Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-18-2013, 04:36 AM   #16
Veteran Member




Join Date: Oct 2011
Location: Aylesbury, Bucks
Photos: Albums
Posts: 492
I've used both. Overall an OVF is (still) more pleasant to use. The TV-watching experience of the EVF creates an additional removal from the scene being viewed. However, when I first used an EVF (a soon-to-be-lost Olympus VF2), my first thought was "that is one big TV screen". So there's one advantage to the EVF: the size of the viewfinder image is not limited by the necessity of using natural light. Reduced need for natural light leads to another advantage: viewfinder brightness is not dependent on the lens mounted. On my new X-E1, I was using 6 and 10 stop neutral density filters and was still able to compose through the viewfinder. That is simply amazing. It is possible that live view would give you the same thing, but since live view on a K-5 is barely usable without an ND filter, I haven't cared to put it to the test.

Back to Sony. The link provided isn't exactly replete with details, but it's safe to say that the SLT format hasn't set the world on fire. All the bulk of a mirror system with a slightly worse viewfinder experience. Phase detection on the sensor makes the entire concept obsolescent, if not quite yet obsolete. Yeah, if I'd bought big into the A-mount, I would now be worrying about its future.

05-18-2013, 09:49 AM   #17
Veteran Member




Join Date: Apr 2010
Location: Tennessee
Posts: 6,617
Original Poster
QuoteOriginally posted by top-quark Quote
Back to Sony. The link provided isn't exactly replete with details, but it's safe to say that the SLT format hasn't set the world on fire. All the bulk of a mirror system with a slightly worse viewfinder experience. Phase detection on the sensor makes the entire concept obsolescent, if not quite yet obsolete. Yeah, if I'd bought big into the A-mount, I would now be worrying about its future.
Sony has been flailing around trying to figure out how to sell cameras. This lack of direction is one reason they have lost sales. When it was rumored they were going away from the OVF.... they lost sales. Had I bought an A900 a few years ago (as planned) I would be pretty unhappy.

I think phase detection sensors on the imaging sensor is a stepping stone technology. As processing power increases and sensors improve CDAF speed will improve. With CDAF systems the image processor has to handle the AF work load. With PDAF systems you have dedicated sensors and processors to handle the AF. Fuji is rumored to be using 2 image processors in the X-Pro 2.

Trying to turn the A-mount system into a mirrorless platform sounds like a K-01 debacle all over again. Maybe Sony can do a better job than Pentax.
05-18-2013, 11:00 AM   #18
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by Winder Quote
This lack of direction is one reason they have lost sales.
I am not sure.

When SLT arrived, I posted and always maintained the point of view that this is nothing but a migration tool for A-mount into E-mount.

Now, as it seems, the full migration roadmap looks as follows:
SLR-Alpha -> SLT-Alpha -> NEX-Alpha -> NEX

(NEX-Alpha is my name for a mirrorless with A-mount).

I see no lack of direction. On the opposite, I see lack of direction from other venders when and how to get rid of the mirror.

It may be Sony takes some hit in sales now. But wait until Canikon loose direction ...

BTW, the rumor has it too that Sony decided to almost stop the release of new cameras for 2013 and to make sure R&D delivers innovative products for a bold move in 2014. I'd like to see some more of Sony genes in the rest of the industry ...

Side note: I consider phase detection sensels on the imaging sensor to be a dead end technology.
05-18-2013, 11:35 AM   #19
Veteran Member




Join Date: Apr 2010
Location: Tennessee
Posts: 6,617
Original Poster
QuoteOriginally posted by falconeye Quote
I am not sure.

When SLT arrived, I posted and always maintained the point of view that this is nothing but a migration tool for A-mount into E-mount.

Now, as it seems, the full migration roadmap looks as follows:
SLR-Alpha -> SLT-Alpha -> NEX-Alpha -> NEX

(NEX-Alpha is my name for a mirrorless with A-mount).

I see no lack of direction. On the opposite, I see lack of direction from other venders when and how to get rid of the mirror.

It may be Sony takes some hit in sales now. But wait until Canikon loose direction ...

BTW, the rumor has it too that Sony decided to almost stop the release of new cameras for 2013 and to make sure R&D delivers innovative products for a bold move in 2014. I'd like to see some more of Sony genes in the rest of the industry ...

Side note: I consider phase detection sensels on the imaging sensor to be a dead end technology.
I don't think Sony has real direction. By their own admission they did not expect such high demand for the NEX-7 or the CZ E-mount lenses. Twice they have altered their plans as they shifted product development away from A-mount to bring more E-mount products to market to meet the unexpected demand.

If Sony releases a true mirrorless FF with a new mount that can use A-mount with an adapter, then I will give them a serious look, but if they are simply going to make engineering compromises to make A-mount a mirrorless system then I doubt I will be interested. If Sony does something radical with the mirror box like develop a system like a 3-CCD camera or new lenses that recess back into the mirror box then that might be worth looking into.

It is hard to say Sony has direction given the number of times the have changed their product plans.

05-18-2013, 11:53 AM   #20
Banned




Join Date: Sep 2006
Location: Millstone,NJ
Posts: 6,491
I just received my Sony NEX-3N today that is the smallest and lightest NEX and the first with zoom lever for the new E mount power zoom lenses for use in video mode just like a camcorder. It is not much larger than my Pentax Q and the 16-50mm (24-75mm FF eqiv) OSS power zoom lens is the smallest and lightest APS-C 3x zoom kit lens I ever seen. It is also the first NEX that supports “TRILUMINOS Colour” that is on the new Sony 4K UHDTVs.

All of the lenses that I use on my full frame mirror less NEX-VG900 work on the NEX-3N that include Sony A mount, E mount and Pentax PK mount. There is also the new PK-NEX Speedbooster adapters that turn F1.4 lenses like my full frame 24mm,35mm,50mm and 85mm into F1.0 lenses.
05-18-2013, 01:52 PM   #21
Veteran Member
philbaum's Avatar

Join Date: Oct 2007
Location: Port Townsend, Washington State, USA
Posts: 3,659
QuoteOriginally posted by Winder Quote
I don't think Sony has real direction. By their own admission they did not expect such high demand for the NEX-7 or the CZ E-mount lenses. Twice they have altered their plans as they shifted product development away from A-mount to bring more E-mount products to market to meet the unexpected demand.
............
It is hard to say Sony has direction given the number of times the have changed their product plans.
In a way, i think they do have direction. Judging from general discussion in a review of their 2012 results, I remember reading that their NEX cameras were far more successful than their A mount sales. The review i read had no specifics, but it appears clear that Sony's priority is with the NEX mount and not with the A-mount. And when they do get back to the A-mount, SLT is not in its future.

Given Sony's financial difficulties, one can hardly blame their camera division from placing priority on their most successful sales area, e.g. NEX. I think there will be a Nex e-mount FF and it will appear in the first half of 2014. Not an unreasonable expectation considering their RX1 achievement.

If Pentax does not produce a FF camera in 2014, then I would expect a number of the Pentax base to buy the Sony Nex e-mount FF, which should be out by the time we get any news from Pentax The Sony is unlikely to provide much automation to Pentax lenses, but will likely provide 2 methods of focusing: focus peaking and the 50% magnified view focusing. (Both those focusing methods work fine when i use my M 50 f1.7 and M 135 f2.8 lenses on my Sony 5n) Price may be the stumbling block for the Sony e-mount FF. I can't see customers paying $2500 for it, but if its down as low as the D600 has been, then maybe so.
05-18-2013, 01:53 PM   #22
Veteran Member
Anvh's Avatar

Join Date: Sep 2011
Posts: 4,616
QuoteOriginally posted by falconeye Quote
A common belief and most probably wrong.

CDAF has more light at its disposal and should outperform PDAF when both systems are driven to the edge of the possible. The embedded PDAF sensels turn useless in low light for a reason, this is a dead end caused by current engineering hurdles to figure out a decent algorithm, to read out sensors fast enough and to shift focus precisely enough.
I can not see how CDAF would be better in PDAF in term of speed in normal condition, the problem is that PDAF can see if the focus is behind or infront of the subject and adjust accordingly, it can even see how much it's out.
CDAF is just adjusting the focus and move the way that increase the contrast and that is simply trail and error.

05-19-2013, 05:44 AM   #23
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by Anvh Quote
I can not see how CDAF would be better in PDAF in term of speed in normal condition, the problem is that PDAF can see if the focus is behind or infront of the subject and adjust accordingly, it can even see how much it's out.
CDAF is just adjusting the focus and move the way that increase the contrast and that is simply trail and error.
That's clearly the reason why people believe PDAF to be superior by principle.

However ...

It is always worth to have some framework to make such judgements.

Both, CDAF and PDAF are cross-correlation maximization algorithms.

In the case of CDAF, it is the auto-correlation of a 2-dimensional region of interest with itself which is maximized.

In the case of PDAF, it is the cross-correlation of a 1-dimensional region of interest (narrow stripe) taken from two separate patches of the lens' exit pupil.

The PDAF case has the known advantage that the free parameter (the shift) can be optimized from taking a single measurement, while CDAF needs a sequence of measurements (3 at least). Say, CDAF needs a factor P more measurements wrt PDAF.

The CDAF case has the less known advantage that it use an order of magnitude more light, say, the amount of light it can use is a factor C over PDAF.

Now, if C>P, then CDAF outperforms PDAF in speed, if the mechanical focus latency can be made small (note that focus travel is the same for both methods).

My claim is that C>P will soon be achieved.

I back up my claim by experience I gathered from the development of a high speed CDAF for an industrial customer. I use an optimization function which allows to estimate direction and amount of shift from just two measurements made in immediate succession. The details are my intellectual property, so I cannot share.

As soon as CDAF outperforms PDAF (i.e., C>P), further advantages come to fruit:
- the measurement and focus travel can happen in parallel, speeding up the operation.
- subsequent measurements can correlate the region of interest first (focus tracking) which leads to a more reliable focus performance with long lenses.
- accuracy is a good as needed, no calibration problems, color shift etc.
- High speed AF.C bursts (24+ fps) may track the focus without further measurements at all, just using the taken photos as measurement. Also needed for video-AF.
05-20-2013, 12:39 PM   #24
Veteran Member
Anvh's Avatar

Join Date: Sep 2011
Posts: 4,616
For static subject i can surely see it happen but not with dynamic ones, there is no differnce in that now and actually CDAF is preffered but for dynmaic subjects...
What we see now is that the sacrifice off a couple of pixels so that the sensor can take a phase measurement.

Here is a good example.

the Panasonic GH3 does this a lot better then the 550D but you still see that it's searching for the focus when the subject gets closer to the camera.
As for using the actual photos as mearement i doubt that will work, you will always be running behind the fact, so if the subject moves constant and predictable that's not a problem but hardly any subject does that and take into account the movement of the camera and it all becomes very complex.

ps.
Two point measurement for estimate direction is pretty simple right, you need to know the time and the positions of the measured points.
05-20-2013, 12:48 PM   #25
Banned




Join Date: Sep 2006
Location: Millstone,NJ
Posts: 6,491
I just compared my new Sony NEX-3N (16-50 power zoom OSS kit lens) with my Pentax Q (8.5mm standard 01 lens) and the Q is not that much smaller. The NEX-3N is the newest and smallest NEX and has built-in flash unlike the Canon EOS M that has a smaller 1.6 crop APS-C sensor and much larger 18-55 kit lens.



Last edited by jogiba; 05-20-2013 at 01:17 PM.
05-20-2013, 01:22 PM   #26
Veteran Member
Anvh's Avatar

Join Date: Sep 2011
Posts: 4,616
Sony is doing very well with their sizes, but when you see the corners of the images
The balance is simply gone with Sony, different sensor technology might improve it though but for now i would prefer a smaller size sensor for small camera's
05-20-2013, 09:46 PM   #27
Veteran Member




Join Date: Mar 2010
Location: SW Washington
Posts: 833
You guys do realize that arguing about the merits of EVF vs OVF on Sony cameras is a moot point right? Sony hasn't released a camera with an OVF since the A580 back in August 2010, and have shown no intention of ever bringing them back. OVF is dead and buried to them.

If they're already determined to stick with EVF, and can get PDAF on sensor, then they might as well get rid of the light-sucking SLT which serves no purpose after that.
05-21-2013, 11:02 AM   #28
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by Anvh Quote
For static subject i can surely see it happen but not with dynamic ones,
...
As for using the actual photos as mearement i doubt that will work
...
Two point measurement for estimate direction is pretty simple right
I am sorry but you didn't take the time to really understand what I said.

You said "I can not see how CDAF would be better in PDAF" and I replied to that. My response was about the respective potential in both methods, not their current incarnations. I have no idea why you then replied by giving current examples

There is no fundamental difference between an actual photo and a CDAF measurement if exposure time is fast and read-out time is low enough. E.g., 24 fps or 40 ms per measurement is fast enough to track a subject. It isn't much more complex than PDAF.C tracking as both will have to predict the focus position at a later moment in time (easy if you model the subject's relative motion in 3D space).

"Two point measurement for estimate direction" isn't what I talked about. I talked about "estimate direction and amount of shift from just two measurements made in immediate succession" which is much different: the 2 points are only minimally apart and the amount is estimated besides direction; in practice actually, the 2 points are so close that you would experience a single measurement when operating the camera, i.e., no wobble or whatsoever.
05-21-2013, 11:55 PM   #29
Veteran Member
Clavius's Avatar

Join Date: Feb 2011
Location: De Klundert
Photos: Gallery
Posts: 4,150
Thanks Falk, for the detailed explanatian... And the reassurance. I love CDAF for it's high accuracy, but the slow speed in the current Pentax incarnation does bother me off course. People have been claiming here and there that CDAF will always be slower then PDAF. Nice to see that those opinions are unfounded.
05-22-2013, 03:21 AM   #30
Veteran Member




Join Date: Jun 2009
Location: Melbourne
Posts: 341
PDAF needs fewer measurements (ideally one) versus CDAF (minimum 3, in reality more). As such, the integration time of the AF sensor per reading needs to be shorter for CDAF in comparison to PDAF, if the allowable AF time is fixed (say it is 1 second, including lens actuation, before AF speed is perceived as inconveniently slow). Therefore even if CDAF sensor is exposed to higher light intensity, the exposure time is shorter, so the light flux seen by CDAF might not be greater.

Furthermore, the amount of light passing through a partially silvered mirror in DSLR AF systems is typically around 30% of the light seen by the imaging lens, with another 5% perhaps lost in AF optics transmission.
So the light intensity projected on the CDAF sensor is ideally 4 times higher than PDAF.
However in reality the resultant S/N ratio for CDAF is not proportionally 4 times larger than for PDAF as the PDAF sensor has the advantages of 1) being monochromatic 2) having larger pixels.

As such, despite CDAF theoretically being subject to higher light intensity, the S/N of a CDAF system is in practice limited by the inherent shorter exposure time (due to more measurements) and limitations of Bayer pixels.

It is hard to say how these differences will pan out in practice but I don't think CDAF will be faster than PDAF in low light in the near future.


QuoteOriginally posted by falconeye Quote
That's clearly the reason why people believe PDAF to be superior by principle.

However ...

It is always worth to have some framework to make such judgements.

Both, CDAF and PDAF are cross-correlation maximization algorithms.

In the case of CDAF, it is the auto-correlation of a 2-dimensional region of interest with itself which is maximized.

In the case of PDAF, it is the cross-correlation of a 1-dimensional region of interest (narrow stripe) taken from two separate patches of the lens' exit pupil.

The PDAF case has the known advantage that the free parameter (the shift) can be optimized from taking a single measurement, while CDAF needs a sequence of measurements (3 at least). Say, CDAF needs a factor P more measurements wrt PDAF.

The CDAF case has the less known advantage that it use an order of magnitude more light, say, the amount of light it can use is a factor C over PDAF.

Now, if C>P, then CDAF outperforms PDAF in speed, if the mechanical focus latency can be made small (note that focus travel is the same for both methods).

My claim is that C>P will soon be achieved.

I back up my claim by experience I gathered from the development of a high speed CDAF for an industrial customer. I use an optimization function which allows to estimate direction and amount of shift from just two measurements made in immediate succession. The details are my intellectual property, so I cannot share.

As soon as CDAF outperforms PDAF (i.e., C>P), further advantages come to fruit:
- the measurement and focus travel can happen in parallel, speeding up the operation.
- subsequent measurements can correlate the region of interest first (focus tracking) which leads to a more reliable focus performance with long lenses.
- accuracy is a good as needed, no calibration problems, color shift etc.
- High speed AF.C bursts (24+ fps) may track the focus without further measurements at all, just using the taken photos as measurement. Also needed for video-AF.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
mirrorless, sony
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sony Full Frame mirrorless prototypes still being tested, coming in mid 2014 jogiba Canon, Nikon, Sony, and Other Camera Brands 17 02-17-2013 10:31 AM
Sony Mirrorless Full Frame coming Winder Canon, Nikon, Sony, and Other Camera Brands 9 12-21-2012 11:05 PM
Sony RX-100 gets DxO'd rawr Canon, Nikon, Sony, and Other Camera Brands 12 09-08-2012 07:17 AM
Going Mirrorless, any suggestion? minahasa Pentax Q 11 09-24-2011 04:51 PM
For Sale - Sold: HUGE Minolta AF/Sony Alpha Gear Collection, Going Cheap :) wallyb Sold Items 2 12-31-2008 05:08 PM



All times are GMT -7. The time now is 07:32 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top