Barmaglot

Interesting - it looks like SeaFrogs offer a port and zoom gear for a 70-200 F4, although they don't specify whether it's a 1st or 2nd generation lens. Looking at wayback machine, this listing appeared quite recently, so it's likely to be the 2nd gen one, and the 4" diameter front element suggests that it should have the full zoom range without vignetting. I have to admit I'm kinda curious as to how it would perform... but not US$2000 level of curious.

I don't have a Divevolk housing to try and answer your first question, but as for the second, these crossbars are available on Aliexpress, example: https://www.aliexpress.com/item/32934231664.html

Looks like third time was indeed the charm - now both ends appear to have a snug but not overly tight fit. As an added bonus, it turns out that if I use my old short macro port (originally meant for 16-50mm kit lens or 30mm macro) instead of the 4-inch dome, I get an absolutely perfect fit with the Canon EF-S 60mm macro lens: null This opens up the possibility of using the 60mm with wet lenses, such as attaching the UWL-09F to widen its field of view a bit for larger subjects on blackwater dives. Time to progress to the waterproof print iteration.

There was a recent thread on the subject in the photography forum: But, what exactly is the challenge with framing on the internal display? Is it the angle? Visibility in sunlight? Size? Some of that can be solved with a much cheaper and less bulky magnifier hood.

FWIW, if you're worried about FX30 overheating, the a6700 is at least as susceptible to that when shooting 4k over prolonged periods, but honestly I can't imagine scenarios where you'd be shooting video for over 20 minutes straight while underwater. FX3 and A7SIII are full-frame cameras, so an entirely different price bracket. FX30 is definitely a more capable video camera than a6700, so if you're looking for a pure video device, it holds the advantage, but a6700 is an excellent hybrid choice, great at both video and stills. Regarding corner quality, some years ago I did a comparative test with my a6300, 16-50mm and 10-18mm lenses, using SeaFrogs 6-inch and 8-inch ports, although the 8-inch port is more like 180mm in actual dimensions. You can see the results here: https://1drv.ms/f/s!AupWSggdlFYKjtRFu-IIxyopM8fvAA?e=oq7ekL - a6700 shouldn't be much different. As I understand it, a6700 and FX30 use the same sensor, but FX30 is purely a video camera, lacking the ability to fire strobes entirely as it has no mechanical shutter. The 1/160s flash sync speed on A6xxx bodies (as well as A7C series) is not nearly as big a deal as some make it out to be. It's about half a stop worth of difference - shooting wide-angle, I'm at considerably slower shutter speed almost all the time anyway to keep background bright enough, and if I want dark background, 1/160s and f/16 at ISO 100 is plenty to make the background comfortably black. The only scenario where it really comes into play is sunballs, and even then, f/22 and full strobe power is generally quite enough. Another point of note: if you want a fisheye for wide-angle, Sony has very limited native choices, and most users adapt Canon lenses (8-15mm on full-frame, 10-17mm on crop), but autofocus options with adapted Canon lenses are limited - at least with my Metabones IV adapter, its green mode works very well in stills, but does not autofocus at all in video, while advanced mode autofocuses in video, but has degraded performance in stills (tested with Canon 60mm macro and Tokina 10-17mm fisheye),

It's 1/160s, not 1/60s. As a long time A6300 user (same sync speed) who just upgraded to a6700, I don't find it particularly limiting.

Took my new a6700 out for the first time yesterday and checked focus with red light. Conditions weren't perfect - the dives were shallow with lots of sunlight, but very poor visibility (tons of particulate matter in the water, lost sight of other divers at 3-5m) but I found some dark crevices, and to my surprise, I was able to see perfectly well on the camera screen while using the red mode on the focus light. I don't know what is it that Sony had changed between a6300 and a6700 in this regard, but whatever it is, it appears to be working very well.

Sorry about cluttering up the thread with only tangentially relevant info - @TimG, maybe it should be split into a separate thread in the DIY forum?

I used a local shop here in Bangkok, and it cost me 650 THB (about US$18) to print zoom gear + the port adapter I'm working on, and another iteration of the port adapter that I just ordered cost me 350 THB (including shipping), so it's comfortably under $10. For the reference, here's the fit of the gear inside the housing: The contact pads are a bit untidy after I filed them down, but they work. The driving cog is on the bottom left, the other two are just frewheeling supports.

I guess I'll have to take my chances and most likely iterate once.

I just made a zoom gear for Tokina 10-17mm fisheye to use with my SeaFrogs a6700 housing, and here's the process that I used. It might look primitive for someone with working knowledge of CAD, but it's what I came up with having had zero experience beforehand. I used the free online TinkerCAD - it runs in a browser, so you don't need to install any software. First, I took a pair of calipers and measured an existing zoom gear for my housing, finding the diameter at the tips of the teeth, at the bases of the teeth, and the width of each tooth. I also counted the teeth, ending up at 72 - a very convenient number, giving me exactly 5 degrees offset between each tooth. Now, in TinkerCAD, I created a cylinder and set its diameter (width x length; TinkerCAD doesn't do 'diameter') to what I measured at the base of the teeth, and its height to the teeth's thickness. Then I added a box, with its length set to the gear's diameter at the tips of the teeth, width set to a tooth's width at the base, and height set to the same as the cylinder. I selected the box and the cylinder and aligned them to each other, so they would be centered. To this, I added four small boxes of the type 'hole' rather than 'solid', each rotated 30 degrees and positioned to intersect with the first long box at its junction with the cylinder. This is basically what I got: Selecting the four small 'hole' boxes and the long red 'solid' box, I grouped them, and got my first two gear teeth: null Selecting this shape, I hit 'Duplicate and repeat (Ctrl+D)' and rotated the duplicated shape by 5 degrees: Repeat 34 more times and I have a full gear of teeth. It's a bit tedious, but not that difficult: Select all, group, and now I have my basic gear as a single piece. Now, put the camera with the lens that I'm making this for into the housing, and use the calipers to measure the distance between the driving cog in the housing and the far edge of the zoom (or focus, if that's the target) ring on the lens - this is going to be the gear's height. Add a solid cylinder of the appropriate height, and diameter to fit within your port's internal diameter, with a few mm of clearance to spare. Add a hole cylinder slightly taller than everything and a few mm smaller in diameter than the solid one - the difference will be gear walls' thickness. Align everything: null Select all, group, and now the gear is hollowed out. Now, the final measurements are the width of the lenses zoom or focus ring, and its diameter. Add a solid box that is slightly longer than the hole cylinder I used to hollow out the gear in previous step, as tall as the target ring is wide, and a couple mm thick, pull it level with the top of the cylinder and align it in horizontal plane: Duplicate, rotate 45 degrees, repeat two more times. Add a hole cylinder of the diameter equal to the target ring's diameter, drag it to overlap the four boxes, then align everything again: null Select all, group, and voila - the design is ready to send off to the printers. Export to STL and get in touch with your nearest printing service. Now, in my case, I had to iterate once - the gear ended up being slightly too small; the teeth were barely catching on the driving cog. I just took the whole thing and increased width and length by 2mm - this got it perfect on the second print. The eight nubs that contact the lens also ended up slightly too long - they wouldn't fit over the lens, even after I expanded that last hole cylinder on the second print. Rather than do a third print, I simply filed them down a little bit, and now they fit perfectly. I went through this whole process start to finish to take the screenshots to illustrate the post, using arbitrary values for sizing, and it took me maybe 30-40 minutes start to finish. Doing the actual measurements would add a little time, but still - this is basically an hour or so long project, not accounting for iterations.

Right, but I'm using a local print shop here in Bangkok to print non-waterproof drafts so that I can dial in my design, with the intent to send the final version to @Isaac Szabo to print it using his tested method. I'm somewhat worried that the design that produces an exact fit using the local shop's printer and settings will produce something else when Isaac runs it through his process.

@Isaac Szabo, a question if you don't mind - how repeatable are parts dimensions across different 3d printers? I just got my second draft of the extension, and now the housing bayonet is perfect, with just the right amount of snugness to rotate with a bit of effort, but the port side is a little bit loose. I have reduced the internal diameter of the opening from 92mm to 91.3mm which, hopefully, will get it exactly right, but I'm a bit worried that when we're talking sub-millimeter tolerances, a different printer, different material and different settings can put it out of alignment.

First draft of the extension/adapter, with the Tokina 10-17mm inside: Got the dimensions a bit wrong - the port side is a little too snug, not leaving space for a o-ring, while the camera side is too loose, not getting any friction. The zoom gear worked on first try but is also a bit too small - the teeth are barely catching. Altered the design, sent it to the print shop for another draft run - fortunately it's quite cheap here in Thailand, less than $20 for both parts.

Strobe modeling light is useful on blackwater dives when you get a critter that is attracted to light - that way you can track it by spinning around, instead of losing it as it swims into your focus light. I'm talking about a different situation though, shooting stuff that will retreat into its hole if you shine a light on it, doesn't matter where the light comes from.

Interesting idea - no, I haven't tried that. The Weefine Smart Focus 1000 that I use does have full and half-power settings in its red mode, but I've only ever used the full. Weather permitting, I'm planning to do some macro dives this coming weekend, so I'll try that out.

Shooting macro, every now and then, I encounter a light-shy critter - saron shrimp, banded pipefish, some crabs, etc. I have a red mode on my focus light, but actually using it is near impossible - at least on my Sony a6300 + 90mm macro lens combo, I basically get a solid red screen with no distinguishing features. Pointing it here and there while mashing the focus-on button I get some darker blotches but nothing identifiable, so getting a shot is basically a matter of blind luck. I'm currently transitioning to a new a6700, and playing with its features, I discovered that if I set the white balance to a custom value off the red light (it ended up reading something like 3600K), then - at least while playing with it dry, in a darkened room, under a red light - the on-screen picture is a lot clearer. The resulting shot is extremely green-tinted, but when shooting raw, it doesn't really matter. However, and this is even more frustrating, this only works when live shoot effect is set to ON - when using settings appropriate for macro (f/16, 1/160s, ISO 100) this produces a black screen, since it previews the exposure as it would be without a flash, and if I set live shoot effect to OFF, then the white balance value is ignored along with everything else for the purposes of framing display, and everything goes super red and indistinguishable again. Are there any ways to make using red focus lights easier on modern Sony mirrorless?

Looking forward to hear of the results. All the reviews of FE 50mm that I've seen were done many years ago, when it was first released, so it'd be good to know what its comparative performance is with newer cameras.

Yes, but @ChipBPhoto is looking for something wider than 90mm to use on his A1. Yes, a 60mm lens on FF would be like using a 30mm on M43 - and as it happens, the 30mm is the most popular blackwater lens on M43 cameras. The questions which stand (and thus far haven't been answered) are: How large is the actual image circle of the Zeiss 50mm in terms of full-frame coverage? Not in APS-C mode, but in full-frame mode. For Canon EF-S 60mm this is known - full coverage at 1:1, slight corner vignetting at longer distances. How good is its focusing speed and reliability vis a vis Sony FE 50mm and Canon EF-S 60mm? And again, the wider AoV is not for shooting larger subjects at a similar distance, but for easier target acquisition at distance before moving in to closer distance for similar framing. The FE 50mm would be the obvious solution for a 'wider than 90mm macro lens', but it suffers from limited minimum aperture (f/16) and has a very poor reputation for focusing speed, which may or may not be alleviated by the newer-generation autofocus in the A1 body, hence @ChipBPhoto looking at alternatives.

How would you rate their focusing speed (traversing between far and close focus, locking on to subject) in air in different lighting conditions, comparing against each other?

Wait wait wait, besides you and me, how many board members were on that trip? LOL, I really should talk to people more... This is actually quite similar to my experience on Sony a6300. Before I used to use a Weefine Smart Focus 1000 for focus lighting, and found that it really struggled to lock focus unless I added the light from both Retra Pro modeling lights and often the spotting torch as well (I'd hold it with my ring and pinky fingers of the left hand, pointing it at the subject, as the rest of the hand was holding up the camera handle). This trip I put my old Archon D36V on the cold shoe, and found that focusing got a lot easier. Still not perfect, but not as much of a struggle as it used to be. It's kinda worrying though, that a brand-new flagship body is exhibiting similar issues to an eight-year-old camera that retailed at a quarter of its cost.

The FE 50mm macro lens suffers from a reputation of being very slow. Whether or not this reputation is well deserved, I can't comment, not having used it, but it's there. It also have a fairly limiting minimum aperture of f/16. When I started doing blackwater, my only macro lens was the 90mm, and on my a6300 body, I quickly found it to be unusable, and my search for a better solution led me to Canon EF-S 60mm macro. That lens focuses quite fast, but tends to hunt in less than very good lighting - I have a Weefine Smart Focus 1000 focus light, and I found that I needed to amplify it with my strobes' modeling lights, and frequently my spotting torch as well. On my last trip, I put a 5400lm video light on the housing cold shoe, and this actually worked quite well - I encountered almost no hunting in one and a half blackwater dives that I did. The Canon 60mm, at least on older bodies, focuses much faster than the Sony 90mm, and its image circle provides almost complete coverage of a full-frame sensor - there's slight vignetting in the corners that disappears as you get close to minimum focus distance. The question is, how does the Sony-Zeiss 50mm compare to it in these criteria - focus speed, and sensor coverage. The newer bodies (A1, A7RV, possibly A7CII/A7CR) have made the Sony 90mm a viable option for blackwater, but plenty of people are still using A7IV and older cameras. The shorter focal length is for easier target acquisition. You have a wider field of view to acquire your subject from a moderate distance, center them in the frame, lock focus, then move in for maximum feasible magnification. The 90-105mm lenses are a bit narrow in this regard; you have to aim them much more precisely for that initial lock-on - many blackwater photographers consider the '60mm on full-frame' to be the sweet spot between magnification capability of 90-105mm lenses, and wide field of view of 30-40mm lenses. I have personally tried Sony 30mm f/3.5 macro on APS-C and found it too wide - it's easy to do the initial acquisition, but achieving proper magnification requires getting in so close that tracking and lighting become an issue.

The idea is to use the APS-C lens in full-frame mode, without cropping. The 90mm has a fairly narrow field of view, making it challenging to use on blackwater dives, and the FE 50mm is slow. The Canon EF-S 60mm when used on full frame only produces slight vignetting in the corners, which is immaterial for blackwater shots, hence my question as to how big is the actual image circle of the Sony-Zeiss 50mm.

How big is the image circle on the Zeiss lens? I know that the Canon EF-S 60mm has only slight vignetting on full frame, but I haven't seen any tests with the Zeiss 50mm.

If you ever end up taking it on a blackwater dive, please sound off with the results. I'm curious as to how it would perform vis a vis an adapted Canon 60mm. I'm currently transitioning from an a6300 to a6700, and while the 90mm has received a very substantial boost in autofocus performance, the Canon 60mm still suffers from hunting, so I'm pondering selling it and getting a Zeiss 50mm instead - while the 90mm autofocus can probably handle blackwater with the new body, its field of view is somewhat too narrow on APS-C when everything is moving around.