.

      • Spy Black

        I don’t know what you’re shooting, but product shots in the studio have pretty significant optical distortion with the 18-140. While you can manually correct for it in C1 (and you’ll need to do this for every zoom setting), CR just needs a click of a box. From a production standpoint it’s a no-brainer. Color rendition is better in CR than C1 as well. I’ve been talking the studio manager to dump the Mk IIs and Mk IIIs for a while now, and it’s beginning to sink in. He’s quite surprised that the crop frame D7200 is dusting the full frame Canons. However I was surprised to find that C1 is also adding to bad renderings.

        • Wildlife, Aerial Photography, some landscape, skiers, athletes, it’s not noticeable to me in general use – product photography, makes sense that you’d find it an issue. I guess I’m used to most of my lenses having some form of distortion or other and I don’t retain any that bug me on that front, so your scenario is more demanding

          • Spy Black

            Yeah, in your kind of shooting you’re dealing more with organic objects and shapes, so no one would ever notice distortions. You’re also probably shooting at a greater distance, which reduces the effect of optical distortions, unless you have any kind of architecture in your frame.

        • peter w

          Color rendition better in CR than COP?
          Strange.

        • Hysz

          Not to sound hostile, but I recently talked to someone about different renditions of sensors, and with all digital data I couldn’t get straight answer, so I will tackle you instead on another ground.

          First off you can ignore this mgs of course, I am not a troll or anything I genuinely want to learn a thing or two, so here is my question:

          Is there any objective [as opposed to ‘look at the pictures’ – subjective] proof of C1 renders superiority? It has different tools, but I tried both [yeah I know, small sample], and I have found zero difference in latitude of colors and dynamic range etc. If you are busy or don’t want to get into debate, can you maybe point me to some objective article that states those things? Again, not being anything but sincere here, and I hear this sentence all over the web, yet no one seems to have any proof, seems to be one of those ‘arguments by repetition’, but I wont say, because I just might be too stupid too see the difference, yet the hard data shouldn’t be too hard to decipher. Thanks in advance.

          • Spy Black

            Well, you’re not picking an argument with me on C1. 🙂 It’s simply a matter of C1 being the studio standard, primarily due to its tethering capabilities, which are better then that found in LR, but also because many studios also have at least one Phase One camera, as this studio does.

            Theparticular problems in this studio are a combination of color issues with the Canon Mk IIs and Mk IIIs, which are further complicated by C1’s rendering of their RAW data. I brought my D600 in one day and performed comparative test, which proved favorable. The company picked up a D7200 as a test mule, and further test revealed better results yet using Camera RAW.

            • Allan

              Is it worth putting up with Capture NX-D’s problems, including slowing down the workflow, to get a better initial rendering of the NEF file?

            • Personally for one off small photo shoots,2-5 images i like to use capture nx. For large photo shoots LR is just a tonne easier to work with even if colours can be off. I think it’s a real shame Nikon can’t produce good software or just sell their raw converter to adobe or others, or give to them for free and demand strict contracts for not spilling the beans.

            • Hysz

              Can Capture NX produce xmp sidecar files? If yes, then you can use LR to cull or sort, and then just copy/paste xmps from NX and you got all colors etc whatever you want + LR sorting/catalogue thing.

            • Thom Hogan

              Short answer: no. And especially no if you’re on a platform where you update your OS regularly due to security or other reasons.

            • Hysz

              I see. This is why everyone has different opinion, and is able to stake his/hers ‘life’ on it. It’s a matter of which camera sensor is rendered in which program. I understand it fully now, if it’s not a matter of full on ‘superior’ product, then I can rest easy. I am very grateful for the information, I will look further myself.

              On a side note, I am sure this is very technical question, but does C1 has GPU acceleration ON EXPORT? In LR GPU acc is only in develop module and using tools, like gradient, brush etc. Some people swear that C1 exports using GPU [when it’s true then it should render those pics 10, 20x faster than CPU only, of course depends on GPU itself]. What is your opinion of this subject? It’s a question which software is faster, but ‘does C1 exports using GPU?’ As usual, you are free to ignore this msg, I am just curious. My trial has shown, that IGP Intel HD530, as well as Gainward 1070GTX does not ‘spike’ on export in C1 version 10 [the latest there was]. Thank you in advance.

            • All you should care about is _results_ — speed, quality, whatever. Whether they use the GPU or not is irrelevant if the quality is worse and/or it’s slower. In general, GPU acceleration should be a good thing, but it’s an implementation detail, not a feature.

            • Hysz

              I know, but if it’s GPU acc, then it is faster. GPU are dedicated to treat pictures like textures and do wonders in really really small amount of time. If I could export 1000 pics in 5 minutes, that would be a good argument against Lightroom. I do stick to LR still, because of the catalog system, but I wonder.

            • Sorry, your reasoning is backwards. If it’s faster it is quite probably GPU-accelerated. But if it’s faster, who cares why it’s faster?

              Although GPUs are great, they also suffer from limited memory and a bandwidth bottleneck. If you’re dealing with a multi-layered image created by stitching D810 RAWs you may be in a situation where the time spent getting stuff onto the GPU and back again becomes a huge issue. Adobe has been working on its tools since computers had 8MB of RAM and a 25MHz CPU if you were lucky, so it’s much more likely to deal gracefully with hardware constraints — but if you’re working on something relatively lightweight, modern GPU-centric applications (e.g. Pixelmator) will be far faster than Photoshop — but throw a big, high resolution image at them and they either grind to a halt or crash.

            • Hysz

              It kinda is, but that was my point. I’d rather know if it is accelerated than if it’s faster in some other mysterious force. LR GPU acc is very pleasant to work with, that’s why the tough entered my brain. Memory limits nowadays are higher, way higher. If I cannot clip LR with my PC, then I can safely say that LR has enough code in it to support large amounts of data. But GPU exporting, this is a novelty. I ran out of my trial, and I am curious AF about that. Sorry if I seem backwards and with left spin, I tend to create those shortcuts, when I think, and I think in another language than I type, sometimes it’s a mess.

            • GPU exporting is more likely to run into memory issues because it has to be full resolution.

            • Hysz

              hd530 inside my CPU – I can grab 1GB from RAM and give it. Dedicated have like 4-6-12GB of VRAM. only bottle neck could be poorly written implementation…

            • > only bottle neck could be poorly written implementation

              Right, which gets back to the original point you seem to have missed. The important thing is how well whatever you’re using works, not the details of its implementation.

            • Hysz

              yeah but I said I saw how AF and Premiere works with high end gpu. So it’s possible. Can a man dream once in a while?

              We all know that LR isn’t the fastest raw converter on planet, but I am optimistic.

            • Hysz

              and I stitched 24+24, yeah for 180×360 pano. My files were 24MPX dng from D600.Worked like charm.

            • Hysz

              GPUs can handle that, slice and dice, put back together, render, like 4096×4096 textures, stich etc.

              There are mods in games that enable usage of textures more than 100MPX in resolution/size. It can handle. GPU architecture changes, but as of previous Nvidia GPUs I am kinda well versed at how they work and what can they do. So no problems, and they are so fast [PCI-E 64GB/s] that they can connect to RAM, CPU back and forth without bottleneck that we would see. I mean there always is bottleneck when something is exported out of CPU [nothing is fast enough to get the data, but that doesn’t apply. Think of real time rendering. Then think of 24MPX [for example] Display, physical one. And the game is: Slideshows. GPU has to render them and put on screen for let’s say 0,2s each [can be thousands of photos] And it would do that, natively. It would have to be about 400$ GPU but still. Prospect of having 1000 of photos in 60 photos per second[like FPS/Hz]? Cmon.

            • Hysz

              And just for the record, After Effects and Premiere Pro can use Nvidia Quadro to basically treat everything as in real-time. It works with Photoshop too, but this guy has some issues that I won’t go into. Google some after effects on Quadro. It doesn’t have to be new one, DX10 one is OK. Render time cut by 40x, and using right codecs, it’s even faster. But this is a side note, I am interested in still picture.

            • Spy Black

              I don’t really know what the multithreaded and GPU capabilities of either program are, actually.

            • Hysz

              OK, thank you kindly.

            • Hysz

              If you ever wondered, LR 6 can use 12 cores, after that another core does less than 3% difference, tested on x99 double xeon thingy [by thingy I mean motherboard with that chipset, 2 CPU 6 cores each etc], It capped at 22GB RAM while working on single files [can spike to higher when HDR/panorama mode is used although I used it [panorama 6x24MPX with 16GB DDR4 ram @3000mhz], and it didn’t choke or anything, it just took a moment.

              Next… It does not really matter how many VRam you have on your GPU, very small amount of GPU is used mainly Open CL [there is wiki]. GPU acc works when you use brushes, circular thingies, gradients an few sliders [exposition I know, rest, didn’t bother to test]

              Finally. 1070GTX was slower than IGP that is within i6700k, because Open CL. All this is testable, and I encourage you to look for answers yourself. I just imagined that it is the least I can do after you replied to me.

              So perfect machine for current LR is: 8 core CPU, BUT LR likes MHZ PER core, so higher one core, the better, that’s why I bought 6700k, easy OC to 4.8GHZ per clock, 4 phys cores 2 virtual in each. Then HDD for archive, and that M.2 memory [512GB is enough] as a OS. Then SSD, you put CACHE there, you put PICTURES as well as CATALOG and ofc whole program on main M.2 memory stick [it’s like stitching from candles to LEDs with that memory. I have installed win 10 pro from boot up to desktop in 1min 18 sec. My ISO was on a 45MB/s SDXC card, nothing fancy. And boot up is non exist-al, you pres butan receive windows. And, for me [d600] 16GB DDR4 3000MHZ RAM. 32GB for 36MPX files.

              Feel free to ask about these things, if I can reciprocate I will feel less like a leech.

            • Spy Black

              I guess I’ll just have to make do with my 4 core i7 with 32 gigs of ram, SSD system and scratch disks, and old 4 gig 980 nVidia system. 🙂

            • Hysz

              It’s perfect. But if you can overclock all cores, as high as they go, along with memory. It really boosts your workflow, really. Mhz per core is the way to go!

            • Hysz

              Another reply, as I found something in your reply. What about tethering makes C1 better? I can only use my D600 and when plugged in I can do everything in LR, even pops out pictures with pre-made presets if I want to impress my clients. I haven’t tested C1 tethering, but many people seem to be impressed by it. Can you elaborate? Again, sorry to bother you, feel free to ignore it, if you don’t feel like answering.

            • Spy Black

              In all honesty, I haven’t checked in LR lately, as I don’t tether in my own studio work, but last I checked you could only trigger the shutter in LR. In C1, as long as you have a supported camera, besides triggering, you have control on camera parameters such as F-stop, shutter speed, ISO settings, etc.

            • Hysz

              I see. Thanks. LR has the same thing. WB, camera settings, live view etc.

          • Thom Hogan

            “Superiority” is a word that almost shouts subjectivity.

            Here’s my take: C1 and ACR use two different color models, and with different consistency across cameras. I find C1 to be a bit more consistent, but ACR to be more neutral (especially with a bit of easily made tuning). Neither would be considered a “superior” approach to me.

            As @spy_black:disqus notes, C1 is probably a better tether companion than LR. Certainly more stable in my experience. But then you are dealing with their color model.

            • Hysz

              Hello, I may have used the wrong word there, I meant something that is better than the other one [I am not native to English, although I like to think that I am fluent, guess not as much as I’d liked to].

              Thank you for your information, rendering seem to be subjective, rather than one program having objectively ‘better’ colors [such as color range, latitude, and dynamic range for that matter].

              I am just trying to debunk some ‘myths’ about those processors, and I find that someone who uses both is not that biased towards any particular, so someone will answer in more ‘objective’ matter.

            • Hysz

              Hello, sorry if I am imposing, but I have a little bit of a different problem, Can I ask your assistance? You are knowledgeable, I am lurking a lot. I have this conundrum. imgur[DOT]com/gallery/mmKrC

              If you would be so kind as to open, you will see 2 pictures, all made using LR [ACR], no other program was used. output is 3000×2000 file that my clients get. The nr 1 file is in MY opinion well exposed, no tints, and great deal of details, second is SOOC raw with basic brush on their eyes [so let that don’t deceive you as to contrast etc], and finally, can you see the greenish tint on nr 2?

              I have printed them both on a nice little [4m x 1m x1.5m carrier] without changing ANYTHING. I get nr 1 to my face and it’s almost identical, I put nr 2 and it’s green all the way. Thing is I made a survey on reddit with over 300 votes on nr 2 and FIVE on number one. As a photographer, how would you solve that? If they take nr 2 and print, it will be a disaster and it will be my fault, even though I have given them ‘versions’ [to potential clients]. I am loss of words here, I could make photo-album with my settings, but I still give them a file and 3 years backup, from which they can take full 24MPX file. And I know my clients, I provide 36 pictures in print, and they print more, and more. I didn’t think about it that much, but person who let’s say gives 200 photos [digital] and that’s all, when I give about 1000pics, both in DVD form and as an album. I make X money of that wedding. He does 200 photos, burns and that’s it. If session, then 10 to 20 pics. And he gets 3X and 4X respectively [without session, and with session]. And that person made picture number 2. I struggle to be as best as possible, if I can do less for more, then yeah, why not. I do have the skill, I mean LR in little pinkie, as well as PS [for photo edit, not for every single thing PS can do]. It all comes down to this… should I use nr 2 and ignore Print, or use mine, that I pick with picker and see 80-85reds etc. I even calibrated my monitor few times, because I was starting to be paranoid. But in print, everything made sense.nr 2 is objectively WORSE, it’s green and both faces are sick [although they are a bit smooth, chap didn’t use curves nor clarity, oh well]. Also if you decide to help me, maybe it would be possible to exchange quicker msgs on some other platform? Just for few hours. I really want what is the best for my clients, but if they like SOOC image with brush smudged on their faces, then Why do I even shoot RAW? Why do I spend at least 10 solid days of 8h work in LR, when I can just give the JPGs [SOOC]. It doesn’t seem right, and in print it will be shown. Can you, will you help me?

            • Thom Hogan

              They both are not neutral, and actually probably far from neutral, just in different directions.

              I suspect that people you ask are reacting to something other than just color, though. The first image puts the skin tones way too high overall. You say it’s properly exposed, but that’s not how most people would see it: they’ll say the guy on the left looks like he has a light bulb for his head.

              What’s really different between the two images is the tonal ramps in the color channels. They max out about the same point in the hottest channel, but they are very different in the other channels.

              For me, I have to see the original information. For example, if the red in that tie is actually red and not orange, the bottom image is more correct. This is why on shoots like this I almost always shoot references in the light, both for color and tonal ramp.

              Can’t speak to all the rest of your comments, though I have a strong suspicion that your monitor is not correctly calibrated or that you’ve got Color Space issues.

            • Hysz

              well hmm, I basically went mad calibrating this monitor, I am 98% sure I got it right. About 10 times it was identical, same light conditions when calibrated, only artificial light, I am asking trivial questions, but I am not a dummy, well not that much of a dummy… Picture is lit only by flash, but bounced, so it could get some color. In LR nr 1 on his head has 82% red 73% G and 65% B. Second is much lower -20 to each with slight variation in green as it is 67%. Ugh, this is a catastrophy then…, but then what about the print? I got them, I even asked my neighbor placing it to my head, for him to compare. I go to stupid lengths trying to please clients in digital and print without making 2 sets of images [because I will give them flash drive and they will print, probably. I never thought of it, no one complained really, business is steady, but I like to go additional mile for them. You confused me to the end. I mean I get your point, I understand you fully, but I cannot comprehend how picker can lie. I mean eyes, screen, print even, but picker in LR is what I check most often when in doubt. Umm… Can I ask you a VERY big favor? Would you grab a nef[dng] and tried to set it for screen and print at the same time?, using the same settings, resulting in one image? It’s OK if you decline or ignore this msg, I know that I am asking for outrageous things.

  • Spy Black

    Although they cheated a bit on the sharpness test, the Laowa 105mm f/2 STF is an interesting lens. It appears that the “Smooth Trans Focus” function is either an exact copy or some alternate variation of the Defocus Image Control you find in the 105 and 135 DC Nikkors. Not sure it’s worth $700, although that may be subjective, but $50 more gets you the impressive Tamron 85mm f/1.8. I think $500 would make the Laowa a lot more attractive. You can get the 135mm f/2 Samyang with Nikon AE chip for $470, and that has proven itself to be a spectacular lens.

  • decentrist

    a new low in wideo review

    • Spy Black

      Did you mean wideo weview?…

      • decentrist

        lenses,bad rewoo,wideo all brought to you by the atoll fillers

        • Spy Black

          Some interesting points were made, inasmuch as it was just a shill.

          • Eledeuh

            Even worse than a shill, it felt pretty much like an ad at this point 😉

      • I think he meant wideo wewiew. Woo woo.

  • MB

    Sigma 85 Art is obviously great lens, but Otus is clearly better at center.
    Well not according to DXo:
    http://b2.ge.tt/thumbnail/6VR5lgi2/0/user-bQvT3T8tPAca3WZuiI09ZlbDG0QXjgpHh1mAI-/0x675.png
    So who do you trust, professional optical bench or DXO mumbo jumbo?

    • Hysz

      Do you know what you are sharing?

      • MB

        Yeap, why?

        • Hysz

          Because they both test in different matter, and comparing results like that is invalid. I am not any fanboy or whatever kids do these days, I am just pointing out, that even MTF charts aren’t comparable between 2 test [or manufacturers]. I have browsed thru DxOs results, and then I have read what Lens Rentals have to say, and in my reasoning they both speak the same thing, as in they had different methods of testing, but both came to the same conclusion. Which when applied scientific method, we can begin to think at least 2 things: one that DxO IS credible, and two that Sigma 85 ART is a very, very good lens. As I said, ‘we can begin to think’, but I have seen previous ‘reviews’ from Lens Rentals and they seem to be on-par with what DxO is showing us.

          • MB

            MTF, OTF and PTF are well and exactly defined and as such should be directly comparable no matter who is doing the measurement.
            Therefor when one evaluates Olaf charts one can now exactly what he is looking at. On the other hand no one really knows, at least in my knowledge, what DxO P-mpix represents so if they results are not in line, as in this case as in quite a few others, with the results of verified measurement equipment I am inclined to say that they are meaningless … and of course you are right, comparing scientifically verified measurements with some “given per se” numbers is invalid …
            I really do not care for any of those lenses, Sigma and Otus are great but way overdeveloped and impractical for use to me, but practice of publishing some imaginary charts not based on some more or less exact measurement but on wishful calculations, as all lens manufacturers are doing at least in my knowledge, or some … mumbo jumbo … is just misleading …

            • Hysz

              dxomark[DOT]com/Reviews/Looking-for-new-photo-gear-DxOMark-s-Perceptual-Megapixel-can-help-you

              This will clear up the P-MPX for you, it’s basically dumbed down MTF chart. In a very oversimplification.

              Can I ask then what is practical for you to use? It doesn’t have to be 85mm, I am just wondering that ‘stats’ do you consider practical for you, for your line of work. I am curious, and mean no harm or anything like that.

              If I didn’t buy 85mm 1.8G, and I would sharpen my teeth on this Sigma, 100%. That’s why I am asking one photographer to another. Thanks.

            • MB

              I am sure people mean no harm here, I mean why would they, this should be fun…
              As for DxO clarification on P-Mpix they are basically saying “P-Mpix is the unit of a sharpness measurement”, no less but no more, what does it represents and what does it measure is unclear to me, it seams more about acutance than resolving but they are not saying. On the other hand DxO is selling this system and I think we should consider DxOMark more as an advertising site than anything else , and I have some doubts about their objectivity when their customers products are ranked, though it is pure speculation on my part, but how can they prove me wrong if they are hiding what exactly they measure …
              And finally if you did buy 1.8G you would be doing the same thing I did because this lense is excellent and the difference between it and much more expensive lenses is negligible no matter DxO or MTF, and it is way more practical for me.
              The thing is it is a bit annoying that some advertisement as DxO is causing so much rave so I just wanted to shake people minds a bit 🙂

            • Hysz

              no, of left side you can see ‘in depth’, and at the end of each thing is a link to research paper. They don’t hide anything. I mean they do, but not things they measure, like field curvature and ‘effective focal length’, as well as few more things, AF speed and such. Those things are part of this thing they are selling Dxo Analyzer:

              ‘DxO Analyzer is the only solution that can fully measure not just the
              quality of captured images, but also artifacts from electronic
              shutters, the effectiveness of 6-axis image stabilization systems, the
              geometry of dual-module cameras for 3D or stereoscopic vision, and the
              dynamic response of imaging devices to changing scenes and light levels.
              Coupled with DxO Analyzer’s ability to measure images the way consumers
              see them, it is possible to fully evaluate cameras and lenses in a way
              that predicts how they will perform in real world situations.’

              see DxO doesn’t tackle things like this, but this soft/hard is. They just test some things and advertise other things. But this isn’t for normal user anyway, it’s not a plot to make us buy things, Analyzer is for big companies.

              Well thiis is poor excuse, but DxO seem to have half of it’s links broken. Please, believe me, when they say about MPX in great details, giving the paper [I was too dumb to follow it, too many equations].

    • silmasan

      Take into account the different samples, it’s possible that DxO (and I think LensTip too) got the ultimate best sample of the Sigma, while with LensRentals we get an average of 10 samples (vs 11 of the Otus iirc). So LR gives you a better idea of what we might actually get.

      • Bob Thane

        Agreed. Also, LensRentals tests at infinity I think, whereas Dxomark does not.

    • decentrist

      you trust your experience

      • MB

        Nope … but we are talking about pretty graphs here and nothing else …

  • IanMak

    So this NEW NEW Nikon D810. Is this a typo or Nikon D810 replacement?

  • Back to top