Sony DSC-R1 - Focus speed and response a weakness?

No I was talking about the EF 50 f 1,4. And there is too many people having conclusion on a yet unseen camera. Only after the first real reviews and galleries we can see how big the R1 will be. If SONY has done carefullyl his homework with this sensor and the NR algorithm it can be really a new trend making camera. Let´s wait for the future.
--
Andre
 
Haven't you seen the various sample pictures? The iso noise is horrible on ISO 800 and up. I know these are pre-production cameras, but I doubt it'll be much different from the final product. We'll see when they are reviewed.
 
I made a simple point of fact comparing the lens on the Sony with
the lenses available from Canon. That's it. I offered no opinion on
how or whether this comparison is useful in practice,
No you didn't. You tried to use the the fact that there is no
equivalent SLR lens as an argument for the R1. I'll quote you if
you don't remember:

"Perhaps you'd care to point me to a Canon fit lens with the same
speed and range for the money?"
Sorry, a point of fact is just that, a point of fact and not an argument. It's certainly possible to use a point of fact as a basis for an argument in support of a hypothesis. I offered neither a hypothesis or argument. You decided to supply both for me and also come to my conclusion.
Now the obvious answer to this is of course that you don't need a
Canon fit lens with the same speed and range for the money, because
unlike the R1 you can combine different lenses and get even better
range and speed. Of course this will be more expensive, but it'll
also have considerably better optics and judging from the previews,
better sensor too.
"You don't need" - now you're presuming to tell me what I need from a camera! In fact you seem to be saying that whether I actually want faster lenses or a bigger range I should be delighted to pay more for them anyway!
If you are a photographer that wants the advantages of the larger
sensor, is happy with the zoom range, and is not so bothered about
action photography etc. then the Sony represents a valid
alternative to a DSLR.
Yes if you'd like to pay $1000 for a DSLR sized camera with P&S
operation.
What's your definition of P&S? The general definition seems to be an automatic camera with no (or very minimal) manual control. The R1 is a fully manual camera, not a P&S.
I suspect there's a reasonable proportion of photographers that
would relate to this. For example, an amateur that enjoys taking
landscapes and wants the level of detail offered by a larger sensor
would, if they chose a Canon DSLR, have to buy an XT/350D plus a
EF-S 10-22mm f/3.5-4.5 USM lens to get the wide angle. This would
work out considerably more expensive than the Sony.
Look, you don't need the 10-22mm for that, and for your information
it would be a 16mm equivalent as opposed to the 24mm from the R1.
If you want to take landscape pictures a zoom lens with 17 or 18mm
on the wide end is just fine on a 1.6X crop sensor. So you'd be ok
with the 350D and kitlens for a lot less money than the R1. Now the
great thing is that if this is not enough you can always buy more
lenses. Yes that'll be more expensive than the R1, but then better
quality and performance usually costs more.
Yes, for your information, you do need a a 10-22mm for that. Have you looked at the Canon lens line-up recently? If you want to get the equivalent focal length of the R1's 24mm on a 1.6 crop DSLR then your cheapest option is the 10-22mm.

The kit lens for the XT is 18mm - on a 1.6 crop this works out as 29mm not 24mm! You might be keen to gloss over this 5mm difference because it doesn't suit your hypothesis, but for a landscape photographer that 5mm at the wide end makes a load of difference.

Oh and you are wrong about the pricing too. Here in the UK the RRP of the XT plus the kits lens is £799 - that's compared to the RRP of the Sony at £699.
Plus the Sony
has 2 more MPs.
Nope it doesn't. It has LESS usable resolution than the 350D. 1,550
lines per picture height vertically, 1,650 lines horizontally as
opposed to 1850x1650 with the dirt cheap 50mm 1.8 on.
http://www.imaging-resource.com/PRODS/R1/R1A.HTM
The Sony has 2 more MPs of output - fact. Thanks for the extra info on "usable" resolution.
I'll quote some more from the preview:

"We found some of its ergonomics decidedly awkward, and both timing
performance and high-ISO noise levels aren't up to the best of the
digital SLRs currently on the market. As such, it's clearly not an
SLR-killer"

At $1000 it better be, or it's just an overpriced and oversized
compact digicam.
And thanks for quoting a small and select part of the review. Of course, the problem with quoting a selected bit from a review is that others may also check out the entire source material:

From the conclusion:

"It offers an absolutely unique focal length range..."

and

"the combination of lens and camera can't be equalled for less than twice its selling price of $999."

Now wasn't that precisely the fact I highlighted in the first place?

Simon C
http://www.eyematter.com
 
Disparaging every aspect of the R1 without any evidence... treating your own opinion like gospel...insulting the intelligence of everyone who disagrees with you...saying stupid things like 'shutter noise just does not matter' (as if it is up to you to decide for somebody else whether it matters...' who's the troll here?
 
Your rude argument.
Hehe that's what I call fuzzy logic.
Which you used in reply #1.
http://forums.dpreview.com/forums/read.asp?forum=1000&message=14957317
The mirror raising and shutter opening is done almost at the same
time, producing a single noise. And no matter how much you try to
argue semantics, it doesn't change the fact that the R1 is slower
than DSLRs. The numbers speak for themselves.
Yes, they do. 7mS shutter lag if prefocused.
So in one single situation that's not very likely to occur, it's
fast. In all other cases it's slower.
Oho, so you mean you just mash the shutter button right down for most shots, increasing camera shake and capturing your subject after some long, variable time determined by the focus and shutter lag?
 
Let's say the camera processes an image in time x and the time between refreshes is y.

Then the first frame is shown after x, at time A.

At time A, the viewfinder lag = x because it is showing an image from x time ago.

At time B, the moment before the next refresh, the viewfinder lag = x+y because the frame has stayed on the screen for y time, and the frame itself was out of date by x when it was first shown.

Then the next frame is shown at the very next moment, at time C.

At time C, the viewfinder lag = x again, as explained above.

This is how the viewfinder lag changes with time: it starts at x, goes linearly up to x+y, then goes immediately down to x again to begin another cycle.

Thus, averaging across time, the average viewfinder lag is x+0.5y.

To take a concrete example, let's say an EVF refreshes every 0.033s and the processing time before an image can be shown on the EVF is 0.01s.

The minimum lag is 0.01s (when a fresh image is shown after 0.01s of processing time)
The maximum lag is 0.043s (the moment before another fresh image is shown)

The average lag is 0.0265s (x+0.5y, or just average of minimum and maximum)
 
Disparaging every aspect of the R1 without any evidence... treating
your own opinion like gospel...insulting the intelligence of
everyone who disagrees with you...saying stupid things like
'shutter noise just does not matter' (as if it is up to you to
decide for somebody else whether it matters...' who's the troll
here?
And on you ramble. Evidence? What about the sample pictures, timing figures, size measurements etc. Or are those not valid because you don't like them? And no, to most people shutter noise does not matter. That it apparently does to you does not make the statement stupid. Rather your assertion that it is.
 
i wonder why canon who's been making prosumer cameras could not
make an APS-C size sensor prosumer camera in the past years if this
technology is so easy to do and not groudbreaking.
Unlike Sony, Canon sells two DSLR bodies in the under $1K territory, making it dangerous to confuse its market by offering a competing configuration that shares some of the key features of a DSLR. Each has its advantages and disadvantages, but the under $1K consumer may not be sophisticated enough to resolve them and they'd end up just splitting the same market between two products. Sony has no DSLR yet, just an announcement about agreement with K-M. This frees them up to experiment, to see what works. Does the DSLR consumer buy one for its resolution vs. noise advantage? It's shallower DoF? Or is it for interchangeable lenses, faster AF (phase detection will always be faster than contrast detection, especially when focus is initially way off), optical viewfinder and absolute optimization of the sensor for image quality alone? If the latter, what if some advantages are tossed in that are not found in a DSLR, such as live preview with either swiveling LCD or EVF and including a live histogram? If cost is an object, is it better to put the money into the sensor, the lens, the buffer/responsiveness or what? Sony gets to find out without harming any of its other products and can then know what to do for the future.

David
 
The EVF has a delay of 33ms. The event happens, 33ms later you see
the event, press the button, the shutter fires instantaneously.
Event to shutter = 33ms.
If the sensor is in the middle of an EVF refresh, there's probably another 33ms before the sensor finishes that, can switch to full-res capture mode and take the shot. This may mean that it's more like 66ms total delay from real-world event to capture.

If focusing is involved, you have to either have the EVF black out to avoid delay or interleave the EVF with at least two snapshots of the sensor around the currently selected AF point, with slight focus motion motion in between to compare AF snapshots. Constrast this with a DSLR requiring a single, very rapid read with no lens motion to determine the present and final focus state. In continuous focusing mode, it can actually move the lens motor while the mirror assembly is moving (once the AF reading is taken, no further readings are necessary to complete the focusing operation), collapsing two delays into one.

David
 
If you are a photographer that wants the advantages of the larger
sensor, is happy with the zoom range, and is not so bothered about
action photography etc. then the Sony represents a valid
alternative to a DSLR.
B&H-ing, I'd say that 350D + 17-85IS at 1320$ or 350D + Sigma 18-125 at 1080$ or 350D + Sigma 18-200 at 1190$ are much more attractive alternatives. And, lighter! Probably, smaller, too!
I suspect there's a reasonable proportion of photographers that
would relate to this. For example, an amateur that enjoys taking
landscapes and wants the level of detail offered by a larger sensor
would, if they chose a Canon DSLR, have to buy an XT/350D plus a
EF-S 10-22mm f/3.5-4.5 USM lens to get the wide angle. This would
work out considerably more expensive than the Sony. Plus the Sony
has 2 more MPs.
I'm sure some people would preffer the Sony. However, your landscape-taking amateur would probably prefere a prime lens. The price might be higher or lower than the 10-22 (ranging from 400 to 1800$). B&H prices.
The Sony does offer an alternative and a comparison is valid,
although certainly not for all potential users.
Agree, comparisons do help. However, such a comparison must be as complete and accurate as possible. One must factor costs, ease of use, ergonomy, weight, flexibility and value. A lens bought for 2000$ will probably sell for about the same price in 5 years. Better value.

And I can go on... external flash placement, LCD which can not be seen from beneath and so on.

And of course, no one saw images yet.

However, I do hope that it will sell good - competition is good for all. And I am very curious about image quality

d/n
 
The EVF has a delay of 33ms. The event happens, 33ms later you see
the event, press the button, the shutter fires instantaneously.
Event to shutter = 33ms.
If the sensor is in the middle of an EVF refresh, there's probably
another 33ms before the sensor finishes that, can switch to
full-res capture mode and take the shot. This may mean that it's
more like 66ms total delay from real-world event to capture.

If focusing is involved, you have to either have the EVF black out
to avoid delay or interleave the EVF with at least two snapshots of
the sensor around the currently selected AF point, with slight
focus motion motion in between to compare AF snapshots. Constrast
this with a DSLR requiring a single, very rapid read with no lens
motion to determine the present and final focus state. In
continuous focusing mode, it can actually move the lens motor while
the mirror assembly is moving (once the AF reading is taken, no
further readings are necessary to complete the focusing operation),
collapsing two delays into one.

David
Hi David,

But when you consider the shutter lag and the mirror movement time with the average dSLR it's about a wash. Perhaps there may be a "measurable" difference in event to acquisition time between a decent dSLR and a fast EVF/fixed lens prosumer like my Sony F828, but in real life operation I see little, if any difference.

With the majority of my mediocre lenses, both with Canon, Nikon and Sigma, autofocus time is the major issue. With slower autofocus lenses such as my Sigma 80-400 OS (which is optically a very fine lens) the autofocus is so slow that my F828 can actually autofocus, acquire, autofocus and acquire two frames while the 80-400 is locking focus.

With my fastest dSLR's (Canon EOS-1D and EOS-1DS) and a really good lens such as my 400mm F2.8, the whole process is somewhere around 58-65 ms with optimal lighting. This is faster than my F828 under idenical circumstances. But with an "average" lens even with my 1D and 1DS bodies I perceive little differences.

On the other hand, with some of my other fixed lens digicams there are annoying delays of up 1 second or even more, the point being that the really fast fixed lens proumers are generally little different in practical use than the average dSLR in terms of autofocs and first frame acquition. Frame to frame - that's another issue unless shooting in the burst mode. Actually my "fastest" camera in the burst mode is a fixed lens digicam which shoots fifteen frames per second at full resolution for a burst of about 30 frames with a good, fast CF card (Olympus E100-S).

Lin
 
With the majority of my mediocre lenses, both with Canon, Nikon and
Sigma, autofocus time is the major issue. With slower autofocus
lenses such as my Sigma 80-400 OS (which is optically a very fine
lens) the autofocus is so slow that my F828 can actually autofocus,
acquire, autofocus and acquire two frames while the 80-400 is
locking focus.
Since the difficulty in locking focus varies not only with technology scheme (i.e. phase vs. contrast detection) but also with lens angle of view, maximum aperture and depth of field (linked to the preceding two), not to mention type of lens motor, your example points out that we'll need to try to form good apples to apples comparisons as these debates evolve (and there'll be many, many more of them over the next several months, I'd imagine!). Perhaps a good vehicle for comparison would be Canon's 17-85mm mounted on a 350D and measured at 17mm and 75mm (two focal lengths that are shared by both cameras). With a good sale, this combo is about $1200, in the R1's ballpark, and the lens covers a very similar FL range (29-136mm equivalent vs. 24-120mm).
the whole process is somewhere around
58-65 ms with optimal lighting.
I'm curious, how exactly are you measuring such short delays with millisecond accuracy? This reminds me of voice mail messages from my mom. "Hi David. It's me - mom. I'm calling at approximately 6:34pm" :-)

David
 
DRG wrote:
snip
the whole process is somewhere around
58-65 ms with optimal lighting.
I'm curious, how exactly are you measuring such short delays with
millisecond accuracy? This reminds me of voice mail messages from
my mom. "Hi David. It's me - mom. I'm calling at approximately
6:34pm" :-)

David
Hi David,

You'll have to ask Canon - it's their measurements based on several dSLR models, not mine :-)

Lin
 

Keyboard shortcuts

Back
Top