[Sandy Hancock]

The DA* 18-85 is on the officially released Pentax lens roadmap. At least it looks like a lens that would match the 18-85 lenses from Canon and Nikon

The lens on the roadmap is a mid-range normal zoom, and as a DA* lens Pentax are probably aiming for it to be be fixed f2.8.

18mm and 85mm are the *extremes* of the possible focal range; that is why the bar starts to fade from about 24mm at the wide end and 75mm at the long end. Those latter numbers should be familiar to many as the exact full-frame equivalent to the current APS-C DA*16-50. The top-end Nikon and Canon zooms are essentially the same.

Expecting anything much beyond that is hopeful at best.

[dmytty]

The lens on the roadmap is a mid-range normal zoom, and as a DA* lens Pentax are probably aiming for it to be be fixed f2.8.

18mm and 85mm are the *extremes* of the possible focal range; that is why the bar starts to fade from about 24mm at the wide end and 75mm at the long end. Those latter numbers should be familiar to many as the exact full-frame equivalent to the current APS-C DA*16-50. The top-end Nikon and Canon zooms are essentially the same.

Expecting anything much beyond that is hopeful at best.

Yes, I am hopeful.

To be clear, this lens would be slightly larger and heavier than the 16-50 f/2.8. It might be around the same size as the Canon 24-70 - and even a touch larger in diameter if Full Frame coverage is desired (it is!).

Regarding size, the DA* 18-85 lens would be smaller than the combined 2 lenses (16-50, 50-135) currently required to cover the most useful focal range for still photography: wide angle (18mm) to portrait sweet spot (85mm). This focal range makes all kind of sense for both Full Frame and 1.5x Cropped Sensor bodies, at least for most shooting styles. It would also make possible a 3 lens 'day pack' consisting of the kits that I mentioned (ie UWA 10-12mm prime, DA* 18-85 f/2.8 zoom, and 55 f/1.4 portrait prime).

A secondary (or even primary) reason to do this is that the same focal range covers most video requirements - and specifically the practical limits for hand-held video. Pentax will have to address the video market sooner rather than later.

The DA* 18-85 f/2.8 combined with a DA* f/1.4 prime portfolio of 12mm, 30mm, 55mm, and 85mm would make for a great Indie video production rig or a stellar kit for almost all common photography scenarios. At ~$5k for 5 lenses and an APS-C body (with FF upgrade for +$2k) the value would absolutely smash anything Nikon or Canon would provide.

So maybe that's asking a lot, but I also have a feeling that Pentax engineers can deliver. Considering that only the UWA 10/12mm would have to be developed from scratch, it's not far fetched either. The 55 is already done, and all of the other prime lenses already exist in a non-weather sealed form. That leaves only a little hope required for Pentax to get the 18-85 done right.

[DSims]

Well Canon introduced the weather-sealed L series 24-70 f/2.8 in 2002 - and that's a very well regarded 3x zoom lens for full frame (at a retail launch price of about $2k). Computerized lens design programs have come a long way in the last 10 years as the computational power has increased exponentially. If the lens was restricted to cropped frame only, or even sacrificed full frame performance in the borders, the ~18-85 4.5x lens would be very doable at a $1k retail price point. In a kit of course, some margin is sacrificed. Pentax needs volume now before it becomes too marginal a brand.

Sorry, not even close on this one. It's a nice dream, however. At this point I'd suggest you need to become a lens engineer and prove it can be done. And then you can sell the world's first 18-85/2.8 DSLR lens for only $1000.

18mm and 85mm are the *extremes* of the possible focal range; that is why the bar starts to fade from about 24mm at the wide end and 75mm at the long end. Those latter numbers should be familiar to many as the exact full-frame equivalent to the current APS-C DA*16-50. The top-end Nikon and Canon zooms are essentially the same.

Expecting anything much beyond that is hopeful at best.

Right. 24-75/2.8 might be doable. Maybe even for close to $1000 if they really wanted to (e.g. like 3rd-party lens pricing), but Canon, Nikon and Sony sure don't! 20-80/2.8 would be absolutely amazing, and way more expensive than $1000 if it could even be done. 18-85/2.8 is simply dreaming.

[DSims]

At the risk of beating a dead horse, I think I found your lens for you!

Fujinon HK4.7X18-F -18.5-85mm T2.0 ZOOM LENS HK4.7X18-F B&H



Go for it if you want! You can actually mount this lens on some DSLRs, such as the 7D, and it's even faster than you asked for! You see, you're not really dreaming if you have the budget!

http://www.hurlbutvisuals.com/blog/2010/05/turning-still-cameras-into-pl-mount-movie-makers/


So now all you have to engineer is the K-mount adapter - the lens is already being made for you!

---

Last edited by DSims; 08-18-2012 at 02:54 AM.

[Sandy Hancock]

At the risk of beating a dead horse, I think I found your lens for you!

Fujinon HK4.7X18-F -18.5-85mm T2.0 ZOOM LENS HK4.7X18-F B&H

Go for it if you want! You can actually mount this lens on some DSLRs, such as the 7D, and it's even faster than you asked for! You see, you're not really dreaming if you have the budget!

$87,000
ROFLMAO

Should be able to do a f2.8 for about $10,000. Should sell like hot cakes, especially as a kit with the K-5 replacement

[DSims]

$87,000
ROFLMAO

Should be able to do a f2.8 for about $10,000. Should sell like hot cakes, especially as a kit with the K-5 replacement

OK, now you got me ROFL too!

[dmytty]

It might be easy to overlook, but that Fujinon specification is a T-stop, which is the film world's designation for Transmission Loss. F Stop is a simple geometry thing (focal length to aperture diameter) while T Stop takes into account the transmission loss of the lenses.

When converted from Tstop, the Fuji is more like an 18-85 F Stop 1.6!

Here's the math and supporting text:

1. If we want to figure out the corresponding F/Stop of the Fujinon T2.0, we can do a comparison with a similar focal length SLR lens. The Canon and Fuji 18-85 are both zooms, so for light transmission purposes we'll say they are valid for comparison.

2. Dxomark gives the T stop of the Cannon 18-85 f/2.8 as T Stop 3.4.

The relationship between FStop and TStop is simple:

SQRT(Light Transmission Factor) * TStop = FStop

Using the same equation, we can solve for Light Transmission:

Light Transmission Factor = (FStop/TStop)^2

In the case of the Canon f/2.8...

Light Transmission Factor = (2.8/3.4)^2 = .67

So about 67% of the light is transmitted.

3. If we figure that the Fuji might transmit a similar amount of light, we can calculate the F/Stop as follows:

SQRT(Light Transmission Factor) * TStop = FStop
...or...
SQRT(.67)*2 = 1.63

If you ask how the Fuji is faster, the answer is that it simply has larger light gathering area. The dimensions of the Fuji are 136 x 352mm (5.5" diameter x 14 inches).

The Fuji is what I call a 'big, expensive' lens for 'big expensive projects'. Then again it's almost 2 stops faster than the 2.8 I'm hoping for and probably has far better performance in regards to chromatic abberation, flaring, contrast, and resolution. In hyperbole terms: The Fuji can clearly resolve a gnat's wing flying into the sun - because that might be the effect that a Hollywood director is demanding. When it comes to performance, I don't think I am as demanding as a big Hollywood director like Peter Jackson (Hobbit) or James Cameron (Avatar) shooting 8k.

To reiterate, an 18-85 f/2.8 with dimensions of around 3.5 inches x 5 inches is not against the laws of physics. As the Fuji demonstrates, a 16 lb zoom can be currently produced with performance about 4x better than the 2+ lb lens I'm hoping for from Pentax.

Regarding cost:
1. Even if lens performance to cost was a linear relationship (it's not!) a 4x decrease in performance would turn the $80k Fuji into a $20k Pentax f/2.8. The reality is that it's exponentially more difficult to engineer and produce an f/1.6 vs an f/2.8, so even disregarding all other factors, the f/2.8 should cost <$10k.
2. The production cost of a lens is not in the raw material (ie sand and films are cheap raw materials) but rather the molds, polishing, film deposition, mechanical assembly and quality control. The bigger the volume, the more this can be automated and the larger the batch processing. Many lens manufacturing processes cost the same whether you're processing 1, 10 or 100 lenses.
3. In higher quantities, the sales, support, and distribution become much lower cost factors. The Fuji f/1.6 might sell <100 units per year. An SLR f/2.8 at $1k would sell in the tens of thousands. Economies of scale baby!

An 18-85 f/2.8 is a question of 'who?' and 'when?', not 'if'. Sure this lens would cannibalize many other zoom lens sales (and even some poor primes), but it's a bold step Pentax needs to make.

Your response?

---

Last edited by dmytty; 08-18-2012 at 12:06 PM.

[dmytty]

$87,000
ROFLMAO Should be able to do a f2.8 for about $10,000. Should sell like hot cakes, especially as a kit with the K-5 replacement

[Snicker] The $10k f/2.8 target was surpassed 4 years ago.

Red has been selling a Full Frame 35mm 18-85 T/2.9 (which translates to about 18-85 f/2.4) for $10k. The dimensions are:

• OD: 142mm
• Length: 11 in [28 cm]
• Weight: 9.9 lb [4.5 kg]

This lens is a bit dated in terms of design (circa 2008) but is also still 1/2 stop faster than the f/2.8 target. Red also has much lower sales numbers (100's of units sold per year) vs. a big brand like Pentax/Canon/Nikon.

Again, if we linearly extrapolate:
The Red 18-85 f/2.4 is 50% faster than an f/2.8, which with 2008 era design tools would allow an f/2.8 lens to be
$5k, 5 lbs, 100mm OD, 5.5 inches in length. I suggest a 10-20% improvement with new design tools, films, and manufacturing processes, which would yield an f/2.8 lens at 80-90mm OD, ~5 inches length, and 4 lbs. Switching from metal to plastics would probably yield a further reduction in weight.

Again, it's not a linear process so an f/2.8 lens could very well be smaller than the target I mentioned. However, to reach the $1k price point, it has to compromise on size and weight.

I should also mention that Pentax should have an mechanical assembly advantage due to the fact that they do not need to include Image Stabilization in the lens. This is no small thing and should shave ~200g (.5 lbs) off the weight, as well as reduce the lens diameter by 10-15% when compared to Canon/Nikon. Cost should also come down by 20% as well.

[Pål Jensen]

The Canon L series 24-70 f/2.8 full frame specs are as follows:

Max. Diameter x Length, Weight 3.3" x 4.9", 2.1 lbs. / 83.2mm x 123.5mm, 950g

That's a 10 year old lens design for $2000. I would say 'large and expensive', but not 'very'.

Designed and produced today, all of those numbers can come down some - even more so if it's an APS-C optimized design. However, I think they slightly improve on those dimensions and still yield f/2.8 in the 18-85mm range.

I said very wide lenses and in an APS context, and 10-12mm lenses were mentioned. 24mm is only moderately wide on APS and is only very wide on medium format!...
10-12mm F:2.8 lenses are going to be very large and expensive. The speed isn't really needed either on such a lens...

[dmytty]

I said very wide lenses and in an APS context, and 10-12mm lenses were mentioned. 24mm is only moderately wide on APS and is only very wide on medium format!...
10-12mm F:2.8 lenses are going to be very large and expensive. The speed isn't really needed either on such a lens...

Tokina makes an 11-16 f/2.8 which is $660 at B&H on Nikon (sorry no Pentax). The performance of the DX zoom lens (FF at 15-16mm) is stellar according to reviews such as Photozone.de And Ken Rockwell even recommends it above the Nikon brand lens (and that is really saying something). Sigma makes a 10-20 f/3.5 and an 8-16 f/4.5-5.6, both around $650.

[Edit]I've had a moving target here, so I'm going to dial it in a little bit with some Full Frame and APS-C figures.

A 14mm Full Frame prime is not out of the question at all - even at larger apertures. The lens diameters would be roughly as follows:

• f1/.4 = 125mm
• f/2 = 106mm

Due to the large diameter (and hence cost), the UWA prime would not make sense for a dual use role (ie same lens for Full Frame and APS-C). However, Pentax could adopt a 14mm Full Frame design to 9mm APS-C quite readily.

An APS-C 9mm prime might have the following diameters:

• f/1.4 = 110mm
• f/2 = 90mm

Respective to f/2 and f/1.4, cost would be a $1-1.5k target for the APS-C, and probably $1.5-$2.5k for the Full Frame. Optically, these lens would handily beat the zooms - and be Weather Resistant as well.

Speed is always needed...think freeze framing a forest during a heavy wind storm...or people moving through a cityscape at night. More so, a fast lens brings some creative aspects to ultra wide angle photography as there is some depth of field to play with (bokeh on UWA?).

If Sigma and Tokina can make awesome UWA zooms at 8-11mm and sell them profitably for ~$600, then a prime is even better and more profitable. Primes are far easier to engineer and produce.

[DSims]

[Snicker] The $10k f/2.8 target was surpassed 4 years ago.

Red has been selling a Full Frame 35mm 18-85 T/2.9 (which translates to about 18-85 f/2.4) for $10k. The dimensions are:

OD: 142mm
Length: 11 in [28 cm]
Weight: 9.9 lb [4.5 kg]

This lens is a bit dated in terms of design (circa 2008) but is also still 1/2 stop faster than the f/2.8 target. Red also has much lower sales numbers (100's of units sold per year) vs. a big brand like Pentax/Canon/Nikon.

Thanks for the laughs, dmytty. If you borrow the lens elements from plastic magnifying glasses I think you can reach your price target.by 2016. Put me first on the list to buy.

While you're wasting your time on this project, I'll be busy designing my 16-300mm f/1.8 lens, which I'll have on the market by the end of 2013 for $800. Today I'm going out and buying a new MacBook Pro Retina. It comes bundled with iWork software, so I should have all the design power I need.

[dmytty]

Thanks for the laughs, dmytty. If you borrow the lens elements from plastic magnifying glasses I think you can reach your price target.by 2016. Put me first on the list to buy.

Here's the timeline of the criticism:
1. Firstly, there was numerous mention that it's just not possible to even make an 18-85 f/2.8.
2. Then I was told that this hypothetical lens would be lucky to get to $10,000 - sometime in the future.
3. Now the criticism is that while a faster lens has been possible since 2008 (for less than the mockingly quoted $10,000), it won't be available for several years at a low enough price point.

As I said in my earlier comments, it's not a question of 'if', but rather 'who' and 'when'.

I'll take that as progress.

To recap the 18-85 lens speed progression:
Fujinon f/1.6 = $80k+
Red f/2.4 = $10k

That's a $70k price drop going from f/1.6 to f/2.4. Another half stop and we're down to $1k territory (btw, I actually figure this lens to be about $1200). Let's see who will be the first manufacturer to sacrifice a few zooms and mediocre primes to grab the crown jewel.

Today I'm going out and buying a new MacBook Pro Retina. It comes bundled with iWork software, so I should have all the design power I need

As far as the Macbook Pro Retina goes, not a bad choice. I prefer a notebook that I can use outside however (Asus NV56). Oh, and the Asus is Windows too, so it runs things like Oslo instead of Iwork.

[JonPB]

Dear Mr. and Mrs. Pentax,

This made me smile. Of course, if you're trying to convince a large company to do something, I doubt implying that they're a sole-proprietor type shop will get you very far. Still, amusing.

...it would seem relatively straightforward to put some seals on the FA Limiteds...

The problem is that Limited lenses are made of metal that expands and contracts enough between usable temperature ranges (well below freezing to hot days in Death Valley) to make adding weather sealing prohibitive.

Of course, I don't see any reason why adding WR-level sealing to the DA35 would be more difficult than the DA18-55. No need to mess with the Limiteds; simply use the existing FA designs. I'd be very happy to see a DA28/2.8 AL WR.

Regarding cost:
...
3. In higher quantities, the sales, support, and distribution become much lower cost factors. The Fuji f/1.6 might sell <100 units per year. An SLR f/2.8 at $1k would sell in the tens of thousands. Economies of scale baby!

While I would be very interested to see Pentax lead the market with an 18-85/2.8 lens, I think the biggest problem is manufacturing. As you are aware, optical aberrations increase exponentially when you either increase the aperture or widen the zoom range. Designing such a lens is, as the Fuji lens demonstrates, perfectly possible; however, the speed and wide range of that lens would require much tighter tolerances because each manufacturing defect is multiplied by the complexity of the lens. Having said that, Canon and Nikon also have to add image stabilizing optics to their lenses, which gives Pentax an advantage.

I think it is more reasonable to expect the wide-zoom DA* lens on the roadmap to be comparable to slower, wider range lenses offered by other companies. Of course, Pentax's 17-70/4 already compares with Canon's 24-105/4 and Nikon's 24-120/4. I think an 18-85/4 or faster DA* lens certainly would be interesting, but I will be surprised if it is f/2.8 unless Pentax is discontinuing the 16-50/2.8 (which isn't too unrealistic, given its poor reputation for SDM failures and the fact that competing off-brand lenses get better reviews at far lower prices).

Then again, one of the reasons we don't see a 24-120/2.8 for a full-frame sensor is that larger image circles also exponentially increase the difficulty of production. Pentax isn't competing in the 135-frame arena, so a 16-80/2.8 could be a flagship lens for an APS-C image circle that other brands don't have any reason to produce... and which would be as unique to Pentax as the Limited lenses. Releasing such a lens, though, might indicate that Pentax is doubling-down on APS-C rather than protecting a future 135-frame sensor camera.

Tokina makes an 11-16 f/2.8 which is $660 at B&H on Nikon (sorry no Pentax).

Tokina doesn't make any lenses for Pentax, probably because the two companies share optical designs for other lenses. My guess is that the "DA Wide Zoom" on the Pentax roadmap is their version of Tokina's 11-16/2.8.

[DSims]

Here's the timeline of the criticism:
1. Firstly, there was numerous mention that it's just not possible to even make an 18-85 f/2.8.
2. Then I was told that this hypothetical lens would be lucky to get to $10,000 - sometime in the future.
3. Now the criticism is that while a faster lens has been possible since 2008 (for less than the mockingly quoted $10,000), it won't be available for several years at a low enough price point.

As I said in my earlier comments, it's not a question of 'if', but rather 'who' and 'when'.

I'll take that as progress.

To recap the 18-85 lens speed progression:
Fujinon f/1.6 = $80k+
Red f/2.4 = $10k

That's a $70k price drop going from f/1.6 to f/2.4. Another half stop and we're down to $1k territory (btw, I actually figure this lens to be about $1200). Let's see who will be the first manufacturer to sacrifice a few zooms and mediocre primes to grab the crown jewel.


As far as the Macbook Pro Retina goes, not a bad choice. I prefer a notebook that I can use outside however (Asus NV56). Oh, and the Asus is Windows too, so it runs things like Oslo instead of Iwork.

Dang! I thought all I needed was a fast computer, iWork, and my crackpot design skills. Still, those idiots at RED, Nikon, Canon, Sony, Pentax, Tokina, Leica, Tamron, Sigma, Panasonic, Olympus, and Fuji have no idea what they're doing. You show 'em!

[dmytty]

This made me smile. Of course, if you're trying to convince a large company to do something, I doubt implying that they're a sole-proprietor type shop will get you very far. Still, amusing.

Just trying to inject some humor into my manifesto (i.e. 'lenses-that-you're-nuts-to-not-make').

The problem is that Limited lenses are made of metal that expands and contracts enough between usable temperature ranges (well below freezing to hot days in Death Valley) to make adding weather sealing prohibitive. No need to mess with the Limiteds; simply use the existing FA designs. I'd be very happy to see a DA28/2.8 AL WR.

I agree completely. Leave the Limited lenses for their Limited usage indoors. It's not too difficult to make a few batches every year, but I wouldn't promote them. In my view, WR should be a feature that is propagated across all new designs. Combined with the primes, and the 18-85 zoom (and maybe the resurrected 80-200 zoom), it's what would make Pentax different.

The Less Limited WR lenses will get all the glory from the Less Limited scenarios they can safely shoot in. I've even coined a slogan: Live Less Limited (copyright 2012).

Personally, I think 30 f/1.4 is where it's at...the resulting 45mm on APS-C is close enough to the 43mm 'natural perspective'; and 30mm is close enough to 35mm for full frame. As to f/1.4...well it's crazy nice on the Sigma 30 at $600, and I think Pentax could even do better at an $800 price point.

A 16-80/2.8 could be a flagship lens for an APS-C image circle that other brands don't have any reason to produce... and which would be as unique to Pentax as the Limited lenses. Releasing such a lens, though, might indicate that Pentax is doubling-down on APS-C rather than protecting a future 135-frame sensor camera.

This view of the chart makes sense: an APS-C at f/2.8 is very doable and far easier than full frame, and 16-85 is a more accurate reading of the depicted range. Hitting the <$1k price point would be a given.

However, the presence of the 18-85 lens type made me consider the idea that this could be a 'sweet spot' focal range for a dual-master lens. Perhaps I'm just trying to hold onto the full frame dream. Does anyone concur in thinking that 18-85 works well for both bodies?

Lenses are not magic. The barriers to a new lens usually fall to determined engineering - both in the optics and manufacturing areas (with mainstream pricing being emphasis on the latter). An 18-85 FF at f/2.8 is just waiting for the right combination of design + manufacturing to yield the right price point. Give or take a few mm, I think this happens sooner rather than later.

Tokina doesn't make any lenses for Pentax, probably because the two companies share optical designs for other lenses. My guess is that the "DA Wide Zoom" on the Pentax roadmap is their version of Tokina's 11-16/2.8.

Yep...agree.