DSLR sensor facts with no acrimony please

joe mama · Mar 16, 2008

Mark Thornton said:
While a 27-200/5-8 for FX would be about the same weight as the
18-135, you can't actually buy one. You have to make do with the
rather larger 24-120 instead.

Yep.

Mark Thornton said:
So in practice you need to determine your needs and the set of REAL
lenses for DX and FX that would meet those needs. In many cases the
DX set will be lighter and perhaps cheaper than the FX set, although
the FX set may over provide against your requirements.

Absolutely correct. The available glass for each system plays a primary role in determining which system works best for you. But so many compare glass in terms of FOV only, rather than figuring DOF/light gathering ability into the equation.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

joe mama · Mar 16, 2008

Both the glass and the sensor play critical roles. Neither by themselves will capture the image. The relationship between glass and sensor needs to be understood.

This might help:

http://www.josephjamesphotography.com/equivalence/#hypothetical

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

John Sheehy · Mar 16, 2008

gaussian blur said:
The pixels on the Canon 5D are 8.2 microns, but the difference is
that the well capacity on the Foveon sensor is divided up into three
slices, with the top layer only capturing about 1/8th as many photons
as the entire 5D pixel, or nearly a 3 stop penalty. Since the top
layer is mostly the blue channel, the sensor tends to be weak in
tungsten light. There's also a lot of overlap in the spectrum of the
three layers, and converting from Foveon-space to RGB adds additional
noise.

If what you say is true, that the top layer of the Foveon 4.6MP only gets 1/8 as many photons as a 5D pixel, then even less of that is available for blue, as the red and green need to be scaled and subtracted from it.

gaussian blur said:
At low ISO, most DSLRs these days are limited by shot noise, not read
noise.

Limited for what? I've heard that statement from Roger Clark, too, and I think it sounds a bit strange, and gives false connotations. Read noise is the biggest problem in low ISOs, not shot noise. Shot noise is very nice when lots of photons are collected, and if a DSLR had only shot noise at base ISO, it would have have about 2 to 3 stops more everyday DR, and several stops more ability in DR with shadows that are low-frequency and high-contrast. Here is a glimpse of what pure shot noise would be like. There are 3 images in this worksheet; the upper right is a synthetic original, it represents 256 linear grayscale levels. The image on the lower left is the image after being processed to receive shot noise with Mathcad's builtin poisson function rpois(). The lower right is the shot-noise-treated image with read noise added to it with rnorm(). The figures I am using here represent a camera with a RAW saturation of 80000 photons at base ISO of 100, 3.5 stops of highlight headroom, and a read noise typical of large-pixel CMOS DSLRs like the high-end Canons and the D3 (about 26 electrons). This would be ISO 204,800:

This would be ISO 3,279,000:

This would be ISO 52,630,000:

--
John

marf · Mar 16, 2008

joe mama said:
Both the glass and the sensor play critical roles. Neither by
themselves will capture the image. The relationship between glass
and sensor needs to be understood.

This might help:

http://www.josephjamesphotography.com/equivalence/#hypothetical

I quote joe mama: "The advantages of the lenses for the F-series system ( a assumed upscaled full frame E-3 ) is that they would be lighter (the two stops smaller aperture affects the weight more than the larger image circle and longer FL) and less expensive (it is much less expensive to make a lens that has double the image circle and is half as sharp). "

I doubt it. You get lenses that are 2 times as long but have the same diameter. So the weigth may be approx 2 times the weight of zuiko 4/3 lenses with the same performance. And the Full frame sensor would not necessary have the same color quality but 4x the sensitivity because its not easy to upscale electronic devices without making compromises ( compromise = less manufacturing steps, because all is extremely much more expensive at the same quality, not just 4x, i assume 20x ). And the body would be approx. 4 times the weight: assume a 100mm diameter of the mount. this is needed at fullframe for high quality wide angle lenses. assume also a huge antishake motor to move the 4x weight of the sensor unit. For this also huge batteries are needed.

regards
Martin

-----------------------------

E - 3 3 0 + E - 5 0 0 + Z U I K O D I G I T A L 8 m m F 3 . 5 F i s h e y e + ZD 1 1 - 2 2 F 2 . 8 - 3 . 5 + ZD 1 4 - 5 4 F 2 . 8 - 3 . 5 + ZD 1 4 - 4 5 F 3 . 5 - 5 . 6 + ZD 4 0 - 1 5 0 m m F 3 . 5 - 4 . 5 + S i g m a 3 0 F 1 . 4

mschira · Mar 16, 2008

Well, the tiltle says it all...
M.

joe mama · Mar 16, 2008

marf said:
marf said:

Both the glass and the sensor play critical roles. Neither by
themselves will capture the image. The relationship between glass
and sensor needs to be understood.

This might help:

http://www.josephjamesphotography.com/equivalence/#hypothetical

Click to expand...

marf said:
I quote joe mama: "The advantages of the lenses for the F-series
system ( a assumed upscaled full frame E-3 ) is that they would be
lighter (the two stops smaller aperture affects the weight more than
the larger image circle and longer FL) and less expensive (it is much
less expensive to make a lens that has double the image circle and is
half as sharp). "

But you leave off the last sentence of the paragraph your quote:

"The disadvantage of the F-series lenses is that they would be physically longer."

So if you're going to purposefully misrepresent what I'm saying, you're already off to a bad start.

marf said:
I doubt it. You get lenses that are 2 times as long but have the same
diameter. So the weigth may be approx 2 times the weight of zuiko 4/3
lenses with the same performance.

Counterexamples:

35-100 / 2 on 4/3 vs 70-200 / 4L IS on 35mm FF
150 / 2 on 4/3 vs 300 / 4L IS on 35mm FF

There are no other lenses that are two stops apart to compare with. But I can cite examples between 1.6x and 35mm FF, if you'd like.

marf said:
And the Full frame sensor would not necessary have the same color quality

Pass the pipe you're smokin' -- or explain, specifically , how that works out.

marf said:
but 4x the sensitivity because its not easy to upscale electronic devices
without making compromises ( compromise = less manufacturing steps,
because all is extremely much more expensive at the same quality, not just
4x, i assume 20x ). And the body would be approx. 4 times the weight:
assume a 100mm diameter of the mount. this is needed at fullframe for
high quality wide angle lenses.

All you do is misrepresent what other people say, and make unfounded assumptions. Why didn't you bother to quote this paragraph:

"Also because the pixels are twice the dimensions as the pixels of the E3 sensor, the glass for the F3 only needs to be half as sharp to resolve them. Thus, to achieve equivalent images with an F3, you would use lenses double the focal length and half as fast with half the optical sharpness, but they would still produce images equally as sharp as the E-system since the glass need not be as sharp to resolve the larger pixels."?

Oh, I know: because you didn't even quote the entirety of the other paragraph, thus committing a lie of ommission.

marf said:
assume also a huge antishake motor to move the 4x weight of the sensor unit.
For this also huge batteries are needed.

You do a fine job assuming without any facts to back them up. And you do a fine job of misrepresenting what other people say. Do you work for the Bush Administration, by chance? If not, you might look into a good paying job with them.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

TrapperJohn · Mar 16, 2008

TaylorHobson said:
Fact 3 viewfinder size.
FF with larger mirror will always be larger and brighter than APS-c
and so forth with 4/3 always having the smallest view using same
priced optics, mirror or prism arrangements.

But they don't. Focusing screen has a lot to do with this. I put a Katzeye on my E1, and it was distinctly brighter. Also gave it a nice retro look, with the split prism and focusing ring.

TaylorHobson said:
Fact 4
The lensmount of the FF and APS-c are same size and 4/3 is probably
in middle, smaller than EOS but larger than Nikon F. Ultimate camera
size is determined by lens mount size so all current dslr formats
will have similar size bodies.

Um, registration distance is a major factor. It is the distance from the sensor the lens must be for optimal performance, which influences body thickness. APS-C uses 35mm registration distance, 4/3 has a shorter one, which is a primary reason the E420 is so small. Side benefit is that you can adapt most any FF legacy lens to a 4/3 camera with good results, the adapter brings the lens back to 35mm registration distance. (except Canon, due to the aperture mechanism) I routinely use a pair of old Nikkors on 4/3, and get superb results, even after manual focus and stopdown metering.

TaylorHobson said:
Fact 5. Why APS-c was introduced

I believe a primary reason for APS-C sensors was soft edge problems encountered with legacy film lenses. Before commercial APS-C, there were pro level dslr conversions from Kodak that used a 24x36 sensor.

TaylorHobson said:
Fact 6 lens ability, size and cost
People think cropped sensors allow you to use smaller lenses for same
focal length equivalent.

In some circumstances, they do. Since APS-C is a crop of a 35mm image circle, some light is thrown away. 4/3, on the other hand, has an image circle optimized for the sensor. My 50-200 F2.8-3.5 is quite a bit shorter and lighter than a Canon 100-400L, and faster as well. Another benefit of the smaller sensors is the increased DOF. For W/A, that's a drawback, but for long telephoto, where you're battling to get enough DOF, it's a distinct advantage.

TaylorHobson said:
Fact 7 New category of cameras will come soon as true alternative to
FF dslrs.

They already exist. They're called APS-C and 4/3 cameras. If you've seen the comparisons of FF vs the best smaller sensors recently, there isn't much difference in IQ and capability any more, at least within practical limits. No, they won't go to ISO12000+, but that's a fairly specialized need. Now, it's down to better DOF control versus better telephoto performance and smaller/lighter lenses for an equvialent DOF/aperture.

And you're not even considering the glass itself. We're just beginning to see digitally optimized FF glass with a near telecentric light path. APS-C crops out the troublesome edges, while 4/3 glass is near telecentric by design.

Theoretically, FF will do better at wide angle. Realistically, there are only two really top rate ultra wide angle lenses on the market today - the Nikkor 12-24, and the Olympus ZD 7-14. They cost about the same - around $1600. Only problem with the Nikkor is you need a $5k D3 to get the full wide angle. My 7-14 sits on a $300 E-330.

Rriley · Mar 16, 2008

in moving out of the film era into the digital age, some principle differences were discovered. It was found that sensors did not respond like film does, that sensors unlike film cannot handle angular light, or alternately they require the light path to be straighter.

Quantitatively this can be said to be extreme in some cases, where RF lenses can be as much as 60 degrees off parallel light path, but more commonly, film based lenses exhibit angles of around 20 degrees off parallel.

Microlenses, an element of glass placed over a photosite receptor are meant to alter this behaviour, but have limits to their operational usefulness, and can work out worse when those limits are exceeded. This would alter the light pattern so that they are directed slightly inward at the edges of the sensor, perhaps 6 degrees. Meaning a sensor that could tolerate light up to 12 degrees from parallel could now tolerate light up to 18 degrees from perpendicular in the corners.

But since it also needs to include the light path from telephoto lenses which do not carry this issue this would mean tele lenses now also carry divergent light of around 6 degrees, but this has been successfully contained below the 12 degree limit. The limits must be contained between UWA and tele lenses or errant behaviour is exhibited.

The costs to this come in some forms of vignetting, in soft edges, in light fall-off, flare and ghosting and decreased contrast. These costs are real and undeniable.

APSC as it uses the centre portion of the older lens suite escapes some of the aberrations, and this in itself is enough to reduce the angle of the light path to a more controllable 12-13 degrees, but this is still on the verge of control for microlenses, and particular cases will have a propensity to step outside the verge of control more often.

4/3rds went another road to tackle this same issue. Olympus noted that APSC also altered the ratio between the sensor diagonal and the distance to the exit pupil (since canon and "" were still using film based lenses this was determined at 52mm) Olympus surmised that if they pushed the exit pupil further away from the sensor to around 85mm, this would have the effect of reducing the light angle. The trapezoid created that describes telecentricity is between the sensor diagonal on one side, the distance from the sensor to the exit pupil to the sensor in length, and the width of the open aperture within a given lens, and this is why stopping a lens down succeeds in limiting these issues.

In the case of 2x crop this is contained to 6 degrees, wide open. The lenses created had the benefit of using more of the available glass with little waste in area, and the ability to shoot wide open while maintaining control of aspects of vignetting, soft edges, light fall-off, flare and ghosting and decreased contrast.

This could also be done in APSC, but that means digital specific lenses for a given format, hence the arrival of EFS, but EFS does not tackle telecentric properties in any deliberate attempt, it merely limits the glass area used so less is wasted. However nonetheless, its a strategy that works because it alters the ratio of telecentricity, (sensor diagonal) but again within limits. Since APSC in either major format doesnt have the widest of UWA lenses, this is of some assistance here too. It is left to the reader to assume why that is.

As we examine the optical efficiency of FF, APSC, 4/3rds, we need to keep in mind that the former are more used to stopping down than the latter, with good reason. To my mind, when this gets all out of shape is when the stop down of the F ratio exceeds the stops difference in sensor size. Where if for instance an UWA lens is stopped down to avoid known issues of vignetting, soft edges, light fall-off, flare and ghosting and decreased contrast on a FF body beyond 2 stops equivalence it is optically inefficient in this use.

It is all well and good to suggest that equivalence holds court over all other methods of comparison, but eventually the simple metering of an exposure that allows for F2 at 35mm EFL, will meter at F2 on any system, and is an operational advantage where DoF 'is at that point an acceptable consideration', and where frame consistency without aberration can still be maintained.

In such a case, F2 does not equal F4, the metering is simply adjusted to assist comparison, and it is not a true reflection of the limits a system is beholding to, when a personal preference dictates that edge sharpness or other aberrations holds no relevance over judgements of DoF.
--
Riley

I'd give my right arm to be ambidextrous (just)

Anastigmat · Mar 16, 2008

TaylorHobson said:
DSLR sensor mania, Just the facts. What I am posting here is
everything I learned on DPreview. You can check the facts yourself by
doing appropriate seaches. Only Fact 7 is made up and represents only
my personal opinion and preference with no claim to being superior in
any particular photo imaging aspect.

The longest threads here are those involving the defense of APS-c and
4/3 sensor sizes versus full frame. Here are some plain facts.

Fact 1 Size.
FF is 2 times larger (area wise) than APS-c. APS-c is 2 times larger
than 4/3. So FF is 4 times larger than 4/3.

TaylorHobson said:
For PS (point and shoots) people who are confused, PS cameras are
compact because FF is 22 times larger than sensor chip used in G9. FF
is 36 times larger than sensors used in typical pocket Powershot PS
types.

Fact 2 light sensitivity
Since all manufacturers try to keep up with the megapixel race so for
same 12 mp sensor constant for each format, a 2x larger pixel in FF
sensor will gather 2x the light of APS-c and 4x light of 4/3. So if
4/3 has max iso of 1600, then APS-c will be 3200 and FF will be 6400
at same noise level using same sensor technology. This is exactly
what you see in today’s crop of cameras

Fact 3 viewfinder size.
FF with larger mirror will always be larger and brighter than APS-c
and so forth with 4/3 always having the smallest view using same
priced optics, mirror or prism arrangements.

Fact 4
The lensmount of the FF and APS-c are same size and 4/3 is probably
in middle, smaller than EOS but larger than Nikon F. Ultimate camera
size is determined by lens mount size so all current dslr formats
will have similar size bodies. Right now FF is much larger but
remember as technology improves, budget consumer FF will come down in
size. Canon’s last budget film FF the rebel Ti is same size as
current Oly E430 claimed to be the smallest dlsr.

Of course a camera cannot be smaller than the lens mount, but camera size has little relationship with sensor or lens mount size. Until recently, 4/3 format cameras are actually bigger than APS-C sized cameras, despite the smaller lens mount and smaller sensor size. The reason full frame models like the 1DS and D3 are bigger is because they are professional models, and for whatever reason these cameras are bigger. The D2X has a much smaller sensor than the D3, but it is about the same size as the D3. The D300 is much smaller than the D2X. There is no reason (other than marketing) why the D3 cannot be as small as a D40.

TaylorHobson said:
Fact 5. Why APS-c was introduced
APS-c came about only due to the very high sensor prices in the
infancy of dslrs. A FF sensor was priced many times higher than an
APS-c or 4/3. When sensor prices fall as they inevitably will, FF
cameras may cost maybe only 10% more than similar megapixel APS or
4/3. Higher pixel count regardless of sensor size will add to cost
because faster chips have to be used throughout to keep fps from
slowing.

Sensors did cost a lot, especially full frame sensors, and they have come down in price due to mass production and the use of cheaper processes like CMOS. But due to marketing reasons, full frame prices may be kept artificially high for some time to come because of lack of competition and because of the desire to force buyers to buy APS-C models.

TaylorHobson said:
Fact 6 lens ability, size and cost
People think cropped sensors allow you to use smaller lenses for same
focal length equivalent. If you make pixel size same which means that
larger sensors will have more pixels, the same lens will produce the
same detail so no detail is gained by the smaller sensor. In any
case, the lens sizes of all three sensor sizes are similar. A typical
50mm f1.4 FF prime is same size or smaller than an equivalent 35mm
f1.4 APS-c prime or a 25mm f1.4 4/3 prime. The cost of lenses in any
sensor size in equivalent focal length are all in same range.

Fact 7 New category of cameras will come soon as true alternative to
FF dslrs.
There will be an interchangeable lens 16:9 hi-resolution (svga
800x600) evf cropped sensor (16mm x 9mm no more than 12 mp) with
choice of 3:2 and 4:3 with new standard lens mount. Since these
cameras have no mirror box, lens can be optimized for thinner bodies
and closer position to sensor. These new cameras will be logical move
up for current PS crowd desiring better quality images and more
flexibility like interchangeable lenses and HD movie taking
capability. No mirror box, smaller sensor than current 4/3, these
cameras will live up to truly more compact and lighter tourist travel
cameras. Canon can make this the G11 with availability of pancake
wide angle prime. Or how about Panasonic TZ9. The current TZ5 already
has a true 16:9 with choice of 3:2 and 4:3. Note the TZ5 uses a lens
circle optimized for 16:9 and 3:2 .16:9 (3712x2088) is wider than 3:2
(3552x2368). Both canon and pano should switch to body based IS like
sony, pentax and oly to keep lens cost down and more rugged.

"Fact 7" is a prediction. And predictions are notoriously unreliable. Do not call predictions "facts."

joe mama · Mar 16, 2008

Rriley said:
The costs to this come in some forms of vignetting, in soft edges, in
light fall-off, flare and ghosting and decreased contrast. These
costs are real and undeniable.

If we compare at the same aperture/DOF, the exteme corners of 35mm FF do take a minor hit. However, this isn't due to the reasons you cite above, it is due the to sudden MTF falloff of UWA lenses near the edges of the image circle. If we compare corners of lenses that do not have this MTF dropoff, we will not see the softer corners. For example, take a look at the MTFs for the 16-35 / 2.8L II:

http://www.usa.canon.com/consumer/controller?act=ModelInfoAct&fcategoryid=148&modelid=14907

Contrast that with the Olympus 40-150 / 4.5.6 and the Canon 70-300 / 4.-5.6 IS, again, at the same aperture (use double the f-ratio on the Canon lens to get the same aperture on the Olympus lens).

Here are the MTFs:

http://www.usa.canon.com/consumer/controller?act=ModelInfoAct&fcategoryid=150&modelid=11922

Here's the review of the Olympus 40-150 / 4-5.6:

http://www.slrgear.com/reviews/showproduct.php/product/1086/cat/15

Here's the review of the Canon 70-300 / 4-5.6 (make sure to see FF results):

http://www.slrgear.com/reviews/showproduct.php/product/253/cat/11

The Canon zoom, with its flat MTF, outperforms the Olympus zoom at the same aperture/DOF, even in the corners (although the two are very close regardless -- I certainly wouldn't choose one over the other on the basis of the test).

Rriley said:
As we examine the optical efficiency of FF, APSC, 4/3rds, we need to
keep in mind that the former are more used to stopping down than the
latter, with good reason.

Yes -- the good reason is to use the same DOF.

Rriley said:
To my mind, when this gets all out of shape is when the stop down of the
F ratio exceeds the stops difference in sensor size. Where if for instance an
UWA lens is stopped down to avoid known issues of vignetting, soft edges,
light fall-off, flare and ghosting and decreased contrast on a FF body beyond
2 stops equivalence it is optically inefficient in this use.

What?! That makes no sense whatsoever. You stop down for different DOFs. You change the ISO for different shutter speeds. For example:

Shallow DOF:

http://forums.dpreview.com/forums/read.asp?forum=1029&message=26773972

Deep DOF:

http://www.pbase.com/joemama/24l

The corners/vignetting mean nothing on the shallow DOF pics, and the corners are perfectly sharp and the vignetting non-existent on the deep DOF pics. The difference is that 4/3 cannot achieve the same shallow DOF as 35mm FF for the same perspective and FOV, but 35mm FF can go just as deep as 4/3 and have sharp corners without vignetting.

Rriley said:
It is all well and good to suggest that equivalence holds court over
all other methods of comparison, but eventually the simple metering
of an exposure that allows for F2 at 35mm EFL, will meter at F2 on
any system, and is an operational advantage where DoF 'is at that
point an acceptable consideration', and where frame consistency
without aberration can still be maintained.

Again, what does that mean? For the same shutter speed, f/2 on 4/3 puts exactly the same amount of light on the sensor as f/4 does on 35mm FF, and the DOFs are the same, too. Assuming equally efficient sensors, then for the same level of detail, the noise will also be the same, despite the fact that 35mm FF has to use two stops higher ISO.

Proof:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=25380951

Lots more proof:

http://www.josephjamesphotography.com/equivalence/#evidence

Rriley said:
In such a case, F2 does not equal F4,

While f/2 does not equal f/4, just as 25mm does not equal 50mm, just as 25mm on 4/3 has the same FOV as 50mm on 35mm FF, for the same persective, FOV, and shutter speed, f/2 on 4/3 puts the same light on the sensor as does f/4 on 35mm FF, has the same DOF, and, assuming equally efficient sensors, has the same noise. And that is all that matters in terms of creating the image.

Rriley said:
the metering is simply adjusted to assist comparison, and it is not a true
reflection of the limits a system is beholding to, when a personal preference
dictates that edge sharpness or other aberrations holds no relevance over
judgements of DoF.

No. Edge performance and vignetting is determined by the MTF dropoff. And the fact that f/2 on 35mm FF puts four times the light on the sensor as f/2 on 4/3, that just gives 35mm FF that much more of an advantage in low light, assuming it is acceptable to sacrifice DOF (and thus edge performance and vignetting) to get it. But it is an option available to 35mm FF that is not available to 4/3.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

jrahadi · Mar 17, 2008

joe mama said:
Both the glass and the sensor play critical roles. Neither by
themselves will capture the image. The relationship between glass
and sensor needs to be understood.

This might help:

http://www.josephjamesphotography.com/equivalence/#hypothetical

In the film days indeed quality lens is much more critical than bodies.
However todays digital realm dictates that lens is only part of the equations:

Subject + Lighting + Lens + Exposure + Stabilization (Tripod/VR/IS/SR) + Sensor + Processing Engine = Image Quality

Deficiency in any of above factor will make IQ suffer.

Nowadays Processing Engine control many factors to IQ: White Balance, Shake Reduction, Noise Removal, Sharpening, JPEG engine, etc.
But future Processing engine may include DXO like corrections & processing.

joe mama · Mar 17, 2008

jrahadi said:
In the film days indeed quality lens is much more critical than bodies.

That's because all cameras used the same "sensor".

jrahadi said:
However todays digital realm dictates that lens is only part of the
equations:
Subject + Lighting + Lens + Exposure + Stabilization
(Tripod/VR/IS/SR) + Sensor + Processing Engine = Image Quality

Deficiency in any of above factor will make IQ suffer.

I prefer to define IQ differently as your definition does not distinguish between image quality and a quality image:

http://www.josephjamesphotography.com/equivalence/#IQ

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Rriley · Mar 17, 2008

i dont care if i get banned for this, this guy is a fake, and here he adds to his own story and people should see it for what it is. I make a simple notation that equivalence is everything in comparisons between systems, and he simply wont accept that, he cannot have that stand, for him equivalence is the only way

he wants you to consider an Olympus kit lens, the 40-150, with the 'theoretically possible' results from Canons own site on the Canon 70-300 / 4-5.6

where everywhere on the net, its well understood that canons UWA are hopelessly outgunned in every possible dimension, this then becomes ' a small hit'
well BS to that

and describing, and i quote myself

joe mama said:
As we examine the optical efficiency of FF, APSC, 4/3rds, we need to
keep in mind that the former are more used to stopping down than the
latter, with good reason.

becomes yet another inclusion of the old DoF tripe

gone are the f aspects of vignetting, soft edges, light fall-off, flare and ghosting and decreased contrast in every phrase he quotes. He wants you to believe, they simply dont exist. Well thats a lie, because they do, and this is irrefutable

and even given a 'satisfied DoF' requirement, F2 still isnt F2, it becomes a discussion about light on sensors, which although true, is not in alignment with the view that glue gun lenses NEED stopping down because of vignetting, soft edges, light fall-off, flare and ghosting and decreased contrast. He wants/needs you to believe that those problems magically go away because they have 4x the light. Well thats rubbish, if you have an UWA lens made by a glue gun factory, its just going to be junk wide open

this guy will fake it any way he can, and he wont give up on it. The inclusion of the kit lens MTF is absolutely the last straw on this guy, its a sign of a man who cannot conceive even a moderate concession on technically valid discussion.

He will trash this to the living end be sure of it, and he will edit out the bits he finds hard to argue thats his pattern, he always does this, but its a gutless attempt to discredit anyone who disagrees with him, or who conflicts with the almighty religion of DoF.
--
Riley

I'd give my right arm to be ambidextrous (just)

joe mama · Mar 17, 2008

Rriley said:
i dont care if i get banned for this, this guy is a fake, and here he
adds to his own story and people should see it for what it is.

You don't have to take my word for it, you can look at pictures:

http://www.josephjamesphotography.com/equivalence/#evidence

Or you can go to http://www.slrgear.com and compare lenses at the same FOV and DOF (35mm FF uses twice the f-ratio for the same aperture/DOF/light-gathering).

Rriley said:
I make a simple notation that equivalence is everything in comparisons
between systems, and he simply wont accept that, he cannot have that
stand, for him equivalence is the only way

Give me an example as to why you would not compare two systems with the same perspective, FOV, DOF, shutter speed, and the same output size. Let me break it down for you:

1) Show me an image where it makes sense to compare corners from two systems at different DOFs.

2) After you've shown me that image, explain to me why you couldn't have compared the images at the same DOF.

3) If you are comparing at different output sizes, tell me why it's unfair to downsample the larger image to the same dimensions as the smaller image, or upsample the smaller image to the same dimensions as the larger image.

4) If you are comparing noise, explain to me why you would compare the noise in images that had different levels of detail, since you could simply apply NR to the more detailed and noisy image to match the detail of the lesser detailed image.

Rriley said:
he wants you to consider an Olympus kit lens, the 40-150, with the
'theoretically possible' results from Canons own site on the Canon
70-300 / 4-5.6

No. I want you to go to http://www.slrgear.com where the lenses were tested and compare them, since they have the same FOV (the max aperture on the Canon lens, however, is twice as big, but that doesn't matter since we are comparing at the same aperture, anyway).

Rriley said:
where everywhere on the net, its well understood that canons UWA are
hopelessly outgunned in every possible dimension, this then becomes '
a small hit' well BS to that

Yet you can't post a single image to support your claim. Here are two fullsize images from a Canon at 24mm f/8 (12mm f/4 on 4/3):

http://www.pbase.com/joemama/24l

and here is a comparison of a FF and 4/3 image at fully equivalent settings (28mm f/11 ISO 800 on 35mm FF and 14mm f/5.6 ISO 200 on 4/3):

http://forums.dpreview.com/forums/read.asp?forum=1022&message=25380951

Rriley said:
and describing, and i quote myself

Rriley said:
Rriley said:

As we examine the optical efficiency of FF, APSC, 4/3rds, we need to
keep in mind that the former are more used to stopping down than the
latter, with good reason.

Click to expand...

Rriley said:
becomes yet another inclusion of the old DoF tripe
gone are the f aspects of vignetting, soft edges, light fall-off,
flare and ghosting and decreased contrast in every phrase he quotes.
He wants you to believe, they simply dont exist. Well thats a lie,
because they do, and this is irrefutable

I have images to back up my claims, linked above. You have your good word -- in other words, nothing.

Rriley said:
and even given a 'satisfied DoF' requirement, F2 still isnt F2, it
becomes a discussion about light on sensors, which although true, is
not in alignment with the view that glue gun lenses NEED stopping
down because of vignetting, soft edges, light fall-off, flare and
ghosting and decreased contrast. He wants/needs you to believe that
those problems magically go away because they have 4x the light. Well
thats rubbish, if you have an UWA lens made by a glue gun factory,
its just going to be junk wide open

Once again, I have images to demonstrate that fact, http://www.slrgear.com bears out that fact, and you have zip, ziltch, nada to support your claim.

Rriley said:
this guy will fake it any way he can, and he wont give up on it. The
inclusion of the kit lens MTF is absolutely the last straw on this
guy, its a sign of a man who cannot conceive even a moderate
concession on technically valid discussion.

Produce an image that demonstrates the contrary. Produce one. But you want to show images at the same f-ratio from different systems, when, as we know, the f-ratio scales in exactly the same manner as the focal length. In other words, "f/2 = f/2" is no more true than "50mm = 50mm".

Rriley said:
He will trash this to the living end be sure of it, and he will edit
out the bits he finds hard to argue thats his pattern, he always does
this, but its a gutless attempt to discredit anyone who disagrees
with him, or who conflicts with the almighty religion of DoF.

Comparing systems with the same perspective, FOV, DOF, shutter speed and output size is "Gutless"? Producing images that demonstrate the claims is "Gutless"? Producing links to lens reviews that support the claims is "Gutless"? And you, just spouting from your mouth, without a single comparison image is what?

Tell me, Mr. Photographer, what do sharp corners mean to you? I mean, what could you possibly care about them:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=27003672

Tell you what -- when you get pics with corners like these:

http://www.pbase.com/joemama/24l

Post'em and get back to me. Until then, you're all talk, and not even good talk at that.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Rriley · Mar 17, 2008

joe mama said:
No. I want you to go to http://www.slrgear.com where the lenses were tested
and compare them, since they have the same FOV (the max aperture on
the Canon lens, however, is twice as big, but that doesn't matter
since we are comparing at the same aperture, anyway).

dont pass it off on them
YOU, are comparing a KIT lens
do you even know how biased and corrupt this argument is?

im not interested in discussing my views with you, its pointless
but people should know a fake when they see one

--
Riley

I'd give my right arm to be ambidextrous (just)

joe mama · Mar 17, 2008

Rriley said:
dont pass it off on them
YOU, are comparing a KIT lens
do you even know how biased and corrupt this argument is?

Choose your lens -- any lens. I don't care. I just chose two lenses with the same FOVs that were tested on 4/3 and FF to compare. Browse the site and find a different comparison. Just do so at the same aperture .

Rriley said:
im not interested in discussing my views with you, its pointless
but people should know a fake when they see one

You mean you're not interested in posting evidence. What's really weird, though, is that for all your talk about how important sharp corners and vignetting are, you don't post a single image where sharp corners or vignetting really matters, whereas I have, and on 35mm FF at full size , and 35mm FF came up roses. Yet you say that the corners of 35mm FF suck, when I've posted direct evidence to the contrary. Again, your pics:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=27003672

My pics:

http://www.pbase.com/joemama/24l

Feel free to post other pics that you feel represent your photography better than that link, though. It's hard to find threads where you've posted pics, and you've no gallery linked. I don't want to unfairly say that the above link is representative of your work and the importance of corners and vignetting.

So, when you can post an image where the corners matter, and 35mm FF could not have gotten the shot, get back to me. I'll even help you out: since 4/3 has in-camera IS, for a low-light pic without a tripod it can use a much deeper DOF without suffering motion blur. A definite advantage for 4/3 -- but because of in-camera IS.

So, c'mon now, sharp corners matter oh-so-much to you. Post a pic. 'Cause sharp corners don't mean squat to me, yet I can get them with my "uselessly flawed" 35mm FF DSLR. Weird, huh?

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Rriley · Mar 17, 2008

joe mama said:
Tell me, Mr. Photographer, what do sharp corners mean to you? I
mean, what could you possibly care about them:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=27003672

yeah sure, what you really need is a gutless attack, i explained that here,
http://forums.dpreview.com/forums/read.asp?forum=1022&message=27015267

"sometimes however, the 'painted light' doesnt quite index properly against the building, creating the impression of camera movement as this closer image from an F20 demonstrates"

joe mama said:
Post'em and get back to me. Until then, you're all talk, and not
even good talk at that.

thats a truckload more than faking it anyway i can

--
Riley

I'd give my right arm to be ambidextrous (just)

Rriley · Mar 17, 2008

joe mama said:
So, c'mon now, sharp corners matter oh-so-much to you. Post a pic.
'Cause sharp corners don't mean squat to me, yet I can get them with
my "uselessly flawed" 35mm FF DSLR. Weird, huh?

two words
kit lens

your not even man enough to admit it
--
Riley

I'd give my right arm to be ambidextrous (just)

shinndigg · Mar 17, 2008

..."4/3rds sucks compared to any/everything else" thread. Sad, really.
--
shinndigg
http://www.pbase.com/shinndigg

shinndigg · Mar 17, 2008

...so, how many 4/3rds sucks threads are there now???
--
shinndigg
http://www.pbase.com/shinndigg

DSLR sensor facts with no acrimony please

Forum Pro

Forum Pro

Forum Pro

Veteran Member

Leading Member

Forum Pro

Veteran Member

Forum Pro

Forum Pro

Forum Pro

Well-known member

Forum Pro

Forum Pro

Forum Pro

Forum Pro

Forum Pro

Forum Pro

Forum Pro

Veteran Member

Veteran Member

Similar threads

Keyboard shortcuts