Jump to content

M.Chapman

Verified
  • Content Count

    2,812
  • Joined

  • Last visited

Everything posted by M.Chapman

  1. Going that way is totally predictable. I wouldn't call it "mismanagement", (assigning the wrong profile or ignoring the profile is mismanagement). In fact this process is exactly what happens when merging two images when one is sRGB and the other is AdobeRGB. Suppose we have an AdobeRGB image that we want to add something from an sRGB image (e.g. cut and paste). PS will give a warning; Paste Profile Mismatch You are pasting content for a document with a different profile Source: sRGB Destination: AdobeRGB What would you like to do? (1) Convert (preserve colour appearance) (2) Don't convert (preserve colour numbers) (1) is normally used for photos (2) maybe useful for graphic designers Mark
  2. I think we agree on the principles. You wondered about the effect of converting and sRGB image to the larger AdobeRGB colour space. Two things happen when doing this. Firstly the sRGB image data in the file is compressed slightly, to leave room for the extra colours in the AdobeRGB gamut (that aren't there in this case). The AdobeRGB profile is then applied which expands the data back out ready to be displayed. The compression and expansion are mathematically inverse processes and nominally perfect (apart from a slight the loss of precision if it's 8 bit data). It's simple to try this for yourself. When you convert an sRGB image to an AdobeRGB image you will see no significant change (it doesn't matter whether you have an sRGB or AdobeRGB monitor). This process is doing exactly what it should, the system is trying to preserve the appearance of the image when converting from one colour space to a larger colour space. Going in the other direction ie. converting AdobeRGB to sRGB is moving to a smaller colour space and, if there are colours in the AdobeRGB image that are outside the destination sRGB space, they will be lost (typically clipped). The precise way the image change in appearance depends on the conversion settings perceptual/absolute colorimetric/relative colorimetric etc. You are also right that profile assignment leaves the image data alone. It simply determines the way the data is interpreted / displayed. If the data was encoded in sRGB colour space and shipped without a profile (as Alamy do) then assigning an sRGB profile is the correct thing to do. No matter what type of monitor you have. But assigning a different profile to one used to encode the image data is going to misinterpret the data. It's a bit like somebody sends you a file of speeds in km/hr. But you interpret the file as if it contains speeds in mph. You'll get the wrong result. There are times when folks will prefer the misinterpreted results (wow the car goes faster - or the colours are richer) .... but it's not a very legitimate thing to do. Unfortunately I suspect quite a few Alamy customers may be doing this inadvertently. If they open an Alamy file (with no profile) in an Adobe product which is set to use a default colour space of AdobeRGB (and they have the missing profile warning dialogs turned off) they will interpret Alamy sRGB data as if it's AdobeRGB and see a more intense image than they should. It's easy to test all these things in PS. You may also find this useful https://turbofuture.com/graphic-design-video/SRGB-AdobeRGB-and-ProPhotoRGB-colour-spaces Mark
  3. Also clicking "Save" when any tags that have been typed haven't yet been added (by clicking the "+" or hitting <CR>) looses the tags that have been typed in without any warning. Mark
  4. Could it be a consequence of out-sourcing various services to India where the Covid-19 disruption is significant? Mark
  5. Give yourself a free copy of the file and the proceeds of any sales.... 🙂 Mark
  6. Pure guesswork on my part, but given the customer will receive a jpg (not a TIFF), I suspect Alamy will have converted the large TIFFs to smaller jpgs and the original filename is just stored in the database for reference only. Mark
  7. Not if the sRGB image is correctly tagged it shouldn't. That's the point of colour management. Now if you ignored the sRGB tag and assigned AdobeRGB or P3 profile instead (easy to do in PS), then the sRGB image will "expand" into AdobeRGB or P3 space. But doing that is "sacreligious"... Mark
  8. I just tried what you suggest. If an sRGB image is correctly tagged, and then converted to AdobeRGB and also correctly tagged, then the appearance is unchanged (all the sRGB colours in the original can be displayed in AdobeRGB space - although some extra banding may appear in subtle gradients due to extra compression). The colours are lost when AdobeRGB is converted to sRGB (that's the most "sacrilegious" bit). I notice in the Apple video it also said P3 Gamut images had to be saved as PNG format for web-use to retain both 16bit and the profile info. I wonder if that's still the case, or does Apple's new HEIC/HEIF format now provide an alternative? (Although I imagine many web bowsers don't support this). Mark
  9. I think the conversion and profile strip is done at the point of receipt, but I'd have to dig back into my notes/emails with Alamy to be 100% sure. I imagine the further compression and generation of lower res thumbnails is probably all done at the same time. That's not to say they didn't keep an offline "archive" copy of what we submitted though.... Mark
  10. +1 Interesting article and video. Although I do think his video examples are massively over-exaggerated. He said they are "renditions" and "somewhat exaggerated" but as a result they are quite misleading IMO. Note how much even the "in gamut" areas of the parrot on the left change. I look forward to seeing a side by side comparison of one of my own images processed in sRGB and AdobeRGB (or P3Gamut) and rendered on a standard and wide gamut display when I next get a chance. The displays in John Lewis etc. look really impressive, but I'm wary that they are also showing images and videos that have been artificially "pumped up" in terms of sharpening and saturation. They are, after all, trying to sell us something (and so are Apple). We're going to reach the state where the real world looks very dull and boring compared to the renditions we see on our screens. Nevertheless, that's the way the market is going, so the ability to edit/produce/sell wide gamut images is going to be more and more important. I think it's something that Alamy urgently need to address. Converting images submitted in wide gamut (e.g. AdobeRGB) to smaller gamut (sRGB) and then shipping without profiles is really not good. I urge them to establish "wide gamut" support and ship with profiles attached. (It's such a shame they have already discarded any wide gamut information in the images they have in the library, so they have to start from scratch*). It's a fiercely competitive market and having a "wide gamut" offering will help them to survive. Wide gamut displays and fully colour managed devices and browsers are the new normal. *Something Alamy could perhaps offer contributors is an easier way to replace/update existing images with wide gamut or freshly edited versions? Mark
  11. For your first submission you are only allowed to submit 3 images which will then be individually inspected by QC. See information here. https://www.alamy.com/contributor/how-to-sell-images/alamy-quality-control/?section=5 Mark
  12. If you're viewing Alamy's webpage in a colour managed browser, using the same monitor, then monitor brightness shouldn't be the issue. However, there are a couple of factors that maybe affecting your view of the images. Alamy's images appear on a white background (which makes them look darker), whereas you may have been preparing them in an editor with a black or dark grey background (which makes them look brighter). Some suggest that images are edited on a white background. Also, if you submitted your images in AdobeRGB format, Alamy will have converted them to sRGB format which can make some colours look duller, if you have a wide gamut monitor. Try downloading a preview image and opening in your editor (make sure you use/apply an sRGB profile - as Alamy will also have removed the profile) and make a side by side comparison with the image you uploaded. Mark
  13. So you did - thanks for reminding me. That's good news. Mark
  14. Absolutely. In line with trying to preserve licence fees, Alamy needs to offer a superior product that makes the best of the latest device displays. Step 1 - Add sRGB tag to ALL existing images in the library Step 2 - Recommend contributors to upload in AdobeRGB and preserve the profile info. Offer customer the option downloading in AdobeRGB (if available) or sRGB format. Conversion of AdobeRGB to sRGB could be done "on the fly" so Alamy wouldn't need to store both versions. Images would be supplied with AdobeRGB or sRGB tag/profile attached. Mark
  15. I must confess to being slightly disappointed that you perhaps think I might not already know that... It's not what the charts are trying to do. They were designed for those with sRGB or AdobeRGB monitors to help check that their setup can render subtle variations in intensity of colours at the periphery of the sRGB and AdobeRGB colour spaces. I think they achieve that quite well. In my case I was also interested to see how my own system fared. Although I have a lower spec monitor, the exact range of colours it can render will depend on the primaries it uses. So this test gives me an insight into which colours in sRGB or AdobeRGB space aren't being rendered by my display. Yes I can't see exactly colours that are missing, but the charts show where clipping is going on. They also show, with some clarity, (by downloading the AdobeRGB version opening in PS and saving as sRGB) the hazards of creating sRGB images in AdobeRGB working space without taking additional steps (e.g soft-proofing). Ignoring the colours you can see (which should be more intense in AdobeRGB on a wide gamut display), I assume that you can clearly see 9 complete 360 deg concentric circular lines on both charts? Agreed. Shame Alamy ships images without profiles.... Yes, I make that comment at the bottom of the webpage. It's one of the reasons I am looking at the implications of updating to a wide gamut monitor whilst still needing to easily produce sRGB images for Alamy and others. Although working in sRGB workspace solves the WYSYWIG problem (on screen in PS will match the sRGB jpeg) , I agree that working in sRGB space makes little sense if investing in a wide gamut monitor. Another solution might be soft proofing, but I can't get it to work properly in PS (it just doesn't give reliable prediction of the end result). Do you know what's wrong with soft-proofing in PS? I'd noticed that LR soft-proofing seems better (at least the histogram changes), but I don't tend to use LR. Your final solution maybe the best idea. So are you able to confirm the following? Assuming you can see 9 concentric rings on the AdobeRGB chart when your monitor is in AdobeRGB mode, what happens when you flick the switch and put it into sRGB mode? Do you see the clipping effect? That's what I'd need. Mark
  16. Thanks. OK that's as expected. At the moment I work in sRGB colour space on an sRGB monitor (simples...). So I can see if clipping is going to happen during the editing phase. If I was working in AdobeRGB colour space with a wide gamut monitor the clipping won't be visible until after I've saved the sRGB jpg and reopened it (which is too late). What would I have to do to see that clipping is going to occur during the editing phase? Does soft proofing work? It's hard to see the effect on my sRGB monitor. Mark
  17. I should have said "Can you see the 9 slightly darker concentric lines?" as it's those that fade way into the coloured "donut" underneath. Mark
  18. OK great. Can those that see all the rings (10 rings, 9 darker concentric circles) try the following? Download the AdobeRGB TIF file from here https://drive.google.com/open?id=14Gyl0pAyrsrIyD43c8AwDX3TB80vX15y. Use the download button at top right of the page that opens. Don't use "right click - save as", or you may get the preview image that Google seem to convert to sRGB. Open in PS or the editor of your choice. You should be able to see all the rings (if your working space is set to AdobeRGB or higher). Now save as an sRGB jpg (as if submitting to Alamy). Now open the sRGB jpg in your editor (to inspect the image Alamy will receive). Do you still see all the rings, or do you now see an area of clipping? If you do, how is it possible (if working in AdobeRGB on a wide gamut monitor) to see that this clipping is going to occur before the AdobeRGB TIF file is saved as sRGB, without actually doing it and then reopening the file? Does turning on soft proofing reveal that clipping is going to happen? I'm just trying to get my head around the implications on my workflow (to produce sRGB images for Alamy and other libraries) of upgrading to a wide gamut monitor. (If it wasn't for the lockdown I'd have gone into the Apple Store and tested it for myself). Mark
  19. As part of trying to get a deeper understanding of the implications of editing in AdobeRGB colour space, when the end result I need (for Alamy and other libraries) is an sRGB file, I have been investigating the capabilities of my own display and effect on my workflow by producing some test images which I have posted online here. Any comments welcome. Can those with wide gamut monitors confirm whether they can see all of the 9 concentric rings (the darker lines) in the AdobeRGB version of the chart? Mark
  20. I contacted Contributor Services on 1st of June with a query about a high value sale, and got an automated response which included the following in bold. Temporarily, we’re unlikely to be able to answer your email due to a reduction in team size whilst we respond to the global Covid-19 pandemic. So it may be best at the moment to delete images manually (it might be quicker in the end...) Mark
  21. It's not a virus, it's (unfortunately) an official part of the login process. It seems random when it appears. Sometimes I don't see it for ages, then it suddenly pops up again. There are several threads about it in the forum - for example; Mark
  22. Alamy rarely confirm anything but it’s simple to test. At the moment it seems that Caption and Supertag have roughy equal weighting, with Tags having a lower weighting. But it does change from time to time. Mark
  23. Why do you want it back the way it was? What relevance would it have to customers? AIM will show them to you in the order uploaded. If you want to control the order in which your images of the same subject appear for a given search term then use the caption, tags, and supertags to help achieve this. It doesn't always work as other factors (e.g. a previous zoom of an image using the same search term) can promote an image and "Date taken" can also demote the position of older images. Mark
  24. It's quite possible that Alamy have changed the meaning of "New" from "Date uploaded" to "Date Taken". They often say that they can, and do, change the search algorithm Using "Relevant" tends to order by date of upload (if everything else is equal). In your case, it's altered because you have put Will Perret Photography as the caption of T9T67Y, T9T844 and T9T7WA and the Caption currently carries a higher weighting than a Tag. If you want to reliably see them in the order they were uploaded, then AIM does this, submission by submission. Mark
  25. I just tried it and all your images I looked at are ordered by "Date taken" (not Date uploaded) I only checked 20 or so, but it seemed to be working and isn't "kaput" Mark
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.