Paperwhite 2018 looks worse than 2013? (Picture inside)

[JSWolf]

Originally Posted by BookCat

This is why I recently bought a used PW3 from ebay. When the PW4 first came out many peeps complained about the poor contrast. I didn't need the waterproofness, so went for the PW3.

The PW2, which I've had for years, has lower resolution (212 dpi) than either the PW3 or PW4 which are both 300 dpi. This is one reason the earlier paperwhite looks a bit 'fuzzy'.

Maybe return the PW4 and buy the Libra or try to find a PW3.

One thing I have noticed is that Kobo has better dithering routines for graphics.

[ilovejedd]

Originally Posted by JSWolf

Interesting. The PW2 looks fuzzy. The PW3 has more contrast with darker blacks and the PW4 is close to the PW3, but not as good.

So far from what I've seen, I think the PW3 has a better screen and/or light then the PW4.

Yes, among the Paperwhites, the PW3 has the best display.

I don't know if the lower contrast is due to the e-ink display itself or if it's because there's an air gap to get flush bezels on the PW4.

Still, I find manga on PW4 perfectly readable. Certainly more legible at small text sizes than the PW2.

Besides, most of my manga get "converted" to CBZ anyway and read with KOReader which, as mentioned, offers contrast adjustment.

[ilovejedd]

Originally Posted by JSWolf

One thing I have noticed is that Kobo has better dithering routines for graphics.

Hmm, I beg to differ.

Granted, reading manga on 7" is more comfortable than on 6". I really wish Kobo had a 32GB version of the Libra H2O. Mind, my experience with 8" devices mostly consist of tablets that are 300+g. At less than 200g, the Forma might actually be feasible.

From left to right: PW3, PW4, Libra H2O

[NiLuJe]

Can't judge dithering given the picture quality, but what I can discern is a different grayscaling process: one of them is gamma corrected, the other is linear (c.f., the 42's tone, probably also affects the contrast between the hair and the bandana).

(That's the *Luma vs. *Luminance algo in IM speak for the -grayscale switch).

So, curious: is it the same source? Is that the native reader in each case? Or KOReader?

[Gametaku]

Still undecided whether to return the PW4. Just purchased the Kobo Libra at 280SGD (around 205USD) and found out that the price is higher in my country

Will wait for the Kobo Libra to reach me before deciding what to do with the PW4.

[j.p.s]

Originally Posted by ilovejedd

Hmm, I beg to differ.

Granted, reading manga on 7" is more comfortable than on 6". I really wish Kobo had a 32GB version of the Libra H2O. Mind, my experience with 8" devices mostly consist of tablets that are 300+g. At less than 200g, the Forma might actually be feasible.

From left to right: PW3, PW4, Libra H2O

Originally Posted by NiLuJe

Can't judge dithering given the picture quality, but what I can discern is a different grayscaling process: one of them is gamma corrected, the other is linear (c.f., the 42's tone, probably also affects the contrast between the hair and the bandana).

(That's the *Luma vs. *Luminance algo in IM speak for the -grayscale switch).

So, curious: is it the same source? Is that the native reader in each case? Or KOReader?

Screenshots would be preferable to photos for comparing dithering. (I understand that getting screenshots from kobos can be problematic.)

[NiLuJe]

@j.p.s: Screenshots won't help on (Mk. 7) Kobo, the dithering is applied by the EPDC & PxP on its own private work buffer, and not earlier by the software. You'll be grabbing an undithered framebuffer .

[ilovejedd]

Originally Posted by NiLuJe

Can't judge dithering given the picture quality, but what I can discern is a different grayscaling process: one of them is gamma corrected, the other is linear (c.f., the 42's tone, probably also affects the contrast between the hair and the bandana).

(That's the *Luma vs. *Luminance algo in IM speak for the -grayscale switch).

So, curious: is it the same source? Is that the native reader in each case? Or KOReader?

Excellent point. I just opened books I had available on both platforms and I didn't stop to consider the difference in source quality.

The KEPUB is 150MB while the KFX is likely around 40-50MB. The KEPUB looks amazing on iPad but it appears the KFX version is more optimized for e-ink. Both were viewed using the native reader. Lol, been too lazy to install KOReader on the Libra as the Nickel has been working quite well for me after a couple of page turn patches.

Alas, I don't know of a way to do a lossless conversion for graphic novels to/from AZW3 and EPUB (such that images display fullscreen same as with purchased content).

Closest I could come up with:
1. Download and transfer via USB
2. Convert to EPUB via KindleUnpack
3. Rename EPUB to CBZ and delete OEBPS related files

This gives an AZW3 for reading on the Kindle native reader and CBZ for reading on Kobo Nickel.

Contrast on PW3 still appears to be better than Libra, though.

[NiLuJe]

Yep, makes sense . I was going to say that if it was a "native" platform-specific format, the Kindle one was likely pre-processed properly (i.e., perfect size for the target device, grayscale, and possibly pre-dithered), while the Kobo one... wouldn't ;p (it's probably even still full color?).

[Quoth]

Screen shots on the Kobo Aura H2O original are basically whatever is in the file, so yes they can be in colour.
I'd only use a full spectrum photography lamp reflected off a neutral surface or a grey overcast sky, not near dusk or dawn, to photograph eInk. Also only with light at minimum or off. Bright enough light so that any minimum LED light on Kindle isn't affecting the image. No reflections.

A "screen shot" can't ever show what ANY technology of screen is like as there is the display characteristics (CRT, Plasma, various LCD, OLED, real LEDs, different eink generations etc) as well as software.

It's really difficult because the camera will not have the dynamic range or gamma of your eyes. You need a calibrated screen and adjustment of the photo and even then obvious differences in real life may be much less or distorted compared to the images you view. Colour is even harder than monochrome due to the spectrum/colour rendition as well as colour temperature of lighting (LED and Fluorescent have per model non-continuous spectrums) and also the method of making the camera do colour. Then the backlight colour temperature / spectrum of eventual viewer's display and how that display does colour, the phosphors on OLED or CRT, the dye filters on LCD (all of which age with resulting colour shift). The gamma, contrast, dynamic range, brightness and ambient light of final display.

[ilovejedd]

Originally Posted by NiLuJe

Yep, makes sense . I was going to say that if it was a "native" platform-specific format, the Kindle one was likely pre-processed properly (i.e., perfect size for the target device, grayscale, and possibly pre-dithered), while the Kobo one... wouldn't ;p (it's probably even still full color?).

Yep, full color and very high resolution.

Tried an AZW3 file (and lossless extracted images as CBZ for Libra). Using native reader, brightness set to 0 (off for PW4 and Libra, very dim light on PW3).

Scanner did a much better job than taking pics with my smartphone. Alas, no RAW mode I know of so this uses 300dpi, JPEG compression followed by JPEG lossless crop.

I think the dim light on the PW3 made it look more washed out compared to the others. The PW3 has better contrast than either PW4 or Libra albeit in this particular example, the PW4 is really close. I find myself preferring the image processing techniques Amazon uses versus Kobo. Makes comics easier to read.

From left-to-right: PW3, PW4, Libra H2O

[NiLuJe]

Yeah, the (stock) CBZ reader is... painful on Kobo ^^.

Plus, 32bpp, so you get a "nice" grayscale -> RGB32 -> grayscale roundtrip for no good reason (well, no, there's a "good" reason, it's called Qt5).

You'll *probably* get something rendering closer to the Kindles on KOReader .

But, yeah, that's much closer to what I'd expect those to look like in person .

(On the scanner sidebar: the flush screen probably helps the PW4, too, bringing the screen "closer". Conversely, the Libra's "handle" probably doesn't help, pushing it a bit farther away).

[ilovejedd]

Originally Posted by NiLuJe

(On the scanner sidebar: the flush screen probably helps the PW4, too, bringing the screen "closer". Conversely, the Libra's "handle" probably doesn't help, pushing it a bit farther away).

The "colors" on the Libra is actually very close to what I see with the naked eye. It really is noticeably lighter compared to the Paperwhites. I find the Kindles do a better job with the grayscale conversion with regards to compensating for the limitations of the e-ink display.

Mind, the native KFX does look a bit better than the AZW3 version.

Will check out KOReader on the Libra later. Lol, was just too lazy to figure out how to assign button mapping on the Libra so both buttons go next on KOReader. Alas, only 8GB storage means frequent content management when dealing with manga.

[Gametaku]

Originally Posted by ilovejedd

The "colors" on the Libra is actually very close to what I see with the naked eye. It really is noticeably lighter compared to the Paperwhites.

But earlier you mentioned the Libra has better contrast than both PWs?

[Quoth]

Libra & PW3 almost the same, and both better than PW4. But original Aura H2O is noticeably the best, though slightly lower resolution.
Compared all four in artificial light and daylight.