Helpful Posts Helpful Posts:  0
Page 2 of 2 FirstFirst 12
Results 21 to 33 of 33

Thread: Experiment in file saving...and by golly, I think I'm on the right track!

  1. #21
    Moderator Dave Humphries's Avatar
    Join Date
    May 2008
    Location
    Windsor, Berks, UK
    Posts
    16,748
    Real Name
    Dave Humphries :)

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by rob marshall View Post
    I think it used to. But in CS5 it saves it as an option. Look at the image EXIF in the shot directly above.
    Hi Rob,

    If you mean Chris's single shot of a tree in front of a fence, it errors and only has file info, no camera info.

  2. #22

    Join Date
    Aug 2008
    Location
    Ariege, France
    Posts
    558
    Real Name
    Paul

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by Colin Southern View Post
    Unfortunately, (and off memory), it also strips out EXIF data, and thus makes it harder to give C&C.

    By the way, I used to use Prophoto, but in reality, if a colour is so "far out" that it can't be represented in Adobe RGB then it's extremely unlikely that it can be printed or displayed anyway, so these days I just stick to Adobe RGB. Personally, I just created an sRGB action and have it as a 1-click button on my list of actions, so with one click I convert to 8 bit & sRGB.
    Depends on the printer really. Mine can (supposedly) support the ProPhoto colour space and there are noticeable differences - slight but still there between sRGB and ProPhoto. A lot of conjecture on this subject but personally I often do a fair bit of editing on some images - tones and colours often get subtly altered to the extent that they go out of gamut. Probably right that they can't be displayed as the colours get 'changed' to fit the available colour space but often an out of gamut warning disappears on conversion to ProPhoto and prints as expected. The EXIF is just overhead that I don't need and one of the things I like about Save for Web is that it is stripped out. For C&C though you have a point.

  3. #23
    rob marshall

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by Dave Humphries View Post
    Hi Rob,

    If you mean Chris's single shot of a tree in front of a fence, it errors and only has file info, no camera info.
    I entered my post at 20:35, and Chris added a shot at 20:42, which made my reference invalid. I meant the Teddy shot, which is now above the tree shot. It has EXIF in it from a save for web.

  4. #24

    Join Date
    Aug 2008
    Location
    Ariege, France
    Posts
    558
    Real Name
    Paul

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by Dave Humphries View Post
    I'm gonna be stubborn

    I have never understood the relevance of dpi and physical sizes when applied to downsizing an image for web use.
    Why juggle with a ratio and physical size just to achieve an answer that's directly available in the droplist of downsize options? Namely Pixels.
    I was scandalized when I read this It's the web, it's got to be 72 dpi - no question. So before sending off a suitable reply I did some research. Some joker has put up a site saying it doesn't matter, with pictures - one at 300 dpi and another at 1dpi (one!). Obviously completely barmy! Several sites claim the same, I'm starting to doubt myself and all the 72dpi claims I downloaded some of these images, opened them up in Photoshop and sure enough there is no difference on the display. File sizes are the same (give or take 4 bytes) because the only difference is in the jpeg header which has the dpi value - totally ignored by the monitor. So, doesn't matter a damn, 1dpi or 300dpi the monitor will just display the image at whatever pixel size you set.
    Another myth broken, I'm crushed. I was about to Google 'Santa Claus exists or not' but honestly I've had enough for one night.

  5. #25
    Moderator Dave Humphries's Avatar
    Join Date
    May 2008
    Location
    Windsor, Berks, UK
    Posts
    16,748
    Real Name
    Dave Humphries :)

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by rob marshall View Post
    I entered my post at 20:35, and Chris added a shot at 20:42, which made my reference invalid. I meant the Teddy shot, which is now above the tree shot. It has EXIF in it from a save for web.
    I agree I can see the EXIF in that one

  6. #26
    rob marshall

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by bambleweeney View Post
    I was scandalized when I read this It's the web, it's got to be 72 dpi - no question. So before sending off a suitable reply I did some research. Some joker has put up a site saying it doesn't matter, with pictures - one at 300 dpi and another at 1dpi (one!). Obviously completely barmy! Several sites claim the same, I'm starting to doubt myself and all the 72dpi claims I downloaded some of these images, opened them up in Photoshop and sure enough there is no difference on the display. File sizes are the same (give or take 4 bytes) because the only difference is in the jpeg header which has the dpi value - totally ignored by the monitor. So, doesn't matter a damn, 1dpi or 300dpi the monitor will just display the image at whatever pixel size you set.
    Another myth broken, I'm crushed. I was about to Google 'Santa Claus exists or not' but honestly I've had enough for one night.
    probably talking out of my rear here, but I always understood that dpi doesn't make any difference to web display. It may affect print, but not web. It is pixels that really count. If you take an original image of 4,000PX long-side, and reduce it to 1,000px, every four pixels have to be reduced to one physical pixel. That results in jagged edges and some problems with colour transition, as you would expect. That's why a 700px image looks worse than a 1024px image at the same quality. The reason I said 72dpi before is that some applications may ask for it, and you may have to set something, and 72 dpi is just a conveniently low number. It's pixels and quality levels when saving that you need to worry about. But there is always a compromise over the final file size.

  7. #27
    Moderator Dave Humphries's Avatar
    Join Date
    May 2008
    Location
    Windsor, Berks, UK
    Posts
    16,748
    Real Name
    Dave Humphries :)

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by rob marshall View Post
    It's pixels and quality levels when saving that you need to worry about. But there is always a compromise over the final file size.
    Agreed and on the latter, I did tests for myself when I started and as a result of those, I use 9 out of 12, or 75/80%, for my jpg saves, anything higher justs wastes web storage space (and bandwidth), go lower and you begin to lose quality, slowly at first, in other words 8 is probably ok too, but I wouldn't go lower than 8.

  8. #28
    rob marshall

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by Dave Humphries View Post
    Agreed and on the latter, I did tests for myself when I started and as a result of those, I use 9 out of 12, or 75/80%, for my jpg saves, anything higher justs wastes web storage space (and bandwidth), go lower and you begin to lose quality, slowly at first, in other words 8 is probably ok too, but I wouldn't go lower than 8.
    Yeah, I found that today when I was trying it. A setting of about 80-85% is fine. Any higher and you don't see any difference, and lower down below 70% looks horrid when seens at 100%.

  9. #29

    Join Date
    Dec 2008
    Location
    New Zealand
    Posts
    17,660
    Real Name
    Have a guess :)

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Quote Originally Posted by bambleweeney View Post
    Depends on the printer really. Mine can (supposedly) support the ProPhoto colour space and there are noticeable differences - slight but still there between sRGB and ProPhoto.
    It's easy to go OOG when printing sRGB*, but I'm referring to Adobe RGB. Basically sRGB covers the gamut of traditional monitors, whereas Adobe RGB expands it to include most CMYK printer gamuts. Prophoto on the other hand is more of a "theory" than a "practice" though. In theory it captures all of what the camera is capable of recording, but in practice we don't have any way to reproduce the extra colours (neighter via monitor or printer), which begs the question of "why use it in the first place" (I'm not saying it's wrong, just saying I think we need to ask ourselves why we we think it's right).

    The downside is that with large colourspaces it's possible to manuipulate an image into areas that can't be displayed or printed - but - the monitor has to display SOMETHING ... unfortunately, it's often not what gets printed because of the different gamuts; so you end up with an image that prints differently to how it displays, and the reason isn't obvious.

    * by OOG I'm meaning it's easy to have colours in the source image that sRGB can't represent.

  10. #30

    Join Date
    Nov 2010
    Location
    Panama City, FL
    Posts
    3,540
    Real Name
    Chris

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    I am going to do this in two posts so you can see both images full size and you should be able to extract the EXIF data. I was almost following you up to a point, but got lost somewhere between it matters and it doesn't matter. The first image is the WEB way and the second, mine. I haven't previewed either one on here, so however it falls, it shall fall.

    Experiment in file saving...and by golly, I think I'm on the right track!

  11. #31

    Join Date
    Nov 2010
    Location
    Panama City, FL
    Posts
    3,540
    Real Name
    Chris

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Second is my way

    Experiment in file saving...and by golly, I think I'm on the right track!

    And, as it would seem, there isn't any perceptible difference which leads me to think (as I always tell my students) operator error. Well, neither is any easier than the other, but one, I am sure is less taxing on your primary server.

  12. #32
    rob marshall

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Chris

    I can't see any difference in the two shots above. The first is 316kb and the second is 287kb. Hardly any difference in size - 10%.

  13. #33

    Join Date
    Nov 2010
    Location
    Panama City, FL
    Posts
    3,540
    Real Name
    Chris

    Re: Experiment in file saving...and by golly, I think I'm on the right track!

    Finally...I can post again...

    Your conclusion is the same one I came to last night when I first posted this. I can only conclude that my original post of the two trees had some kind of operator (me) malfunction in setting up the post. Using the Save to Web is certainly easier - well, no harder and if it kepes the server from overload, I am all the more for it...cheers.

Page 2 of 2 FirstFirst 12

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •