E-P1 + New Samsung = Death of the SLR?

First of all with a reflex viewfinder you are not using a direct through the lens view like when looking through a telescope or binoculars. You are looking at a transmitted view projected onto a ground glass/stippled plastic screen.
I know that. So what?
A lot of people posting here are claiming that reflex viewfinders are only limited by the resolution of the human eye. Once again, we are looking at a view projected onto a ground glass/stippled plastic screen and if the resolution of our eyes was that good, this is what we would see.
I have 20/13 vision with my glasses. Yet, using a viewfinder magnifier definitely shows me more detail than without. Thus, the basic viewfinder is exceeding my visual acuity.
Also it is implied that there is almost zero lag with optical systems unlike EVF systems.
Which is true.
Both photons and electrons are limited by the speed of light as to how fast they can travel. What is more that mirror has to flip up for the exposure to be made and that is not zero lag when it comes to exposures.
But I didn't claim that. I claimed the viewfinder has zero lag. I didn't claim zero shutter lag.

--
Lee Jay
(see profile for equipment)
 
I'd never use the thinnest DOF on EVERY shot, but more options is ALWAYS better.
On the other side, when you want a large DOF, the 4/3 systemn has an advantage, specially with Olys built in IS. On a FF you may have to step down your f stop so you end up with a to long exposure time for handheld photos.
Your half an eyeball in focus example only shows someone who needs to learn how to use the DOF.
With a fast price you can still get some good isolation. There does come a point when you don't need a narrower DOF.

i.e. only half an eyeball in focus??

Daniel.
--
http://www.ohb.no/foto
************
Torstein
 
On the other side, when you want a large DOF, the 4/3 systemn has an advantage, specially with Olys built in IS.
No, that's a myth, and IS lenses are available for full-frame.
On a FF you may have to step down your f stop so you end up with a to long exposure time for handheld photos.
So up the ISO - the larger sensor inherently has higher ISO capability because it's larger.

If DOF and shutter time are fixed, neither system has an advantage over the other. The large sensor low-light advantage is only present if you can allow DOF to get shallower. On the other hand, the smaller sensor never has an IQ advantage because it's smaller.

--
Lee Jay
(see profile for equipment)
 
First of all with a reflex viewfinder you are not using a direct through the lens view like when looking through a telescope or binoculars. You are looking at a transmitted view projected onto a ground glass/stippled plastic screen.
I know that. So what?
It makes a big difference to how much that image can be magnified.
A lot of people posting here are claiming that reflex viewfinders are only limited by the resolution of the human eye. Once again, we are looking at a view projected onto a ground glass/stippled plastic screen and if the resolution of our eyes was that good, this is what we would see.
I have 20/13 vision with my glasses. Yet, using a viewfinder magnifier definitely shows me more detail than without. Thus, the basic viewfinder is exceeding my visual acuity.
This is just a matter of magnification, or rather the lack of it. With magnified live view you can see detail that you could never see with a reflex viewfinder because if you magnified it that much you would be examining the structure of the stippled focusing screen.
Also it is implied that there is almost zero lag with optical systems unlike EVF systems.
Which is true.
Both photons and electrons are limited by the speed of light as to how fast they can travel. What is more that mirror has to flip up for the exposure to be made and that is not zero lag when it comes to exposures.
But I didn't claim that. I claimed the viewfinder has zero lag. I didn't claim zero shutter lag.
The viewfinder is simply an aid to taking photographs.

I wasn't just responding to your comments, but those made here and elsewhere. Whilst I'm well aware of the current limitations of EVFs they will probably get a lot better. Using live view for critical focus in field macro photographs has really brought home to me the limitations of reflex viewfinders. Yes until the issues of lag etc have been sorted out I will still be using an SLR for most stuff. However, we are taking photographs with a digital camera that records detail on it's electronic sensor and we primarily view these photos on a computer monitor. Therefore, when EVFs become better they will give a far better representation of what is being captured than an OVF.

Also whilst PDAF is currently faster than CDAF, this might not always be the case. In an interiview a while back it was clearly stated that Olympus saw no reason why CDAF could not eventually made to work just as fast, if not faster than PDAF.
 
First of all with a reflex viewfinder you are not using a direct through the lens view like when looking through a telescope or binoculars. You are looking at a transmitted view projected onto a ground glass/stippled plastic screen.
I know that. So what?
It makes a big difference to how much that image can be magnified.
Most of the time, most people don't magnify it at all.
A lot of people posting here are claiming that reflex viewfinders are only limited by the resolution of the human eye. Once again, we are looking at a view projected onto a ground glass/stippled plastic screen and if the resolution of our eyes was that good, this is what we would see.
I have 20/13 vision with my glasses. Yet, using a viewfinder magnifier definitely shows me more detail than without. Thus, the basic viewfinder is exceeding my visual acuity.
This is just a matter of magnification, or rather the lack of it. With magnified live view you can see detail that you could never see with a reflex viewfinder because if you magnified it that much you would be examining the structure of the stippled focusing screen.
That doesn't negate the fact that the TTL OVF is providing more resolution than my eye can see.
But I didn't claim that. I claimed the viewfinder has zero lag. I didn't claim zero shutter lag.
The viewfinder is simply an aid to taking photographs.
It's the way we frame photographs, and it's the way we hold focus points on subjects. On fast-moving subjects, I find that EVF lag means I can't even get the subject in the frame, much less shoot it.
I wasn't just responding to your comments, but those made here and elsewhere. Whilst I'm well aware of the current limitations of EVFs they will probably get a lot better.
They'll still use power, prevent PD AF and have lag. As far as I'm concerned, those last two are unrecoverable death blows.
Using live view for critical focus in field macro photographs has really brought home to me the limitations of reflex viewfinders.
EVFs and LCDs have use in some specialized circumstances. I use them when I have a camera mounted to my telescope, for instance. However, for more than 99% of my shots, they are very drastically inferior inherently .
Also whilst PDAF is currently faster than CDAF, this might not always be the case. In an interiview a while back it was clearly stated that Olympus saw no reason why CDAF could not eventually made to work just as fast, if not faster than PDAF.
Current PD AF systems can run at hundreds of Hz and only need one sample to determine proper current focus position, two to determine velocity and three to determine acceleration. CD AF systems need tens of samples just to get the current proper focus position. I don't see how taking tens of samples from a big, slow sensor will ever be as fast as taking one sample from a small, fast sensor.

--
Lee Jay
(see profile for equipment)
 
Honestly, I prefer the EVF of my FZ50 ( much worse than the G1 one) to the OVF of the E3, that I still cannot use to manual focus properly anyway.
I've tried the FZ50 (and 30, and 20) and the Canon S1, S2, S3, S5 and SX10 and own the S3. I've shot thousands of shots with it. I can tell you that I prefer even the Mickey Mouse pentamirror system in my Rebel XT over all of them by a factor of about, oh, a million or so.
Tastes, tastes... VFs are meant to take pictures, not to view the real world. In that respect, EVFs are infinitely more flexible.

I would prefer an EVF worse than the one of the FZ50 over the OVF of a D3, if only they eliminate any perceptible lag and add resolution. And this is not the future: this was already done in the G1.

The possibility of reviewing the pictures and histograms without taking the eye of the VF alone suffices to me. And there is so much more you can add (zooming on the VF for manual focus of a distant bird, for example), that possibilities are endless. Live view, IMO, is unusable except when using tripod.

And we have seen just the first EVIL ever!! Imagine in 5 years, for example... OVFs are just a legacy of the past. They just died.

Cheers,
L.

--
My gallery: http://w3.impa.br/~luis/photos



Oly Ee3 + 12--60 + 50--200 + EeC-14 + Oly EfEl50R
Pany FZee50 + Oly EfEl50 + TeeCon17 + Raynx 150 & 250
Nikn CeePee4500; Cann SDee500
 
Current PD AF systems can run at hundreds of Hz and only need one sample to determine proper current focus position, two to determine velocity and three to determine acceleration. CD AF systems need tens of samples just to get the current proper focus position. I don't see how taking tens of samples from a big, slow sensor will ever be as fast as taking one sample from a small, fast sensor.
Interesting argument: first time I see it exposed so clearly. Lag is also a non starter, especially in bursts.

As for the other they have been said over and over, but what can you say to people set on destroying mirrors? :)

They are preaching a Revolution, but I have serious doubts that the top brands will listen. As you, they know the physics of photography. Won't sacrifice their high end products.

Perhaps some will make toys for consumers, if they find they can mass produce them cheaper. And they will be peddled as revolutionary. So everyone will be happy.

Am.
--
Photostream: http://www.flickr.com/photos/amalric
 
Current PD AF systems can run at hundreds of Hz and only need one sample to determine proper current focus position, two to determine velocity and three to determine acceleration. CD AF systems need tens of samples just to get the current proper focus position. I don't see how taking tens of samples from a big, slow sensor will ever be as fast as taking one sample from a small, fast sensor.
Cheers Lee,

The EVF on the G1 already refreshes at 180Hz. Some people credit it's relatively fast CF-AF speed to the EVF. There are rumors talking about 250Hz for the Samsung NX. If you accept 0.2 seconds as the maximum lag that would give you 180/5 = 36 measures on the G1. 36 measures already looks overkill for a binary search, but I know almost nothing about CD-AF algorithms, so it's possible that even more could be needed. Either 10 years of P&S development weren't enough to refine these algorithms, or the communication with the lens isn't fast enough, or the bottleneck is in the lens focusing speed .

This said I would like to know: how fast can a G1 focus any of the Olympus SWD lenses ?
The way I see it there can only be developments coming our way.
--
Duarte Bruno
 
I would prefer an EVF worse than the one of the FZ50 over the OVF of a D3, if only they eliminate any perceptible lag and add resolution.
They can add resolution, but they can never eliminate lag. Adding resolution adds lag and adds power consumption. Think about what has to happen. Photons enter the lens and strike the sensor. The sensor has to accumulate enough of them for an image. The sensor must be read out. The raw data has to be demosaiced and processed (gamma, color, downsampling, etc.), and then that image has to be passed to the microdisplay. That display then has to display it. You can't do all of that without lag. You might be able to reduce that lag to tens of milliseconds but even that is too long for fast-moving subjects.
And this is not the future: this was already done in the G1.
No, it wasn't.
The possibility of reviewing the pictures and histograms without taking the eye of the VF alone suffices to me.
I don't review my pictures while I'm taking them, even on my EVF camera. Doing so means missing shots because I'm looking at the image I just took instead of observing the scene for additional shooting opportunities.

--
Lee Jay
(see profile for equipment)
 
Current PD AF systems can run at hundreds of Hz and only need one sample to determine proper current focus position, two to determine velocity and three to determine acceleration. CD AF systems need tens of samples just to get the current proper focus position. I don't see how taking tens of samples from a big, slow sensor will ever be as fast as taking one sample from a small, fast sensor.
Cheers Lee,

The EVF on the G1 already refreshes at 180Hz.
I suspect the sensor is not sampled that fast, and the images are not processed that fast, and they certainly aren't in low-light. Only the display refresh is that fast, and that is irrelevant to lag time.
There are rumors talking about 250Hz for the Samsung NX. If you accept 0.2 seconds as the maximum lag
I don't. I would accept 6ms.
that would give you 180/5 = 36 measures on the G1. 36 measures already looks overkill for a binary search, but I know almost nothing about CD-AF algorithms, so it's possible that more could be needed. Either 10 years of P&S development weren't enough to refine these algorithms, or the communication with the lens isn't fast enough, or the bottleneck is in the lens focusing speed .
It's not the lens. DPreview's "demonstration" of how fast the G1 focuses was a good demonstration of why CD is so slow. The lens is fast, the camera overshot the right point, then had to go back to find it. PD doesn't do that.

--
Lee Jay
(see profile for equipment)
 
I would prefer an EVF worse than the one of the FZ50 over the OVF of a D3, if only they eliminate any perceptible lag and add resolution.
They can add resolution, but they can never eliminate lag. Adding resolution adds lag and adds power consumption. Think about what has to happen. Photons enter the lens and strike the sensor. The sensor has to accumulate enough of them for an image. The sensor must be read out. The raw data has to be demosaiced and processed (gamma, color, downsampling, etc.), and then that image has to be passed to the microdisplay. That display then has to display it. You can't do all of that without lag. You might be able to reduce that lag to tens of milliseconds but even that is too long for fast-moving subjects.
The point is not to make zero lag, but to make the lag to be imperceptible. Your eye-brain combo is a sensor itself, and has some of the problems as an EVF has. In particular, your eye already has lag (and for the very same reasons you described above!). You just don't notice it.
And this is not the future: this was already done in the G1.
No, it wasn't.
Come on, it was. Read the reviews. True, not perfect, but indeed almost. And take into account that this is the first EVIL ever! I would have never expected the first EVIL to have such a great EVF in the first try. And all reviewers were equally surprised. Pany did such a fantastic job, that simply everyone was surprised. Simply a spectacular job.
The possibility of reviewing the pictures and histograms without taking the eye of the VF alone suffices to me.
I don't review my pictures while I'm taking them, even on my EVF camera. Doing so means missing shots because I'm looking at the image I just took instead of observing the scene for additional shooting opportunities.
I suppose it depends also on what you shoot and which are the conditions. My experience is exactly the opposite. I am tired of throwing away series of several pictures after taking them, when I review them in the LCDs and see the mess, and the bird was gone forever. With an EVF, you take ONE bad picture, the next one will be corrected thanks to the autoreview in the EVF (and if you don't want autoreview, just dissable it). And for the third pic, you're right, you don't need to review the image, but touch the shutter and the review is gone. This is seamless already in P&Ss.

Waiting for the soon to come EVIL 4/3rds flagship!! It's as inevitable as a non-gasoline car... it will just come much sooner. :)

Cheers,
L.

--
My gallery: http://w3.impa.br/~luis/photos



Oly Ee3 + 12--60 + 50--200 + EeC-14 + Oly EfEl50R
Pany FZee50 + Oly EfEl50 + TeeCon17 + Raynx 150 & 250
Nikn CeePee4500; Cann SDee500
 
The point is not to make zero lag, but to make the lag to be imperceptible.
That will require it to be in the single digits of milliseconds. Not gonna happen.
Come on, it was. Read the reviews.
I have. The text and the samples are in direct conflict with each other. The thing performs no better than most of the better P&S's.
Waiting for the soon to come EVIL 4/3rds flagship!!
Three contradictory terms.
It's as inevitable as a non-gasoline car... it will just come much sooner. :)
The non-gasoline car will be far more useful.

--
Lee Jay
(see profile for equipment)
 
The point is not to make zero lag, but to make the lag to be imperceptible.
That will require it to be in the single digits of milliseconds. Not gonna happen.
This claim is just religion.
Come on, it was. Read the reviews.
I have. The text and the samples are in direct conflict with each other. The thing performs no better than most of the better P&S's.
LOL!!!

:) :)

--
My gallery: http://w3.impa.br/~luis/photos



Oly Ee3 + 12--60 + 50--200 + EeC-14 + Oly EfEl50R
Pany FZee50 + Oly EfEl50 + TeeCon17 + Raynx 150 & 250
Nikn CeePee4500; Cann SDee500
 
that is completely right, but 4/3rds has a considerable cost and some weight advantage against 35mm....
Agreed (in some cases), but that wasn't his claim.

--
Lee Jay
(see profile for equipment)
 
The same lens 8*heavyer on FF
double the image circle 2*2*2
Uhhh...no. Image circle size often doesn't change the size of the lens at all, and where it does it's by a modest amount.

You're confusing aperture with image circle.

--
Lee Jay
(see profile for equipment)
 
The image circle FF 43,27MM vs 4/3 21.64 MM
so for the same lens all must be twice W,H,D
If something is twice the size its weight is 8 times
Berl
 
First of all with a reflex viewfinder you are not using a direct through the lens view like when looking through a telescope or binoculars. You are looking at a transmitted view projected onto a ground glass/stippled plastic screen.

A lot of people posting here are claiming that reflex viewfinders are only limited by the resolution of the human eye. Once again, we are looking at a view projected onto a ground glass/stippled plastic screen and if the resolution of our eyes was that good, this is what we would see.

Also it is implied that there is almost zero lag with optical systems unlike EVF systems. Both photons and electrons are limited by the speed of light as to how fast they can travel. What is more that mirror has to flip up for the exposure to be made and that is not zero lag when it comes to exposures.

Having said all this I still prefer the reflex viewfinder for now, except for tripod macro work where I now mainly use live view. The simple fact is that for critical focus magnified live view is in a different class to an optical view finder and you can get precision focusing that could never be achieved with an OVF. EVFs will improve and the example I gave above is clearly an example of how live view is already superior in some situations to an OVF.
My favourite analogy on this question. Would you prefer high resolution monitors instead of windows on your houses? With gain for dark days and information about temperature, weather forecast etc?

--
http://dslr-video.com/blogmag/
 

Keyboard shortcuts

Back
Top