Covermakers Chat Thread

[msfry]

I can't wait to see some examples.

As the PDF I use has two images, the first one right side up, the second one upside down, I presume you'll extract the first one. Both images inserted are 600 x 600 (5" x5"), at 120 ppi resolution, but the Word program scales the first one down to fit into actual CD Case size shown on the page, which is 4.74" x 4.74", or 95% of original size. So I guess that means you scale my pdf covers back up to 600 x 600? The upside down picture is actually 600 x 600 and does not scale down. Both are sized to suit the job they are for -- hard case -- origami case.

Does higher resolution improve quality? Artwork snipped off a webpage using Word's Snipping Tool saves images by default at 96 ppi. I increase my resolution to 120 ppi in Photoshop before I begin working on a cover. Not sure what other covermakers do.

[annise]

I don't fully understand this

Some PDF's have the text as overlay over the image making the image extract lose that information. To make a rule that I think consistently creates a good cover candidate I have done the following:

But I suspect it may be just a difference use of the words
I don't remember seeing a pdf with text as an overlay - coverart is mainly made by finding a picture of the author or a place or something and overlaying text showing title, author, reader, and LibriVox which then becomes the image(picture in my pdf program) which is used in the PDF

Anne

[Availle]

Anne, if you make the pdf using MS word, you can just add an image underneath the text of the word document. I presume that's what Bengt means with "text overlay".

See for example my pdfs for the "Science History":
https://librivox.org/the-science-history-of-the-universe-vol-1-by-francis-rolt-wheeler/

The main cover is made with gimp and holds all the information including the text (meaning, it's really an image), while the back of the pdf is the text created with word with the full photo underneath. (Hope this makes sense... )

[annise]

So it has the "standard version, it's just prettying up the "back page" ? Nice idea but beyond my pdf software I suspect.

Anne

[Availle]

Yes, kinda.

As for "beyond the software", I don't think so.
You can insert any image into a word or open office document (that's how you get the front cover in there in the first place).
And then you click on the image for the settings - "Wrap -> in background" allows you to have the text you write in the document on top of the image. (That's for Open Office, but word is similar. )

HTH.

[msfry]

Based on my experience in Word, you can create a text box, making sure the "background" and "border" are set to "none", then drag the textbox over an image. You can apply all the standard effects to the textbox in this position. However, when you save the document as a pdf, it blends together into one image that can no longer be separated.

[BengtW]

I don't fully understand this

Some PDF's have the text as overlay over the image making the image extract lose that information. To make a rule that I think consistently creates a good cover candidate I have done the following:

But I suspect it may be just a difference use of the words
I don't remember seeing a pdf with text as an overlay - coverart is mainly made by finding a picture of the author or a place or something and overlaying text showing title, author, reader, and LibriVox which then becomes the image(picture in my pdf program) which is used in the PDF
Anne

An example:
https://archive.org/download/safar_nameh_2309_librivox/persianpictures_2309.pdf

The image extracted is just the picture itself, the text, the Librivox logo and white boxes are all separate objects and will not be part of the image but as separate images which cannot be reassembled as the position information is lost. Luckily this image extract is disqualified due to not being square. The image is indeed 1497x1498 and thus discarded. I can imagine this can cause problem in the future.

So far all look well and the video covers are much sharper using the new method.

[carthinius]

An example:
https://archive.org/download/safar_nameh_2309_librivox/persianpictures_2309.pdf

The image extracted is just the picture itself, the text, the Librivox logo and white boxes are all separate objects and will not be part of the image but as separate images which cannot be reassembled as the position information is lost. Luckily this image extract is disqualified due to not being square. The image is indeed 1497x1498 and thus discarded. I can imagine this can cause problem in the future.

Since this is coincidentally a cover I did, I know now that I need to change my workflow for your efforts!
I can include an extra step to reduce the cover image and all other elements into one image.

On a sidenote: Would it be helpful for you to get a seperate file in a higher resolution or a 16:9 image like this one (https://1drv.ms/i/s!Ajiz4PA9i4S7hO8jeQUa_0OQ1W2coA?e=3bc0bw)?

[BengtW]

Wow that cover is beautiful but I do not want you to do any additional effort for the sake of the YouTube channel. I just want to make the best out of what is already available. I am actually very how things are right now and my contribution here is only to make your cover justice by sing the best available source already available. But as the channel (creation of video, title, summary, playlists, etc) is 100% automated I need to use methods that consistently works. The summary part for example has been a nightmare to consistently generate with the chapter creation, unicode characters and the limited characters you can use on a video.

[carthinius]

It certainly sounds like a nightmare! I will from now on include a single-layered 300 dpi version of the cover in the PDF so that your automation will get the best possible outcome without to much trouble.

[annise]

Bengt
Thank you - I thought you felt that way.
Internet Archive is a precious asset seeming sometimes to be in an ocean full of sharks and LV is grateful to be a very small part of it and so we try not to use up too much of their storage on window dressing.
And Bengt I'm sure doesn't want to spend all his days on this.
If I didn't think covers were worth having I wouldn't make them - and I feel annoyed when I see them distorted to fit a rectangular space. But they are PD and I keep making them.
But we are not going to use up IAs storage for higher quality images.

Anne

Cross post - thanks carthinius

[TriciaG]

It certainly sounds like a nightmare! I will from now on include a single-layered 300 dpi version of the cover in the PDF so that your automation will get the best possible outcome without to much trouble.

You mean 600 dpi, right?

[carthinius]

You mean 600 dpi, right?

No, 300 dpi is a print resolution and corresponds roughly 1417 x 1417 pixels at the cover's measurements of roughly 4,7 by 4,7 inches – which means it should be more than enough for BengtW AND look nicely if printed!

[TriciaG]

Aha! Got it!

[msfry]

Internet Archive is a precious asset seeming sometimes to be in an ocean full of sharks and LV is grateful to be a very small part of it and so we try not to use up too much of their storage on window dressing.

But we are not going to use up IAs storage for higher quality images.

Anne

I am curious about these comments.
First, are our images stored on IA, not on Librivox?
Second, what is the "ocean full of sharks" comment?
Third, is IA complaining about a lack of storage space? Have we approached them about how much storage LV can use? Are we limiting ourselves needlessly?
Fourth, why can't we store the 600 x 600 images on LV servers until BengjW has uploaded them to Youtube (who stores them independently from us and from IA), then periodically delete them from our servers as we do other extraneous files? It would save him a lot of conversion work, it seems, and not cost anyone at LV more than a few clicks to clean house once a month or so.
Fifth, is Anne's the only voice allowed to weigh in here? From her comments above, it seems so.