April 19, 2014, 11:14:54 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Lawliet

Pages: 1 [2] 3 4 ... 22
EOS Bodies / Re: New Full Frame Camera in 2014? [CR1]
« on: March 28, 2014, 12:05:53 AM »

I'm not sure they have had the tech for more DR and it didn't seem like they cared enough early on. A

What could stop them from reading the individual sensor cells of a dual pixel at different amplifications? Blending the exposures isn't exactly beyond high school math either. I.e. thats what you can get via Magic Lantern, just without the potential resolution loss, because you don't have to fall back to sensels that where meant to provide spatial data.

EOS Bodies / Re: New Full Frame Camera in 2014? [CR1]
« on: March 27, 2014, 11:00:38 PM »
I'd like to see quad-pixel technology come out... with a sensor where you can bin all four sub-pixels together and have a 10Megapixel FF sensor with kick-ass low light capacity or access the sub-pixels and have a 40Megapixel high resolution sensor....

Phase One beat you on that idea!  :P
Admitted, going from ISO800 to ISO3200 that way won't get anybody from FF/APS-C-Land excited, but the idea is there.


That's a very informative chart ... surprising that Nikon charges as much as Leica for repairs ... also Nikon's repair time is pretty pathetic only topped by the scandal hit Olympus.

It doesn't tell you what was repaired. For example both Canon and Nikon have quite high repair costs. Is their service generally more expensive then the other brands? Or do for example the (assumed) often rented out for comparable rough use  superteles skew the result?

So I'm genuinely shocked that in my ignorance I believed there was a proper step up in sensor real estate to be lusted after, alas, er... this.

You always could go for backs like the IQx60/IQx80 or some of the newer HBlads - those have sensors that are roughly  the size of the usable area of 6x4,5.
But then there is more then raw sensor size, like design goals or the lenses. A bit of extra resolution, but mostly the lenses and the leaf shutter for me.

EOS Bodies / Re: New Full Frame Camera in 2014? [CR1]
« on: March 27, 2014, 06:46:52 AM »

1920*3=5760 which is exactly what 5D3 got.

That would be a 2:1 aspect ratio, not much love here. And It would require odd blending/skipping patterns to capture standard video. Not much love either.

The first fitting, in tradition of the 5760px of the 5D3, resolution would be 1920*2*2 - to get full RGB for 4k-video and 39.3MP stills. going for the slightly larger 4096 flavor would net you 44.7MP. Current OTS cores can handle that resolution up to about 30fps; in a dual config. that should work even with continuous AF. no line skipping also implies less moire and less noise at higher ISOs. Dual photodiodes allow for higher low ISO DR. Who could argue against such a machine?

EOS Bodies / Re: New Full Frame Camera in 2014? [CR1]
« on: March 27, 2014, 03:32:50 AM »
I would think the opposite! A mid priced Merc would sell well IMHO.

But only if they
a) manage to keep the level of quality&performance up despite the lower price, otherwise expect backlash.
b) keep the air of exclusivity&luxury that makes the car covetable in the first place.
..think NEX in Hasselblad clothing, though the current state of affairs in that department could be worse.

Something to keep in mind about the 5D800-story: is it just a question of sensor resolution? Or rather a matter of being in an existing ecosystem, or the added effort that has to be made to make actual use of the potential resolution? And then there is the field of pictures in motion, if thats even part of your job description Canon wins more or less by default.

EOS Bodies / Re: New Full Frame Camera in 2014? [CR1]
« on: March 26, 2014, 10:19:44 PM »
I have looked closely at the A7r for that reason but there seem to be three big question marks - light leak, AF performance and IQ when using an EF adaptor.

The light leak story is imho overblown, thats unless you regularly shoot with dense ND filters. Mileage varies with lens choice - how tight are the tolerances of the mount and how susceptible to stray light effects is the lens as such?

IQ-wise the adapter contains no optics, its just an extension tube. While the additional tolerances introduced by the second mount are measurable most of the time just about every other source of error will dominate.

AF? I got the A7r for its manual focusing facilities. :)
Esp with the TS-E lenses AF is...well...you'd be waiting for eternity. With native lenses its ok, with promising developments(see a6k), adapted its more a proof of concept to me.

Am I the only person looking at the sensor dimensions and thinking "they get away with calling THAT medium format these days?... and people pay what?!"

Guess that depends on whether you look for the numbers or just for the best data you can get.
Most people who pay for dMF are in for the second reason.

They havent been able to make much inroads into the Nikon/Canon upper end of the DSLR market let alone MF.

According to their roadmap they're rather trying to approach that from the lower end, go for the upper echelons once they have a base. Looking at their market share in asian regions, the absolute performance of lets say the a6000 and A7r. the recent announcement of a CPS/NPS-like service and such the next 18 months or so might be interesting and prod Canon into actually doing something more exciting then slapping new labels on entry level bodies.

but the sharpness "issue" is too overemphasized IMO. 

Some people seem to forget that close to wide open the areas affected by lens defects tend to be out of focus anyway. And once you stop down to apertures that result in sufficient DoF the lens is at its sweet spot again.
Now reliable focusing , as well as more subtle rendering characteristics, are more relevant topics, at least away from shooting charts or similar.

EOS Bodies / Re: Canon's Plans For NAB 2014 Next Month [CR1]
« on: March 20, 2014, 11:15:09 PM »
Instead of a 7DC, they release a 5DC... 24MP Dual Pixel & 4K. Game over.

Even better: double the 4k resolution to get 4:4:4 output from a bayer sensor or at least mostly artifact free results. Preferable with full sensor readout for additional sensitivity & signal quality. For 30fps that wouldn't even be that much more workload then Nikons new compact handles, not to talk about using two processors. Throw in the standard visual aids for video.
All the video, landscape& studio crowds would love that.

EOS Bodies / Re: Evolution or the murder of art?
« on: March 18, 2014, 11:47:45 PM »
I'm not saying it will happen quickly, but if technology advances to where the frame rate can be 24 fps and a full RAW image can be captured for each frame, we'll be there.

Well, Nikons V3 can capture 60 full res raw files, i.e similar 18MP to the 1Dx or the run of the mill APS-C cam, per second, or 20fps with tracking AF. So much for "quickly"  ;D

But how much effect that can have on photography is very genre dependent. For sports its quite an advancement. For anything planned not so much. And once you bring flash into the game its at least massively expensive (get me some Scoros, and that's w/o requiring high power) or way beyond that. Replace that <10kg mobile unit or some flashguns with multiple trailers filled with gear... 8)

As for picking frames from a video...both technical details like motion blur and artisitic&narrative decisions are often mutually exclusive, or at least not exactly helping each other. That's for absolute minimal budget/low quality, otherwise definitely unavailable material or fools.


I've browsed through most of the thread and I'm surprised so many are taken aback by this guy having his assistant (s) set the camera up on a landscape picture.

We could take a look at the credit list on those moving picture thingies that came up recently - everybody knows names like, lets say Kubrick, Spielberg or Tarantino. Not that much work on the physical cameras from those I'd wager, yet they're the most strongly associated with the final product. Even their DoPs have a different job description then the camera crews. Just because one can do multiple jobs on smaller scale productions it doesn't make micromanaging trivialities mandatory. Doing a models makeup on my own would end in "epic fail" anyway...  :-[

Canon General / Re: Review: Canon Speedlite Transmitter ST-E3-RT
« on: March 17, 2014, 06:40:31 PM »
I haven't had the same unreliable connection with my YN-E3-RT as you guys have, but It definitely feels cheaper, and didn't do well when I had it outdoors for a -30C shoot (I can't comment on if the Canon could have fared better).

As for reliability: I'd say the Canon is more reliable at being reliable - less "It works, most of the time at least". Either black or white, no greys in Canonland. Might just be a more demanding setting for the frequency evaluation program? Who knows?

At -30C most batteries go to sleep, and that small thing hasn't enough mass to provide thermal capacity or insulation to keep at least some warmth in and the chemistry going. In such circumstances Lithium cells work best, either the ~1.6V-1.7V AA-sized ones or if you want to go the distance replacing two AAs in series with a single CR123A primary might be an option. Those still work in conditions that would probably have killed your gear a while ago.

EOS Bodies / Re: EOS 7D Replacement Mentioned Again [CR1]
« on: March 17, 2014, 06:26:32 PM »

We aren't talking cell phone distances here. Most EVFs are recessed at around a quarter of an inch or so, pretty much always less than an inch. At around an inch, 3000-5000ppi would do it, but at a quarter of an inch, people with better than 20/20 vision are going to see pixels.

Your whole chain of argument is consistent in missing a vital point: there is at least one optical element in from of the display. I.e. absolute and apparent  size, resolution, distance to the eye,... of the display are decoupled, making the whole story somewhat amusing but nonetheless moot.

Also seeing those discreet pixels would require them to have sharp borders, absolutely no light bleeding to neighbor cells, nothing like the trinitron tube of old times. No effective AA-filterlike construct to blend colors.

Having pixels in the VF has some benefits on the one side, on the other: in favorable conditions now I see the structure of the matte screen. That's not really much better then it's electronic counterpart in the first place...

Pages: 1 [2] 3 4 ... 22