Page turning latency benchmarks?

[Renate]

Originally Posted by Hyphen-ated

You could also have it go "I have detected the beginning of a touch. This might end up being a tap or a swipe, so I'll speculatively start any work necessary to respond as quickly as possible once I know which it is"

Yes, but all such work should be done before you even touch.

Originally Posted by Hyphen-ated

How many different devices have you tested with an oscilloscope in this way?

Two. Not enough. Not even the devices that I own.
You need to connect to the E ink data ribbon (or a test point).
I can even do this with a broken/missing screen.
If my mail connectivity were better, I'd ask people to send me devices that are broken.

[JSWolf]

Originally Posted by Hyphen-ated

Unfortunately, "my page turned faster" is much less useful than actual numbers for anyone who wants to make purchasing decisions informed by page turn speed

How would I time the page turn? I just know my H2O turns the page noticeably faster then your Aura HD.

Actually, page turn speeds for Readers that are no longer for sale are meaningless. Mine turns pages faster then yours. I would expect the current devices or some of them to turn pages even faster. So if you are looking to buy a Reader, slow numbers from an old Reader with a slower processor and slower screen are meaningless.

[Hyphen-ated]

Yes, but all such work should be done before you even touch.

Right, emphasis on "should". Reality may be different.

Originally Posted by JSWolf

How would I time the page turn? I just know my H2O turns the page noticeably faster then your Aura HD.

Have you got a smartphone? If you take a video of some page turns like the ones I posted, I will time it

Actually, page turn speeds for Readers that are no longer for sale are meaningless.

Buying pre-owned devices is a thing

In an ideal world there would exist a comprehensive database of latency measurements for all readers. This would also be useful for someone thinking about upgrading off an old device to see what kind of improvement they might see.

[Hyphen-ated]

Originally Posted by JSWolf

How would I time the page turn?

Or if you'd rather do it yourself than have me do it, what you do is:

* take the video. make sure to remove your finger from the screen as quickly as possible after each tap.

* open the video in a player that allows you to advance forward or backward by a single frame. for example, mediaplayerclassic (where the keybinding is ctrl-left and ctrl-right). it also needs a high precision timer, which you can do in MPC by right clicking the timer in the lower right

* go to the first frame where your finger touched the screen. note the time

* go to the first frame where the new text is fully readable and all the old text is gone. note the time.

* repeat for all 10 presses or however many you did

* paste results into a spreadsheet, calculate the average difference

[Renate]

Originally Posted by Hyphen-ated

In an ideal world there would exist a comprehensive database of latency measurements for all readers.

Including:

• Time to open a book and display first page
• Time to go to next page in normal operation
• Time to go back a page during normal operation
• Time to jump to an "unexpected" page

Things that affect this:

• Single core vs. multiple core, speed
• Size of display in pixels
• E Ink interface 8 bit vs 16 bit
• OS, app

FWIW: B&N Glow2, 3 are 8 bit Eink, Glow4 is 16 bit

[Quoth]

Page turning latency is only an issue on scanned pdfs, not real ebooks, even on ereader HW that's over 10 years old. Unless it's got a lot of big images.

You can narrate even from a DXG or KK3 without a gap. Ereaders without page turn buttons (Kindle Basic, Kindle PW, Kobo Aura etc) are slower than a Kobo Libra due to needing a swipe. Taps can inadvertently highlight or if on the wrong place bring up a menu. Processing a tap or a swipe is slower on IR than capacitive and both add latency.

[meeera]

Originally Posted by Quoth

Ereaders without page turn buttons (Kindle Basic, Kindle PW, Kobo Aura etc) are slower than a Kobo Libra due to needing a swipe.

I've had a lot of buttonless Kobos, and none of them yet have needed a swipe.

[Renate]

Originally Posted by Quoth

Page turning latency is only an issue...

Ereaders have not hit 100% acceptance by the public.
I certainly know people who have tried them and given them up.
What role does speed of page turn play?
(It doesn't matter that an ereader page turn is faster than a physical page turn.)

My old B&N Glows' page turns didn't kill me, but I'm happier with my Poke3.
YMMV

[ottischwenk]

I am interested in how fast pages are turned at most with cbr or cbz files - this can take a relatively long time with low computing power.

[Quoth]

Originally Posted by meeera

I've had a lot of buttonless Kobos, and none of them yet have needed a swipe.

I meant swipe or tap rather than button press. Sorry.

[Quoth]

Originally Posted by Renate

Ereaders have not hit 100% acceptance by the public.
I certainly know people who have tried them and given them up.
What role does speed of page turn play?

Most people haven't tried one. There is also a massive difference between a 6" @ 167 dpi eink with older screen and no front light and a 6.8" to 7" ereader with current generation screen, 227 dpi to 300 dpi and front light.

Most people reading eBooks use a tablet or phone app.

Many people prefer paper even if they do read eBooks.

Many people think eink is only 6" kindles, if they even know about eink at all.

This forum is of course totally unrepresentative of fiction readers in general.

I've only ever met people complain about PDF or comic page turn speed, never real ebooks of mostly text that are novels.

[JSWolf]

I prefer a tap. That way if I accidentally touch the screen I won't turn the page. I've been swiping ever since the Sony Reader PRS-650.

[Hyphen-ated]

Originally Posted by Quoth

Page turning latency is only an issue on scanned pdfs, not real ebooks, even on ereader HW that's over 10 years old. Unless it's got a lot of big images.

this is completely wrong, sorry. did you not see the videos I posted demonstrating horrific latency of well over 1 second for normal text in an epub? (on 8 year old hardware to be fair. but it's mostly the software's fault)

I've only ever met people complain about PDF or comic page turn speed, never real ebooks of mostly text that are novels.

congratulations on your new experience today!

[Hyphen-ated]

Originally Posted by ottischwenk

I am interested in how fast pages are turned at most with cbr or cbz files - this can take a relatively long time with low computing power.

I'd run a test. Do you know a good public domain test file?

[PeterT]

For cbr / cbz files I expect a big factor will be the resolution of the individual image files compared to the screen resolution.

Sent from my Pixel 4a using Tapatalk