Vero 4K H265 movies (8-bit encoded only?) - blacks show as grey, colours washed out (reduced colourspace)

Mine just says ‘ok’ for HDMI4 (ARC)

Black clipping displays as prompted, but I also see a slight flicker of bar16.
White clipping displays like it should I guess, only bar253 isn’t flashing, else it is really hard to see.

Hmm, interesting. So yours apparently works as it should.
May I ask your settings on your Sony TV? Which HDMI input is the Vero plugged in?

Also, can you pls. let me know what do you have:

  • Under Settings / Display / Picture / Picture mode (e.g. Cinema Pro, Cinema Home, etc.)
  • Further down going into Advanced settings, under Brightness, what is your Black level setting?

You mentioned an AVR, I guess Vero HDMI is going into the AVR first, and the AVR going into the TV, correct?

Thanks in advance!

Run them, returned parsing and ok. Shall I also re-check the EDID data after this?

Does it stay on parsing; or does it eventually switch to OK?

It switches to ok right after parsing (no visible delay).

Ok – that’s what I’d expect.

White clipping is irrelevant on modern day TVs. Peak light output for 235 or 255 is the appropriate measure to be used. As I said before, full range YCbCr with BT.709 colorimetry is more than likely to be ignored by most displays. If you have a high sensitivity tristimulus colorimeter or a spectroradiometer, you can easily check this out. Unfortunately, I won’t be able to do any testing. I sold my house and everything is in storage. It will take 2 months or so before I can move into my new house.

RGB Full with appropriate Q0,Q1 quantization bits is what would work for OP’s case.

What makes you think that displays are expanding 16-235 to 0-255?

They have to to make use of the full range of their pixel brightnesses.

What full range of pixel brightness? You may have to explain your rationale. I have never heard this in the industry before. If this was indeed true, then display calibration for limited range is pretty meaningless.

Maybe I’m not using the right words - I’m not in the industry. All I’m saying is if you have something producing a red, green or blue spot on a screen, you want it to be off to convey black and as bright as it can be to convey the brightest white, however many nits that is. So if you send a signal at level 16/255 that maps to off and one at 235/255 should map to ‘fully on’. Otherwise you are wasting the capabilities of the display.

I assume that’s why @retroresolution’s clips which only have values between 16 and 235 display correctly if we send them with ‘limited’ in the InfoFrame.

Hi @sam_nazarko,

Not wishing to derail the investigation of @grahamh’s issue, but can you tell me if the fix for my original colourspace issue (confirmed as working in the test build) is now in staging?

I ask as A) the Vero is reporting updates as available, but I don’t wish to install them if the fix will be erased, and
B) I’m slightly concerned that a fix for @grahamh’s issue may conflict with my original issue from May.

Many thanks,
RR

I don’t think it is in staging yet. I don’t understand why updates are available - suggest you hold fire.

If the tweak I just made is not going to help @norbertlang I will take it out again. Otherwise you would have to set limited range manually on vero or your TV for those vids to render properly.

Ah, sorry, got the names confused Graham. Of course the issue under investigation is @norbertlang’s, not yours.

Neither my tv nor monitor have any settings controlling full or limited range. All my equipment is old and cheap…

Further to last weeks note regarding issues experienced with the test kernel, specifically YouTube crashing the vero and requiring a full power cycle, I’ve now experienced 3 or 4 instances of a YouTube video freezing mid way through playback.

On the last instance, a circular icon containing ‘0’ was superimposed on screen. In all instances, the vero is completely unresponsive, however after 2 or 3 minutes the system finally closed YouTube, and returned to the main menu.

I’m certain no background restart occurred, as the Favourites menu (where i have linksto the youtube app and various channels) was selected and open.

Again, not sure if this is a kernel issue, or an add-on issue (youtube is at the current newest version).

Should I leave fill debug log enabled at all times (currently all logging is disabled)?

I will respond when I get back home, tonight or else tomorrow night, but I just used settings from a topic in AVForums, will get back to you!

Can you point me to that report, please? Either I missed it or forgot it.

Hi,

Sure, no problem:

Thanks

Sorry, I hadn’t made the connection between YouTube and that post.