Hall of Fame

Introduction

The Delta Hall of Fame grades for sharpness 200 enlarger, projector and industrial lenses at two working distances.

Lenses that score 90%+ are awarded Gold – reference-quality.
Lenses that score 80-90% are awarded Silver – serviceable for professional use.
Lenses that score 70-80% are awarded Bronze – borderline suitable or obviously compromised.
Lenses that score below 70% are classified ‘Soft’ – for vintage effects only.

The relevance of working distance and magnification cannot be overstated. Delta is NOT ranking these lenses as macro optics – an already well-documented topic. The objective is to discover how they perform as taking lenses for typical stills and video usage. The ‘Near’ metric assesses performance at working distances of 50-100cm, which corresponds roughly to their use case as enlarger lenses and a broad range of close-up applications. The ‘Far’ metric assesses performance at working distances of 7-15m, which derives from their use case as projector lenses and points to their behaviour as general and portrait lenses.

The ‘Near f5.6-f8’ metric is a reliable guide for those seeking a ranking of enlarger lenses in their native application. The ‘Far Wide Open’ metric is a similarly reliable guide for those seeking a ranking of projector lenses. However, because of the relative ease with which geometry and vignetting can now be corrected for capture applications, such properties do not figure in the rating. The only aspect of lens performance here rated is ‘Sharpness’ – which we give a somewhat subjective, but practically-oriented, definition that deserves explanation.

The industry standard is a series of charts mapping MTF, distortion and aberration data. The problem is that the industry standard isn’t adhered to. Each manufacturer presents data differently, which makes published information incompatible. Over time, there isn’t even consistency within the literature published by a given manufacturer. There are many independent monolithic, empirical surveys – such as those conducted by LensRentals, LensTip and PhotoZone, each of whom have different methodologies and priorities that deliver slightly different verdicts, and which have varied over time in a way that makes even internal comparison unviable. There are other useful resources such as Digital Picture where data is supplemented by visible results, addressing inevitable questions about tangibility: what does a 0.1 point increase in tangential resolution at 30lp/mm look like? To which we might add – on what imaging platform?

Most photographers take the view that only what works matters. Many are barely interested in the zoomed-in minutiae of resolution but care greatly about global rendition properties – the look. Some rather well known creatives don’t even care whether a lens is ‘sharp’ – whatever that means. Delta offers the definition of sharpness as a combination of resolution, contrast and freedom from aberrations at a given point in the image circle: the central 6mm diameter (Zone A), the band of 6-13mm diameter (Zone B, including APS-C corners) and the band of 13-18mm diameter (Zone C, including full frame corners).

A 35mm sensor with a pixel pitch of 5.5µ has been used as a baseline for comparison. However, the finally awarded ‘mark out of ten’ has in many cases been adjusted after re-testing using a Micro Four/Thirds sensor with a more revealing 3.3µ pixel pitch. The awarded mark is comparative to the unattainable ideal of a 10/10 lens that perfectly resolves our target sensor at all points in the image circle. Naturally, we didn’t have one of those for comparison, so all lenses were referenced to a number of benchmark designs. For instance, after long debate we plucked from the air a judgment that the central resolution of the Sigma 105/1.4 at f2.8-4 scores 9.7 – very high, but not un-improvable. Over time and through hundreds of tests, we gradually built a matrix of such reference points against which any lens could be compared.

The results offer a reliable picture of resolving capability that enable us to a) discuss the relative merits of enlarger, projector, repro and taking lenses on a level playing field, b) to group as uncontroversially as possible the spectrum of Delta-catalogued lenses into four categories, and c) arrive at the above ‘Hall of Fame’.

Anyone serious about testing, or even reading about, lenses knows that absolute definition is impossible. How many samples must be tested before being arriving at the best one? Should results from the best sample even be permitted – should sample variations not be averaged? How do you interpret or distill a complex set of data (for instance, in a zoom lens, continuously varying resolution across the frame at continuously varying working distances at continuously variable focal lengths) into a simple, yet accurate, take-away? Is chromatic aberration a more or less important factor than geometric aberration? Indeed, is lens resolution more fairly measured before or after geometry correction? How can conflicting results generated by different test methods be reconciled?

Therefore, we’ve not aimed a spuriously ‘accurate-sounding’ grading system: marks out of ten were awarded for satisfactory-looking sharpness in two Zones, averaged and itemised. Wherever possible, several samples were tested several times. Wide open performance is treated as its own metric, despite different apertures being compared, to demonstrate how projector and enlarger lenses compare. The f4-8 performance metric stretches an enlarger lens into territory dominated by taking lenses, whereas the f5.6-8 ‘typical working aperture’ average shows what they were designed to do best, especially at near distance.

As the largest survey of its type so far undertaken, we’re confident that it gives a fair perspective: no lens with a Gold award will disappoint and every lens with a Silver rating will be useable as ranked: sharper than any Bronze-awarded lens; not as good as the best.

General Takeaways

  1. None of the lenses in Delta match the performance of a reference prime at long distances and/or wide apertures. Little lenses suit little apertures.
  2. Most late five and six-element enlarger lenses score high Silver (85-89%) awards for f5.6-8 average sharpness at close range, and middling Silver awards (82-86%) for long-distance. For comparison, the Panasonic 20-50mm f4.5-5.6 scores 82.3% for distance and 79% close-up. The reference Sigma primes score 93-96% in all distance categories, and drop just below 90% at their minimum working distances. Therefore: at long distances, most good enlarger lenses are better than kit zooms, but not as good as reference primes; at short distances, they are typically as good, or better, than reference non-macro primes.
  3. Generally, shorter enlarger lenses are optimised for higher magnification (= higher resolution). A given manufacturer’s 50mm lens is usually sharper (in Zone 1, at least) than its c.100mm model. But as you approach infinity focus, many 50mm enlarger lenses are close to the coverage limit of a 35mm sensor: a 43mm diameter image circle. Therefore corner performance suffers. Optical vignetting kicks in to some extent at even middle distances for many 50mm enlarger lenses, which is why we see Zone 3 performance degrade obviously in that instance relative to their close-range behaviour. Therefore: at > 1m working distances, shorter lenses resolve better centre frame but longer lenses render more consistently across the frame.
  4. Few cine projector lenses and fewer slide projector lenses render Zone 3 to a degree of professional decency – particular at close range (< 1m) and long distance (> 20m). However, in their mid-range sweet spot, and in Zone 1, the best of breed is capable of fully resolving modern sensors. Therefore: for best results, only use projector lenses for central framing where peripheral aberrations add to the effect of subject isolation.
  5. Just as the broadsword of the big prime cedes supremacy to the dagger of the enlarger lens at short range, so at 2:1 and higher magnifications, the enlarger lens gives way to the laser scalpel of the industrial lens. Heading into the domain of life-size and larger, scanner, microscope and lithographic lenses – increasingly fine-tuned to given repro ratios – are masters of their realm. However, although Delta catalogues these specialised creatures, we do not test them very often – although we have dragged one or two out of their comfort zone to play with our 50-80cm test target. Suffice it to say that plenty of good work has been done by others in this area.

We pay attention to drawing styles and creative choices elsewhere, but the Hall of Fame addresses sharpness and the pursuit of technical perfection for a broad range of ‘normal’ use-cases. Like me, you may have favourites not on the list, but these are the best enlarger, projector, industrial and repro lenses for analog or digital applications.

But which are the best?

Best at what? If by ‘best’ you mean rated Gold at f5.6-f8 at all working distances – ie, the alt-lenses that behave most like ordinary taking lenses – the following shortlist take top honours:
Fujinon EX 50/2.8
Meopta Meogon 80/4
Rodenstock Apo-Rodagon N 50/2.8, 80/4 and 105/4
Schneider Apo-Componon 60/4

If by ‘best’, you mean sharpest at f2.8 at near distance, joint honours go to the Rollei AV-Apogon 90/2.4 and Minolta CE 50/2.8 for centre-frame (Zone A) performance, but the Minolta has sharper corners, and neither resolve as well in Zone C as the EL-Nikkor 50/2.8.

If by ‘best’ you mean highest resolution at any aperture, the following shortlist are capable of scores of 9.2 or better at their peak (Zone A only):
EL-Nikkor 50/2.8, 63/2.8, 80/5.6
Fujinon EX 50/2.8, EX 75/4.5
Schneider Componon-S HM 50/2.8
Schneider Apo-Componon 60/4, 90/4.5
Rodenstock Apo-Rodagon N 45/2.8, 50/2.8, 80/4 and 105/4
Meopta Meogon-S 50/2.8, 80/2.8, 80/4
Agfa Color-Solagon DI 70/4.5
Leitz Focotar 40/2.8
Taylor Hobson Ental II 50/3.5

If by ‘best’ you mean ‘best corners at distance’, the shortlist is quite different to ‘best corners close up’, and so on. For full details of how a lens performs, each of the lenses tested has its own mini review. Please read for details.