14-45 OIS and 14-42 OIS, MTF and lens diagrams

Started Mar 8, 2010 | Discussions
Ehrik Veteran Member • Posts: 8,014
14-45 OIS and 14-42 OIS, MTF and lens diagrams

Blue represents contrast, green resolution. Image centre on the left, corner on the right. Dashed/solid line is tangential/radial direction. Wide open performance only. Higher MTF(%) is better.

It seems at 14mm , the new lens is sharper and more contrasty in the centre but the old lens is more consistent towards the outer parts of the frame.

At 42/45mm , the old lens is better over the whole frame , but the difference is small except perhaps for centre resolution (but I'd say more than 12Mp would be needed to fully exploit that).

Remember that flaws will be reduced when stopping down, we can't say how much from these diagrams.

Both have 12 elements in 9 groups with one aspherical element. Biggest difference (and cost saver) is the size of the front group. The design is seemingly very similar, except for the rearmost three elements.

14-45:

14-42:

Knight Palm Veteran Member • Posts: 3,723
Which lens element moves with the O.I.S.?

A single one, the sixth maybe?
The newer 14-42 lens is larger than the older 14-45.

It makes sense to optimize the manufacturing cost for this kitlens as its bundled with each cameras. If it also comes in lighter like here, that's a bonus.

-- hide signature --

Ludwig Wittgenstein; British philosopher born in Austria; a major influence on logic and logical positivism (1889-1951):

“What can be said at all can be said clearly; and whereof one cannot speak thereof one must be silent”

 Knight Palm's gear list:Knight Palm's gear list
Olympus XZ-1 Olympus E-3 Olympus E-300 Olympus E-400 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD +4 more
OP Ehrik Veteran Member • Posts: 8,014
Re: Which lens element moves with the O.I.S.?

14-45:

14-42:

Knight Palm wrote:

A single one, the sixth maybe?

I don't think it would work well to just shift one element. There would be nothing to compensate the aberrations. I've seen about 6-8 lens diagrams with the stabilisation unit denoted and as far as I can recall, there are at least two, sometimes three, elements moving for OIS.

We know with closeup lenses that the single-element ones, sold cheap in +1+2+4 sets, have problems with CA, whereas the more expensive ones are achromats with at least two elements made of different materials.

For OIS efficiency the design attempts to make the OIS group as light as possible which means it will be placed somewhere in the middle or rear. (The Nikkor 80-400VR is an exception, with the VR group near the front, but it's also the stabilised lens that's tested the worse, less than one stop.)

I think second or third group from the rear is quite common.

If I had to guess, I'd say in the 14-45 it's the group with the aspherical element or the group to the left of that one. In the 14-42 it's probably the group just to the right of the iris or possibly the rearmost group but that would be a first (to my knowledge).

The newer 14-42 lens is larger than the older 14-45.

It seems to trade front-group diameter for length. The diameter difference is bigger than the 45/42 ratio. Perhaps the entry pupil is located closer to the front on the new lens.

It makes sense to optimize the manufacturing cost for this kitlens as its bundled with each cameras.

Yes, the added length costs virtually nothing and the smaller front group saves a good deal. It may also speed up manufacture, letting them make more lenses per month using the same production line.

If it also comes in lighter like here, that's a bonus.

Yes. The added length and on balance somewhat less impressive MTF graphs may be less welcome, though. (But good enough for a kit lens.) Hopefully they will still continue to make the 14-45 but with limited capacity and many lenses waiting to be made that can't be taken for granted.

viztyger
viztyger Veteran Member • Posts: 3,280
Re: Optical Performance and Cost Cutting

Thanks Erik, that's a very useful analysis. It seems that on average optical performance has suffered a little at the expense of cost cutting. I don't have any kit zooms anymore as I tend to sell them with the body when I upgrade. I prefer switching to the 20/1.7 or the 45/2.8 instead. The extra lens brightness makes a huge difference, although switching lenses can sometimes cause you to miss a shot. Maybe one day Panasonic will introduce an f/2.8 zoom, then I'll be tempted.
--
Björn

http://www.bmupix.com

 viztyger's gear list:viztyger's gear list
Sony RX1 Sony a7R II Voigtlander 20mm F3.5 Color Skopar SL II Sony FE 35mm F2.8 Sony FE 55mm F1.8 +3 more
OP Ehrik Veteran Member • Posts: 8,014
Re: Optical Performance and Cost Cutting

viztyger wrote:

Thanks Erik, that's a very useful analysis. It seems that on average optical performance has suffered a little at the expense of cost cutting.

Yes, at the two points (full wide and tele, at infinity wide open) that we get to know about at this stage. Boke, flare, AF speed, etc. is unknown and may be improved.

As we remember from the comparison between the two 7-14 lenses, even measurably lower MTF in the corners may still produce extremely rewarding photography.

It seems, going by info Dcuk got from a Panny rep, that the 14-45 will continue to be sold with the G2 ; only the G10 will get the new lens. Which I consider good news.

I don't have any kit zooms anymore as I tend to sell them with the body when I upgrade. I prefer switching to the 20/1.7 or the 45/2.8 instead. The extra lens brightness makes a huge difference, although switching lenses can sometimes cause you to miss a shot. Maybe one day Panasonic will introduce an f/2.8 zoom, then I'll be tempted.

Lots of trade-offs: zoom range, speed and size. My Tamron 17-50/2.8 is small for a constant f/2.8, but would probably be unbalanced on the G1. So question is how much smaller EVIL and 4/3" would allow a lens to be. And how much OIS would add back. For a standard one-lens walkaround zoom, I'd prefer it to start at 12mm [24].

viztyger
viztyger Veteran Member • Posts: 3,280
Re: Optical Performance and Cost Cutting

Ehrik wrote:

Yes, at the two points (full wide and tele, at infinity wide open) that we get to know about at this stage. Boke, flare, AF speed, etc. is unknown and may be improved.

As we remember from the comparison between the two 7-14 lenses, even measurably lower MTF in the corners may still produce extremely rewarding photography.

That is very true. I've never used the Zuiko 7-14 but the Lumix 7-14 is excellent. A very good reason to stick with Micro Four Thirds...

It seems, going by info Dcuk got from a Panny rep, that the 14-45 will continue to be sold with the G2 ; only the G10 will get the new lens. Which I consider good news.

The Lumix global site as well as the images of the G2 in the dpreview preview show the G2 with the new 14-42 lens. But who knows, certain markets might see the G2 bundled with the original 14-45 lens.

Lots of trade-offs: zoom range, speed and size. My Tamron 17-50/2.8 is small for a constant f/2.8, but would probably be unbalanced on the G1. So question is how much smaller EVIL and 4/3" would allow a lens to be. And how much OIS would add back. For a standard one-lens walkaround zoom, I'd prefer it to start at 12mm [24].

I would accept some concessions to achieve a constant f/2.8 in a small(ish) form factor. Used in conjunction with the 7-14, a modest 2x, 18-36mm zoom range would be very useful. An 18-36 would be ideal for my event photography, since I almost never need to go wider.

I agree that for a general walk-around zoom, a starting focal length of 12mm is ideal. I imagine it being quite a bit larger though if it zoomed out to 36mm while maintaining a constant f/2.8.
--
Björn

http://www.bmupix.com

 viztyger's gear list:viztyger's gear list
Sony RX1 Sony a7R II Voigtlander 20mm F3.5 Color Skopar SL II Sony FE 35mm F2.8 Sony FE 55mm F1.8 +3 more
saskalex Junior Member • Posts: 40
Re: 14-45 OIS and 14-42 OIS, MTF and lens diagrams

From the diagram it would seem that the rear element (exit pupil) of the new 14-42mm lens is also smaller than that of the "old" 14-45. Light hits sensor at a flatter angle, which does not do much for image quality.

OP Ehrik Veteran Member • Posts: 8,014
Re: 14-45 OIS and 14-42 OIS, MTF and lens diagrams

Thanks for your comment.

saskalex wrote:

From the diagram it would seem that the rear element (exit pupil) of the new 14-42mm lens is also smaller than that of the "old" 14-45.

Good catch. The rear element is just slightly smaller but it's farther away from the sensor (at least at the depicted zoom and focus setting; it might move). The exit pupil isn't the same thing as the rear element, though. It's the image of the iris when looking from the rear. It's somewhere on the front side of the rear element (i.e. farther from the sensor).

Light hits sensor at a flatter angle, which does not do much for image quality.

Without knowing the refractive properties of the individual lens elements and doing a ray tracing based on that and curvatures indicated by the low-resolution image, or being able to investigate the lens in real life, we don't know the location of the rear pupil so we can't really draw that conclusion.

Keyboard shortcuts:
FForum MMy threads