I recently moved to a new home – half the size of the one I was leaving. In a big home, you have lots of storage, a basement, an attic, a large garage, rooms you never use, and you tend to keep stuff. Lots of old stuff. Like old computers.
But when you move, you can’t leave anything behind. You check all the hidden corners of your home, and you find your old iMac. Second generation, officially known as the iMac G4 700 MHz. And often nicknamed Sunflower, because of its very original shape.
I had bought it in 2002. My first Mac. My first Apple widget.
The machine did age very well. It had been solidly built, with good quality components. It boots, and makes all the right sounds and moves. You log-in, and find a desktop which is not really different from what you see on a modern Mac (this machine is running OS X 10.5 Leopard – the last OS X version to run over PowerPC processors). Your spirits are high. You’ve been reunited with an old friend.
On paper, the specs of an old iMac – a 700 MHz 32 bit processor, 512 MB RAM, WiFi, Firewire – are not totally ridiculous if you compare them with a recent entry level machine. A modern Macbook has a 64 bit dual core processor, generally running at 1.3 GHz, with the ability to burst at 3 GHz. It does not look like a huge difference. Admittedly, a modern laptop also has much more RAM (8 GB) and more storage (256 GB SSD as opposed to a 40 GB spinning drive). But in the real life, it results in a huge gap in speed.
The iMac is incredibly slow at performing tasks as simple as ripping a music CD. And the worst is not the speed, it’s the inability to do anything really useful with the machine because the hardware and OS X Leopard can’t cope with modern security protocols:
you can’t connect to a wireless LAN, because the iMac and OS X Leopard support at best WPA (original version) and not WPA2 which is the standard today.
you can’t browse any modern Internet site, because the browser does not support the recent encryption layers of https, and every site worth its salt defaults to https now.
you can’t access any of the on-line services requiring a fat client (the Apple Music store from iTunes, for instance) because nobody’s accepting connections from such an old thing (probably another TLS vs SSL issue)
Software from 2009 still works (things like Photoshop Elements 2 or Microsoft Word 2008) but obviously recent versions of popular software are written for Intel Macs, and don’t run over Power PC processors (not even software as basic as a browser).
Basically, it could work if you downgraded your home WiFi network to WPA (or even worse, WEP), and only browsed sites whose servers have not been updated since the end of support of OS X Leopard (in 2009, I believe). That’s a scary perspective.
So, what will happen with this Mac? I will donate it. To people who still have room in their large home, where it will join a burgeoning computer museum. Until they decide to downsize of course.
An anecdote – the iMac was the first Apple computer I bought, but not the first I used. In the early eighties, I even had the privilege of being trained on Apple SOS (nicknamed Applesauce, of course), the operating system of the Apple III. It makes me a true geek.
For the fans of all things Apple, and among all the bios of Steve Jobs, I recommend the excellent “Becoming Steve Jobs“, by Brent Schlender and Rick Tetzeli.
I don’t want this blog to turn into a Nikon fansite. But Nikon related pages are now the most read: the Nikon D700 and FE2 entries have been the two most visited pages lately, leapfrogging the pages related to the Angenieux 28-70 f/2.6 zoom, which had been the readers’ favorite for years. And I can’t hide that Nikon film cameras are those I prefer, and that I’ve put my money where my mouth was.
Interesting things are happening at Nikon’s. On August 23rd, they will unveil a new full frame mirrorless digital system, launch a new lens mount and at least one lens.
The new lens mount will be typical of modern mirrorless cameras (short flange distance, and, I assume, no mechanical interface at all – autofocus and aperture control being all electric ), but its diameter will be unusually large – much larger in any case that the Sony E lens mount.
Over its 59 years of commercial life (so far), the Nikon F mount has gone through many revisions to support successively aperture indexing, automatic aperture indexing, matrix metering, auto-focus, silent wave auto-focus motors, and more recently, electronic diaphragm control.
Because Nikon has made a core business principle to guarantee at least a modicum of compatibility between its older lenses and its newer generation of bodies (particularly for high-end cameras sold to professionals), the new full frame mirrorless body will accept Nikon F lenses, via an adapter. But Nikon has not shared any detail about this adapter yet.
The adapter could be made simple, with no electrical contact and no mechanical linkage to the lens. Generally speaking, mirrorless cameras are not dependent on the automatic aperture pre-selection capabilities of the lens, so it’s likely that any Nikon F lens old enough to have an aperture ring will not only physically mount on the adapter, but will somehow work when the camera is set to semi-automatic exposure and manual focus mode. But recent lenses deprived of an aperture ring (or with an electronic control of the aperture) would not work with such a simple adapter. Which would go against Nikon’s tradition of preserving compatibility in priority for recent and/or expensive pieces of equipment.
The adapter could be made very complex. Sony supports Minolta/Konica-Minolta/Sony A mount lenses on its E Mount mirrorless bodies thanks to two models of adapters. The most complex of the two, the LA-E4, has its own autofocus motor in order to provide support and adequate AF performance for screw-drive autofocus lenses (which still constitute the majority of the Series A lenses offered by Sony today). Sony’s adapter also has a Phase Detection AF module, probably because its A series lenses were not designed for the contrast detection auto-focus system of its NEX mirrorless bodies.Nikon’s original AF and AF-D lenses (the screw drive lenses without an auto focus motor) could be supported using a similar setup if Nikon really wanted to, but I doubt they’ll have any appetite for such a solution (one of the reasons being that professionals have been buying AF-S lenses with a built-in auto focus motor for almost 20 years now – and probably don’t use many screw-drive auto-focus lenses anymore).
Nikon’s now defunct One series (J1 to J5 viewfinder-less cameras and V1 to V3 SLR like models) could accept F mount lenses thanks to an adapter. With the FT1 adapter, auto-focus lenses with a built-in auto-focus motor (AF-S lenses, with or without an aperture ring) are fully supported (all auto-exposure modes, vibration reduction and auto-focus, of course).
Older auto-focus lenses (the AF and AF-D lenses) can be used in all the auto-exposure modes but don’t auto-focus. Lastly, AI and AI-S manual focus lenses will only be usable in Manual or Aperture Priority Auto Exposure modes.
Nikon FT1 adapter (Nikon F to Nikon One lens mount adapter) – the adapter is seen from the front (where the F lens will be mounted). Source: Adorama
My bet is that the new adapter will offer the same functions as the FT1. It will fully support any lens introduced in the market since the last years of the XXth century (AF-S, AF-S G, VR, AF-P), and with reduced capabilities, most of the older lenses.
Will there be a penalty in terms of auto-focus performance for users of AF-S lenses ?
That’s the real question.
First answers on Aug. 23rd…
Update: Aug 24th, 2018
So.. Nikon had a busy day yesterday: they launched a new Z series of bodies with 2 cameras, the Z6 and the Z7, 3 lenses of a new S series, and a F to Z adapter.
In the picture above, the Nikon F to Z adapter looks very similar to the FT1 adapter – no built-in auto-focus motor, no auto-indexing mechanism to support full aperture metering for AI or AI-S lenses, but “a mechanical actuator to operate the aperture on the lens you mount to it” (according to DPREVIEW)
As I expected last week, lenses released during the last 18 years (AF-S, AF-S G, VR, AF-P) are fully supported, and the VR lenses gain 5 axis image stabilization in the process.
Older lenses (AF, AF-D) will not auto-focus but will still access all the auto-exposure modes of the Z bodies. Older manual focus lenses will mount but will have more limited exposure control capabilities.
For more (and in particular an opinion about the auto-focus performance), you can check DPREVIEW’s very interesting first take on the F to Z adapter.
The city, located in the desert between the Red Sea and the Dead Sea, was a very prosperous trade hub in the Antiquity, but lost of its importance in the Middle Ages to the point it was uninhabited and totally forgotten until its ruins were rediscovered by an archeologist in the early 1800s. The city had been built at the far end of a narrow canyon, and is famous because of its temples directly carved in the walls of rock forming the canyon. It was recently used as the lair of the bad guys in Indiana Jones’ “Last Crusade”.
Getting there is not exactly easy – a long flight to Amman or Eilat, followed by a long bus drive through the Wadi Rum desert (of Lawrence of Arabia fame), followed by a walk under an excruciating heat. And at the top of that, the access fees are exorbitant. But the place is absolutely unique, and the end of the approach in the narrow canyon is really magic.
Petra is one of those places that always look good in pictures – and I absolutely wanted to bring back images I would be pleased with. So why did I bring to Jordan a 40 year old camera I had never used before?
Normally, before an important trip, you’re supposed to test the camera in advance: you change the batteries, you expose a roll of film, you have it processed, and you look at the pictures it produced very carefully, before you finally declare the camera fit for service.
That’s the process I followed, with a Fujica AX-3 I had earmarked for this trip. But it did not pass the test. When I downloaded the scanned images, only a few days before I was due to the airport, 30% of the images were severely under-exposed and I could not see a pattern (it looked like random). I had just moved to a new home, my trusted cameras were still in a storage facility and too difficult to access, so I decided to trust Nikon, and brought with me a Nikon N2020 (aka F501) I had just bought for a miser a few days before, and only briefly examined.
To be honest, it was not that big of a risk. I had bought the Nikon from a second hand camera dealer of good reputation. I live in the 21st century and I have a good digital camera, and can use an iPhone as a backup. I decided that on this trip, on a given day, I would shoot digital for a few hours, then switch to the film camera. This way, even if the Nikon severely malfunctioned, I was not going to come back with no image at all.
At the top of that, Nikon cameras from that era are reliable. From all the cameras I have used over the years, Nikons are the only ones that have never let me down:
Fujica and Pentax cameras from the seventies have all sorts of mechanical problems (with the shutter, in particular). Cameras from the early eighties also suffer from relatively troublesome electronics (capacitors, stabilization circuits).
I owned Minolta Maxxum and Vectis cameras and Minolta AF lenses in the nineties, and they were not trouble free when they were in their prime (the only lens that ever broke in my photo equipment bag was a Minolta Vectis zoom). I have no recent experience with those cameras, but time generally makes reliability worse, not better.
The Olympus OM cameras I’ve used have been solid and reliable, but some models (the OM-2 Spot Program in particular) tend to go through their batteries with an alarming voracity, which could be an issue on a long trip.
Canon A series tend to develop a well documented shutter problem over time. I can’t use my Canon A-1 until I have it fixed.
My Canon T90 has been flawless (and a pleasure to use), but the model has a reputation for being a ticking bomb (from a reliability point of view) because of issues with the magnets used to control the aperture, and because of capacitors and batteries soldered to the camera’s integrated circuits.
On the other hand, even Nikon cameras I bought in bulk in antique shows or from thrift stores have been easy to bring back in service – generally the only thing missing was a good battery. They have a very reliable shutter and an accurate meter, and no light leak issue. Some Nikon cameras develop some annoying issues (the rubber grip on modern Nikon digital cameras, the LCD display in the viewfinder of the F4), but nothing that would prevent you from taking good pictures.
As a conclusion
I received the scans a few days ago. The exposure was a bit off (over-exposed by 1/2 stop in average – it’s likely that the camera had not been calibrated by Nikon for such a luminous landscape), but nothing that could not be adjusted in Adobe Lightroom in a couple of seconds. There’s still life in those old cameras.
The N2020 (F-501 outside of the US) was Nikon’s first mass market auto-focus SLR. It was an upgrade of the N2000 (F-301 “in the rest of the world”), Nikon’s first SLR with an integrated motor.
On this trip, I used it as a manual focus camera, with a very compact Series E 35mm f/2.5 lens. The ergonomics is still very conventional (dials and rings instead of menus and LCDs), it simply needs four AAA batteries that you can find anywhere in the world, and it’s a pleasure to use.
More about the Nikon F501/N2020 in a few weeks.
I initially wrote that the building shown in my photos and drawn by Herge in “Coke en Stock” was the “Monastery”. It was wrong. In fact, it’s known as the “Treasury”. And it was neither a monastery or a treasury, but the mausoleum of King Aretas IV, who ruled the region in the 1st century AD.
And I watched “Indiana Jones and the Last Crusade” again – in the movie, the “Treasury” is not the liar of the bad guys, it’s the place where the Holy Grail and its guardian reside.
The F4 was the last Nikon camera to enjoy (for a few years) an undisputed supremacy on the professional market. Successor of a long line of modular pro cameras, the F4 uniquely combines conventional ergonomics with a modern feature set.
A typical Nikon pro-camera
Now that we have discussed in detail the F4’s auto-focus system, let’s spend a few cycles on what makes it such a great camera.
The F4 is a true Nikon F – a modular body with an extremely high built quality, compatible to some degree with almost anything Nikon has ever made. With 24 different focusing screens, 4 models of viewfinders, 3 grips and 3 backs to choose from, it can be customized to almost any specific requirement.
It has been designed so that it will never let the photographer (or an unexperienced assistant) waste a roll of film or botch a shooting session:
there is no menu, just knobs and switches – just look at the camera, and you visualize how it’s set up.
there are locks on the film door, on the shutter speed knob, on the ISO dial, on the exposure mode switch, on the on/off switch, and a flashing red LED will warn the photographer if the film is not properly loaded in the camera.
It oozes build quality: it’s a full metal precision instrument.
Another Rosetta Stone
Nikon has been using the same bayonet lens mount since 1959, but even if its basic dimensions have stayed the same, it has evolved to the point where old lenses could damage some of the recent bodies, and recent lenses can’t be reasonably used on most of the older cameras.
In the Nikon line-up, the “pro” or “advanced-enthusiast” cameras are compatible with more generations of the Nikon F bayonet than the amateur oriented models, and a few bodies are exceptionally good at bridging the generations. The F4 is one of those, and deserves its “Rosetta Stone” nickname.
High level, the F4 can work with almost any manual focus lens (pre-AI made from 1959 to 1976, adapted AI, AI and AI-S), thanks to an array of pins inherited from the Nikon FA (like the FA, it offers matrix metering with manual focus lenses). Old ultra-wide angle lenses protruding deep in the mirror chamber can also be accommodated: the mirror can be locked in the raised position. And it has the electrical contacts needed by more recent AF lenses.
Of course, not all the exposure modes are available (pre-AI lenses have to be used stopped down, and only the Aperture Priority and Manual mode work with AI and AI-S lenses – but with the benefit of matrix metering and focus assist).
With Nikon’s AF and AF-D lenses, everything works. AF-S (motorized) lenses work also, as long as they still have an aperture ring. The AF-S G lenses (deprived of an aperture ring) can only be used in Shutter Priority and Program Mode.
Only very recent lenses (the AF-P with an electronic control of the lens’ iris) can’t be used at all.
The F4 is not known for any major issue (Nikon was wise enough to abstain from using magnets as a substitute to springs in the shutter mechanism or soldering capacitors or batteries to the mother board).
In the eighties, LCD displays were a new thing, and vendors were often adding warnings or disclaimers in their user manuals, because they did not know how LCDs would age. In fact, they seem to have aged pretty well with most cameras, with the exception of the Nikon F4.
The camera has two different LCDs: a small one at the top left of the viewfinder, showing the number of remaining images on the film roll, and a large display at the bottom of the viewfinder, showing to usual exposure parameters (metering mode, shutter speed, aperture …).
The top LCD is in fact part of the body of the camera itself (a small prism redirects its image so that it’s visible at the top of the viewfinder’s focusing screen). On the camera I purchased, it looks like it was covered by a dark green spider web. It’s distracting, but not a big issue, as the information displayed is not essential (I’m considering covering the LCD with a piece of black tape).
The issue of the “leaking” main LCD is more serious. In the most acute cases, the LCD display at the bottom of the viewfinder is all black (as if it was covered with black ink), and totally useless. This LCD is part of the viewfinder. Replacement DP-20 viewfinders are easy to find, in the $50.00 to $70.00 range on eBay, but even copies deemed “perfect” show small signs of leakage at the edge of the LCD. I don’t know whether they’re now as bad as they will ever be, or if they will further degrade over time.
How does it compare?
The F4 is unique – no other camera offered such a combination of modern features (matrix metering, multiple automatic exposure modes, auto-focus) in a somehow conventional modular body.
In the Nikon family, the F3 is a much simpler camera. Its High eye Point (HP) viewfinder is better suited to manual focus operations (longer eye point and higher magnification) – it is easier on the eyes, but gives very little information (the shutter speed selected by the camera in Auto exposure mode, augmented by a + or a – in semi-auto mode). If you only intent to operate in manual focus mode, don’t need spot or matrix metering and don’t mind the weight (the F3 is a surprisingly compact camera, but it’s very dense), the Nikon F3 is very good pick.
The N90s (F90x) is not a modular camera, it is not as solidly built and is more plasticky (with a peeling film door). It is designed around a modal interface (press a button, turn the control wheel to change shooting parameters). In terms of everyday performance, it’s more capable than the F4. Its viewfinder shows a larger image (0.78x), its auto-focus is much faster, and although large and heavy in its own right, it’s not as voluminous as the F4. But because it does everything so well (in the background) and only communicates with the photographer through LCD screens, it feels more robotic than the F4.
When the F4 was launched in 1988, Canon was starting its transition to the new EOS camera generation. They still had two manual focus professional models – the modular and very conventional New-F1 of 1981, and the T90, with modern ergonomics but without matrix metering. Canon would not offer a real pro auto-focus camera until the EOS-1 of 1989 and they had no trans-standard constant aperture USM zooms until 1993, leaving a few years of reprieve to Nikon. The EOS-1 had a fixed viewfinder, but could be fitted with battery grips of various capacity. Nikon’s F4 successor, the F5, followed Canon’s example: modern ergonomics, and less modularity than its predecessor (the battery grip was integrated to the body and could not be changed). In that sense, the F4 is the last really conventional and modular 35mm SLR.
The F4 is not a rare camera. It may have struggled at the end of its sales career, but during its first three years, it sold in large quantities (270,000 copies sold between 1988 and 1990).
Fully functional cameras with some issues (cosmetic or leaky LCD display in the viewfinder) sell for a bit more than $100.00, with faultless copies fetching twice as much.
To a large extend, the value of the F4 is related to the accessories it’s coming with – non standard grips, viewfinders and focusing screens were not sold in large quantities (and sometimes only in select geographies), and sold separately, they can be worth more than the camera itself.
As a conclusion
The F4’s mission #1 was to never let a photographer miss a picture inadvertently – all settings are visible on clearly identified knobs and dials, there are locks everywhere, and even a warning LED if the film is not properly attached to receiving spool – it’s the play-it-safe tool for a pro who does not want the embarrassment (and the potential loss of revenue) of failing to deliver.
Technically, if offers what was the best in 1988 and still is pretty good now (large viewfinder, 1/8000 shutter, OTF Flash with 1/250 sync, matrix and spot metering). Its auto-focus system, while not as capable as a modern camera’s, is still good enough for any scene where you have the time to focus first and then re-frame.
Like the F3, it’s a camera that grows on you. I was not crazy about it when I started using it (its size, its weight and a narrow auto-focus zone were a serious let down), but I rapidly came to appreciate it. It does not feel like a computer or a robot taking all the decisions for you (like a N90S would), it’s still an analog camera with conventional commands, with an auto-focus system that forces you to pay attention to what you’re doing.
It will give its best results if you have the deliberate approach that makes shooting with conventional film cameras so rewarding.
A simple semi-auto 35mm SLR, the Fuji STX-2 is a typical learner’s camera. It is better at this exercise than the over simplified and somehow antique Pentax K1000, although it does not benefit from the huge supply of Pentax K compatible lenses on the second hand market: it uses Fujica’s proprietary X-mount bayonet.
Over the course of the last 18 months, I’ve purchased and tested half a dozen Fujica and Fuji SLRs from the seventies and early eighties.
Some I really liked – the ST801 (1972) is one of the very best m42 (universal screw mount) semi-auto cameras ever built. The AX-3 of 1979 is a very competent aperture-priority camera designed for enthusiast amateurs. Some I did not particularly like – the AZ-1 of 1978, an automatic camera deprived of a semi-auto override and of the digital numeric display of the ST901- or the AX-5, too close to its entry-level siblings in spite of its impressive specs sheet. The worst of the Fuji 35mm SLRs was also the last one, the AX-Multi Program of 1985, with very limited capabilities and a questionable build quality.
When Fuji launched their X-mount bayonet mount in 1979, they presented three new models all based on the same new chassis (the program mode only AX-1, the AX-3 and the top of the line AX-5), as well as an entry level semi-auto camera, the STX-1. The first two letters of its name were telling the whole story: it was a close derivative of the ST series, where the m42 lens mount had simply been replaced with the new Fujica X-mount bayonet. In 1982, its matching needle meter was replaced with 3 LEDs, and it became the STX-1n. In 1985, in parallel to the AX Multi, Fuji also launched an updated version of the semi-auto STX-1n, the STX-2. High level, it is a plastic bodied version of the STX-1n, with a shutter upgraded to go up to 1/1000 sec.
Contrarily to the AX Multi, the STX-2 is not an over-simplified camera. It’s a true semi-auto camera, with a depth of field preview, a split image telemeter, a big needle showing the pre selected shutter speed in the viewfinder, and a mechanical timer for…selfies, all features that the prototypical learners’ camera, the Pentax K1000, is missing. Its metering system is based on a silicon cell, which controls a set of 3 LEDs at the right of the viewfinder. TTL (Through The Lens) metering not only works with the Fujica’s X-Fujinon lenses, but also, thanks to an adapter, with almost any m42 screw mount lens (stopped down).
The average weighted metering system seems reasonably accurate, and only requires two very common LR-44 batteries (alcaline) or SR-44 (silver oxide).
It can not be equipped with a winder, does not show the pre-selected aperture in the viewfinder, but on an entry level camera it’s not a big issue.
The biggest disappointment is the viewfinder. It’s narrow, dark, and lacks contrast. It is significantly worse than the viewfinder of the AX-3 and AX-5 cameras (which is somehow OK without being great), and horribly worse than the viewfinder of their common ancestor the ST801, which is at the same time wider, brighter and more precise.
The other issue, of course, is the scarcity of X-Fujinon lenses on the market place. The Fujica STX and AX cameras came generally bundled with one of the multiple variants of a 50mm f/1.9 standard lens, which are still abundant today, but very few photographers bothered to buy anything else. Those who did mainly purchased the 43-75mm zoom or the 135mm tele, with the 28mm wide-angle lens finding a few takers. Other lenses (while nominally on Fuji Photo Film’s large catalog of lenses) were probably never stocked by retailers, and are nowhere to be seen today. And the situation is not really better with the independent optical companies: with the exception of Tamron (which had an Adaptall 2 ring for the Fujica X mount), none of the big brands seem to have made lenses for Fuji’s bayonet.
The STX-2 does not look as bad as the AX-Multi, but only by a small margin. The fit is correct (no gap between parts, no loose part) but the finish disappoints: black plastics body is dull and easily scratched, and the dials and knobs leave an unmistakable feeling of cheapness.
As a conclusion
Because it’s a simple mechanical camera closely derived from a long line of m42 screw mount semi-auto SLRs (from the ST701 to the ST605), the STX-2 does not seem to have suffered from the reliability issues of the AX series. It has not been spec’d down to give a false impression of simplicity like the AX-1 or the Multi-AX, and on paper, it has everything a beginner eager to learn the basics of photography will need. Because it’s mainly built out of plastics, it’s also a very light camera (510g), and you will forget it’s in your backpack when you’re hiking. With an adapter, it can use almost any m42 screw mount lens (Fujinon or other) in addition to the difficult to find X-Fujinon lenses.
That being said, it’s also an ugly camera with a bad viewfinder. If I had a large collection of m42 screw mount lenses, I would rather use them with a nice Fujica ST801. The 801’s shutter is faster (1/2000 sec, which is sometimes useful now that 400 ISO is the new normal film sensitivity) and its viewfinder much more comfortable to use.
If I had a few really good X-Fujinon lenses (there is a 50mm F/1.2 EBC I’d like to find one day), mounting them on an Fujica AX-3 would make more sense to me: it’s a fully featured aperture priority camera, with a good semi-auto mode, a decent viewfinder and a nice finish. I would simply bring along a STX-2 as a backup, in case the electronics of the AX-3 decides to go on strike.
If you look at the prices on eBay, the STX-2 is almost the most expensive of the Fujica and Fuji bayonet SLRs: only the rare top of the line AX-5 will cost you more. I can understand why it could be worth more than other Fujica and Fuji AX cameras: it’s the best spec’d of the mechanical semi-auto cameras, and it’s simpler and probably more reliable than its siblings of the AX series. But if I was on the market for a learner’s camera, I would also consider cameras such as the Nikon FM or the Olympus OM-2000: they may be slightly more expensive, but the FM is more solidly built, the OM-2000 has a faster shutter, and both have a better viewfinder and a much wider choice of lenses on the second-hand market. Not to mention that they also look much nicer.
In my opinion, unless you’re a passionate collector of all things Fuji, buying a STX-2 only makes sense if you can get it for real cheap (a few US dollars), or with one or two lenses to sweeten the deal.
the ST, STX and AX 35mm SLR cameras were manufactured in Japan by Fuji Photo Film, and sold under the Fujica brand until 1985, when they were simply rebranded as Fuji. The company is still in business today, and operates under the Fujifilm name. The lenses were sold as Fujinon (m42 mount), X-Fujinon and X-Fujinar (X- mount). Fuji also had lenses branded as X-Kominar in their catalog (from Komine, an independent optics company also working for Vivitar).
Fuji film’s current bayonet is also named the X-mount, their current lenses are named Fujinon-XC or XF, but there is absolutely no compatibility between the old and the new generation.
Fuji was (and still is very) proud of its lens treatment process, and the lenses that benefit from it are recognized as Fujinon EBC lenses.
Fuji also sold its camera and lenses in Germany and Central Europe through the Porst retail network. I don’t think that the STX-2 was every sold by Porst, but the STX-1 was sold as the Porst CR-1, and Fuji’s lenses as Porst UMC X-F lenses.
Today, Fujifilm is known for its digital cameras (APS-C and medium format mirrorless systems), and still makes serious money in the film business, with its instant film and cameras (Instax). The majority of the revenues of the company come from their document printing equipment business, and to a lesser extent, from their activities in the chemical and cosmetic industries.
The F4 is at the same time the last Nikon pro-camera with conventional/analog controls, and the first offering most of the functions that modern bodies have made us familiar with (auto-focus, matrix metering, PASM multi-automatism, motorized film advance and rewind). It was also the last Nikon camera to enjoy (for a few years) an undisputed supremacy on the professional market.
In 1988, auto-focus was still in its infancy. Minolta had launched the Maxxum 7000 in 1985, and Nikon had followed with their first auto-focus SLR, the F501 (N2002 in the US) one year later. The F4 was the first implementation of auto-focus on a professional SLR.
Compared to what was available in 1988, the F4’s auto-focus was not bad at all, but the arrival of the Canon EOS-1 one year later made it look slow and primitive, and today, the performance of its auto-focus system is what makes photographers think twice before using or buying an old F4.
The F4 is built around Nikon’s second generation auto-focus module, the “Advanced AM200”. Its single horizontal AF zone is composed of 200 CCD sensors, and the focus zone is very small (probably 3mm wide and 1mm high – if the engraving on the viewfinder’s focusing screen is any indication). The module is shared with other Nikon auto-focus SLR bodies of the same period (F801/N8008, F601/N6006 for instance) but the F4’s auto-focus motor is stronger and the use of a 8 bit microprocessor makes the whole setup faster than on lesser cameras.
On the viewfinder’s focusing screen, at the place where the split image rangefinder would be on a manual focus camera, the AF zone is signaled by two small brackets engraved on the glass of the focusing screen. The markers of AF area are very difficult to see if the subject is poorly lit, and it can be difficult to visualize exactly where the camera is focusing (on modern cameras the active auto-focus zone is often surrounded by red LEDs. Nothing of that sort here).
So, is the auto-focus really that bad?
The F4 was famously launched at the Seoul Summer Olympic games. Today, nobody would use a F4 to shoot sporting events with long tele-zooms, but in all fairness, with lenses of shorter focal distance, the auto-focus is reactive and accurate. Its biggest weakness in my opinion is not its slowness, or that it tends to hunt if the scene is poorly lit. It’s that the AF zone is so narrow that you really have to pay attention to what part of the scene the AF sensor is pointed at. You can not simply point the camera loosely towards the subject, and expect the auto-focus system to locate it somewhere in the central area of the frame. You have to aim precisely at it, right at the center of the focusing screen, let the system focus on it, then keep the shutter release button half pressed (to lock the focus) and re-frame your picture. In that sense, it’s more a focus-assist system (Nikon write about an “electronic rangefinder” in their user manual) than an auto-focus system as we know it now.
The N90S (F90X outside of the US) that came a few years later (1994) benefits from an auto-focus system of a newer generation: it still has a single zone auto-focus module, but it’s much wider (7 x 3mm), and is shaped like a cross (it can focus on subjects presenting vertical lines as well as horizontal lines), even in low light and with slow (f/5.6) lenses.
The user experience is very different. The camera easily finds the subject even if not perfectly centered in the frame, and it focuses rapidly and accurately, even on mobile subjects moving erratically. In that regard, the N90S is already a modern camera.
Using the F4 as a manual focus camera
A common opinion in the forums is that you should simply forget about the auto-focus system, and use manual focus lenses on the F4. Even if it’s technically possible – the F4 accepts almost any manual focus lens ever made by Nikon (with very few limitations) – I’m not sure it’s a great idea: the viewfinder of the camera has not been designed for that.
In the days of manual focus cameras, manufacturers had to make focusing as easy as possible for the eyes of the photographer, and as result, they used to design viewfinders with a high level of magnification (0.9X) and they spec’d focusing screens to be as precise as possible, even at the cost of looking at bit dark and coarse. As soon as auto-focus systems became the norm, manufacturers adopted viewfinders with much lower magnification factors (0.7X), in conjunction with much brighter and smoother focusing screens, which had a more limited ability to show small differences in focusing. [more about the evolution of the viewfinders of SLRs over time]
The F4’s default viewfinder (the DP-20) and its default focusing screen (the BriteView Type “B”) are typical of an auto-focus camera. The DP-20 is a long eyepoint / low magnification viewfinder (22mm and 0.70X respectively), and the “Type B” focusing screen is bright but not very precise. The Type B provides none of the focusing aids you find on a manual focus camera (no micro prism, no split image rangefinder). It’s difficult to set the focus with the naked eye when you can’t or don’t want to use the auto-focus system.
Optional focus screens optimized for manual focusing were available from Nikon and from third party vendors (Nikon’s Type “K” and Type “P” focusing screens have a ring of micro prisms around a split image rangefinder spot, the Type “J” just has a central micro prism focusing spot). Equipped with such a focusing screen, the F4 could be used as you would use a F3 or a FE2 – with the additional benefit of matrix metering. Those focusing screens are pretty scarce now, and are sometimes offered for more than $150.00 (they’re specific to the F4 – you won’t be able to focus accurately with a focusing screen designed for the F, F2 or F3).
On the F4 I bought, the Type B focusing screen had been replaced with a Type E. It’s a variant of the Type B with horizontal and vertical lines etched in the glass. If I mount a manual focus lens on the camera, I just use the auto-focus system as an electronic rangefinder – when the big green dot is on in the viewfinder, it’s focused. This “electronic rangefinder” is as good as a split image rangefinder – you just have to trust it.
More in a few weeks, with my take on compatibility, reliability and ….
Early Nikon AF cameras and their AF modules
On a camera, the overall performance of the auto-focus system (and the quality of user experience) depend on much more than just the AF module.
An auto-focus system is composed of hardware (the AF module with its CCD sensors, the microprocessors) and of software (the AF algorithm). It is part of an even larger system (the camera itself with its other subsystems and its lenses) with which it has to integrate smoothly.
As an example, an explanation of the AF performance improvements between the F90 (N90 in the USA) and the F90X (N90s in the USA), both equipped with the same CAM246 AF module (from a very interesting blog entry in 678vintagecamera.ca :”Improvements over the F90 included faster AF speed and tracking. Nikon claimed a 2x increase in CPU speed, a 25% faster AF lens drive speed over the F90, and a new AF algorithm.”Same AF module, dramatically improved performance thanks to hardware and software improvements. (source: https://www.678vintagecameras.ca/blog/forgotten-film-warrior-the-nikon-f90x-aka-n90s)
It is well known that Angenieux – before retreating to less cost sensitive markets like the movie industry and the military – made a final attempt to secure a place in the consumer market, with a now legendary Angenieux 28-70 F/2.6 AF zoom, launched in 1990. In 1994, the company was sold to its current owner, Thales, and left the consumer market for good. Tokina launched its own 28-70 f/2.6-2.8 soon after, and is widely rumored to have purchased the blue prints of the lens from Angenieux.
But in France, there are collectors who believe that Tokina had played a role in the design and the manufacturing of the Angenieux 28-70 lens. For them, Tokina did not have to buy the blue-prints, because the Angenieux lens was itself the result of a cooperation between Angenieux and the Japanese optics company.
After all, Angenieux had no prior experience with auto-focus lenses, and may have seen a cooperation with Tokina as a way to accelerate the product development and reduce the costs.
How similar are the various Tokina AF 28-70 lenses to the original Angenieux AF 28-70 F/2.6?
Just looking at the characteristics, the 28-70 f/2.8 lenses can be grouped in 3 generations:
1988 – 1994 – the 28-70 F/2.8 AT-X (the non-PRO model) predates the Angenieux by two years. The optical groups share a similar high level design: 16 elements organized in 12 groups. Apart from that, the Tokina and the Angenieux look very different: the Angenieux carries the brand’s very distinctive design language, and is beefed up in every dimension. It’s longer, wider and heavier. The AT-X requires 72mm filters. The Angenieux and all the other Tokina 28-70mm F/2.8 lenses need 77mm filters.
1994 – 1999 – the AT-X PRO 28-70 AF/2.6-2.8 (and PRO II) lenses are the ones whose main specifications are the closest to the Angenieux. The main difference between the Tokina and the Angenieux comes from the ability to disengage the auto-focus on the lens itself on the Tokina (by pulling the focusing ring): Angenieux never had such a feature.
On the Japanese Domestic Market, the Tokina were sold as F/2.8 lenses (no reference to F/2.6), but in the rest of the world they were marketed as f/2.6-2.8 zooms. [The Angenieux was sold as a F/2.6 constant aperture lens – in theory, a lens opening at f/2.6 lets 10% more light go through than a lens opening at f/2.8 – the difference is largely symbolic.]
The PRO II of 1997 was a significant upgrade over the first AT-X Pro, and benefited from one “High Refraction Low Dispersion” (HLD) optical element, from a better multi-layer coating, and from a faster focusing mechanism. The lettering on the lens’ body still reads AT-X PRO. The easiest way to recognize the PRO II: a bayonet hood mount that replaces the screw-on mount of the previous Angenieux and Tokina models.
2000-2002 – the 28-80 AT-X “PRO” of Year 2000, and the “Special Value” AT-X PRO SV 28-70 AF F/2.8 of 2002 are clearly 2 variants of the same model, but seem to have little in common with the earlier AT-X PRO models and with the Angenieux: significantly different dimensions, different minimum focusing distance, two aspheric optical elements, internal focusing.
Tokina AT-X 270
Tokina AT-X Pro
AF 28-70 F/2.6
AF 28-70 F/2.8
AF 28-70 F/2.6-2.8
Elements / groups
closest focusing dist.
Tokina AT-X Pro II
Tokina AT-X Pro 280
Tokina AT-X 287 PRO SV
AF 28-70 F/2.6-2.8
AF 28-80 IF f/2.8
AF 28-70 IF f/2.8
Elements / groups
Special glass elements
2 Aspheric, 1 SD
2 Aspheric, 1 SD
closest focusing dist.
Using one of those lenses today?
With the full frame digital cameras becoming more affordable, there has been a renewed interest in the Tokina 28-70 f/2.8 AF lens family in the recent years. They’re a far cheaper alternative to current luminous trans-standard zooms from the big camera makers. What do you lose if you use a lens from the nineties?
compatibility: the lenses were designed for 35mm film, and are only a good fit with full frame digital cameras (on cameras with an APS-C sensor, their angle of view is similar to a 43-105mm zoom on a 35mm camera). The Angenieux was available in Nikon AF and Minolta AF mounts (both of the screw driver AF type), and in a Canon EF variant, with an integrated auto-focus motor. In addition to the mounts of the big three, the Tokina models were also available for the Pentax KAF mount.
As far as I know, Canon and Minolta-Konica-Sony have never altered the bayonet mount of their auto-focus lenses, and Sony Alpha bodies still have the motor required to focus automatically with a “screw–drive” lens: any of the Tokina AF lenses should work on a Canon or Sony camera. The case of Nikon is more complex. All those lenses behave like Nikkor AF lenses of the first generation, which means they won’t auto-focus on Nikon bodies deprived of an auto-focus motor, such as the D3x00 and D5x00 series, as well as the new D7500 (there should be no issue with Nikon’s full frame cameras, as they still have an in-board auto-focus motor).
Performance of the 28-70 f/2.8 lenses compared to modern offerings
It’s difficult to assess – few of the tests conducted by paper magazines in the nineties are still available today (most of the magazines are gone, and the online archives of the survivors don’t often go that far back). Shutterbug is a good source of information: they’re one of the few surviving US photography magazines, and the articles they published in the nineties are still available on their Web site.
Tests were made with film cameras, in reference to equivalent zooms from Canon, Minolta and Nikon. A few tests published on the Web 10 years ago were conducted with digital cameras with a smaller APS-C sensor and a lower resolution, and are of little value with today’s high resolution full frame sensor cameras.
on the forums, as usual, there is a lot of hear say, wishful thinking and self re-inforcing opinions, but little in terms of facts or serious tests.
That being said,
there is not much information about the original 28-70 AT-X (non-PRO) of 1988 on the Web. It got a good review from Ken Rockwell in 2011 (obviously as a “used” lens option for cost conscious buyers). He liked its price, its relatively small size and weight, and its sharpness at the center, even at full aperture. In his opinion, you needed to stop down to f/8 to get to excellent levels in the edges or at 70mm.
the 28-70 AT-X PRO 2.6-2.8 (the so-called “Angenieux” design) generally got great reviews. Reviewers were impressed by its built quality and its sharpness (with a few restrictions): the review of Peter Burian on Shutterbug is very positive (it was originally published in 1999, and Peter was shooting with film). He rates the image quality as exceptional in the center of the frame at full aperture, and excellent even in the edges at F/4 and above, with a sweet spot in the 35 to 60mm range between F/5.6 and F/11 where it’s as good as a prime lens.
Maybe because he tested the lens more recently on a digital full frame camera (a Nikon D700), Eric Tastad in ERPhotoReview is not as enthusiastic. In his opinion, the lens is very sharp in the center at 28mm and full aperture, but needs to be stopped down to F/5.6 to reach excellent levels across the frame and above 50mm. And it will remain relatively weak at 70mm even when stopped down. To summarize, you could say that in his opinion the lens should have been sold as a 28-60 f/2.8-4.
the 28-80 AT-X Pro F/2.8 IF – there does not seem to be a definite opinion about this lens – according to some of the reviews, it’s inferior to its predecessors, while others (Peter Burian at Shutterbug in particular) say it’s the best of the bunch (which would be logical considering it’s more recent, and that it does not seem to have been designed with aggressive cost cutting in mind). The fact is that it had much serious competitors than its predecessors: lenses such as the more recent Nikon 28-70 F/2.8 AF-S are significantly better at full aperture, and focus faster and silently thanks to an integrated auto-focus motor.
the 28-70 Pro SV is a budget version of the 28-80, and is generally considered inferior in performance to the 28-70 AT-X Pro II.
The Angenieux is a collector’s item. Its value on the market has little to do with its usage value. It can not be found for less than $1,500 – much higher than more recent Canon, Minolta or Nikon 28-70 f/2.8 lenses.
The price of the Tokina lenses does not necessarily reflect the reputation (good or bad) of a specific 28-70 model – poorly regarded AT-X PRO SV lenses are often proposed for prices as high as the AT-X PRO II ($250 to $275 for perfect copies). The AT-X Pro 28-80 tends to be even more expensive (up to $400.00), but at this price it’s getting dangerously close to the cheapest lenses f/2.8 zooms of the Big Three (the Nikon 28-70 f/2.8 AF-S zoom can be found at $500.00).
More about the Tokina 28-70 AT-x PRO II f/2.6-2.8 in a few weeks…