ttbek

Rather than being a criticism I'm just saying that information is sparse and video doesn't mean much in terms of optical quality. My experience with these has as I said been with cheaper ones like the Opteka, Polaroid, and no-name ones floating around. I assume one that they charge a thousand for will be better. I just don't know how much better. You don't need to try and convince me, I will see if I can search out sample raws or the like. This kind of thing is also difficult because I don't always know the lens used, I gave your adapter the benefit of the doubt since I know that lens isn't good to start out with wide open, so to me the performance is still a bit of an open question. On the Canon front, I have acquired a 400 f/2.8 already (used, for significantly less than the 300 f/2.8 you mentioned). Even at no loss of light the Samsung is an f/4.2 equivalent, so with the adapter it works out as a 405mm f/4.2 equivalent, right between my f/5.6 and f/2.8 lenses at just about the same focal length. Which is quite nice so long as the quality of the adapter is as good as I would like, the price also falls in between. I could try it, I guess the only risk is the shipping cost back to B&H or the like if I don't find it satisfactory. I don't really need it though (aforementioned Canon lenses). Might be worthwhile for Marco depending on the how good it is.

The other thing that always makes me worry is that most of the people reviewing are doing video, and the video shown was 720p. 1080p is only ~2 MP.... how can I review critical sharpness at that? 4k, is still only ~8 MP, my intended output in this case is 28 MP... so I think you see my reason for skepticism when I have nothing to evaluate these properly. For lenses DXO usually gives me at least a relative idea if I can't find one in store, but discussion of these adapters is still too sparse for my liking. Might bite anyway eventually.

And hella expensive, lol. I've always been a bit skeptical of these front mounted ones. It's true there is no loss of f-stop though you will lose some t-stop (hopefully quite minimal). I guess just most of the ones I've seen are pedalled cheap on Amazon and E-bay as a long range panacea but are actually of horrible optical quality. I know a few of the camera brand names have made some that are purported to be quite good, but they weren't in the filter sizes I wanted as they were made for particular cameras, sometimes even being the same size but with a different thread pitch. It's not often that I come across any concrete information on these so it has been hard to separate the wheat from the chaff. This product and the Raynox DCR 2025 Pro both seem to be quite well reviewed though. What makes the HDP 9000EX so much more expensive? Is it almost entirely due to the size (fitting 62mm threads for the 2025 and 72mm for the 9000) and lower demand? I'll look into these Raynox ones a bit more but will be continuing my other project anyway, I'm having fun with it and I already have a bunch of Canon lenses.

Ah, I misinterpreted your video, I thought you had it working like the metabones adapter. I'm going to disagree here though, my goal is to have true AF by the camera body controlling the lens. I agree that the other project is great for video shooters, probably even much better for them than what my goal is, but I am primarily a stills shooter. I'm kind of tempted to order one anyway, this is the conversion kit alone right? So one would need to buy the Apurture DEC separately, correct? My Italian is non-existent :P If you could manufacture a smart adapter it would let me focus on the firmware, as is the crudeness of my adapter takes too much time from coding. I don't have any nice tools for making it, relying on epoxy and soldering from parts entirely taken and sawed apart from macro extension tubes putting them into a dumb adapter for now. Actually, the part from the NX macro tube fits beautifully into the dumb adapter, but the Canon side is problematic. Just having something with contacts on the NX and Canon side, with corresponding contact points on the outside of lens such that a logic analyzer can be place in between would be pretty awesome, even more so if it can be done to an NX-L. I put in for two NX-L adapters, how should I send you my shipping info and phone numbers, I didn't see a place to enter it, just send it to your email listed there?

Hey Luca, I have some requests for this since your pace is much faster than mine. When you put out the adapter, can it have external connection points for monitoring the communication? Can it have a standard port over which the firmware can be updated (Canon does update EF protocol sometimes). And can you open source the firmware? Even if it is still paid firmware it would be good if users can make their own changes to keep it working with newer EF lenses if Canon makes minor changes that make it stop working (as they have stung some of the 3rd party lensmakers with in the past). I don't know what sort of chip your are using, hopefully it is something not too tough to program for ^_^. Your timing is too good, you will retire the NX-L before I got around to ordering mine :P I want to support this though so I will still order, and if it turns out to be working well enough for BIF with the long primes then I'm going to owe you more T.T :P I will try ordering NX-L today, please include my phone number in the address area (otherwise the post here will probably ship it back to you). Orange and red shirts, hurrah!

Actually, according to Samsung they were the same, then they flipflopped at the last moment just before release. I'm skeptical that they would have had time to disable all those cores at that point, they were probably shipped to retailers already. So either Samsung was right or Samsung is right :P As a programmer you will usually query the OS about these things and it will report back on the processor as it sees it. On Linux (Tizen in this case) everything is a file per se and we can literally check this stuff out in /proc If it is disabled, then it won't be showing up. True if they have indeed limited the capabilities of the Vs, but I'm not yet convinced that they have. Maybe Samsung reps were initially confused because the processors were coming off the same line but binning for which to have more cores and which less, but for now I suspect they are the same.

Luca: Don't tease us man, are you declaring victory, progress, or extremely early prototyping code that's been tested with only a handful of lenses? In my own case I managed to get electronic control of the aperture on a Canon lens today, the 40mm. IS.... seems to want to run all the time the camera is receiving commands on my lenses (e.g. 28-135) that have it, even when the switch is off on the lens. Probably the lens passes the switch state to the camera but the camera makes the ultimate decision, is my guess so far. Progress has been slower than I hoped in part because i was too cheap to buy the logic analyzer.. hard to debug early stages here. My colleague is working on getting the nano to act as a logic analyzer, probably going to be faster than waiting for one to ship from China. Anyway, give us the gritty details! On a side note, this video is super cool:

Eeeh? Even for Metabones 10k is chump change. 10k won't get you a single good developer for 6 months. I'm certain that at least a few people worked for quite some time on that. That said, progress would be much faster if I had a 40 hour week to work on this instead of maybe a handful of hours every other weekend or so.

Just two or three, on the edge there, so it's not a big deal for me to do it more than once.

Not good news :P First I tried with Canon 40mm pancake, the lens wasn't responding at all, I suspect that the protocol has expanded a bit. Spent that weekend trying to get any kind of response to no avail. Good-ish news, during the week I tried the 28-135 and there is definitely movement in the lens, but I've had no time to explore further. It seems like I might need to get a logic analyzer to have a shot at it working with newer Canon lenses. If anyone wants to leapfrog me, there is someone that claims to have made such a working adapter for basic manual control using buttons on the adapter for focus and aperture, and was actually using it on NX. Scroll way down to the post by a Peter S on January 2, 2012 at 14:15 https://pickandplace.wordpress.com/2011/10/05/canon-ef-s-protocol-and-electronic-follow-focus/ Need to adapt that code to Arduino Nano when I get a chance, very similar microcontroller used in his project. http://blueringlab.com/2016/03/canon-ef-lens-aperture-control-test/ It's too bad he didn't post that code, the answer to his unresolved precise aperture setting is in Peter's post. http://howiem.com/wordpress/index.php/2016/07/07/motion-control-canon-ef-lens-hacking/ Is an informative post, he also has code available, but it is a bit different, being driven from a connected computer rather than a microcontroller, etc.. but still possibly useful. In my searching about, it seems that almost everything the public knows about the EF protocol originates from this German thread: http://www.dslr-forum.de/showthread.php?t=649529&page=61 Which is unfortunately a jumbled mess and all the useful compilations of the info some members made seem to be defunct. Still, it's worth combing through. The next step for me I think is to re-verify the physical integrity of my connections, then start working with the 28-135 instead of that obnoxious 40mm. Not sure 100% about the logic analyzer yet... but since I'm thinking it will be needed eventually I'm starting to feel like I should order one soon. The NX side is easier and harder. I think I have quite a bit more concrete information on the protocol, thanks in part to rockymountain's blueringlab site, but this side likely involves fooling the camera about certain aspects of the lens and spoofing lens responses. Luca: If I want more than one focal reducer should I donate that number of times? <- in bold because the site is merging my posts and I thought this should be separate.

kidzrevil: What makes you say that? Looks to me like what it is, about a 1.1x crop, far closer to FF than super 35.

Hey Caleb, I think they ship regardless, but I think Luca will clarify. I think the earlier ship dates are less expensive to encourage earlier participation and try to build momentum.

Hey Luca, I did look at the raws a few days ago. It's solid, but not amazing, I see some loss of sharpness and other minor aberrations. Maybe Darktable isn't doing the exposure quite right, or the adapter doesn't have great transmission, the brightness was increased by around .35 stops. Since it's about a 1.1x crop instead of 1x a full stop isn't to be expected, but I was expecting more like .6 or .7 stops. Still, it's pretty decent, I think I'll get one soon.

Luca: Sorry, very busy week at work, there's a competition we're participating in so we've been trying to meet those deadlines. I would look now but I don't have have the software for it on the WIndows side of this computer, promise I'll look before I sleep tonight though. carlic: I'm going to try, that's not a question of funding, merely of my free time, my colleague's, and if we have enough info to pull it off. We finally have the parts to get started but he is sick so I'm not sure if we will take a look tomorrow or put it off until next weekend. If all people need is aperture control, that is a much easier proposition than the full deal and I believe implementation is available online as code for something like an Arduino. I have no intent of doing any manufacturing, if I am successful I will post the code, a full component list, recommended tools, etc... so that others can easily do it themselves. I intend to modify a dumb Canon to NX adapter using contacts from the AF supporting macro extension tubes for both NX and Canon, so a screwdriver, small handsaw, some kind of hotglue or epoxy and those 3 cheap adapters should be all that is needed mechanically. The whole thing should be doable with < 100 USD even if you don't own any of the tools at the start. Miniaturizing components or packing it all neatly into a custom adapter body is a different story, maybe Luca would be interested in that portion if we reach that point. Don't get your hopes too high though, we haven't even really started and it may be quite difficult, we'll see. It's also possible that we face lens by lens challenges, in which case it's not exactly feasible for us to go through every lens on or own, so we would post what lenses it is known to work with and put the code on Github or the like so that others can contribute parameters for their lenses if they figure them out.

Thanks Marco, but it has only one SPI? I think this project needs two off the top of my head. Also my friend already has owns a Nano, which is why we went for it for now. Finally, the Bluering code is written on the Nano and gives us a slight head start.

Hey Brian, how much opportunity cost are we talking here? I'm not going to back down from my 10k offer though I suspect that wouldn't exactly cover things, just a feeling, lol. On the other hand it would be a bit irritating to pay it out and then find out it wasn't so hard after all (possibly, I'll have fun taking a shot either way). I'm referring to the electronics of course though rather than the lenses. Also, a question about that for you, is something like an Arduino Nano micro controller really fast enough to handle the spi at native speeds? I'm going to prototype with one in either case, but if it's not then I may move to FPGA later if the Nano is functional and just sluggish. Luca, I would prefer to be one of those supporters, but I guess I'll have to wait for those first shots from others if you're not feeling up to posting two raws, same lens, same framing (tripod), same focus, same aperture. I know you have video samples, but since I'm not much of one for video that's just not my evaluative area.

Inazuma, in regards to that, at the time Metabones added that to their FAQ many of us thought it looked like bullshit because they didn't think there was enough of a market in part. And the other part is that they had some qualifiers, something about not getting quite the quality they wanted, I suppose in a 1x crop situation for the adapter (disingenuous because they offer non 1x adapters for some other cameras). I think Luca also could not do that as he has gone with what I think is a 1.1x crop that is using pretty much all available depth.

Hey luca, I've been absent a while. Could we see two raw files with the same lens (one with your NXL and one with just a glassless adapter). Also, is there space in your current design for pass through electronic pins? I do intend to take a (long) shot at the electronic communication and am wondering if I could work on one of these directly. Waiting still for some parts from China to make my attempt at doing it with a glassless adapter.

I'm struggling to think of many examples for any system that cover that range. 24-70, plenty, 24-105, a decent number, 28-135 they're around, 24-150!?! There are some 18-135 lenses that would hit that as an equivalent range, but you say you're looking for that as the APS-C range... so we're looking for a FF lens that's 42- 225ish. The only things that come to mind are the other superzooms, none of which are optically stellar. I guess there is the Sony 24-240 FE lens, only a 10x zoom rather than the more typical 20x+ of 18-200. Fun fact, the 18-200 is the only NX lens that saw production in Japan, makes one wonder if there was any foul play, lol. It's not missing focus so much in the bright sunlight as it is just that soft at the longer focal lengths, manual focus it, it isn't any sharper. At 18 the lens is quite decent, but if you only needed the wide end then there's lots of other options. Go with two lenses, even like the 16-50 PZ and 50-200 combo work out well and are cheaper together. If you're using it for video it's quite decent, and I guess it's video where having that range in one lens is most useful. Anyway, the 18-200 is the only lens I have bought for which I'm unsure why I still own it. Sometimes I think I'll avoid the hassle of taking more lenses and just take the 18-200, give it another chance on a walk... 10 minutes later I'm back home to take two other lenses instead.

For Sandro, the 18-200 is the problem. I think f/6.3 is perhaps too dark for the on sensor PDAF points to be effective. The lens focus speed itself is pretty wicked fast on that one, but if the camera doesn't know what to set it to.... you'll notice it sometimes jumps and that jump is super fast. And yeah, there are situations where NX is a bit dumb, if the subject is small in frame the NX cameras love focusing on the background instead. There are some issues like that, but I still find the NX1+50-150 quite capable of BIF and sports. The 18-200 was exceptionally bad among NX lenses. I can shoot a reasonable amount of BIF with the 50-200, the 45 and 85 primes are very fast focusing, just too short for BIF, but just fine for sports. If you need to use the NX1 in truly low light, that green beam really works quite will in even absolute darkness to a decent distance. Without it, yeah, it doesn't make it as low as a fair number of others. Jefbak and Marco: To be fair, they were speaking in the context of maintaining their current quality, but yeah, it could just have been an excuse because they didn't feel demand was high enough for NX. After all, some other mounts were able to have them at reduced levels of focal reduction. Don't remind me of the 300 f/2.8... I still want it badly. If quality samples are looking good then I would be in for an EF version even without aperture or electronic controls so long as the price is in line with that. I do have a number of FF fully manual lenses in EF mount that would be even nicer on NX that way (e.g. Samyang and Laowa). I could chip in some for a production run, but that 10k is reserved for a fully functioning near native AF electronic adapter.

Could you show an equivalent sample later on a tripod/set down on something solid? Much appreciated. It's looking pretty good.

There are some potentially useful files for the AF in the NX30 open source firmware, just do a search of the extracted files for "lens" to find some interesting stuff. By my reading of the agreement when downloading the files I can't just tell you what's in there. I'm not sure if it tells us much more than ihkim's work, but it might fill in some values. If anyone doesn't know where to download those yet: http://opensource.samsung.com/reception/receptionSub.do?method=sub&sub=T&menu_item=photography&classification1=digital%20cameras I don't remember if these were still in the NX1/NX500 released files or not, that firmware is on my laptop at work and I don't want to download it again at the moment. Regarding my interest only in the Canon side of things, I own 0 Nikon lenses but quite a few Canon lenses, including now some long glass (400 f/2.8). Any optical design for the Canon will work for Nikon with only some length extension to the housing, but not vice versa if the Nikon design used that space. There are more Canon lenses that need the aperture control. So yes, a bit of a selfish interest, but that's how it is.

ihkim's site is unfortunately not very intelligible for me through google translate. While his English isn't perfect it is way way better than google translate and he seems to have set up another site (that is linked to from his blog) where he goes through what he knows of the protocol in English: http://blueringlab.com/ Unfortunately I have no interest in a speedbooster for Nikon lenses on NX. For Canon lenses, yes, and such a design could be shared for Nikon lenses. On the other hand though, a design for Nikon lenses that uses the extra space would not then work for Canon lenses.

Hmm, Metabone's complaint about the space isn't totally invalid. Their claim is that they couldn't make it fit and also be at the same quality as their other adapters (and I assume they mean while keeping down manufacturing costs as being sort of implied). Anyway, I'm very interested in this, If we can get one allowing good AF with EF lenses.... speedboosted or even just adapted I would be willing to put 10,000 USD towards a manufacturing run. In either case I would need to see the good performance in the prototypes first of course, for AF and optically. The size of the prototype doesn't bother me. For instance, if on the prototype an Arduino and external power, etc... are needed, so long as in the final production design it is constrained to being contained in the adapter (and here I don't mind quite a thick adapter). Others can tell you I'm good for my word, I funded most of the burner NX500 for the firmware hacking. SMGJohn, we do know what tools they used, but knowing the tools doesn't give us their source code. You can find the source for the camera OS and build instructions here: http://opensource.samsung.com/reception/receptionSub.do?method=sub&sub=T&menu_item=photography&classification1=digital%20cameras but it doesn't have the source for the camera app that runs at boot and really does everything camera related. This kind of change and repackage the firmware approach has been done before on the NX300 (someone added encryption) Vasile's approach is much safer since once the base firmware is modded all other changes are done outside of that and should be reversible with no chance of bricking. Vasile, I'm just afraid if I fund your lens you will lose your motivation, or worse, spend all your time taking photos :P