OCRd PDFs - slow - what to do?

[LivresInOz]

(This is in NO WAY a criticism of an app I love and use daily. Its just an enquiry seeking help)

OCRd PDFs on Koreader can be very slow:
- opening
- turning pages
- registering gestures (which causes the most difficulty)

Yet I have no choice but to OCR some docs.

Why is this so slow (I'd like to learn about this)
and
more importantly, any suggestions how I can improve the situation, either:
1) Using settings in Koreader
2) How I create the PDF?

___________________________
PDF converting/SAVING options

Re (2), please see page 16 of this guide, under "Adobe PDF Tasks", showing the three main options including:
a) Text and pictures only
b) Text over the page image
c) Text under the page image

https://static1.abbyy.com/abbyycomme...sers-guide.pdf

[sgmoore]

Anything other than a) Text and pictures only could produce a pdf which is much larger and hence take longer to load , require more memory and take much, much longer to display.

Don't know about the kobo, but any ereaders I have seen have been optimised for reading text and hence have slower processors than general purpose phones/tablets and computers and so don't handle graphics well.

For example, I borrowed a really old book from archive.org which was obviously ocr-ed with the page images left in it and it was about 50 times the size I would expect and so slow it was unusable on my kindle, but fine on my pc.

Of course, omitting the page images won't work very well unless the ocr works perfectly in the case of technical documents or almost perfectly in the case of novels.

If you do need the images, you can try lowering the image resolution to see how low you can go.

[Frenzie]

Yes, I don't think the mere addition of OCR should make a noticeable difference in performance. Are you saying these files are unproblematic prior?