HDTV 4K

[David White]

I am making AVs to play on my Samsung QLED 14K HDTV and up to now have set my basic picture definition to 1920x1080 pixels. I understand that  these TVs are capable of displaying 3840x2160 pixels. Furthermore the TV technology enables much higher dynamic range, colour saturation and other advantages.

Would displaying my images at this higher resolution gain any advantage in picture quality?

[tom95521]

Yes. 4K resolution videos containing 4K media should be sharper on a 4K HDTV. As far as I know PTE is limited to 8 bit color range so HDR may not be noticeable. If you play media from a USB stick color dynamic range might be higher from image than video.

https://www.digitaltrends.com/home-theater/what-is-hdr-tv/

Tom

[David White]

Thanks Tom,

I'll see if I can set up an A/B comparison test using both resolutions and see if my aging eyes can tell the difference!

[jmG-06100]

To me the difference is most noticeable when getting really close to the TV screen, and, then, yes, it makes a difference.
Try looking at a suspension bridge and inspect the individual cables.
BTW, I was astonished to read that you had a "14K TV", since my high-end TV is only 4K and I know about 8K TV, nothing (yet) beyond that.

 

[David White]

Quote

I'm astonished too! Don't know when the extra 1 crept in - our TV is, of course 4K.

When we first got it the TV it came with a series of demonstration images one of which was a test chart showing a chequer box pattern of squares, each containing alternate vertical and horizontal lines at maximum resolution. Neither my wife nor I could differentiate them from further than about 50cm from the 52inch screen. I've just repeated the experiment and watched it again which confirmed the previous result. It also contained two identical video sequences one in 4K definition and the other in 1080x1920 standard definition. I could not tell the difference however close I stood!

I've also run a short A/V alternating the same image in both formats. Again neither my wife nor I could tell which was which except by looking at the captions I inserted. They both looked pin sharp along with apparently no difference in colour rendition. My Granddaughter will be here over Christmas, if she can't tell the difference it will confirm my belief that the technology has gone beyond the ability of humans to benefit from it.

I think I shall stick to importing my images at 1080x1920; if nothing else it will save a lot of memory!

 

[jmG-06100]

For a limited time, you can download three photos of Bridges with fine details in the suspension cables (and one road sign), so you can compare different definitions on your TV screen, and also for "pixel peeping" on your computer.  Download from:
https://www.dropbox.com/sh/5f2h7tdk73vrbeu/AABAdUz9MXGjtBZFdfsvGyYza?dl=0

These JPG files are at full camera definition (20 Mpix), 4K (8 MPix), FHD (2 Mpix) and 800 pixels wide (0.5 MPix).

You can create a 4K slideshow incorporating these three samples at four definitions to check the visible difference, as well as check each image directly on your TV (I use a USB memory stick for that).
Let us know your findings and what younger eyes (your granddaughter) can see.

Of course, the ultra high definition (UHD = 4K) will be noticeable only if the picture is visible for a minimum length of time, not for a typical 5 seconds during a slide show.

[jmG-06100]

Here is a sample image at 4K (slightly reduced to fit within 8 MB), with focus at about 3/4 of the distance to the feet of my grandson.

[David White]

Quote

I have tried to download these images but I can only view them at thumbnail definition - the one of your grandson for example is only 563 x 750.

How can I download the full definition versions please?

I clicked on the "Download" button but nothing happened.

 

[David White]

Hi jmG,

I think I've cracked it! I right clicked each image and it gave me the opportunity to Save As and the definition is as stated so I'll now use these great pictures to see how they look on the TV.

For the record I'll take then directly into PTE and publish them As HD & 4k videos and view via a memory stick on the TV.

I will also try them as straight jpg images to asses if there is any difference in quality.

Thanks very much for the comprehensive material.

[jmG-06100]

David, while you kept searching, I started this reply: <<if you double-click on the image included above you can Copy it and then if you paste it into Photoshop (or save it to disk) ;  you should get an image that is 3840 pixels tall (and 2880 pixels wide (4/3 ratio)).  That's a start.  For the collection of different sizes on my DropBox account, try and look at your usual (default) Download area in MS/Windows after your "download all" >> (next time).

I do not have access to my files the same way as you, so cannot tell you exactly where that "button" is.

Anyway, you have the files now.

You can view them directly, one by one, on your TV.  You can also make a single MP4 animation (in 4K) with PTE to see how the different definitions display (and insert the file name as a header to each view, for ease of comparison).  If you make a FHD slideshow, you should see no difference ikn the display of the FHD and the 4K files.  The 800 k files will probably look "acceptable" if you do not scrutinize too much.

Let us know.

[jmG-06100]

In "Project Options", "Default" tab, you can insert a "Text comment for new slides", then "Insert Template" and choose "Picture name".

And "Apply to all slides", which should provide:  <%MainImg.FileName%>.

Hoping that helps.

[David White]

Hello Again,

Hope you had an enjoyable Christmas.

I've now completed a series of tests using your Suspension Bridges files.

In the end I made three sequences - one for each bridge in which I compared the Highest definition pictures with lower definition.

I did not bother with the 800x600 as they were obviously inferior as you would expect.

The comparisons  of the others were as follows:

Rope bridge 1920x1440 against 2880x3840 

Pont Luiz 1920x1440 against 5184x3888

Cantilever Iron Bridge 1920x1440 against 5184x3888

In PTE I set all slide lengths to 2s with zero transition and repeated the pair 7 times.

I published them as MP4 and transferred it to a memory stick which I played on our TV.

This gave 3 videos that made 2 second A/B comparisons (similar to flicker tests that astronomers use to detect star movements).

The results were that even close to the screen no one who watched them could see any differences in the clarity or colour of the pictures.

The only exception was of Pont Luiz where right t the top of the image at very close viewing you could detect a slight change in the shadow position of the fencing and the definition of the people standing there.

I then went back to your original jpeg file of this bridge and the definition on my PC was pin-sharp I could clearly see the fencing rails and the people standing there. So I viewed it on the TV where it was just as sharp!

Finally I compared the two MP4 images, and much to my surprise, the lower definition (1920x1440) image was the most accurate.

I assume this is because 1920px is the same resolution as the line-scanning of the TV.

Obviously I shall now stick to 1920x1080 images unless I intend to zoom out significantly for particular effects.

Thanks once again  and have a prosperous New Year

David

 

[jmG-06100]

Thank you David for your detailed report.

Just a simple question to be sure: was the published MP4 file created by PTE at the definition of "3840x2160" and "high quality"?  I presume so.
My TV screen is slightly bigger than yours, which might explain why I see a difference (Sony KD-55" A1 OLED)

Just for the record, the three bridges beeing scrutinized are the following:

2017-Q606-5121 Ponte de Dom Luís I, Douro River, Porto, Portugal
41°8'25.233" N 8°36'34.446" W
https://en.wikipedia.org/wiki/Dom_Luís_I_Bridge

2017-R907-3658 Bridge of the Gods, Columbia River Gorge, Stevenson, State of Washington looking towards the State of Oregon,  USA      
45°39'46.819" N 121°54'18.87" W
https://en.wikipedia.org/wiki/Bridge_of_the_Gods_(modern_structure)

2019-X803-6889 Rope Bridge, Plopsaland amusement park, De Panne, Belgium    
51°4'52.121" N 2°35'52.271" E
https://www.plopsalanddepanne.be/en/attractions/suspension-bridge

 

[David White]

Ahh - that's solved it!

I assumed that the because the quality box contained the 1920x1080 figures thiey just reflected the aspect ratio and had no other effect.

I did as you suggested and lo and behold the picture quality immediately improved and I could also see the difference between the lower and higher definition of the two images of the Dom Luis Bridge.

There is a noticeable improvement in the video I am renovating as well.

I did try an even higher quality to match the definition of your highest quality samples - 5184x3888 but my video card is apparently not good enough to render the images properly so I abandoned that as one step too far!

[jmG-06100]

Hi David.  I am glad I asked the question ; the reason for no improvement in  your tests becomes obvious, once you know why...

Should you want to zoom into the image, then more pixels are beneficial but your TV cannot show more than its 4K definition.  Enjoy the realm of Ultra High Definition (UHD, aka 4K).