rscheffler wrote:
For other VM40 Heliar owners, are you seeing some field curvature at farther distances? If aiming for optimal across-frame sharpness, is it better to focus in the midfield instead of the center?
OK, just had the chance to try the Heliar in a more controlled setting and better weather. Target subject was a row of houses across the street. Observations: during my initial use yesterday it looked like the lens back focused with correct rangefinder alignment. And this remains true for the very center of the image. The house directly across the street is slightly soft in the center of the image at wider apertures (2.8-4) but sharpness is good in the rest of the image along the same plane. Stopping down to f/5.6-8 masks this somewhat. At f/5.6, slightly front focusing appears to be a reasonable compromise to achieve across-frame sharpness. I tried very small increments of RF misalignment. The lens appeared to be quite sensitive to overshooting this. Slightly too much and midfield and edge sharpness noticeably dropped compared to center, even stopped down.
I haven't yet done a 'find edges' field curvature test but the Heliar reminds me of my LLL 35 Cron replica copy's field curvature behavior, though less extreme. With the LLL there's focus shift and I intentionally front focus the RF patch to get good 'averaged' stopped down across-frame results (my copy might be more extreme than others). The Heliar doesn't appear to show focus shift, at least at far distances. I haven't yet used it enough at nearer distances.
I still have mixed feelings about the Heliar. It's not quite as 'simple' as I was hoping. In some ways it's almost too similar to the quirks of the LLL.
Here are a couple GIFs showing field curvature with this copy on an M240. Focus was on the near bottom left part of the soccer goal post positioned at frame center. As can be seen in the wide open image, focus is slightly behind the intended point, but means that the overall plane of focus is in the intended plane.
I'm seeing slight focus shift between f/2.8 and f/4, after which DOF seems to mostly mask it. TBH, this is reminding me more and more of my LLL 35 Cron replica.
For reference, here is the FC Fred posted earlier in this thread:
rscheffler wrote:
Here are a couple GIFs showing field curvature with this copy on an M240. Focus was on the near bottom left part of the soccer goal post positioned at frame center. As can be seen in the wide open image, focus is slightly behind the intended point, but means that the overall plane of focus is in the intended plane.
I'm seeing slight focus shift between f/2.8 and f/4, after which DOF seems to mostly mask it. TBH, this is reminding me more and more of my LLL 35 Cron replica.
For reference, here is the FC Fred posted earlier in this thread:
Hey, my bad! I totally forgot to check out the CV 40/2.8 for FC. But if I remember correctly, it performed really well in the center and mid-field when shooting wide open at infinity distance. But, your example shows that it definitely has that wavy effect.
IQ at infinity for me looks identical to Fred's review tests, so there doesn't seem to be a performance penalty for adapting it to the SL2-S.
The rendering of this lens reminds me so much of the 28 Summaron. Since they share the same filter size (as Fred pointed out), I think these two lenses would be a perfect pair for a small/light M kit.
highdesertmesa wrote:
First few shots with this lens on the SL2-S
IQ at infinity for me looks identical to Fred's review tests, so there doesn't seem to be a performance penalty for adapting it to the SL2-S.
The rendering of this lens reminds me so much of the 28 Summaron. Since they share the same filter size (as Fred pointed out), I think these two lenses would be a perfect pair for a small/light M kit.
Looks good! Are you seeing good across-frame sharpness at infinity when focused at the center? Wondering if mine is a bit off, or if I'm expecting too much from it. It also has quite strong vignetting with cool/blue color shift. On my M240, using the 35 Cron pre-ASPH lens code fully improves color quality in the area of the vignette.
rscheffler wrote:
Looks good! Are you seeing good across-frame sharpness at infinity when focused at the center? Wondering if mine is a bit off, or if I'm expecting too much from it. It also has quite strong vignetting with cool/blue color shift. On my M240, using the 35 Cron pre-ASPH lens code fully improves color quality in the area of the vignette.
Thanks! If you’re seeing uneven sharpness across the frame, I’d try another copy. That sounds like my first 50 f/1, which would look unsharp in seemingly random places and had excessive focus shift at infinity.
I don’t remember seeing softness anywhere but the corners on mine. And the corners improve with every f-stop until f/8 is nearly perfect except for the farthest bit of the corners.
Regarding color shift, on my LCD, the JPEGs look like there is a magenta shift at the edges, but it looked normal once I got it into Capture One. Cool blue doesn’t sound like a color shift, just darkening from vignetting. Have you tried removing the vignetting then boosting the saturation to see if the edges retain the color shift? Have you made sure Lightroom isn’t applying a lens profile from the DNG? Sometimes Lightroom introduces all sorts of weirdness if lens corrections are applied.
So far I've only reviewed SOOC jpegs (from DNG + jpeg pairs) in Photo Mechanic. There's definitely a difference in the vignette color between uncoded and coded as 35 Cron pre-ASPH by my M240. It's not a huge difference but it's very similar color difference to what I get with the LLL 35 Cron replica if uncoded vs. coded with the same code.
I think what's bothering me most with this copy is its apparent field curvature and the negative impact it has on sharpness at center when optimizing for across-frame performance. I'm not convinced this copy is optimal. Given the store where I purchased it has a short return window, I'm going to do so while I still can.
rscheffler wrote:
So far I've only reviewed SOOC jpegs (from DNG + jpeg pairs) in Photo Mechanic. There's definitely a difference in the vignette color between uncoded and coded as 35 Cron pre-ASPH by my M240. It's not a huge difference but it's very similar color difference to what I get with the LLL 35 Cron replica if uncoded vs. coded with the same code.
I think what's bothering me most with this copy is its apparent field curvature and the negative impact it has on sharpness at center when optimizing for across-frame performance. I'm not convinced this copy is optimal. Given the store where I purchased it has a short return window, I'm going to do so while I still can....Show more →
I need to go back and look at my JPEGs, but on the LCD the color looked off in the corners with no profile assigned on the SL2-S. Never seen that before where the JPEG looked wacky but the DNG was good.
For sharpness, what happens when you optimize for center sharpness? What are you seeing in the rest of the frame? The corners never get fully sharp at f/8 if focused at the center. I’m ok with that as even the slightest crop at f/8 removes the those farthest corners. I haven’t checked midframe closely, but across the centerline from edge to edge on the long side should be sharp by f/4-5.6.
p.12 #10 · Voigtlander 40mm f/2.8 Heliar VM Review
rscheffler wrote:
For sharpness, what happens when you optimize for center sharpness? What are you seeing in the rest of the frame? The corners never get fully sharp at f/8 if focused at the center. I’m ok with that as even the slightest crop at f/8 removes the those farthest corners. I haven’t checked midframe closely, but across the centerline from edge to edge on the long side should be sharp by f/4-5.6.
With my copy, I need to front focus slightly compared to ideal rangefinder focus to achieve best central sharpness. It pulls the plane of focus forward enough that the midfield and periphery of the image drop in sharpness along the image plane that corresponds to focus at the center. If I focus for correct rangefinder patch alignment, the center is a little back focused (more noticeable up to f/4 and can still be noticed with pixel peeping at f/5.6) but the rest of the image is correctly focused. Setting focus between these extremes (which actually is not much in terms of focus ring rotation and is a bit annoying near infinity due to the infinity lock) and shooting at f/5.6 seems to be the necessary compromise for sufficient DOF coverage of the desired plane of focus across the full image. f/8 gives a bit more leeway. I mean, it's usable but this is also exactly what I currently do with my copy of the LLL 35 Cron replica. There are aspects I prefer of the LLL over the VM, hence why I will return the VM.
p.12 #12 · Voigtlander 40mm f/2.8 Heliar VM Review
I returned my copy after completing the review, but now I regret it because it performed optimally. After receiving Ron's report, I realized that I should have tested it for field curvature, which I had assumed to be minimal based on its performance at infinity.
p.12 #13 · Voigtlander 40mm f/2.8 Heliar VM Review
rscheffler wrote:
With my copy, I need to front focus slightly compared to ideal rangefinder focus to achieve best central sharpness. It pulls the plane of focus forward enough that the midfield and periphery of the image drop in sharpness along the image plane that corresponds to focus at the center. If I focus for correct rangefinder patch alignment, the center is a little back focused (more noticeable up to f/4 and can still be noticed with pixel peeping at f/5.6) but the rest of the image is correctly focused. Setting focus between these extremes (which actually is not much in terms of focus ring rotation and is a bit annoying near infinity due to the infinity lock) and shooting at f/5.6 seems to be the necessary compromise for sufficient DOF coverage of the desired plane of focus across the full image. f/8 gives a bit more leeway. I mean, it's usable but this is also exactly what I currently do with my copy of the LLL 35 Cron replica. There are aspects I prefer of the LLL over the VM, hence why I will return the VM....Show more →
Definitely not normal. Infinity hard stop should be the best center focus position wide open. This is especially important for a lens with an infinity lock.
Speaking of the infinity lock, it makes this lens frustrating to use at infinity when adapted to another platform since most adapters are slightly too long by design (like my Leica M-Adapter-L). This means that with a normal adapter, to focus at infinity one has to hold exact focus while the focus knob is only partway engaged into the infinity locking mechanism. If one releases the knob, it moves by itself to snap to one side or the other, losing focus. And since it's under tension, even manually holding it in place can be precarious. My solution for adapting this lens is to use a macro adapter with adjustable infinity position. I calibrate it on a known infinity-perfect lens like my 28 Cron at f/2, then lock it down in that position. Once I have infinity set on the adapter, all my M lenses hit infinity at the hard stop just like they do on an M body. But any old macro adapter will work even if it doesn't have adjustable infinity position – just put the lens to the infinity lock hard stop, set the aperture to wide open, then adjust the helicoid to achieve infinity focus through the EVF. Once you have that set, you can use the lens normally since most helicoid adapters are very stiff and not accidentally moved.
p.12 #14 · Voigtlander 40mm f/2.8 Heliar VM Review
Desmolicious wrote:
Ron, you can try out mine if you want. Just don't tell me it sucks until I am out of ear shot.
Thanks, it will be worth trying to see how it differs, if at all, with the one I got. I don't think my copy sucked, in general. I just didn't like that I was apparently dealing with field curvature similar to, though somewhat less extreme, than what I experience with my LLL. Since I already have the LLL, I don't really need to add another similar focal length with similar behavior. Maybe I just got a bit unlucky with the VM?
p.12 #20 · Voigtlander 40mm f/2.8 Heliar VM Review
Which 40mm Voiglander would you then choose for landscape in f/5.6-11 most of the time if you consider field curvature, distortion, colour rendering, smooth light transition and sharpness?
