Ad
Ad
Ad
Pages: « 1 ... 5 6 [7] 8 »   Bottom of Page
Print
Author Topic: 1 Feb, 2008 - Quality vs. Value - When is Enough Enough?  (Read 33792 times)
jani
Sr. Member
****
Offline Offline

Posts: 1604



WWW
« Reply #120 on: February 15, 2009, 01:05:48 PM »
ReplyReply

Quote from: Panopeeper
Immaterial for what? Blooming is not immaterial, because the effects range from ignorable to devastating. However, it is immaterial for the dynamic range, because the DR ends exactly at the pixel saturation at base ISO. At higher ISOs the range is even more limited simply by the numerical range of the pixel values, thus saturation is not a factor at all.

Back to your question: if you think blooming affects the dynamic range, then pls explain, how.

The statement I reacted to wasn't the notion that you now present, but this:

Quote
The fact is, that exclusively the darkest shadows are interesting.
In the cases where you have over-saturation of the sensor, which bleeds into neighbouring sensor elements, this statement is no longer true; you have exceeded the brightness of the camera's dynamic range so much that there are very obvious and very real problems.

Such an image is therefore not suitable for a shadows-only analysis.

My objection wasn't one made in a world made only of black or white.
Logged

Jan
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #121 on: February 15, 2009, 09:05:39 PM »
ReplyReply

Quote from: jani
In the cases where you have over-saturation of the sensor, which bleeds into neighbouring sensor elements, this statement is no longer true; you have exceeded the brightness of the camera's dynamic range so much that there are very obvious and very real problems.

Such an image is therefore not suitable for a shadows-only analysis
That post of mine was an answer to

if you want to demonstrate the dynamic range differences between two cameras, you need to photograph something with a wide range of subject brightness

An image causing sensor blooming sensor and containing deep shadows would be just "something with a wide range of subject brightness", and I stated, that that is not necessary; thus sensor blooming is ab ovo not a consideration for me.

In fact, when I am asking for raw files for this purpose, I always say "two-three stops underexposed" (among other criteria). Finding out if an actual image is suitable for such measurement takes sometimes more time than the measurement itself.
Logged

Gabor
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #122 on: February 15, 2009, 09:30:29 PM »
ReplyReply

Quote from: Panopeeper
That post of mine was an answer to

if you want to demonstrate the dynamic range differences between two cameras, you need to photograph something with a wide range of subject brightness

An image causing sensor blooming sensor and containing deep shadows would be just "something with a wide range of subject brightness", and I stated, that that is not necessary; thus sensor blooming is ab ovo not a consideration for me.

In fact, when I am asking for raw files for this purpose, I always say "two-three stops underexposed" (among other criteria). Finding out if an actual image is suitable for such measurement takes sometimes more time than the measurement itself.

Perhaps you should clarify this concept, Gabor. I can understand, if you know how much an image is underexposed, then that factor can be taken into account when assessing DR for comparison purposes. I presume that, theoretically, you could compare two shots of a white brick wall containing a certain amount of detail, each shot being precisely 10 stops underexposed. The camera which shows the most detail in the white brick wall might be described as having the greater DR. Is this correct?
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #123 on: February 15, 2009, 09:44:31 PM »
ReplyReply



Quote
There's nothing preventing you from using a cropped portion of your FF 5DMk2 where you would have used the 50D, and you should get the same IQ - if not - better - as you would from a 50D. There's a substantial improvement of DxO rating for a 5DMk2 sensor compared with the 50D sensor.

Mark,
I don't quite know why I'm not as excited as perhaps I should be regarding the 5D2. The lack of full manual control in video mode disturbs me, as well as an apparent waterproofing problem. I think these two issues will be addressed in the next upgrade.

I find that I'm so often caught by surprise with Canon's upgrade path. If I'd succumbed to the excitement and praise of their first DSLR, the D30, and bought one at an unprecedented price (for me) of A$6,000, I would have been kicking myself when the D60 was announced a few months later.

6 months after I bought a 20D, Canon announced the 5D. I feel too much of a materialist when I keep upgrading at this rate. The 40D was a mistake in view of the soon-to-be-released 50D. My fault for buying on impulse.

When you examine the 'on screen' results at DXOmark, comparing the 20D with the 5D2 (at the pixel level), it's surprising how close those results are in respect of SNR, DR, Tonal Range and Color Sensitivity. The DR of the 5D2 is very marginally better. On the other hand, the color sensitivity of the 20D is marginally better than that of the 5D2.

Basically, it looks as though the 5D2 is a full frame sensor with 20D pixels and an extended ISO range.
I therefore find it difficult to believe that a cropped portion of the 5D2 sensor (equivalent to a 20D sensor) would give me the same image quality as the 50D with almost double the number of pixels (85% more, to be precise).

In other words, the reasons for preferring a 50D image to a cropped 5D2 image (with long telephotos), are more compelling than the reasons for preferring the higher resolution pontential of the 5D2 compared with the 5D1. (85% increase is better than 65% increase, no?)

Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #124 on: February 15, 2009, 10:12:20 PM »
ReplyReply

Quote from: Ray
I can understand, if you know how much an image is underexposed, then that factor can be taken into account when assessing DR for comparison purposes
The concept of "underexposure" has no use in the evaluation. Underexposed compared to what? Manual metering? Evaluative metering? Center metering?

When I ask for underexposed shot, I have only one thing in mind: that there be patches suitable for measurement in the deep shadows; therefor I ask for smooth, unicolored, uniformly lit areas in different shades. An underexposed gray wedge is very good, a transmission wedge is the best.

Pls look at my post #85 on Page 5 of this thread. It does not matter how much the transmission wedge had been underexposed; the only important thing is, that there are steps in different shades down to the uselessly dark. Thus I can select from many different shades, exhibiting different degrees of noisiness. I don't believe in DR expressed by a single number. For example from that shot I measured dozens of patches; here is a small crop of the result, from the end of that list:

EV   -> noise

9.28 -> 32.5
9.32 -> 35.1
9.41 -> 34.9
9.51 -> 41.0
9.59 -> 40.0
9.69 -> 43.2
9.75 -> 43.9
9.78 -> 44.7
9.89 -> 46.4
9.95 -> 51.7

The first number shows the intensity of the patch from saturation downwards, the second number is the degree of noise, as the standard deviation in percentage of the average pixel values in the selected patch. 51.7% is roughly SNR=2 at -9.95 EV. This means, that if you accept SNR=2, then the DR is 10 EV (this is the 40D at ISO 200). If you accept only SNR=3, i.e. 33% noise, the DR is somewhere between 9.3 and 9.4 EV.

I know the highest possible pixel value from any shot, which contains anything causing raw clipping. (This is often ISO dependent, and sometimes the channels are not equal.) I determined those levels once and now I do not need anything in the higher ranges. The "intensity position" of the selected patch can be calculated from the clipping level without having anything near clipping in the shot.

I don't need any actual measurement on the highlights, for I know that the noise is always lower in the highlights. Why would I care for the noise in the top stop of the DR? The interesting issue is the noise in the very shadows, isn't it?
Logged

Gabor
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6970


WWW
« Reply #125 on: February 15, 2009, 10:20:27 PM »
ReplyReply

Ray,

I agree - buying a camera can be as frustrating as buying a computer. As soon as you get comfortable with it, the next model makes it obsolete and that happens very fast. But we must be hardened to that reality and only buy when we really think it worthwhile in terms of the bird in hand.

Now, on that note, as the birds sing, the 5DMk2 has 36% larger pixel pitch than the 50D and it's low light ISO is 2.1x that of the 50D. Along with the higher resolution those must be reasons why you should expect better image quality from it.  

Cheers,

Mark
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Herkko
Newbie
*
Offline Offline

Posts: 20


WWW
« Reply #126 on: February 16, 2009, 04:22:36 AM »
ReplyReply

Quote from: Ray
When you examine the 'on screen' results at DXOmark, comparing the 20D with the 5D2 (at the pixel level), it's surprising how close those results are in respect of SNR, DR, Tonal Range and Color Sensitivity. The DR of the 5D2 is very marginally better. On the other hand, the color sensitivity of the 20D is marginally better than that of the 5D2.

When I examine the 'on screen' results from my 20D and 5DMKII files I can see a country mile difference in image quality. Good old lenses like EF 50/2.5 Macro or EF 300/4 IS are utilized properly, at last. Even when I crop out 8mpix portion of 5DMKII the color accuracy, lack of noise artifacts etc. are better than 20D, not to talk about the whole 21mpix...

Ability to crop has got a new meaning when I have been looking my recent shots. Definitely better with less artifacts and closer to natural surface colors at 100% than my ex. 5D had.

Just a casual ice surface shot from yesterday: from this I can extract unsharpened 100% like this. If I'm desperate I can even save heavily cropped finished picture like this.
IMO, very nice and accurate surface details + tone graduation compared to my similar in 5D and especially in 20D pictures. (And more observant of you can also notice that I haven't bothered with the fw upgrade yet ;)
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #127 on: February 16, 2009, 06:06:17 AM »
ReplyReply

Quote from: Herkko
Even when I crop out 8mpix portion of 5DMKII the color accuracy, lack of noise artifacts etc. are better than 20D, not to talk about the whole 21mpix...

Well, of course I'm not referring to the whole 21mp. I'm talking only about an 8mp crop from those 21mp. There is no question that 21m 20D pixels are better than 8m 20D pixels.

However, you are just the person we need on this site. There's been a lot of criticism of the meaningfulness of the DXOmark figures. Some people seem to think they are just plain wrong, but no-one seems to provide any evidence that they are wrong or inaccurate. The criticisms seems all hot air.

You are now in the privileged position of being able to provide such evidence, since you own both a 20D and 5D2   .

Can I suggest the following methodology? Shoot an appropriate subject with varied tonality, fine detail, lots of color and contaning a few smooth and dark surfaces. Use the same lens with both cameras at the same aperture and ISO, and at the same distance to subject. Use tripod, MLU and remote cord or self-timer. Bracket all shots +/-1/3rd stop. The 20D, according to DXO, is slightly more sensitive than the 5D2. That means at the same ISO and shutter speed, the 20D will receive a slight overexposure compared to the 5D2. However this might not be apparent until you have cropped the 5D2 image to the same FoV as the 20D image because the 5D2 scene is larger. Do not try to compare exposures until the 5D2 image has been cropped in ACR to the same FoV as the 20D image. The histogram will change as you adjust the crop.

Another concern, use a lens that you know autofocuses accurately on both cameras. The Live View of the 5D2 should allow for totally accurate focussing. We don't want distractions such as one image having slightly different focussing than the other. Post results here. Okay!  
« Last Edit: February 16, 2009, 06:09:00 AM by Ray » Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #128 on: February 16, 2009, 06:21:42 AM »
ReplyReply


Quote
I agree - buying a camera can be as frustrating as buying a computer. As soon as you get comfortable with it, the next model makes it obsolete and that happens very fast. But we must be hardened to that reality and only buy when we really think it worthwhile in terms of the bird in hand.

Mark,
I only upgrade my computer when I find it really worthwhile. My first computer took a whole 2 minutes just to open an 18mb PhotoCD file. That was intolerably slow and there was great need to upgrade. I've been using my current Win 64 bit system with 6GB of RAM and a WD Raptor 10,000 rpm hard drive solely as a scratch disc, for over 3 1/2 years now and feel no urge to upgrade because it's still a reasonably fast computer and gets the job done.

Quote
Now, on that note, as the birds sing, the 5DMk2 has 36% larger pixel pitch than the 50D and it's low light ISO is 2.1x that of the 50D. Along with the higher resolution those must be reasons why you should expect better image quality from it.

Let's have a look at some hard facts from the DXOmark website. The 5D2's SNR, DR etc are approximately a whole stop better than those of the 50D. For example, at ISO 3200 the 5D2's DR is 9.09 EV, the 50D's DR at ISO 1600 is 9.19 EV.

A similar situation applies comparing the 50D at ISO 400 with the 5D2 at ISO 800, and the 50D at ISO 800 with the 5D2 at ISO 1600, with regard to all the parameters of noise, DR, tonality and color sensitivity.

However, oddly enough, between ISO range 100 to 400, the DR of the 5D2 is only marginally better than that of the 50D, at both the pixel level and the normalised print size; that is, less than a stop better.

Now such an improvement certainly appears to be worthwhile. However, when one brings into the equation the sort of lenses one might use with each camera, that apparent advantage might sometimes be non-existent.

For example, the lens I would use most with a 5D2 would be my Canon 24-105/F4 IS. For very wide angle requirements I would use the D700 with Nikkor 14-24/F2.8. For long telephoto work, I would prefer the 50D with my 100-400 zoom.

But supposing I compare the 50D and 17-55/F2.8 combination with the 5D2 and 24-105mm? The EF-S 17-55 doesn't have quite the same range as the 24-105, being only 27-88mm full frame equivalent, but it's reasonably close. Whatever aperture I choose for the 5D2/24-105, I can choose one aperture wider using the 50D/17-55. The EF-S 17-55 is F2.8 across the whole FL range, and the 24-105 is F4 maximum across the whole FL range.

If I want extensive DoF and decide to use F11 with the 5D2/24-105 at ISO 400, I can use F8 or even F6.3 with the 50D/17-55 at ISO 125-200. If I'm shooting in the street at night without flash and need to use the 5D2/24-105 at F4 and ISO 3200, I can use the EF-S 17-55 at F2.8 and ISO 1600 with the 50D.

In this situation, I can't see any advantage for the 5D2. With another combination, comparing the 5D2 and 85/1.2 with the 50D and 50/1.4, for example, I can see there would be an advantage in terms of sharper and cleaner images with a shallower DoF, but not necessarily at smaller apertures. I think the 50D with 50/1.4 at F6.3 would be very close to the 5D2 with 80/1.2 at F11, bearing in mind that the resolution advantage of the 5D2 sensor is a mere 18% greater than the 50D.

Have I just talked myself out of buying a 5D2?  

Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #129 on: February 16, 2009, 07:00:03 AM »
ReplyReply

Quote
It does not matter how much the transmission wedge had been underexposed; the only important thing is, that there are steps in different shades down to the uselessly dark. Thus I can select from many different shades, exhibiting different degrees of noisiness. I don't believe in DR expressed by a single number. For example from that shot I measured dozens of patches; here is a small crop of the result, from the end of that list:

Gabor,
I'm having trouble with your logic (probably my fault). On the one hand you say you want steps in different shades down to the uselessly dark, yet in the same sentence you say that it doesn't matter how much the transmission wedge has been underexposed. Surely it clearly matters from the perspective of the person taking the shot. If the shot of the different shades doesn't extend to the uselessly dark, then the exposure was too great. Is there another explanation?

Quote
The concept of "underexposure" has no use in the evaluation. Underexposed compared to what? Manual metering? Evaluative metering? Center metering?

Didn't you answer your own question? Underexposed compared to your own standard you've expressed as follows: "I know the highest possible pixel value from any shot, which contains anything causing raw clipping. (This is often ISO dependent, and sometimes the channels are not equal.) I determined those levels once and now I do not need anything in the higher ranges. The "intensity position" of the selected patch can be calculated from the clipping level without having anything near clipping in the shot."

Finally, if we go to the source of this little contretemps, my statement:  "...if you want to demonstrate the dynamic range differences between two cameras, you need to photograph something with a wide range of subject brightness", I would just repeat that my emphasis was on demonstration, not measurement, and the context was a finished 'picture' along the lines of Michael's forest shot comparing the Canon G10 with the Phase P45+.
« Last Edit: February 16, 2009, 07:02:53 AM by Ray » Logged
Herkko
Newbie
*
Offline Offline

Posts: 20


WWW
« Reply #130 on: February 16, 2009, 07:04:37 AM »
ReplyReply

Quote from: Ray
Can I suggest the following methodology?
....
Post results here. Okay!  

I have to politely decline that request  First I want to show real life examples for what I bought that camera. And that is not performance testing but nature photography or similar activities.

IMO, based on my normal 'real' shooting, 1DMKIII and 5D are embarrassing good cameras image qualitywise. Enough for me to submit files to demanding customer or printing 12x18". Then 5DMKII is even better, especially with good glass and tripod. My primary photography concern is not image quality now, but things like generating ideas, being at location in peak light, determining focal length, working hard enough for composition.. The things that should been the primary components for good photography.

I would sometimes talk about other camera properties than ultimate image quality, it seems to be a bit overrated topic nowadays. How about battery durability, bigger lcd, live view or viewfinder? Have they improved one's photography in any way?

To follow the topic, I feel having now enough camera for my applications.
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #131 on: February 16, 2009, 08:21:37 AM »
ReplyReply

Quote from: Herkko
I have to politely decline that request  First I want to show real life examples for what I bought that camera. And that is not performance testing but nature photography or similar activities.

Pity! I guess I'll just have to continue believing the DXOmark results until someone presents images demonstrating a discrepancy. Of course, most of us are mainly interested in real life photography, but it sometimes helps to know your equipment. Fortunately, there are organizations like DXO, Dpreview and Photozone who take the trouble to test cameras and lenses, and as a consequence I'm saved the trouble. However, if or when their results are seriously inaccurate, I'd certainly like to see the evidence rather than just read about vague and subjective opinons.

My own tests of equipment which happens to feature on the DXOmark website, seem to correspond quite well with the DXO results; not exactly, but close enough for their results to be credible to me.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6970


WWW
« Reply #132 on: February 16, 2009, 10:44:56 AM »
ReplyReply

Quote from: Ray
Mark,
 However, when one brings into the equation the sort of lenses one might use with each camera, that apparent advantage might sometimes be non-existent.

For example, the lens I would use most with a 5D2 would be my Canon 24-105/F4 IS. For very wide angle requirements I would use the D700 with Nikkor 14-24/F2.8. For long telephoto work, I would prefer the 50D with my 100-400 zoom.

But supposing I compare the 50D and 17-55/F2.8 combination with the 5D2 and 24-105mm? The EF-S 17-55 doesn't have quite the same range as the 24-105, being only 27-88mm full frame equivalent, but it's reasonably close. Whatever aperture I choose for the 5D2/24-105, I can choose one aperture wider using the 50D/17-55. The EF-S 17-55 is F2.8 across the whole FL range, and the 24-105 is F4 maximum across the whole FL range.

If I want extensive DoF and decide to use F11 with the 5D2/24-105 at ISO 400, I can use F8 or even F6.3 with the 50D/17-55 at ISO 125-200. If I'm shooting in the street at night without flash and need to use the 5D2/24-105 at F4 and ISO 3200, I can use the EF-S 17-55 at F2.8 and ISO 1600 with the 50D.

In this situation, I can't see any advantage for the 5D2. With another combination, comparing the 5D2 and 85/1.2 with the 50D and 50/1.4, for example, I can see there would be an advantage in terms of sharper and cleaner images with a shallower DoF, but not necessarily at smaller apertures. I think the 50D with 50/1.4 at F6.3 would be very close to the 5D2 with 80/1.2 at F11, bearing in mind that the resolution advantage of the 5D2 sensor is a mere 18% greater than the 50D.

Have I just talked myself out of buying a 5D2?  

As soon as you bring lenses into the resolution equation, which one must, the choice of aperture becomes a critical factor because of diffraction limitation. If you haven't focused much on this topic I recommend Cambridge in Colour Diffraction. If maximum resolution is the objective, that will end much of the discussion about which f/stop to use. If you wish to sacrifice resolution, you can maximize DoF with any of the lenses you discuss, but then it gets down to lens quality and physics.

27~88 is considerably less focal length range than 24~105. Let's not minimize that. And what is the quality of the 17~55 lens compared with the 24~105 L lens? As for aperture, one lens gives you one stop wider aperture than the other, good for lower light conditions where you want adequate shutter speed for hand-holding (at least 1/25th of a second with IS on and real steady hands) and one stop lower ISO. I don't see any other advantage.

And the fact is that DxOMark still shows a substantial improvement of sensor quality moving from the D50 to the 5DMk2 (from about 63 to 79).

The incremental quality and features of the up-grade are there. Whether you really need them is a subjective issue depending on your images and the quality differences which you can (a) see and (b, matter to you or your clients.
« Last Edit: February 16, 2009, 10:46:11 AM by MarkDS » Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
jani
Sr. Member
****
Offline Offline

Posts: 1604



WWW
« Reply #133 on: February 16, 2009, 04:17:57 PM »
ReplyReply

Quote from: Panopeeper
I don't need any actual measurement on the highlights, for I know that the noise is always lower in the highlights. Why would I care for the noise in the top stop of the DR? The interesting issue is the noise in the very shadows, isn't it?
Wouldn't colour accuracy be interesting? Highlight roll-off?
Logged

Jan
Ray
Sr. Member
****
Offline Offline

Posts: 8900


« Reply #134 on: February 16, 2009, 05:40:16 PM »
ReplyReply



Quote
27~88 is considerably less focal length range than 24~105. Let's not minimize that.

Mark,
There is a slight disadvantage there, I admit. Travelling with a D700 and 50D, if I need less than 27mm I would have to switch cameras and use the Nikkor 14-24. There's a gap of 3mm. Not really significant. If I need a slightly longer FL than 88mm, I can use my EF-S 60/2.8 macro which becomes effectively 96mm on the 50D and is a very, very sharp lens, according to Photozone.

The gap problem would really be between the 96mm of the EF-S 60/2.8 and the 160mm of the 100-400 zoom. For this reason, for travelling purposes I'm considering getting the highly regarded Canon 70-200/F4 IS which I would carry in place of the 100-400. I would then have a (35mm equivalent) focal length range from 14mm to 320mm with just a few small gaps. The 70-200/F4 IS might also be good enough to use with my 1.4x extender which I find pretty useless on the 100-400 zoom.

Quote
And what is the quality of the 17~55 lens compared with the 24~105 L lens? As for aperture, one lens gives you one stop wider aperture than the other, good for lower light conditions where you want adequate shutter speed for hand-holding (at least 1/25th of a second with IS on and real steady hands) and one stop lower ISO. I don't see any other advantage.

The quality is very good indeed. The EF-S 17-55/2.8 appears to be sharper than the 24-105 by a significant margin, according to Photozone whose charts I show below. Even at F2.8, the EF-S 17-55 appears to be at least marginally sharper than the 24-105 is at any aperture. The borders of the 24-105 might appear to be marginally better, but that's deceptive because the 24-105 is being tested on an APS-C format. One expects the borders to be relatively good. On any full frame DSLR, including the 5D2 those borders are inevitably going to be worse.

Hope you don't think I'm boasting   .

[attachment=11565:Canon_24..._v_17_55.jpg]


Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6970


WWW
« Reply #135 on: February 16, 2009, 07:03:04 PM »
ReplyReply

Ray, good to see the lens quality is excellent. That really helps.

As you are in the business of selling images and I am not, I can afford to really be minimalist when travelling. With all the nonsense we're facing from the airlines and the security these days, I've reduced my travel equipment to the 1Ds3, a 17-40 f/4 L, the 24-105 F4L and the 70-300 IS DOf/4~5.6. This last lens is not one of Canon's shining stars, but it's workable and very compact; so I have a continuous range from 17 to 300mm using little space and weight. Believe it or not, my camera bag dates from the early 1960s - an extremely light-weight soft-sided (some kind of nylon or polyesther) rectangular box below the economy class dimension limits, with a minimum of dividers - holds everything and all is reachable because the whole top flips open. Can't find anything like this on the market these days - it's not fancy enough, but is it ever practical. I also pack a Lumix LX-1 as a back-up camera. So far I've been fortunate with my 1 series cameras. They just work, and I hope that continues to be the case. One can get so cluttered with stuff it becomes hard to manage on the go. Of course there are obvious compromises here, but honestly, I use the 24~105 for over 90% of my images. The next thing I'm looking at is a Gitzo traveler tripod. A fabulous piece of equipment - so light and small and totally sturdy - but the price is also fabulous (like in the land of fables) so I haven't plunged yet.

Anyhow, this is all getting OT, except to say all these choices reflect quality:value trade-offs and so far to this point enough is enough.   "Lean and Mean" is the name of the game.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #136 on: February 16, 2009, 07:30:03 PM »
ReplyReply

Quote from: jani
Wouldn't colour accuracy be interesting? Highlight roll-off?
Neither has anything to do with dynamic range or noise. Particularly, there is no "highlight roll-off" on the level of raw data; that is a function of the raw image processing, not of the sensor.
Logged

Gabor
jani
Sr. Member
****
Offline Offline

Posts: 1604



WWW
« Reply #137 on: February 17, 2009, 04:06:01 AM »
ReplyReply

Quote from: Panopeeper
Neither has anything to do with dynamic range or noise.
Both are of practical interest to photographers, however, and do limit the usable dynamic range of the system.

Quote
Particularly, there is no "highlight roll-off" on the level of raw data; that is a function of the raw image processing, not of the sensor.
So now you're saying that the Super CCD HR sensor doesn't exist, and that sensor bloom is a function of the raw image processing? That's interesting. How do I go about fixing it with, I presume, dcraw?
Logged

Jan
Rob C
Sr. Member
****
Offline Offline

Posts: 12213


« Reply #138 on: February 17, 2009, 02:17:22 PM »
ReplyReply

[quote name='MarkDS' date='Feb 17 2009, 02:03 AM' post='260562']
Ray, good to see the lens quality is excellent. That really helps.

Believe it or not, my camera bag dates from the early 1960s - an extremely light-weight soft-sided (some kind of nylon or polyesther) rectangular box below the economy class dimension limits, with a minimum of dividers - holds everything and all is reachable because the whole top flips open.


Mark

Couldnt resist posting this, which was shot today for exactly the same reason for another little site The only change from the NEF is that its converted to Jpeg and reduced in size - no corrections even attempted - all for speed, to catch the moment!

Rob C

EDIT: Lots of good stuff from the 60s!
EDIT 2: This venerable old bag is retired too; an aluminium floozy has usurped its place.
« Last Edit: February 17, 2009, 02:33:51 PM by Rob C » Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6970


WWW
« Reply #139 on: February 17, 2009, 03:52:55 PM »
ReplyReply

Well, ya know, it kinda looks like that, except mine is grey nylon fabric - not venerable old leather. That one is a collector's item!  Treasure it
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Pages: « 1 ... 5 6 [7] 8 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad