The funny thing is that The SL is a compromise body to begin with. You overpay for a very average body in the hopes that Leica will make some amazing glass to pair with it. This proposed combo is essentially the worst of both worlds. I like a lot of Canon glass, but if I would ever be ridiculous enough to pay $7500 for $1800 worth of camera, I wouldn't immediately turn and cast aside the primary advantage for that over-investment.
Does the firmware update make the sensor less crappy?
Jekabs: Better, but still long way behind the sensors in D7200, NX1 and probably A6300... At least the noise pattern is pretty random without obvious banding, so should clean up in post pretty good.Still, good news for customers to see some progress from Canon.
Let's be fair, while it lags some behind D7200 and the ilk, this is a major step up for Canon and it's something hopefully they can build on.
AngryCorgi: So many errors in this announcement its hard to choose a place to start!
All that said, though, the bodies look pretty great, really. If it didn't have such convincing imagery associated with the PR, i'd almost be fooled into thinking it was vaporware. Such poorly written mfg PRs are the stuff of vaporware of yore and even modern crowd-funded unicorns too.
Here's just a few: 45MP effective doesn't make sense. If you assume they are using a 4:1 ratio of resolution and following the same pattern as the non-H model, then that is 25.5MP + 6.4MP + 6.4MP = 38.3MP (effective), not 45MP. If the crop for the H model (H being obviously a reference to the APS-H format), then it's not APS-C, so those fields don't carry over and the table is wrong. I doubt the raw photosensor count and actual effective photosensor count for the APS-H sensor are identical, surely related to the previous error. The copy-pasted Sigma data calls the EVF a 2.3MP EVF, when in reality its a 2.3Mega-DOT (3 dots = 1 pixel) EVF. I wouldn't note this, since its common, except that this is Sigma and they make Foveon sensors that in the past have recorded full color at each photosite, not Bayer sensors which are RGBG requiring 4 photosites to record 1 pixel of color, so the ironic error is pointed out!
So many errors in this announcement its hard to choose a place to start!
I wish blackmagic and Panasonic/Olympus would get together and make a mirrorless camera that combines the best of the blackmagic global-shutter sensors and cinema dng with the handling/AF/still capability of the M43rds cams. That would really be the closest thing to a perfect all-around semi-pro camera.
AngryCorgi: I don't understand what's going on at Leica. They keep offering these amazing cameras at disposable camera prices. Another day brings another announcement by Leica of a camera that's a total steal! Insanity! This camera would normally go for $10,000 easy! I'm floored!
If I think they should normally be worth 10,000 but they are selling them for 3,000, you then surmise that me "putting my money where my mouth is" would be buying your used one for 90,000? I knew Leica fanboys were crazy, but now I see the problem stems from a complete misunderstanding of economics.
I don't understand what's going on at Leica. They keep offering these amazing cameras at disposable camera prices. Another day brings another announcement by Leica of a camera that's a total steal! Insanity! This camera would normally go for $10,000 easy! I'm floored!
AbrasiveReducer: A (relatively) simple optical viewfinder would have made this a cult camera and distinguished it from the GR and $300 Coolpix A.
Oh I see. Not your fault, it was mine. I forgot I wrote in response to another.
Gentlemen, I never commented on the sharpness as a function of the sensor. I don't know where this tangent started. I'm merely saying there isn't much to improve on against the GR and A.
AngryCorgi: Wow. That 35/2 ASPH is a steal. Talk about "priced to move"!
Nope. They are usually a little pricey, but these are real bargains!
I paid $350 plus $99 for a brand new Coolpix A and matching Nikon VF back in late 2014. You can still buy a Coolpix A for less than $600 NEW and the new GR2 is also less than $600. Both cameras have insanely sharp lenses and good sensors. The 14-bit files from my Nikon are every bit as malleable in post as my old D7000 was (in other words, fantastic DR). I just don't see that being the case with the files from the Fuji and it'll be hard for them to produce a lens as sharp or sharper than the almost perfect lenses on the other two cameras. The only way I see this camera making sense is if they have a long-term plan to expand the X70 ecosystem to include various focal ranges. Then it almost becomes part of a collection of cameras that meets a variety of needs and all have the same/similar handling.
How do these compare, optically, with the 1st gen ones? I was not as impressed with those as I was with the metabones.
Wow. That 35/2 ASPH is a steal. Talk about "priced to move"!
HowaboutRAW: "much like a first-generation Foveon sensor"
So what about Foveon gen 2 and gen 3.
Sorry not seeing this loss in the Quattro either, The Quattros don't appear to be doing interpolation either.
No, but it is a bit of a grey area, because, again, it's an approximation, albeit based on more information than above 19M. I think up to 19MP it looks pretty good, but I also think Merrill images look even better (low ISO only).
The light passes thru the blue in four areas overlaying the green and red who cover the same area, but only address the 4-blue-zone as a single zone. So the luminance channel is recorded in the blue, but while the blue gets 4 samplings of color resolution, the green and red only get 1 sampling for the same sensor area. The luminance data is used to approximate fluxuations in the green and red channel through some proprietary algorithm. Anything above 19.6M is just straight up dumb interpolation.
Quattro doesn't do interpolation when shooting 4.9M images. It features a loss of color resolution (luminance is recorded at 19.6M as is the blue channel, but the red and green channel are only good to 4.9M) when shooting 19.6M images. All of it's higher resolution modes ARE straight interpolation.
HowaboutRAW: Mr Baskin:
You like the EVF in the NX1 more than that in the Leica SL?
No, the SL has a better EVF.
You've crossed the line between fanboyism and insanity. Enjoy your new surroundings!
Personal preference is always open to "feel" evaluation. Thus, it's not a debate, it's an opinion. The writer has an opinion, yet you want to shout his opinion down. Unreal arrogance on your part, but par for the course when reading your posts.