Unfortunately, this is probably going to be contentious.

I have 2 questions, 

1. Does anybody know the criteria for the screeners to approve or decline a scan.  I understand that it is subjective, but you would think that they would take votes into there decision.  I have seen 100% yes votes get declined, 25% no votes get declined, 100% yes votes get approved, 0 votes both ways.  It seems that there is no uniformity.

Number 2, when users vote on cover scans, do they pull out there original cover to compare, or do they base there decision upon what they prefer?

Again, I know scans are subjective.  I have seen scans that the contrast was tweaked to enhance features that are subdued (usually look washed out).  I have seen excuses like "I  prefer the original" or " Loss of detail". 

When reds are off, I can understand, or when black level are off, or poor cropping, or rerelease cover, these I do understand.

I have a different opinion than Skip has. In my opinion, nearly every "improved" scan I've seen moves well away from looking like the original, with visible increases in contrast (thus losing a detail in both the light areas and the darks) and with the artificial enhancement of added "sharpness" which adds non-existing colors on contrasting edges. Both move good images into eyepopping but inaccurate reproductions. The current contribution of Edward Scissorhands is a good discussion point. Both front and backs have the contrast increased, with loss of detail in both the light areas of Depp's face on the front and in the shadows. The back is mixed: there is a small loss of these details, but there are far fewer of them to lose. On the other hand, the graininess of the back cover is significantly reduced by reducing the saturation of the colors. So, (again, in my opinion) the front cover submitted is worse, and the rear cover is mixed better and worse, but I prefer it to what is currently in the database.

But most of our voters prefer unreal exaggerated images to accurate ones, and often the screener agrees.

Charlie, I also have the impression it's not always a clear-cut case.
I've submitted scans that got very little votes at all and they all got accepted.
I once submitted a scan that got 20+ yes votes and only 2 no votes, where I explain in my notes that I believe my scan improves the darker tones (the original didn't show some details in the black areas) and it got denied.
Sometimes it's really a matter of personal preference I guess, but I do get the impression that no votes on cover scans carry a lot of weight, meaning that as soon as there is one no vote, screeners tend to be a lot more critical towards a cover scan.

Agree with Katatonia: what you can do to try and make the process less random is to state explicitly in your contribution notes in what way your scans are better than the existing ones (e.g. better cropping, sharper, higher resolution). You might also point at certain areas (faces, patterns in clothing etc.) where the improvement is particularly visible.

Quoting VibroCount:

Quote:
I have a different opinion than Skip has. In my opinion, nearly every "improved" scan I've seen moves well away from looking like the original, with visible increases in contrast (thus losing a detail in both the light areas and the darks) and with the artificial enhancement of added "sharpness" which adds non-existing colors on contrasting edges. Both move good images into eyepopping but inaccurate reproductions. The current contribution of Edward Scissorhands is a good discussion point. Both front and backs have the contrast increased, with loss of detail in both the light areas of Depp's face on the front and in the shadows. The back is mixed: there is a small loss of these details, but there are far fewer of them to lose. On the other hand, the graininess of the back cover is significantly reduced by reducing the saturation of the colors. So, (again, in my opinion) the front cover submitted is worse, and the rear cover is mixed better and worse, but I prefer it to what is currently in the database.

But most of our voters prefer unreal exaggerated images to accurate ones, and often the screener agrees.

I agree with this completely.  I see so many images being submitted that have the contrast cranked up, the blacks cranked down and saturation cranked up to make the the image more "stunning", but unfortunately nothing like the actual cover image artwork.  These invariably get a vast majority of "yes" votes, because when you compare the two images on the screen, the "enhanced" one definitely "looks better".  I, personally, do not believe very many people actually pull out the actual cover and compare the images they just look at the two images on the screen and decide which they like better.....totally subjective.

Quoting Taro:

Quote:
You might think I'm crazy, but it has happened on some occasions when voting on a cover submission, that I've held my own cover next to my computer screen to check if the colours were spot on or not ... but I only do this when in doubt.

I always compare submitted covers  with the real thing held up to the screen.

Like Skip, I keep my my screen calibrated for photographic use, using Spyder 3.  Mostly I find that the new covers tend to be sharper, but the colours are off. Which then leads to a dilemma as to whether the new covers are "of significantly higher quality than the existing online images". There's a good example of this dilemma at the moment, where the submitted cover is larger, sharper, much less artifacts.....but the colour is way off. Indeed the colours of online are closer to the real thing than the submission.

I agree with Cliff (Vibro) all too recently there are far too many images coming through that are being "Photoshopped" and don't represent the actual covers due to over cranking of this or that.

Yet we state that credits should be as on screen, yet cover scans are Photoshopped to hell.

Take the image as scanned with DVDP I'd say, no enhancements just the Cover. When Photoshopped we are losing too much detail. As was recent seen in Aeon Flux the colours where cranked right out of proportion, cover should have been Blue Metallic with the new image almost black (it was withdrawn when pointed out: Thank you submitter). But some of these coming through lately are a waste of time IMHO way too dark, all details gone.

Steve

Steve:

I don't agree with Cliff. I recognize his many years in the printing business, but the fact remains is that eyes cannot calibrated. He sees things differently than the martian and i see things diferently from you.The only way we could really create a truly accurate representation of a cover is if we had some sort of notes from those that produced the cover  and i mean specific notes.

The same is to som edegree when "tuning" a house for Audio, no two ears perceive sound the same way, some owners prefer the sound to be brighter, some like deeper bass and so forth.

We could have this same argument over restored films. Sometimes the restoration team has specific notes as to how the film was to appear and the color base and so forth. I think the current version of GWTW is absolutely gorgeous, however I could argue as well that GWTW is not representative of the original film on one basis alone, the original Technicolor negatives were out of registration by FIVE pixels, this was corrected several years ago which resulted in literally a picture which had never been seen before, let alone in 1939, and presented the sharpest and clearest image ever seen of this classic film.

The biggest problem I see in images are images with have bad black levels, or have they brightness levels turned up to the point that the image looks as though it has been smeared with milk and so forth. The biggest problem I see with colors are usually in the reds and the blues, with a tendency to (by my eyes) to sometimes weaken the reds to the point where they are very weak or almost pastel and in blues and bent towards the violet end of the blue spectrum.

But Like i said, I recognize with my years in home theater installation that sound and colors are both highly subjective in both Audio and Color setups, dependent upon the eyes of thehome owner, not the installing technician, and i have seldom if ever seen these differences to major, they are usually very subtle. I think we have all seen what happens when one of the primarys goes out on a color...I have NEVER had a customer that he liked his image that green for example, or that they liked their registration to be out so far that it looked like you were watching a 3D image without glasses.  So many times you really are spitting hairs and you can't calibrate the human visual or audio tools, everyone is going to be different and we won't have access to the notes from the creative eyes that generated a given cover to tell us exactly which colors he used, or which other settings he used in creating the cover. We can only do the very best we can, and like others I sometimes hold the ACTUAL up next to the image, but that is no guarantee that i perceive the cover the same way as the Contributor. I give as detailed voting notes as I can in the limited space I have to try and describe any shortcomings I might perceive in a given cover.

Skip