Ad
Ad
Ad
Pages: « 1 ... 7 8 [9] 10 »   Bottom of Page
Print
Author Topic: Canon 50D @ 15MP  (Read 100303 times)
DarkPenguin
Guest
« Reply #160 on: September 03, 2008, 11:00:32 AM »
ReplyReply

Have you seen the 50D samples at DPreview?  If that is the level of jpeg sharpening I feel pretty confident that the RAW files will have none applied.
« Last Edit: September 03, 2008, 11:02:02 AM by DarkPenguin » Logged
ejmartin
Sr. Member
****
Offline Offline

Posts: 575


« Reply #161 on: September 03, 2008, 11:47:09 AM »
ReplyReply

Quote
While I know that RAW is considered...well...the raw data, I seriously doubt that the data is taken without any processing at all....just encapsulating the data into a file format, applying metadata and saving the file to compact flash would be some minimal processing task for the DIGIC...that doesn't include the exposure curve and white balance that is applied to the RAW file so that it doesn't look like cr@p when we view it. 

It wouldn't be surprising if some very conservative, baseline sharpening and noise reduction was happening too.
[a href=\"index.php?act=findpost&pid=219193\"][{POST_SNAPBACK}][/a]

There is no exposure curve or white balance applied to the raw data; that is the purpose of the metadata -- to specify the chosen picture style (tone curve) and the white balance determined by the camera's metering, for the raw converter to make use of (or not).  There is also no sharpening or NR applied to Canon raw data unless one turns on long exposure NR.

This is not to say that there isn't processing of the pixel data before a RAW is written; it's just that the processing is at the level of individual pixels (for instance, the correlated double sampling that removes pixel reset fluctuations from the pixel values) rather than the sorts of manipulations one typically does in post-processing.
Logged

emil
Ray
Sr. Member
****
Offline Offline

Posts: 8944


« Reply #162 on: September 03, 2008, 06:42:08 PM »
ReplyReply

Quote
LOL, that's what I meant. I don't need Noah's rule when I have a built-in metering and histogram.

I wasn't suggesting you do. I was addressing the negative attitude that a higher pixel count might be disadvantageous because to realise the potentialy higher resolution one would need to use faster shutter speeds. My sunny 16 rule was an example of how fast shutter speeds can be when the light is good.

Quote
Such general statements are not of much help. Anyway, irrespectively of the total dynamic range of the scenery, two stops more or less make a huge difference for me (that is the difference between ISO 200 and 1600 of the 40D's DR).

The dynamic range of subject brightness in the real world can be huge; far greater than any current camera can capture in a single shot. I imagine that the DR of the 40D at ISO 1600 is at least as good as that of slide film in the days of yore.

I think we might have to wait for a shift in paradigm and/or some new breakthrough technology before we see a substantial improvement in the DR of digital cameras.

One rather dismaying fact about the Bayer type sensor is that it necessarily has to block out about half the light it receives, perhaps as much as 2/3rds depending on scene content. There's at least a whole stop of DR thrown away as a result of the need for color filters.
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #163 on: September 03, 2008, 07:37:14 PM »
ReplyReply

Quote
I wasn't suggesting you do. I was addressing the negative attitude that a higher pixel count might be disadvantageous because to realise the potentialy higher resolution one would need to use faster shutter speeds. My sunny 16 rule was an example of how fast shutter speeds can be when the light is good
You have a sure way to turn anything into an umitigated rubbish.

Quote
The dynamic range of subject brightness in the real world can be huge; far greater than any current camera can capture in a single shot. I imagine that the DR of the 40D at ISO 1600 is at least as good as that of slide film in the days of yore
See above.
Logged

Gabor
Ray
Sr. Member
****
Offline Offline

Posts: 8944


« Reply #164 on: September 03, 2008, 09:05:02 PM »
ReplyReply

Quote
You have a sure way to turn anything into an umitigated rubbish.
See above.
[a href=\"index.php?act=findpost&pid=219311\"][{POST_SNAPBACK}][/a]

You sure have a strange definition of rubbish.
Logged
ARD
Sr. Member
****
Offline Offline

Posts: 296



WWW
« Reply #165 on: September 04, 2008, 03:05:41 AM »
ReplyReply

I have just spent the last few minutes reading this entire thread. Some excellent content and information from both sides of the argument.

I do wonder though, if people go into such depths about pixel count, diffraction etc and so on, does it take over from the ultimate goal which is taking photographs - it's almost like having the ultimate high performance car, but constantly taking the engine out instead of driving it.

Still a good thread though  
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #166 on: September 04, 2008, 03:22:15 AM »
ReplyReply

After we get the DPreview of the technical capabilities of the 50D, I was thinking about selling my 5D and getting the 50D. The problem is, if it produces images as good as the 5D in tonality and noise, who would buy a 5D?

On the upside, the body is only 1400.00US so that's good news.

I also wonder if it's weather sealing is the same as the 1D series cameras, or if it's a lesser sort?

Other than being a crop frame camera, if the sealing, shutter release time to failure, and image reproduction are as good as the 5D, plus you get 15MP--what a great camera for the price.
Logged

DarkPenguin
Guest
« Reply #167 on: September 04, 2008, 09:41:08 AM »
ReplyReply

http://forums.dpreview.com/forums/read.asp...essage=29193081
Logged
ejmartin
Sr. Member
****
Offline Offline

Posts: 575


« Reply #168 on: September 04, 2008, 11:10:34 AM »
ReplyReply

Quote
After we get the DPreview of the technical capabilities of the 50D, I was thinking about selling my 5D and getting the 50D. The problem is, if it produces images as good as the 5D in tonality and noise, who would buy a 5D?

[a href=\"index.php?act=findpost&pid=219371\"][{POST_SNAPBACK}][/a]


The best sensor in photon collecting efficiency per unit area is currently the D3; it beats the current Canons (1D3, 1Ds3, and 40D) by about 15%, probably due to its gapless microlenses, and it beats the 5D by a factor 1.5.  So, if the 50D with its gapless microlenses achieves the sensitivity of the D3, it still loses by a factor 1.75 in light-collecting ability relative to the 5D because the APS-C format is so much smaller than full frame (less than 40% of the area).  So unless the new sensor is 2/3 stop more efficient than the 5D (a true technological breakthrough), it will not be even close in image quality to the 5D.
Logged

emil
Slough
Guest
« Reply #169 on: September 04, 2008, 01:42:54 PM »
ReplyReply

Quote
You have a sure way to turn anything into an umitigated rubbish.
See above.
[a href=\"index.php?act=findpost&pid=219311\"][{POST_SNAPBACK}][/a]

Amen.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8944


« Reply #170 on: September 04, 2008, 08:13:53 PM »
ReplyReply

Quote
You have a sure way to turn anything into an umitigated rubbish.
See above.


Quote
Amen.
[a href=\"index.php?act=findpost&pid=219477\"][{POST_SNAPBACK}][/a]

In my experience, I have found when people resort to such abusive language it is usually because they either have not been able to understand the argument, have been able to understand the argument but don't like it even though the argument is sound, or are simply not man enough to admit they might be wrong.
Logged
JohnKoerner
Guest
« Reply #171 on: September 09, 2008, 10:42:28 AM »
ReplyReply

I haven't been able to read any comments as to any improvements in the weather sealing of the 50D.

I think one of the main failings of the 40D compared to the D300 was quality of craftsmanship. If two cameras are relatively equivalent in output, the better-built camera becomes the more attractive buy.
Logged
DarkPenguin
Guest
« Reply #172 on: September 09, 2008, 11:02:45 AM »
ReplyReply

Quote
I haven't been able to read any comments as to any improvements in the weather sealing of the 50D.

I think one of the main failings of the 40D compared to the D300 was quality of craftsmanship. If two cameras are relatively equivalent in output, the better-built camera becomes the more attractive buy.
[a href=\"index.php?act=findpost&pid=220357\"][{POST_SNAPBACK}][/a]

I'd argue that the cheaper one becomes the more attractive buy.

It is my understanding that there are weather sealing improvements.  Check the dpreview preview.

Phil Askey: "Apparently (we've been told verbally) the 50D's body offers better environmental sealing than the EOS 40D thanks to a slightly different construction and tighter seals. We'll get more information on this as soon as we can."
Logged
DarkPenguin
Guest
« Reply #173 on: September 15, 2008, 10:32:15 AM »
ReplyReply

http://forums.dpreview.com/forums/read.asp...essage=29327398
Logged
fike
Sr. Member
****
Offline Offline

Posts: 1377


Hiker Photographer


WWW
« Reply #174 on: September 15, 2008, 12:47:27 PM »
ReplyReply

Quote
http://forums.dpreview.com/forums/read.asp...essage=29327398
[a href=\"index.php?act=findpost&pid=221562\"][{POST_SNAPBACK}][/a]

Those look very good!  The picture of the billboard shows really good transitions into pure black shadows.  There isn't too much detail in the shadows, but that shot was taken at 1/125 at f/4.5, so it isn't like they could have pulled much more detail out of the shadows by changing the aperture or shutter speed--at least while remaining at handheld speeds.
« Last Edit: September 15, 2008, 12:48:35 PM by fike » Logged

Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.
DarkPenguin
Guest
« Reply #175 on: September 15, 2008, 01:49:25 PM »
ReplyReply

Quote
Those look very good!  The picture of the billboard shows really good transitions into pure black shadows.  There isn't too much detail in the shadows, but that shot was taken at 1/125 at f/4.5, so it isn't like they could have pulled much more detail out of the shadows by changing the aperture or shutter speed--at least while remaining at handheld speeds.
[a href=\"index.php?act=findpost&pid=221583\"][{POST_SNAPBACK}][/a]

I still want to see raw files.  But all in all it is looking pretty good.
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #176 on: September 15, 2008, 02:22:39 PM »
ReplyReply

It is amazing. I don't mean the sample shots, but how people interpret images, which are totally meaningless regarding the camera's capability. Nothing but absolutely nothing can be judged based on these shots.

I analyzed a few raw files, which were shot in a photoshop; the shots suck, the photog did not have any comparison in mind. However, some are suitable for measuring the noise. It is on par with the 40D.

In other words: the technology is better than that of the 40D, Canon could have increased the DR, but they decided to go for the pixels. Sux.
Logged

Gabor
fike
Sr. Member
****
Offline Offline

Posts: 1377


Hiker Photographer


WWW
« Reply #177 on: September 15, 2008, 02:30:19 PM »
ReplyReply

Quote
It is amazing. I don't mean the sample shots, but how people interpret images, which are totally meaningless regarding the camera's capability. Nothing but absolutely nothing can be judged based on these shots.

I analyzed a few raw files, which were shot in a photoshop; the shots suck, the photog did not have any comparison in mind. However, some are suitable for measuring the noise. It is on par with the 40D.

In other words: the technology is better than that of the 40D, Canon could have increased the DR, but they decided to go for the pixels. Sux.
[a href=\"index.php?act=findpost&pid=221604\"][{POST_SNAPBACK}][/a]

You are right, in the balance of tradeoffs, they chose the easy way to market the camera--with simplistic megapixels. The general public likes a simple story.  Perhaps they could have further improved DR or high-ISO noise levels, but what they did achieve is that they didn't backslide towards worse noise-levels or lower DR--in fact it will probably be slightly better.  I can agree with you that perhaps I would choose 14MP or 12MP with better DR and noise characteristics, but they didn't do that and that is that.  I think you are beating a dead horse.  The new camera seems to be better than the old--maybe not in the way that gabor would like--but better.
Logged

Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.
DonWeston
Full Member
***
Offline Offline

Posts: 148


« Reply #178 on: September 15, 2008, 02:47:09 PM »
ReplyReply

Quote
It is amazing. I don't mean the sample shots, but how people interpret images, which are totally meaningless regarding the camera's capability. Nothing but absolutely nothing can be judged based on these shots.

I analyzed a few raw files, which were shot in a photoshop; the shots suck, the photog did not have any comparison in mind. However, some are suitable for measuring the noise. It is on par with the 40D.

In other words: the technology is better than that of the 40D, Canon could have increased the DR, but they decided to go for the pixels. Sux.
[a href=\"index.php?act=findpost&pid=221604\"][{POST_SNAPBACK}][/a]


Just out of curiousity, Panokeeper, how did you examine these files??? Everyone has their own interpretation of things. I downloaded the Canon images, especially the landscape one, as that is what I shoot mainly. Resized it in CS3 and printed out 36" crops. To my eyes, [20-20] looking at 12+" this file looks great. Not quite the smoothness of my 5D images but on printed image, not bad at all. Certainly a good step over the 40D images I had. ....

Everyone has their own standards, but I find actual print the most fair and accurate to go by in deciding my own needs....fwiw, if the aps-c chip and processor is this good, I think this bodes well for the new FF with digic 4.....ymmv...jmho
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #179 on: September 15, 2008, 05:16:03 PM »
ReplyReply

Quote
how did you examine these files???
I am measuring the noise (as standard deviation) on smooth, unicolor, evenly lit areas, on the non-demosaiced raw data. Any image is good, which contains such spots in the very dark shadows.

The images I received are not very suitable for this purpose, but I did find a few spots good enough to measure.

I am working on a collection of such measurements with different cameras; I will post it when it is so far. Perhaps more suitable raw files from the 50D will appear in the meantime.
Logged

Gabor
Pages: « 1 ... 7 8 [9] 10 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad