[Anode]

Not sure where to post this but am hoping I might get some direction in this forum.

I have the standard basic Directv receiver- model# D11. Its hooked up to the TV via the VCR. All connections are S-video. TV is a Sony 27" CRT.


I am seeing color bleeding on almost all channels. Its most noticeable on Red foreground against a White background, and in the vertical direction.

If I can describe it - I am seeing the color bleed over into the horizontal lines above the actual boundary of the color. So if there is a red patch against a white background, I would see reddish shades on the area right above the patch.


To test what component (TV, VCR, Sat, Cables) is contributing to this I disconnected the sat receiver and hooked up my DVD player. Everything else remains same i.e. video still goes thru vcr via s-video. I did not see any color bleed. This is why I feel its the receiver.

Or it could be the programming/broadcast quality from DirectTV itself ?!


Any suggestions or ideas of what I could do to fix this ? Its annoying now that it has caught my attention.

 

[Chad B]

It sure sounds like convergence error to me, except that it is not on other sources.

What you describe wouldn't be caused by the receiver, and I see Directv programming all the time at customer's houses and I don't see that.

It could be that the TV is being overdriven (contrast too high), which is exaggerating a convergence issue in the TV. Maybe the Directv receiver is just a bit hotter with it's output than the DVD player; ie the DVD player's 100% white is not quite as powerful as the Directv receiver's 100% white (this could be caused by a number of things including any picture controls in the DVD player), or the movie you played on the DVD player to test is recorded darker than average sat programming. Regardless, I would put up a crosshatch pattern from a test DVD up and double check for convergence errors. I bet you'll find it on the DVD player also.

One last idea would be that maybe putting the video through the VCR is causing or exaggerating it. Try putting the s-video directly into the TV.

 

[Anode]

The reason I rule out convergence is because the color bleed is in the same proportion in any part of the screen. Whether its a channel logo in the corner or big red letters in the middle, the amount of bleed on the top of those objects is pretty much same.

Also, the convergence test cleared fine when I recently ran a calibration (dve & avia) on the tv. Nothing more than a very slight color bleed in the corners that you can see only when up close (2-3 feet) from the tv.

Yes I have tried hooking up the receiver directly to the tv to rule out the vcr but could not find any difference.

 

[Anode]

just an update and a question while I am waiting for the new receiver to show up -


I hooked up my minidv camcorder to the same input, same way (through the vcr), and same cables, after recording some material that has a lot of red foreground with white background (thats when the red color bleed is most obvious), and I couldn't see any problem at all. I could not see the color bleed. So that tells me that it is definitely the receiver.


I was kind of suspecting the broadcast/programming too because sometimes it does seem like its more apparent on some channels and not others. Maybe its a combination of both. Hopefully the new receiver will fix this annoying issue.


In the meantime, irrespective of what the cause of the color bleed is -

Why is that only some and not all colors bleed ? The most obvious is red foreground against a white background. Colors like yellow and blue are the next ones that show the color bleed, but why is it not so obvious on many other colors - like white foreground against black background or vice versa ?



And why is the bleeding in the vertical upwards direction ? The color bleed is mostly seen on the tops of objects. And on the sides. But why not under those same objects as well ?



Just trying to understand this..er..colorful phenomenon...

 

[BertM]

Do a search for 'chroma upsampling error' on the net.


A wiki page here: http://en.wikipedia.org/wiki/Chroma_subsampling


Your satellite receiver probably handles MPEG the wrong way. And as colour is coded as a separate signal in the mpeg stream, black and white (or 'luma') isn't affected.

 

[Anode]

I know what is chroma subsampling and upsampling, and also about CUE.


But isnt CUE about upsampling from 4:2:0 to 4:2:2 (or 4:4:4), and wouldn't standard(digital) TV broadcast be 4:2:2 always ? Would there be a need to 'upsample' a tv broadcast signal ?

 

[ChrisWiggles]

That does not sound like CUE to me, which occurs when deinterlacing and outputting progressive, and is a sideways (mainly) artifact mostly visible on diagonal edges or on heavy motion. It really just sounds like a bad sat box, have them replace it with a new one. I haven't seen this particular problem on sat boxes, but I've seen other output problems, they should just replace it with a working sat box.

 

[ChrisWiggles]



Quote:
Originally Posted by BertM /forum/post/15900491


Do a search for 'chroma upsampling error' on the net.


A wiki page here: http://en.wikipedia.org/wiki/Chroma_subsampling


Your satellite receiver probably handles MPEG the wrong way. And as colour is coded as a separate signal in the mpeg stream, black and white (or 'luma') isn't affected.

It actually is very unlikely, since satellite is a video source, so the chroma is actually per-field, unlike DVD where it is per-frame but often wrongly handled as per-field which is the CUE error when deinterlacing.

 

[BertM]

Makes sense it isn't CUE... however I've seen very bad colour shifting in ATI hardware decoders, probably not CUE but simply an error in decoding (the HD 2400 and 2600 cards suffered from the colour displacement in the first few months of their existence)


Thinking a bit further, it might be bad insertion of the colour signal on the s-video signal where the colour is a bit delayed? Which means the error shouldn't occur when using the RGB or component output of the decoder.

 

[Anode]

sad thing is component output is available only on "HD" receivers and not on their standard ones.

 

[ChrisWiggles]

Just have them switch the box. I"ll bet it clears it right up.

 

[Anode]

I hope so too but I am skeptical since they are sending the same model(D11) and not their latest one(D12). Well it will be here Friday so will you know how it goes.

Here's a couple of shots I took of the screen to demonstrate what I am talking about. Not very clear but still gives you an idea what I am talking about. Look for the red around the "4" in the channel logo.

 

[Anode]

swapped receivers. but no significant change.


could it be even possible that the output level(voltage) of the s-video output is higher than the 'standard' level ? Is it worth trying to put an attenuetor (or just a resistor in series!) to see if that fixes it or not ?

 

[ChrisWiggles]

that looks like poor beam focus to me. Are you sure it isn't present on other sources? It would be if it is.