In the last couple of posts I’ve bashed on the Fuji 18-55 zoom. From my point of view it’s not really bashing it but instead it’s just a touch of objectivity in a sea of gushing praise on the ”Fuji can do no wrong” as well as the ”Half of the 18-55’s are bad copies” nonsense found when looking for information out there. Today’s topic is a sprinkling of stuff about the 18-55 as well as the Fuji XT-1 but as a whole it’s more about really getting to know your gear than anything else.
A few updates ago a reader left a comment or a twitter response or something to the effect:
So you’ve spent like forever with the Nikon Df and a couple of days with the XT-1. Of course it’s going to feel funny.
I absolutely agree holistically but not about the couple of handling specific things I was complaining about in that post. Patrick LaRoque gave me a couple of super helpful setup tips that solved a few of my specific handling issues but on the whole my opinion is about the same. Thank goodness most of them have no effect on things I give a hoot about. I only bring them up for other’s consideration. As for getting to know your gear I typically shoot StupidCrap™ that I don’t care about so I don’t look like an idiot when I’m using cameras with real people. That also gives me some broad idea about what works/what doesn’t, some of the limitations of dynamic range, etc.
Last weekend I shot a little over 1,000 shots and almost every one of them was not only to make photographs but also testing to one degree or another. Testing in the sense of specifically trying variations and experiments that would give me experience and results to become intimate with the XT-1, X-TRANS RAW files, and two of the lenses that I’ve not used extensively. There were four things I wanted to be far more comfortable with.
- The XT-1 4.00 firmware features.
- The 18-55 XR in real use.
- The 14mm 2.8.
- X-TRANS RAW files in real world light.
The Fuji 18-55 XR
In my last post I summarized general findings but I didn’t go into any specifics or what I looked at, or how I tested, or even my own bottom line on if I like the lens or not. At the top is a BTS shot of one of my lens testing setups. My version of extreme corner testing.
I shoot using large-ish apertures, slow-ish shutter speeds, with subjects in motion to some degree. I did mention that all of those things will trump lens resolution in most cases if we’re measuring in some scientific way. My own measurements are more along the lines of; but how does it look? In semi-psudo-technical-qualitative terms here’s what I found.
- First up I’m not completely impressed by the 18-55’s performance at the long end. Especially used really close up.
- In terms of overall look I’m on the positive side of the 18-55 looks good. In-focus to OOF transitions look good. Slightly OOF backgrounds look good. Contrast is okay. Slightly OOF hair looks pretty decent (a big thing for me, consider this bokeh that I give a shit about see above out-take).
- I mentioned that the technical performance is mostly on-par with other crappy-ass $100 kit lenses which aren’t so bad either. They are but I like the way the Fuji looks a little better. One way I like it a little better is how it looks where it’s not so great in technical performance. Do I like it as much as the 23/2 welded to the X100? No but but I like it better than say the Nikon 18-55 VRII.
- Where it is absolutely fantastic? The 18-55 sings somewhere in the neighborhood of 25mm to 35mm around f/5.6. Anecdotal evidence suggests the range 27mm to 30mm but lets call that within experimental error tolerance. At f/4 it’s probably in that singing place around 27-30mm too. Even in the extreme corners and close up (see below).
- I’m still not completely confident with the VR/OIS in terms of what’s absolutely reliable vs what’s iffy. I get inconsistent results that are perplexing. Still have to play with that a bit more.
- This lens has delayed my purchase of the 35mm/1.4. Still on the list but I’m in no rush. I kinda want to see a version II of this lens. Maybe an f/1.0? Maybe with a better/quieter motor?
- Bottom line is this is about the only X-series zoom I like a whole lot. I am not a big fan of the size of the other zooms in this range. I much prefer primes on either end. The 18-55 is tiny and gives up not a lot. Is it as fan-tabulous as my 14mm or 56mm or 23mm at the long or wide end? No but it gives me a viable 28mm and 50mm equivalent in a tiny package that I will use a lot. Here’s a question; Why the hell don’t they use that motor in all of the lenses? I hope there’s a reason and would love to know what it is. It’s not just the speed. I love how silent and smooth it is. Every one of my other lenses feels a lot rougher and louder.
The XT-1 New Firmware 4.00 Features, Etc.
I’ve never been one that’s fond of messing around with focus point selection on the fly. On the fly as in between shots of a any given subject in a fluid situation. It just doesn’t work for me on any camera I’ve ever owned for what finds it’s way in front of my camera. Combine that with the hateful tiny buttons on most X-series bodies and it’s not something that I can do as part of the flow for a dynamic situation. Dynamic as in you’re looking thru the viewfinder of a fluid situation and see something and want to compose a bit differently so you flip the focus point somewhere else… oooops well that moment’s gone.
I’m much better at manual focus, being prepared and knowing I’ll be okay based on how far away I am, my aperture, my focus plane relationship to where the subject was/is… I’ve ranted about how manual focus is in many cases a far better strategy for those situations than attempting to monkey around with focus point positions. The one thing that interested me new with firmware 4.0 was the new group focus combined with one of the advantages of EVF/mirorless/contrast detect AF. The fact I can put focus points anywhere instead of a more limited range near the middle. The 3x3 group focus in both single and continuous focus modes would theoretically allow me to use AF for a ton of stuff I shoot with compositional flexibility that I need without forking around with focus point position every other shot.
How did this play in reality? I have to say it worked great. So ,so, so well that I’m going to stress the limits of group focus under much more demanding circumstances next time I shoot. There was only one instance in hundreds of shots where part of the 3x3 group was off the subject and the camera grabbed something in the background vs the closest stuff. I was wary of this based on my experience with previous X-series AF propensities for doing just that. Seems cured. I’m even going to try tracking with the 56mm wide open and up close and see what happens.
X-TRANS RAW Flexibility, Metering, Etc.
Let’s talk about real world light for the thousandth time. If I care to control every aspect of lighting and fill ratio and all that I can get absolutely pristine results from just about any digital camera ever made. Hell I could do that with my D2H back in 2003 when I brought a bunch of gear to the table. SOOC JPEGS looked friggin’ fantastic. Who cares, it’s not those conditions that have bugged me about digital ever, ever, ever. Honestly I don’t see what gets people that shoot in those conditions so excited about anything in the last 10 years.
One of my overly optimistic assumptions about digital is that any minute now we’ll get the same flexibility, malleability, and ability to shoot real world light that photographers have always had with black and white negative film and since the late 80’s with color neg. I love shooting negative film in real world light. It looks great, always, always, always. Sure high ISO has come a long way in giving you grainless/noiseless results but in not-to-challenging conditions lets take a look at typical digital and real world light. Take my word for it all numerical dynamic range indicators are not created equal. Again, it’s ”but how does it look”. The best camera I’ve ever used at a wide range of ISO’s compared to negative film in real world light has been the Nikon Df. So the big question is how does X-TRANS do? Even bigger is how do I have to shoot it and how does it look when I do…
Looking at the BTS shot at the top again (see how efficient I am, testing multiple things at once) you can clearly see the setup here and I’ll walk you thru what’s going on lighting wise.
- There’s a big tall window to camera right and in front of Mary. Let’s call that the main light.
- Another window exactly the same but blocked off about 3 feet on the bottom by an air conditioner is to camera right and to the rear of Mary. Let’s call that the subtle kicker.
- I moved Mary way back from those windows, about 8 to 10 feet so that there wasn’t fall off from head to toe. That also increased the fill ratio created by the bounce-back of the large room as well as the 4x8 white board you can see in the upper left of the frame.
- Now let’s take a look at that fill from a big white card just out of frame for a full length shot. You can clearly see the effect by comparing the shadow on Mary’s rear most leg cast by the closer leg to the shadow cast by Mary’s arms on her stomach. The shadow on the rear leg is easily a stop darker as it’s not being lit by the fill from the white board. You can also see how much that card is adding by the rim light from that direction on her raised leg on the shadow side.
- So we’ve added fill, at least a stop’s worth — probably more even at a point that’s relatively far away from that reflector. You can see based on that shadow cast on Mary’s abdomen that shadows get real dark, real quick.
- Purely for testing purposes and the fact that I’ve got a thing for white, the challenge is shooting a not so challenging real world lighting setup.
I want white whites but not ugly blown out whites. I want mid-tones, especially upper mid-tones like skin to actually be upper mid-tones and shadows to be ummmm, not black. How hard can this be? Well with digital if I put the skin where I want it guess what happens — yep every bit of the whitest parts of the white on both shirt and shorts gets all rammed up against the right edge of the histo-gram and almost everything white is flashing at me when I chimp. Shooting this on negative film is like falling off a log and it will look great. Not sure of exposure, first time out of the gate with something like this? Great just add a stop to the exposure, hell, add two.
Digital? Well, be berry berry careful… one false move and crap-o-la you loose. Common wisdom suggests that we be conservative and shoot it so no flashy shit is showing on those large white areas. Great, looks like shit in camera with skin way to dark and shadows all muddy. We’re shooting RAW so who the hell cares right? Wrong! You know what pushed shadows look like when you go that route? Very commonly they look really ratty with crappy gradation. The big question in getting to know one’s gear is just where you make trade-offs, how much highlight head-room you’ve really got, how ratty lower-mids/shadows look when you increase exposure in post by how much at what particular ISO’s.
In reality I’m not all that concerned about the deepest darkest shadows. I really only care about where I can and where I must put mids and upper mids in relation to highlights that I care about (as opposed to specular reflections that I don’t care about). I also care how that plays in relation to ISO’s that I commonly use in real world lighting. My most used ISO’s are 400, 800, and 1600 so that’s what I care about. The answer that I arrived at was that the X-T1 has at least a stop less headroom in the highlights across the board with more than a stop at 1600. What this means to me is that I have to expose about a stop lower than I would with the Df to get the same highlight renditions when then push the mids up in post via exposure or whatever. So how ratty do the shadows look doing this… well ratty-er than I would like but acceptable up to 1600. Past that and they really start to get nasty.
The shot above is ISO 1600 pushed 2/3 stop in post with shit-tons of highlight compression via VSCO 400H preset on import. The shot below is also ISO 1600 with a slight aperture/shutter speed adjustment and pushed 1 stop in post with the same import preset to give semi-normalized results.
At 1600 I’m teetering on the edge of completely gone in the whites for the first shot pushed 2/3. No margin for error and maybe even gone here or there but hidden in the compression and the VSCO under 255 cut-off point. In the next shot the highlights are pretty solid but teetering on the edge of any more pushing up of the mids and ratty-ass-city. So I’ve pretty much found where this camera hits the wall. Will I go beyond the wall or just live with denser mids? Sure because in many many cases you don’t need or want a higher key look in real world light or dealing with super bright whites. It sure is nice to have the flexibility though. Hence I’m still waiting for that negative film response with all the other stuff digital gives us. The Nikon Df (and D4) are pretty close. This kind of thing is way, way more important to me than some tiny bit of less noise next year. Also one of the reasons the Df is so great.
Can I live with this? First off shooting this at 1600 wasn’t really required, it’s just an ISO I use a lot. I could have easily shot this at 800 or even 400. I’m surprised that I don’t get a lot more headroom in the highlights at 800 and 400 though. I can also live with this partly due to the smaller sensor. I can shoot scenes that I wanted at f/8, f/5.6 or f/4 at a larger aperture and get equivalent DOF. What you are looking for and what I’m looking for might be quite different. Heck, I don’t really care that much about noise in and of itself, I left the fake grain in the VSCO processed shots above. The point is that really, really getting intimate with how what looks in under what conditions is absolutely essential when out in the field in the heat of the moment as opposed to making arbitrary decisions. All this crap wasn’t specific stuff you can take to the bank as far as settings and numbers, it’s far more to give a general idea of what kind of things I do when getting to know a camera and gear.
Now onto the XT-1 specific stuff, far quicker and more specific conundrums for fellow XT-1 shooters out there. Consider this part both observations of mine as well as questions regarding woking method for all of you and how you deal with it. Here are a few vexing things that just bug the shit out of me.
So let’s talk about the wonders of the EVF. You know the WYSIWYG mantra of see what you are getting before you shoot it and that live histogram… Why oh why does the live histogram smash into the edge before the actual JPEG file does when you chimp what you just shot? Why? For gods sake if I have a live histogram I want it to show me exactly when I’m hitting the edge not sorta-kinda. I can get really close by frigging guessing. I just don’t get this. I thought it might be me and anecdotal so I tested it. Yep live histogram smashes into the edge before the actual JPEG does if you chimp the histogram right after you shoot.
Okay so combine the above with another conundrum. This isn’t really Fuji specific but I shoot RAW+JPEG for two reasons. One is that you sort of have to shoot RAW+JPEG to view it at 100% when chimping critical shots. The other reason is so that I can send out nice looking JPEG’s on the fly prior to uploading all the RAW’s etc… Obviously both the live histogram as well as the chimped histogram show highlights smashing into the edge way before the RAW files in LR do. What do you guys do? Just memorize what JPEG setting give you how much highlight headroom in lightroom? Mess around with JPEG custom settings so that you get terrible looking JPEGS but the histograms match RAW files when imported? Then use the in-camera RAW processing to send out a JPEG here and there? Let me know what you do if you can…