Standardized response time test for PDF ebook readers

potatis · 08-02-2017, 08:17 AM

Hi
Im in the buisness of buying a big e-ink device but i have a hard time comparing the screen update speeds. I.e. the time it takes to flip pages using different devices. There are video reviews of some devices, but i havent been helped much by these. The videos sometimes have an angle that makes it hard to see when the page flip buttons are pressed, thus making it hard to decide response time. Furthermore, the videos are sometimes cut in a way that makes the loading speeds seem faster.

By making a standardized test we could have a more informed discussion on which device is faster. I am looking for a test with high ecological validity, but focused on page flipping speed.

How could a test be designed to test the speed of e-ink devices in a more standardized way?

My idea would be to:
1: Film diagonally showing the next page button as well as the screen.
2: Have focus on both screen and the button so any artifacts can be identified.
3: During button presses, move the finger in from the side. Making it apparent when the finger presses the button.
4: For devices with multiple display settings (some onyx models have a "fast" setting with lower image quality), show which setting is online.
5: No cuts in the video.

Perform tests A and B.
a) Using each of three openly accessible text files. (5mb PDF, 100mb PDF, 2mb epub). Open the file. Go to page 50. Flip 10 pages as fast as possible. Measure the time from the first page flip to the last.
b) Using each of three openly accessible text files. (5mb PDF, 100mb PDF, 2mb epub).

This test gives information on:
Speed of opening different formats and file sizes.
Speed of "go to page"
Speed of 10 consecutive page flips.
A metric of page flipping speed.

This test controls for:
Using a fast or slow file for reviewing purposes.
Cutting the video to give a more favorable impression to the less meticulous viewer.

jhowell · 08-02-2017, 09:15 AM

I doubt that you will get much traction on this effort. Personally I don't find page flipping speed an important issue with any of the devices/software I currently own. They are all fast enough for me.

You appear to be trying to prevent trickery, but there are still many ways to game the test that your proposal doesn't prevent. For example, capturing the test at a lower than normal frame rate so that everything appears quicker when played back or by pre-rendering the test page images.

Flipping to the next page as fast a possible might tell you something, but it isn't a fair test of actual page flipping in normal usage. Many devices prepare the next page for viewing while the current page is being read and your test would effectively disable that feature.

Cinisajoy · 08-02-2017, 11:35 AM

The best way I know to do this is buy all the devices and test them. Then you will know exactly how long it takes you and the device to flip a page. Now if you want someone else to do the work for you, I don't think you will get many takers.
I think most if not all refresh fast enough to not distract you from reading. Less time than it takes to turn a page in a physical book.

shalym · 08-03-2017, 01:27 PM

Quote:

Originally Posted by Cinisajoy

The best way I know to do this is buy all the devices and test them. Then you will know exactly how long it takes you and the device to flip a page. Now if you want someone else to do the work for you, I don't think you will get many takers.
I think most if not all refresh fast enough to not distract you from reading. Less time than it takes to turn a page in a physical book.

True ebook formats like epub and mobi may refresh quickly, but this isn't necessarily true for a 100 mb pdf file, which is one of the things the OP is looking to test.

Shari

Cinisajoy · 08-03-2017, 03:34 PM

Quote:

Originally Posted by shalym

True ebook formats like epub and mobi may refresh quickly, but this isn't necessarily true for a 100 mb pdf file, which is one of the things the OP is looking to test.

Shari

Ok so then my advice is spot on. Only the OP can determine what is best for his needs.

JSWolf · 08-03-2017, 03:40 PM

If you want something for PDF that's eInk, the Sony 13.3" PDF Reader DPT-RP1 is what you want. Otherwise, an iPad or iPad Pro will be what you want.

HarryT · 08-03-2017, 03:48 PM

Quote:

Originally Posted by JSWolf

If you want something for PDF that's not eInk, the Sony 13.3" PDF Reader DPT-RP1 is what you want.

What do you mean "not eInk", Jon? The RP1 assuredly does have an eInk screen!

Cinisajoy · 08-03-2017, 04:01 PM

Quote:

Originally Posted by HarryT

What do you mean "not eInk", Jon? The RP1 assuredly does have an eInk screen!

I think he put one too many words in that sentence lol.

08-02-2017, 08:17 AM	#1
potatis Junior Member Posts: 2 Karma: 10 Join Date: Jul 2017 Device: Onyx Boox Max Carta	Standardized response time test for PDF ebook readers Hi Im in the buisness of buying a big e-ink device but i have a hard time comparing the screen update speeds. I.e. the time it takes to flip pages using different devices. There are video reviews of some devices, but i havent been helped much by these. The videos sometimes have an angle that makes it hard to see when the page flip buttons are pressed, thus making it hard to decide response time. Furthermore, the videos are sometimes cut in a way that makes the loading speeds seem faster. By making a standardized test we could have a more informed discussion on which device is faster. I am looking for a test with high ecological validity, but focused on page flipping speed. How could a test be designed to test the speed of e-ink devices in a more standardized way? My idea would be to: 1: Film diagonally showing the next page button as well as the screen. 2: Have focus on both screen and the button so any artifacts can be identified. 3: During button presses, move the finger in from the side. Making it apparent when the finger presses the button. 4: For devices with multiple display settings (some onyx models have a "fast" setting with lower image quality), show which setting is online. 5: No cuts in the video. Perform tests A and B. a) Using each of three openly accessible text files. (5mb PDF, 100mb PDF, 2mb epub). Open the file. Go to page 50. Flip 10 pages as fast as possible. Measure the time from the first page flip to the last. b) Using each of three openly accessible text files. (5mb PDF, 100mb PDF, 2mb epub). This test gives information on: Speed of opening different formats and file sizes. Speed of "go to page" Speed of 10 consecutive page flips. A metric of page flipping speed. This test controls for: Using a fast or slow file for reviewing purposes. Cutting the video to give a more favorable impression to the less meticulous viewer.

08-03-2017, 03:40 PM	#6
JSWolf Resident Curmudgeon Posts: 74,704 Karma: 130140792 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	If you want something for PDF that's eInk, the Sony 13.3" PDF Reader DPT-RP1 is what you want. Otherwise, an iPad or iPad Pro will be what you want. Last edited by JSWolf; 08-03-2017 at 11:28 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Response time on the Dasung Monitor?	healinggreen	Alternative Devices	1	02-23-2017 02:22 AM
Response time of the Onyx Boox Max (vs Dasung monitor)	healinggreen	Onyx Boox	0	12-16-2016 02:32 AM

08-02-2017, 09:15 AM	#2
jhowell Grand Sorcerer Posts: 6,574 Karma: 84812983 Join Date: Nov 2011 Location: Tampa Bay, Florida Device: Kindles	I doubt that you will get much traction on this effort. Personally I don't find page flipping speed an important issue with any of the devices/software I currently own. They are all fast enough for me. You appear to be trying to prevent trickery, but there are still many ways to game the test that your proposal doesn't prevent. For example, capturing the test at a lower than normal frame rate so that everything appears quicker when played back or by pre-rendering the test page images. Flipping to the next page as fast a possible might tell you something, but it isn't a fair test of actual page flipping in normal usage. Many devices prepare the next page for viewing while the current page is being read and your test would effectively disable that feature.

08-02-2017, 11:35 AM	#3
Cinisajoy Just a Yellow Smiley. Posts: 19,161 Karma: 83862859 Join Date: Jul 2015 Location: Texas Device: K4, K5, fire, kobo, galaxy	The best way I know to do this is buy all the devices and test them. Then you will know exactly how long it takes you and the device to flip a page. Now if you want someone else to do the work for you, I don't think you will get many takers. I think most if not all refresh fast enough to not distract you from reading. Less time than it takes to turn a page in a physical book.

Advert

Advert