(sigh), Kindle/KPF format Blues...

[Tex2002ans]

Originally Posted by jhowell

What are the dimensions of the two images you are using?

Why are you even using two different images instead of using “height=# width=#” to scale the larger image down to fit the older format?

That's what I'm scratching my head about.

Use a high resolution source image if you can (300+DPI).

Example:

1200x400

1. Scale the KF7 version down by some fraction width/height:

<img src="example.png" width="600" height="200" />

(so it looks okay on a Kindle DX, etc.)

2. Give the KF8 and higher the original resolution (or % width).

<img src="example.png" width="100%" />
<img src="example.png" class="fleuron" />
<img src="example.png" />

KF8 can still handle the higher resolution images... there's no need to try to create some sort of "middle resolution" image.

* * *

Amazon is recommending using at least 300DPI images or higher, where possible...

Like the fantasy map examples JSWolf always talks about. These publishers should insert the full resolution map in the ebooks (or at least 300DPI), not give us some low resolution, scaled-down, pixelated, unreadable mess.

As jhowell quoted above, high DPI will allow pan/scan, and actually be able to read the higher-quality maps (even if the image is larger than the device screen itself).

But there's no way they can make that mandatory. There's too many books out there (like book scans) which don't have such high resolution source images.

Again, it seems like a recommendation, and letting you know what's possible with Enhanced Typesetting, not a requirement.

[slowsmile]

jhowell...

"In my research into KFX format, which implements Enhanced Typesetting, I have come across a different technical meaning of HDV. It seems that images that exceed 1920 pixels on a side are allowed in reflowable books in KFX, but not KF8, and books with those images are tagged internally as having yj_hdv 1.0."

Yes I would agree with you about the above. But if you format your epub for KF7, KF8 and KFX, then mobi7 format will be used on your KF8(without ET). That also means that your smaller images for mobi7 will be in absolute pixels might be OK. But this is really all dependent on how you store your images in the epub. Do your store your smaller images for KF7 in the epub or do you store your larger images at the HD sizing for KFX? If you store them as for the smaller KF7 sizings the display on the KFX device will be degraded because the images will be too small -- not to the HDV spec. And if you store your images in your epub at the larger HD sizings then KFX displays will be OK but the displays on KF7 devices and KF8 devices(which also use mobi7 image format) will probably be much larger than screen width so they will just default to max screen width by default on both KF8 and mobi7 devices.

[slowsmile]

@Tex2002ans...

1. Scale the KF7 version down by some fraction width/height

You can't use percentages values for scaling in mobi7 because it will ignore that styling. Remember this is OLD mobi7. The only way you can style and size a mobi7 is using absolute pixels in "height" and "width" declarations within the image tags. So forget about percentage formatting for mobi7 because it's not allowed. If you don't believe me then best ask Hitch.

2. Give the KF8 and higher the original resolution (or % width).

If you increase the size of an image using percentage values and you go beyond its actual pixel size that's when you will get blurriness and pixelation problems due to extrapolation -- which is just a guessing game at best. Not advisable.

[Tex2002ans]

Originally Posted by slowsmile

Remember this is OLD mobi7. The only way you can style and size a mobi7 is using absolute pixels in "height" and "width" declarations within the image tags.

... That's exactly what I wrote.

Originally Posted by slowsmile

If you increase the size of an image using percentage values and you go beyond its actual pixel size that's when you will get blurriness and pixelation problems due to extrapolation -- which is just a guessing game at best.

Which is why you use high resolution sources. Much less chance of blurriness/pixelation due to upscaling.

Again, why they recommend 300+DPI source images.

Back to the fantasy map example. Better if it was downscaled to half size on your Kindle and "unreadable", you click on it, and now you can see the full resolution image and pan/scan.

If you preemptively included a lower resolution map, users have an unreadable mess, AND there's nothing they can do about it.

[Hitch]

Originally Posted by slowsmile

jhowell...


Yes I would agree with you about the above. But if you format your epub for KF7, KF8 and KFX, then mobi7 format will be used on your KF8(without ET).

Wait. Where did you get this conclusion, please? Because you think that everything is either KFX or KF7, is that right? So...you think that the KF8 "version" of the MOBI that Tex, JHowell and I are contemplating as existing, won't exist, is that right?

BUT...if we make the images the recommended size, KF8 coding or not, won't they simply slide into the KFX file?

What am I missing here?

That also means that your smaller images for mobi7 will be in absolute pixels might be OK. But this is really all dependent on how you store your images in the epub. Do your store your smaller images for KF7 in the epub or do you store your larger images at the HD sizing for KFX? If you store them as for the smaller KF7 sizings the display on the KFX device will be degraded because the images will be too small -- not to the HDV spec.

No offense, but I think that this is where the trolley is leaving the tracks. You have multiple images for each, yes? The HD and the KF7 version. You can opt to hide the KF7 for the ePUB, no? Or am I losing my mind?

And if you store your images in your epub at the larger HD sizings then KFX displays will be OK but the displays on KF7 devices and KF8 devices(which also use mobi7 image format) will probably be much larger than screen width so they will just default to max screen width by default on both KF8 and mobi7 devices.

Okay--are we talking ePUBs? Actual ePUBs? Or are we talking about Kindle files? Can we specify so that I can be sure that when you say ePUB, you mean an ePUB on an ePUB device? Or "ePUB-source" when you mean the souce material for a mobi? When you say ePUB, I think that you mean a file being read on iBooks or B&N, so...I think I'm reading something different than what you mean when you type it.

Thanks.

Hitch

[jhowell]

Originally Posted by slowsmile

You can't use percentages values for scaling in mobi7 because it will ignore that styling. Remember this is OLD mobi7. The only way you can style and size a mobi7 is using absolute pixels in "height" and "width" declarations within the image tags. So forget about percentage formatting for mobi7 because it's not allowed. If you don't believe me then best ask Hitch.

No one is saying to use percentage for MOBI7.

I will try a detailed example. I have an image that is twice a wide as it is high and I want to display it at 80% of the screen width.

As per the publishing guidelines, I size it to the minimum for 300 PPI at 960x480 pixels. (I could make it larger if I want.)

I know that on an old Kindle device with a 600x800 pixel display I want it to be rendered at 480 pixels wide (80% of 600).

Using media queries I cause one of the following to display and the other to be hidden based on the format delivered to the customer device:

not amzn-mobi:
<img class="show-for-non-mobi" src="myimage.jpg" style="width: 80%;">

amzn-mobi:
<img class="show-for-mobi" src="myimage.jpg" width="480" height="240">

[slowsmile]

OK, here are the results of my comparisons between a KPF file containing the same image as an epub file(for upload to KF7, KF8 and KFX).

And @jhowell and Hitch are both right. When I compared these two images in KP3 they looked OK and quite passable. So my humble apologies for causing any unnecessary stress or spontaneous heart attacks about this. But yesterday, during my testing, I was getting poor images during test for the epub images when using my media query plugin. I guess I was doing something wrong in my preparation. I'm also quite relieved because this result means that my media query plugin for Sigil is OK for a few more years yet. I'm also not going to post the images for comparison, there seems to be no point now.

And, just for interest sake, here's the dual format code that is automatically produced for my epub images using my media query plugin for Sigil:

<p class="P38"><img alt="Word same as upload.jpg" class="mobionly" height="189px" src="../Images/image15.jpg" width="506px"/></p>

<p class="P38"><img alt="Word same as upload.jpg" class="kf8only" src="../Images/image15.jpg" style="width: 75%;height: auto;"/></p>

The plugin has been working well, without any apparent problems, for several years.

[Hitch]

Originally Posted by slowsmile

OK, here are the results of my comparisons between a KPF file containing the same image as an epub file(for upload to KF7, KF8 and KFX).

And @jhowell and Hitch are both right. When I compared these two images in KP3 they looked OK and quite passable. So my humble apologies for causing any unnecessary stress or spontaneous heart attacks about this. But yesterday, during my testing, I was getting poor images during test for the epub images when using my media query plugin. I guess I was doing something wrong in my preparation. I'm also quite relieved because this result means that my media query plugin for Sigil is OK for a few more years yet. I'm also not going to post the images for comparison, there seems to be no point now.

And, just for interest sake, here's the dual format code that is automatically produced for my epub images using my media query plugin for Sigil:

<p class="P38"><img alt="Word same as upload.jpg" class="mobionly" height="189px" src="../Images/image15.jpg" width="506px"/></p>

<p class="P38"><img alt="Word same as upload.jpg" class="kf8only" src="../Images/image15.jpg" style="width: 75%;height: auto;"/></p>

The plugin has been working well, without any apparent problems, for several years.

Phew! I was starting to have a heart attack, lol. You know what, though, William--I'm okay with these kinds of discussions. I truly am. This is how we all cut through the BS and find out what really works. And discussions like this help me, personally, clear my head about what Amazon's doing and all that.

At my shop, as I mentioned, since CS went the way of the Dodo, we're *&^%$#@ busy. to the point that we might--gasp!--start a WAITING LIST. I mean...a waiting list????? What that means is, I don't get the chance, as I used to, to sit around and play with coding ideas, to see if this works or that does. When Amazon came out with the latest KP3, I was having kittens over the whole "export a KFX" file thing.

Plus, those of us in the biz don't really get a chance to do testing through the PW, as well. We are, to some extent, at the mercy of our customers. (For example, the customer that prompted this thread removed his KPF from sale. I can't buy or download a copy of it. I can't test it. He refuses to talk to Amazon...talk about grinding your teeth!).

So...I don't mind, AT ALL, hashing out this sort of stuff. I don't.

One last question--Can you confirm the original size of the image in your sample, in the last post, please?

Hitch

[slowsmile]

@Hitch...

"One last question--Can you confirm the original size of the image in your sample, in the last post, please?"

I used a banner-like image, that had text in it, which was 450 x 168 pixels in size with a resolution of 300 ppi. This absolute sizing in pixels wasn't exactly 75% of screen width. It was actually slightly smaller but I wasn't so concerned with exact size -- I was more concerned about the image quality for the KFX display. And if you want a more accurate image percentage sizing for all KFX images, you can do fine width adjustments by increasing or decreasing the baseline width value in the dialog in my queries plugin. To reduce the width of the image, increase the baseline value and to increase the size then just reduce the baseline value in the dialog. You could also change image width size of individual images by simply altering the percentage value in the image styling in situ.

"At my shop, as I mentioned, since CS went the way of the Dodo, we're *&^%$#@ busy. to the point that we might--gasp!--start a WAITING LIST. I mean...a waiting list????? What that means is, I don't get the chance, as I used to, to sit around and play with coding ideas, to see if this works or that does. When Amazon came out with the latest KP3, I was having kittens over the whole "export a KFX" file thing. "

Yeah, me too. My answer to the KFX thing was to learn as much as I could about Kindle Create and thereby learn KFX as well. I know that you probably don't have time to learn KC but I've often wondered what you would say to a customer who brings you a single KPF file and then tells you that he/she accidentally deleted their master Word doc AND master KCB project file and can you help?(There have also been several instances of such problems on the KDP forum). There is an easy solution to that problem when you realize that that the KPF file is just a zipped version of your KC project file -- that contains both your final master KCB project file and original master Word doc. And I stumbled upon that solution completely by accident. It was one of those "I wonder what would happen if I try this..." moments. There's nothing written anywhere on the web about how to recover your master docs for Kindle Create. I found the solution completely by 'accident', which is really just another example of how poor the Kindle support documentation is for KC.

"Plus, those of us in the biz don't really get a chance to do testing through the PW, as well. We are, to some extent, at the mercy of our customers. (For example, the customer that prompted this thread removed his KPF from sale. I can't buy or download a copy of it. I can't test it. He refuses to talk to Amazon...talk about grinding your teeth!)."

Yes, I can appreciate your frustration -- he's no help at all !! I'm just glad that I'm not a professional formatter who must argue against or succumb to every customer's whim and fancy. Honestly, that would drive me up the wall for sure.

"When Amazon came out with the latest KP3, I was having kittens over the whole "export a KFX" file thing. "

To be honest about it, with this new KFX environment which is so poorly documented, I'm have a tough enough time just formatting my ebooks and books just for me...lol.

[Hitch]

Originally Posted by slowsmile

@Hitch...


I used a banner-like image, that had text in it, which was 450 x 168 pixels in size with a resolution of 300 ppi. This absolute sizing in pixels wasn't exactly 75% of screen width. It was actually slightly smaller but I wasn't so concerned with exact size -- I was more concerned about the image quality for the KFX display. And if you want a more accurate image percentage sizing for all KFX images, you can do fine width adjustments by increasing or decreasing the baseline width value in the dialog in my queries plugin. To reduce the width of the image, increase the baseline value and to increase the size then just reduce the baseline value in the dialog. You could also change image width size of individual images by simply altering the percentage value in the image styling in situ.

Okay, thanks. I'm going to TRY to find some time to play with this.

Yeah, me too. My answer to the KFX thing was to learn as much as I could about Kindle Create and thereby learn KFX as well. I know that you probably don't have time to learn KC but I've often wondered what you would say to a customer who brings you a single KPF file and then tells you that he/she accidentally deleted their master Word doc AND master KCB project file and can you help?(There have also been several instances of such problems on the KDP forum). There is an easy solution to that problem when you realize that that the KPF file is just a zipped version of your KC project file -- that contains both your final master KCB project file and original master Word doc. And I stumbled upon that solution completely by accident. It was one of those "I wonder what would happen if I try this..." moments. There's nothing written anywhere on the web about how to recover your master docs for Kindle Create. I found the solution completely by 'accident', which is really just another example of how poor the Kindle support documentation is for KC.

Oddly enough--I actually do know that. It's about ALL I know about KC--but I too learned it completely by accident. How? A customer sent me his .kpf file and without thinking, I dl'ed it through my download accelerator. Just as it does with .pages files, it "thought" it was (yup!) a compressed file and it saved it as a .zip. I wasn't paying attention and double-clicked it--and sure as shoot, ta-da!, there was the archive! TOTAL accident.

Yes, I can appreciate your frustration -- he's no help at all !! I'm just glad that I'm not a professional formatter who must argue against or succumb to every customer's whim and fancy. Honestly, that would drive me up the wall for sure!! To be honest about it, with this new KFX environment, I'm have a tough enough time just formatting my ebooks and books just for me...lol.

Yeah....it's not easy. We deliberately moved to more actively marketing our print layout services, which I'd said we'd never do, when all the doofy and DIY tools came along. (I've been asked to betatest at least 10 different startup pieces of software/websites, all intended to be insta-eBook services, and I have yet to see one that's actually easier than "use styles, export HTML, open in Sigil," or even--ignoring the CSS--open your Word file in Atlantis, push button, but..)

I mean, TBH, I'd said since '09 that the day would come, that pretty much "any" book could be made by pushing a button (with certain assumptions, of course, like styles, headings, yadda). That day isn't here yet, but the reality of "type word file, make eBook" is pretty well here, now, if you JUST use Styles and Headings. If you have non-fic, well, then it's a bit more work. So...I always expected that the business would taper off and then effectively die out, leaving some MUCH smaller number of commercial formatters in existence.

What I hadn't foreseen was the vast percentage of the market that Createspace had. It's boggling, when I look at it now. Sure, I knew that they were doing a lot of books, but I never understood just HOW MANY. I know for a fact that at least one formatting biz that's been around a long time also was downright saved by the demise of CS and even we are busier than EVER and I mean, EVER.

And, of course, there are tens of thousands of more-complex books, that publishers and authors never thought that they'd be able to make, that they are now, which keeps businesses like mine busy.

But still...that day WILL come, eventually. Of course, I keep thinking that someone will come along and buy my shop, which would allow me to retire (woo!), before that day. :-)

Hitch

[slowsmile]

@Hitch...

"Yeah....it's not easy. We deliberately moved to more actively marketing our print layout services, which I'd said we'd never do, when all the doofy and DIY tools came along. (I've been asked to betatest at least 10 different startup pieces of software/websites, all intended to be insta-eBook services, and I have yet to see one that's actually easier than "use styles, export HTML, open in Sigil," or even--ignoring the CSS--open your Word file in Atlantis, push button, but..)"

Well, if we're talking about tools, I've recently dumped Scrivener as my favorite converter and gone back to using Jutoh again and have grown to love that app. With its new updated and easier-to-use user interface, it's the best of the bunch now IMHO because converting to epub using Jutoh also leaves me with the least number of tasks to do in Sigil. Jutoh now gives you a huge amount of fine control over your output epub. It's document cleanup is also unparalleled because it's cleanup goes well beyond just cleaning the cruft from your file(Jutoh cleanup now has 37 options).

[slowsmile]

"But still...that day WILL come, eventually. Of course, I keep thinking that someone will come along and buy my shop, which would allow me to retire (woo!), before that day. :-)"

Well, for my own reasons I think that your business will be busy for at least the next 50 years. One of the main reasons for saying this is that I very much doubt that Amazon will make their applications too easy. Similarly, Amazon makes a fair bit of dosh allowing The Big Five publishers to use their "special services". So if Amazon, via their free apps, makes it too easy for a person off the street to easily produce the perfect ebook or book that is the same quality as those produced by the Big Five, then Amazon would not get that extra money stream(which is considerable) from the Big Five for all those "special services" that we can't access. That's why I think your biz is as safe as houses for decades to come.

"But still...that day WILL come, eventually. Of course, I keep thinking that someone will come along and buy my shop, which would allow me to retire (woo!), before that day. :-)"

Well, heck that's everybody's dream, right? And that's a great thought. Just think -- you could retire to somewhere like Hawaii!! And you could even comtinue formatting books for people as a small side-line. You will need that extra dosh to pay for all the sun tan oil that you'll be using everyday...

[Tex2002ans]

Originally Posted by slowsmile

[...] I very much doubt that Amazon will make their applications too easy. [...] if Amazon, via their free apps, makes it too easy for a person off the street to easily produce the perfect ebook or book that is the same quality as those produced by the Big Five, then Amazon would not get that extra money stream(which is considerable) from the Big Five

This is absurd. Revenue isn't the only side the equation.

On the production side:

If these tools were so spectacular... there's nothing stopping "The Big Five" from using these tools to create 'the perfect ebook' either.

(And we all know, "ebook quality" isn't the only thing separating trade publishing from self-publishing.)

On the costs side:

Making it easier+higher-quality saves manpower all around:

• Less complaints from customers (higher quality books)
• Less customer support for authors (it converts 'perfectly')
• Less hours needed to tweak/convert

On the conversion side:

There will always be books that can't fit the "easy conversion" mold:

• Cookbooks
• Math
• Scans
• [...]

There will always be more format wrenches that get thrown in the mix:

• Google Docs
  • Have you seen some of the abominable code coming out of Google Docs?
• Microsoft Office (2010, 2016, 2019, 365)
  • Different versions of Word produce vastly different innards.
• ancient versions of software
• different OSes/browsers
• Desktop/Mobile

Side Note: It reminds me of this recent LibreOffice talk discussing copying/pasting between browsers/OSes... and the absolute hell it is:

LibreOffice Conference 2019 – Online: copy/paste

And on the user side:

You know that famous saying... creating smarter/easier tools creates dumber users.

These people will come out of the woodwork and break any tools, no matter how easy/simple, with even more absurd files (remember footnoteception?). :P

* * *

I would highly recommend watching this talk from ebookcraft 2018:

End of the Conveyor Belt: Quality control, support improvements, and user feedback

it discussed the QA backend in Kobo.

Probably the closest you'll get to hearing what actually goes on behind-the-scenes at Amazon.

[slowsmile]

@Tex2002ans...I disagree with most of what you say mainly because you haven't scaled in Amazon's greed factor for profit. Most people know that the Big Five publishers already get special treatment by Amazon which they will happily pay for but the ordinary indie publisher cannot afford. And I hardly think that Amazon would reveal their "special treatment" of the Big Five at conferences. That would create quite a stir wouldn't it, with indie publisher's angrily screaming "Unfair advantage!!". Like me you are just the little man. Live with it, my friend, because you have no other option and no real say concerning Amazon's well-known business tactics.

[Tex2002ans]

Originally Posted by slowsmile

Most people know that the Big Five publishers already get special treatment by Amazon which they will happily pay for but the ordinary indie publisher cannot afford.

What's the "special treatment"?