Pages: 1 2 [3] 4   Go Down

Author Topic: Leica Q2  (Read 3175 times)

hogloff

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 641
Re: Leica Q2
« Reply #40 on: March 15, 2019, 07:11:41 pm »

There is a Fuji X100F for you for 1/4 price, smaller size and with a hybrid viewfinder. Too bad it's not weather sealed, maybe next generation.

I have the original X100... Sits on my shelf with little use. I can take my A7r / A7R2 with both a 35 2.8 and 28 2.0 and it provides way more flexibility. Typically my travel kit is 25mm, 35mm and 85mm along with the A7R2 and at a price likely less than the Q2. Much more flexibility...and I bet no one can tell the difference in prints from the Leica and A7R2.
Logged

faberryman

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 679
Re: Leica Q2
« Reply #41 on: March 17, 2019, 01:12:29 pm »

Much more flexibility...and I bet no one can tell the difference in prints from the Leica and A7R2.
The Leica comes with a better menus tax.

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3173
    • Photos
Re: Leica Q2
« Reply #42 on: March 17, 2019, 02:08:14 pm »

I have the original X100... Sits on my shelf with little use. I can take my A7r / A7R2 with both a 35 2.8 and 28 2.0 and it provides way more flexibility. Typically my travel kit is 25mm, 35mm and 85mm along with the A7R2 and at a price likely less than the Q2. Much more flexibility...and I bet no one can tell the difference in prints from the Leica and A7R2.

I was tempted several times to get a version of X100 but knowing my old X-E1 is almost the same size, particularly with a slightly longer 27 F2.8 (42 equiv), made it difficult to justify. Newer phones getting better doesn't help either. Still, the simplicity is tempting.
To make the case for Leica though I find it even more difficult because of options that offer most things that Leica offers on paper while being much cheaper.

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6384
Re: Leica Q2
« Reply #43 on: March 19, 2019, 07:06:03 am »

iPhones have a camera lens that corresponds to the 26mm full frame lens. Discuss.
There is a trend—due to sensors with far more resolution than most “display intents” need—for single focal length cameras to go wider, relying a bit more on cropping options. The Q2 offers framelines for a 30MP  crop to “35mm equivalent FOV”, and a 1.5X crop emulates an 21MP APS-C sized sensor at “normal” FOV, like 42mm in 36x24mm format.

Not that this matters for me; I’d never pay even $1000 for a single focal length camera—even if it also made phone calls!
Logged

KLaban

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
Re: Leica Q2
« Reply #44 on: March 20, 2019, 06:43:04 am »

I'm awaiting the introduction of the SL2 with interest. Hopefully it'll have the Q2 sensor, will behave with my existing M lenses and offer the alternative of an EVF platform as an addition to my existing M rangefinders.
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 901
Re: Leica Q2
« Reply #45 on: March 21, 2019, 02:05:12 pm »

Would the Panasonic S1R meet your expectations for a SL2? Different user interface, but otherwise seems to do what you say...
There are rumors around that Panasonic may slap a Leica label on the high-end FF gear - which might come with Leica no longer developing the SL series?
Logged

KLaban

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
Re: Leica Q2
« Reply #46 on: March 21, 2019, 02:54:52 pm »

Would the Panasonic S1R meet your expectations for a SL2? Different user interface, but otherwise seems to do what you say...
There are rumors around that Panasonic may slap a Leica label on the high-end FF gear - which might come with Leica no longer developing the SL series?

They are different beasts. The S1R is too large and heavy for my needs and as far as I know is not optimised for M lenses.

I understand the SL2 is in development and doubt Leica would enter into the alliance only to cede the camera to Panasonic. Time will tell.
Logged

faberryman

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 679
Re: Leica Q2
« Reply #47 on: March 21, 2019, 03:58:28 pm »

There are rumors around that Panasonic may slap a Leica label on the high-end FF gear - which might come with Leica no longer developing the SL series?
I think the market would be small for a rebadged SR1. Leica is going to have to actually do something.

Christopher

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1411
    • http://www.hauser-photoart.com
Re: Leica Q2
« Reply #48 on: March 21, 2019, 04:46:42 pm »

I think the market would be small for a rebadged SR1. Leica is going to have to actually do something.


Well at the end the sl2 is a rebadged sr1.. different design, but the sensor will probably be exactly the same...


Sent from my iPhone using Tapatalk
Logged
Christopher Hauser
[email=chris@hauser-p

KLaban

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
Re: Leica Q2
« Reply #49 on: March 21, 2019, 04:54:11 pm »


Well at the end the sl2 is a rebadged sr1.. different design, but the sensor will probably be exactly the same...


Sent from my iPhone using Tapatalk

You must be the first to have seen the SL2 and have the specifications.
Logged

Christopher

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1411
    • http://www.hauser-photoart.com
Re: Leica Q2
« Reply #50 on: March 21, 2019, 05:24:06 pm »

Perhaps I have, but it doesn’t really matter. Leica will surely have a Second completely different sensor for the sl2... and why shouldn’t they?

It makes perfect sense to develop something on their own for one specific camera, especially because it means you can spend even more money on development instead of using something devolved my a partner company.


Sent from my iPhone using Tapatalk
Logged
Christopher Hauser
[email=chris@hauser-p

KLaban

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
Re: Leica Q2
« Reply #51 on: March 21, 2019, 06:21:48 pm »

Perhaps I have, but it doesn’t really matter. Leica will surely have a Second completely different sensor for the sl2... and why shouldn’t they?

It makes perfect sense to develop something on their own for one specific camera, especially because it means you can spend even more money on development instead of using something devolved my a partner company.


Sent from my iPhone using Tapatalk

Manufacturers use essentially the same sensor in very different cameras.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6384
Re: Leica Q2
« Reply #52 on: March 21, 2019, 06:30:54 pm »

Manufacturers use essentially the same sensor in very different cameras.
Indeed: as surely as film SLRs using the same "sensors" (36x24mm frames of film) varied greatly in price and performance.
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 901
Re: Leica Q2
« Reply #53 on: March 21, 2019, 09:27:42 pm »

Leica doesn't have the volume to get custom sensors (why they buy from TowerJazz(?) instead of Sony is a mystery to me).

Canon and Sony build sensors (technically, Sony Imaging buys them from the related Sony Semiconductor).

Nikon buys (usually) Sony sensors, but has enough volume to order at least semi-custom chips.

Fuji buys Sony sensors, sometimes with custom color filters (the X-Trans is a catalog model Sony sensor with a unique filter - not sure whether Sony or Fuji applies the filter). Fuji medium format sensors are catalog Sony chips with standard filtration.

Pentax, Phase One and Hasselblad  buy stock Sony sensors from the catalog.

Panasonic and Olympus buy Micro 4/3 sensors from somewhere (Sony? TowerJazz?). If the 20 MP Micro 4/3 sensor is a Sony, it's a couple of generations behind the current models - no BSI, no copper wiring, etc.

Panasonic and Leica buy TowerJazz (?, but the high resolution sensor is from not Sony) sensors for the L-mount

Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6384
Re: Leica Q2, and Leica and Panasonic sensor sources
« Reply #54 on: March 21, 2019, 10:50:21 pm »

Leica doesn't have the volume to get custom sensors (why they buy from TowerJazz(?) instead of Sony is a mystery to me).
I am not sure at that: the 45x30mm sensors for the "R" bodies (one CCD, then two CMOS models) are custom designs to a degree, but maybe based on a standard design for the individual photosites. And I have read that the same has been true for various sensors used in digital M models. What Leica lacks in volume it can sometimes make up for with its ability to pass the higher price of custom sensors onto its customers!

Also, does anyone know the status of Panasonic's plan of getting back into the sensor business?  It was producing and even selling its own "Live MOS" sensors for a while, then withdrew for that sector and then indicated that it was getting back into sensor making.
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 901
Re: Leica Q2
« Reply #55 on: March 22, 2019, 11:58:25 am »

Panasonic has been talking about an organic sensor that may have some interesting properties for a few years now, but (as far as I'm aware) we haven't seen even a working prototype.

The very best current sensors seem to be bumping up against the limits of conventional sensor design (if we define conventional as 14 bit Bayer (or X-Trsns, which is a variant of Bayer) sensors using standard materials).

The low-hanging fruit may be getting more dynamic range by moving to a 16-bit readout while improving noise/photosite capacity farther. The best current sensors have meaningful information in all 14 bits, so more information would need more readout bits. A few medium format sensors (and a RED sensor or two?) already use 16-bit readouts.

You can't go too much farther with resolution before getting bitten both by diffraction and reduced photosite capacity (a 60ish MP 24x36mm sensor should work - both APS-C and medium format sensors are at that density and working well). The 20 mp Micro 4/3 sensor is a comparatively weak performer - perhaps because it's too dense (that would be about 36 MP APS-C, 80 MP 24x36mm).

Organic materials may offer more photosite capacity, allowing more DR at the same photosite sizes,  (needs 16-bit readout), smaller photosites without sacrificing DR (watch out for diffraction), or both.

Multi-layer sensors could offer interesting improvements, especially in color rendering. So far, Bayer and X-Trans sensors have been able to beat any multi-layer sensor for most images, simply by offering enough more resolution and DR that they capture most or all of the color information that the multi-layer sensor gets while also getting much more luminance and spatial resolution. The latest multi-layer sensor that has been announced is Sigma's L-mount 20 MPx3 design (next year). It will offer 20 MP each of red, green and blue - but only 20 MP of spatial resolution. It will probably lag the best of the competition in DR - Foveons always have. A 45 MP Bayer sensor actually offers more green resolution than the Foveon (22+ MP), which the eye is most sensitive to, while lagging in red and blue. However, it offers more than twice the spatial resolution and a couple extra stops of DR. A 45 MPx3 Foveon would almost certainly outperform 45 MP Bayer (substantially), especially if the DR was the same. However, the Foveon process has never been at a stage where it offers similar spatisl resolutions.

Logged

faberryman

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 679
Re: Leica Q2
« Reply #56 on: March 22, 2019, 12:30:03 pm »

You can't go too much farther with resolution before getting bitten both by diffraction and reduced photosite capacity (a 60ish MP 24x36mm sensor should work - both APS-C and medium format sensors are at that density and working well).
If you take a 24 x 36 section out of the 40.4 x 53.7 IQ4150 sensor, you get 60MP for full frame. That should be the next jump for Sony.

Telecaster

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3690
Re: Leica Q2
« Reply #57 on: March 22, 2019, 03:41:55 pm »

Panasonic used to have a large stake in TowerJazz…maybe still does. That likely explains Leica's use of TJ sensors.

My interest in sensor tech at this point is about finer tonal gradation. Don't care about more spatial detail for its own sake. It's generally IMO a Good Thing, though, if sensors outresolve lenses.

-Dave-
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6384
Re: Leica Q2
« Reply #58 on: March 22, 2019, 06:30:08 pm »

The low-hanging fruit may be getting more dynamic range by moving to a 16-bit readout while improving noise/photosite capacity farther. The best current sensors have meaningful information in all 14 bits, so more information would need more readout bits. A few medium format sensors (and a RED sensor or two?) already use 16-bit readouts.
That 16 bits might be about as fas as it needs to go; that is enough to count 65,536 photons exactly, which is about enough to cover full well capacity of current good sensors, so enough even if read noise is completely eliminated. If anything the FWC is likely to go down rather than up as resolution increases, and it might be that the way forward is more 14-bit pixels (or even many more 12-bit ones). Most images are well handled by a 14-stop DR, and the extreme cases that need more can be handled by dithering/downsampling/binning to improve the output dynamic range. That also allow more parallelism in ADC, for a possible frame-rate advantage.

You can't go too much farther with resolution before getting bitten both by diffraction and reduced photosite capacity (a 60ish MP 24x36mm sensor should work - both APS-C and medium format sensors are at that density and working well). The 20 mp Micro 4/3 sensor is a comparatively weak performer - perhaps because it's too dense (that would be about 36 MP APS-C, 80 MP 24x36mm).
Neither diffraction nor "per pixel dynamic range" are really the impediments that they are sometimes made out to be:
- At worst, diffraction means that the full resolution increase benefit will only be achieved at lower f-stops, while at higher f-stops, the gain will be less or even negligible, but resolution will never be worse than at a lower pixel count. That will only fail when the f-stops that gives a worthwhile resolution benefit are so low that lens aberrations become a significant impediment, and we are nowhere near that level yet — not even in phone-cameras!

- Per pixel DR is an unfair and somewhat irrelevant measure when comparing different pixel counts. For one thing, downsampling (or on-sensor binning) to a lower pixel count improves the per pixel DR, and once read noise is low enough, the result comes close to recovering the per pixel DR than one would get with fewer, large photo-sites. As an extreme case note that the "photo-sites" of traditional film (silver halide crystals) are 1 bit; either exposed or not. All tonal gradations, dynamic range and such come from the dithering of many billions of such 1-bit signals.
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 901
Re: Leica Q2
« Reply #59 on: March 23, 2019, 02:50:17 pm »

Phone cameras are already unable to use all their resolution due to diffraction (a faster lens would help - so it's not quite to BJL's "no possible lens" point). iPhones (just to use a popular example) use a 1/2.5" sensor and a ~4.25mm f1.8 lens. The diffraction limit on that combination is around 6 MP (assuming the lens was perfect). A lens around f1 or f1.2 might resolve all the pixels, and it's not necessarily impossible, especially because it doesn't have to cover a large sensor.

Also note that the f1.8 aperture is not what it seems, due to the tiny sensor. The aperture equivalent on a full frame camera is above f8, closer to f11.

An iPhone (or any other phone that isn't too thick for most people to want) is basically a 3-4 MP (those lenses aren't perfect, so you don't reach the 6 MP diffraction limit) f8-f11 wide angle camera with around 7 stops of DR. Other than the enormous amount of post-processing the phones do, it's almost exactly equivalent to a very early DSLR like a Canon EOS-D30 from 2001 (with a lens that is wider, but dimmer than anything one could economically fit to a D30 at the time).

The only practical way to build a phone camera better than that (not "get a more pleasing image through automatic post-processing", but "collect more real data") without making the darn phone thicker would be to use multiple cameras and merge the images. Light tried to do that, and the first generation wasn't great, but someone may get there.

Until then, the laws of physics keep phone cameras stuck right around the image collection capabilities of 20 year old DSLRs, although with incredible post-processing.
Logged
Pages: 1 2 [3] 4   Go Up