then let’s make a petition to envato. we all agree that ratings dont help anybody, even worse, they can kill the project. i think my project is very good, but i have only a one star since a happy customer moves on to other things and dont comes back just for a review. and if the project works smoohtly, it wont be any customer that will contact me in search for help, that i could ask nicely for a good review
When I’ve deployed the new watermark in FCPX I’m advised that it is too bright.
Your colleagues at Support seem confident that if there are any difficulties in utilising the new watermark unaltered in FCPX that “VideoHive team is on it”; so they’ve pointed me back to this forum.
Since I’m trying to replicate this example (supplied by you) of the watermark in action, perhaps you would simply indicate the precise settings you used to achieve this in FCPX? A screenshot of your FCPX settings would definitely help eliminate any possible confusion.
Thanks in advance.
felt_tips saidHave to agree with this since envato has really strict quality control.
The solution is to remove the star ratings. They achieve absolutely nothing. They do not represent the quality of the item. T
If this marketplace didn’t have reviewers then the star-rating system would have ment way more.
And since the marketplace does have reviewers, it follows that star-ratings are in fact an appraisal of their efforts too.
…unless what you are viewing is the combined alpha for the watermark and any underlying footage you have, in which case it should be a solid white (i.e. opaque). The alpha should look roughly like the watermark looks overlaid on black.
You’re right, solid white was due to viewing combined alpha for watermark and underlying footage. Alpha for the watermark alone does indeed look like the watermark overlaid on black.
Thanks for the alternative formats – but some quick tests yielded no improvement.
Searching the interwebs for watermark use in Apple Compressor or FCPX yields numerous examples where opacity is set in the software; none involve a watermark with preset opacity like the VH one. I’ll keep searching.
Whichever way, you should report back with your findings to Support / Mark Brodhuber.
Let’s hope Mark is still following this thread, because last time I contacted Support (July) it took them 7 days to respond. But I’ll reiterate – it’s inconceivable that the new watermark wasn’t extensively tested across multiple platforms prior to release. Envato prescribed the watermark – they will have the answer and will be able to advise an interim solution until they roll-out automatic watermarking.
Does this help?
I can see that the colour correction is off now, but what really puzzles me is how in a Normal mode compositing overlay, the 25% opacity version should have half the luminance of the 100% version. It should have one quarter of the luminance. Something is amiss here.
The only thing I can think of is that there’s a general colour correction layer over the whole thing or that FCPX is interpreting the footage with a rather extreme colour profile. Perhaps it thinks that this is linear footage. Still doesn’t quite explain the behaviour though.Are you able to view the alpha channel separately from the RGB values. That could be revealing. I’m afraid I don’t have FCPX, so I can’t test it. Perhaps you could ask advice on an FCP forum.
If I view the alpha channel separately from the RGB values, all I get is a totally white image. BTW when the .png is imported, FCPX automatically sets ‘Alpha Handling’ to ‘Straight’ which I believe is the correct setting (the other options being ‘Premultiply’ and ‘None/Ignore Alpha’).
Thanks for your input on all of this felt_tips
What’s “Correction 1”? Some kind of color correction? It appears to be present and active on all of the PNGs.
I’ve just checked out the new PNGs. There’s nothing wrong with them. Perfectly standard stuff. The new watermarks consist of a pure white fill, that has an alpha value of 20%. If you composite it against pure black, that should give you an RGB value of [51,51,51] measured in 8bit color.
The watermark also has a thin mid-grey line, around 1 pixel wide around the edge. This has an alpha value of somewhat less than 20%... around 15%. Composited against pure white, the dark regions should give you an RGB value of [235,235,235] to [245,245,245]. In other words, it’s very faint and the watermark will show up much more strongly against black than against white.
In your first image, although it appears to be at an Opacity of 25% and is over a near-black background, the color being created by the logo is [61,61,61], this is much lighter than it should be. This would indicate that the translucent areas of the watermark are being interpreted as almost opaque [244,244,244] instead of [51,51,51]. The only thing I can assume is that “Correction 1” is some kind of color correction that you have applied to your alpha channel. Could that be? What happens if you remove it?I’d be kind of surprised if FCPX could get it so wrong without completely misinterpreting or ignoring the alpha altogether. My money’s on user error.
Does this help?
Above are three screen grabs taken of FCPX in action. The first is an approximate match of the ideal previously demonstrated by MarkBrodhuber. To achieve that, the opacity of the layer containing only the watermark has been set to just 25%. The second and third are of a lighter background with the watermark layer opacity set to 100% and 25% respectively.
Regarding Blend Mode, only changing to ‘Hard Light’ makes a marginal improvement.
I admit that automatic preview image and thumbnail image creation might not be such a good idea (nor a simple one). However, the automation of the watermarking is now straightforward, especially because we aren’t allowed to tweak it in any way anymore, so why bother with manual labor on it at all?
ok so the watermark was really wrong after all
No, yes and no.
It was very definitely the new watermark, the size was slightly too large, but otherwise it was deployed as supplied and recommended.
In other of my clips the brightness/opacity is fine; in this dark clip it seems too bright. Is that an FCPX issue? Quite possibly.
Maybe other non-AE authors would like to contribute with their experience? Study of other previews would suggest the same watermark is producing brightness variations with different users.