| Author | Thread |
|
|
04/26/2004 11:12:42 AM · #1 |
Thanks!
Message edited by author 2004-09-09 00:00:54. |
|
|
|
04/26/2004 11:19:50 AM · #2 |
Interesting questions.
I have the same camera you do, and I've been shooting in large-fine as well. I have printed out a couple of shots in 8 x 10 and they looked great to me. So I would say that you could go up to at least 8x10 without any loss in picture quality.
I know next to nothing about the raw format. |
|
|
|
04/26/2004 12:34:25 PM · #3 |
There is no inherent "dots per inch" to a photographic image.
You have a 3076x2052 pixel image. You can print it at 10 dots per inch and have an image that is 307 inches by 205 inches (but very "grainy" and pixelated looking). Or you can print it at 300 dots per inch and have an image that is 10.25 inches by 6.84 inches (lots of detail, even under close scrutiny). The imaging data didn't change at all, just the relation of the input pixel to the output pixel.
The 180 value is just an arbitrary value assigned; it can be changed to anything without the need for resampling with programs like Genuine Fractals, etc. The only time you need to resample is when you don't have enough pixels and you need to create more. (Of course, resampling can only get you so far! You can't just magically create more detail!)
I always shoot in raw, simply because of the extra latitude you get in post-processing.
In addition, raw is compressed in a lossless fashion. JPEG is a lossy compression scheme, and can result in compression artifacts, although I've never noticed any when using the "fine" setting.
Message edited by author 2004-04-26 12:43:02. |
|
|
|
05/06/2004 08:21:03 AM · #4 |
I have the same question, regarding resolution and RAW. I'm using the softare that came with my digital rebel to send my files to photoshop 7. I'm not sure what to set the resolution to prior to tranfer.
How is this determined and what impact does it have?
Thanks |
|
|
|
07/01/2004 05:04:05 PM · #5 |
I am new to digital photography and have the same camera as hgordin and breckinshire (EOS-300D Rebel).
What I am wondering is how do I setup raw format on that particular camera? I curently use highest quality jpeg but can't see the setting for RAW format. Or is it something that is setup in Photoshop CS?
Thanks
|
|
|
|
07/01/2004 05:42:29 PM · #6 |
| I'd like to know what format you save in once you've processed your RAW file. TIFF, JPEG, or PSD? I've heard several different arguements on the subject but a definitive answer would be helpful. Thanks |
|
|
|
07/01/2004 05:53:40 PM · #7 |
Originally posted by digistoune:
I'd like to know what format you save in once you've processed your RAW file. TIFF, JPEG, or PSD? I've heard several different arguements on the subject but a definitive answer would be helpful. Thanks |
TIFF prior to printing, JPEG for web and projector display, PSD for intermediate/ work in progress with layers.
|
|
|
|
07/01/2004 06:26:28 PM · #8 |
Originally posted by hgordin:
I have the same question, regarding resolution and RAW. I'm using the softare that came with my digital rebel to send my files to photoshop 7. I'm not sure what to set the resolution to prior to tranfer.
How is this determined and what impact does it have?
Thanks |
It really doesn't matter and has no impact. That is an arbitrary value which can be changed in photoshop... It is up to you to determine how many of those existing pixels you want per inch. That will depend on your use (e.g. web or computer monitor display vs. prints).
|
|
|
|
07/01/2004 06:50:32 PM · #9 |
Originally posted by Gordon:
Originally posted by digistoune:
I'd like to know what format you save in once you've processed your RAW file. TIFF, JPEG, or PSD? I've heard several different arguements on the subject but a definitive answer would be helpful. Thanks |
TIFF prior to printing, JPEG for web and projector display, PSD for intermediate/ work in progress with layers. |
Thank you, Gordon!
For answers to questions like these and digital workflow/imaging in general, what is the best book on the subject in your opinion?
Message edited by author 2004-07-01 18:52:00. |
|
|
|
07/01/2004 08:09:33 PM · #10 |
Originally posted by digistoune:
Originally posted by Gordon:
Originally posted by digistoune:
I'd like to know what format you save in once you've processed your RAW file. TIFF, JPEG, or PSD? I've heard several different arguements on the subject but a definitive answer would be helpful. Thanks |
TIFF prior to printing, JPEG for web and projector display, PSD for intermediate/ work in progress with layers. |
Thank you, Gordon!
For answers to questions like these and digital workflow/imaging in general, what is the best book on the subject in your opinion? |
Take a look at this thread for a discussion on books for PhotoShop CS.
David
|
|
|
|
07/01/2004 08:12:59 PM · #11 |
Originally posted by Britannica:
Take a look at this thread for a discussion on books for PhotoShop CS.
David |
Which thread? |
|
|
|
07/01/2004 09:24:35 PM · #12 |
Originally posted by Acier:
I am new to digital photography and have the same camera as hgordin and breckinshire (EOS-300D Rebel).
What I am wondering is how do I setup raw format on that particular camera? I curently use highest quality jpeg but can't see the setting for RAW format. Or is it something that is setup in Photoshop CS?
Thanks |
RAW is only available in the "creative" modes: P, Tv, Av, M, and A-DEP. The "basic" modes are designed to be easy to use, and RAW isn't. |
|
|
|
07/01/2004 09:30:13 PM · #13 |
Originally posted by digistoune:
Originally posted by Britannica:
Take a look at this thread for a discussion on books for PhotoShop CS.
David |
Which thread? |
Oops! Is my face red!
This thread.
David
Hanging head in shame. :(
|
|
|
|
07/01/2004 09:45:25 PM · #14 |
Reading Jelloeye's original question, I realize it wasn't really answered and now I'm curious as well. Let me try to rephrase it and see if anyone has any insight.
Shooting in RAW results in an image with the same resolution (for a given size) as a JPEG image, but with a greater pixel depth (12 bits/channel instead of 8). Does this make any difference for resampling? Can the resampling algorithm use the extra pixel depth to make higher quality interpolations? |
|
|
|
07/01/2004 10:45:38 PM · #15 |
What the 12 bits per pixel gives you is about + and - 2 stops in converting your picture from raw to 8 or 16 bit tiff using the software that came with your camera. You can also purchase, for $99, Phase One C1 light software, that allows you to apply levels and curves to the raw exposure allowing larger gradients in shadows or highlights or both. This capacity is available in Photoshop CS as well at a higher price. If you keep the image in 16 bit tiff format, you will have greater range in the colors represented, but only a display medium capable of displaying 12 bit colors will be useful for this purpose. Most graphic cards and most computer monitors are 8 bits only.
You can also get larger gradients shadows and highlights by only using the software that came with your camera. Transfer three images from raw to photoshop, one overexposed, one correctly exposed, and one underexposed. Place the correctly exposed on the base layer, the overexposed on the next layer, and the underexposed on the top layer. Now select the layer containing the overexposed and multiply blend at about 50% to bring the details out of the shadows, then select the layer containing the underexposed and screen blend at about 50% to get the details out of the highlights. Then, when satisfied flatten the image. This is the cheapest option, it works well, but is much harder than using either Phase One C1 Light or Photoshop CS.
Take a look at the two pictures in my portfolio of cormorants. The one that has detail in shadows and highlights was done with Phase One C1 light, but I did the equivalent using the levels blending also.
|
|
|
|
07/02/2004 10:34:23 PM · #16 |
I have been able to double and triple my print file size by increasing horizontal or vertical dimension by 10% at a time in photoshop without introducing any artifacts. I have done this with jpeg files, although you could try doing it with 16 bit tiff files and then converting to jpeg when you are done.
|
|
|
|
07/02/2004 11:31:10 PM · #17 |
Extra precision pixels don't actually get you any more dynamic range. The start and end points are still the same - just the number of potential steps in between is increased. The tonal changes are more precisely (not accurately) recorded. The big advantage to this is that when you make changes - particularly large changes like lightening the entire image by 2 stops or doing dramatic curve adjustments, that all that extra precision can be spread out to give a reasonably pleasing result, still maintaining smooth tonal transitions. When you do this with an 8 bit per channel image, you end up spreading the data out across wider step intervals - typically leading to banding and generally dissappointing results.
So the extra precision of high bit files allow you to make more dramatic edits - if you don't do the editing it doesn't gain you anything. You will at some point convert your 48 bit per pixel, 16 bit per channel images down to 24 bit per pixel, 8 bit per channel final images - at which point the extra precision is thrown away and you are down to 255 levels per channel.
e.g., the decimal equivalent would be to consider numbers that were always rounded to one decimal place, compared to numbers that are rounded to 2 decimal places.
If we assume the dynamic range is always between 0 and 1 - with one decimal place you can have the following levels
0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1
And that's it. You can't have a 0.15 level - if you move images around, or under expose them - all the data bunches up in those fixed bands - if you apply a sharp curve to recover an underexposure - you can spread it out, but there isn't a whole lot of subtle detail left. If you under exposed - and all the detail was between 0 and 0.2 - you'd have an image with exactly 3 possible levels.
The two decimal place version lets you have levels like
0.0, 0.01, 0.02 0.03, 0.04, 0.05 and so on - 10 times more detail, and 10 times more levels, just between each point on the original scale - now if you under expose and bunch all the data down in the 0 - 0.2 range - there is 20 levels of detail there that you could pull out later.
but in both cases the highs and lows are in the same place - just the number of steps in between changed.
Message edited by author 2004-07-02 23:32:10.
|
|
|
|
07/02/2004 11:48:46 PM · #18 |
| Gordon, that is the most elegant explanation I've ever seen for this issue. Thanks! |
|
|
|
07/03/2004 05:15:35 PM · #19 |
The four extra bits give you 2048 rather than 256 choices for each of the three colors in each pixel. They do not increase the number of pixels. However, when increasing the number of pixels by resampling, using software such as photoshop, this is done by interpolations. The higher the precision for each original pixel, the better the interpolations will be. If you increase the number of pixels by resampling, you will not gain any resolution, but you will have a smoother transition with the added pixels with the better interpolations. It seems clear that it would be best to convert raw to 16 bit tiff and then increase by resampling the tiff prior to converting to jpeg.
|
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 04/02/2025 10:24:53 PM EDT.
