HUH! Video Patterns GONE! (Checkerboard/Zigzag)

[Phire]

I made a great discovery today, i was recently plaugued by the checkboard/zigzag pattern over my 704x480 captures. Today I just recieved my brand new VCR the JVC HR-S9600U one of the most expensive consumer non-professional VCRs available if not the most. It had an S-Video Output so I decided to use it, miraculously when recording off of it there are NO patterns, completly clear picture. It also played back tapes perfectly so I wasnt getting that flickering or the bending at the top of my video. So what does this mean? Does it mean the Breakout Box is defective and does not recieve composite video properly? Or is it the card that is interpreting the video signal wrong? All I know is S-Video works. Matrox should really do more of an investigation into this though.

Heres a screenshot:

[Walrus]

Phire: Very interesting results. It could have to do with the way composite mixes the chroma channels together into the same signal. On the other hand, it certainly should rid me of a desperate need to get a new capture board. Maybe I should try to get me a composite->s-video connector to see if the problem is indeed with the composite connector or not.
Thanks for your investigative work!

[Michel Carleer]

Phire,
Did you try the composite out of your new recorder? What are the results?
Michka

[mgu2]

I had a similar problem with recording from the internal TV tuner. When I bought the best coax cable I could find, the patterns were gone. It looks like even a very small interference makes the work very hard for the JPEG compressor as it has to deal with abnormally "high frequency" image. The result is the ugly pattern. It seems your VCR has good S-video output or, more likely, you have a good S-video cable so it is immune to external interference. Is your S-video cable shorter than the composite one?

[Phire]

Walrus, yeah I suggest you get some device that has an S-Video output and see if its gone for you also. Micheal, yeah the pattern returns with composite cables, I also borrowed a friends thick insulated gold tipped cable, makes no difference.

"It seems your VCR has good S-video output or, more likely, you
have a good S-video cable so it is immune to external interference. Is your S-video cable shorter than the composite one?"

hehe well, the VCR did cost 500$ and I used the S-video cable supplied with it, it's extremely short, I almost couldnt make it reach to my BOB. its about twice as short as the composite cable.

[Michel Carleer]

Remember, in a composite signal the chroma info modulates a 3.59 (NTSC) or 4.43 (PAL,SECAM) MHz subcarrier which is mixed to the luma AM signal. S-video transmits luma and chroma separately, the signals don't get mixed. Once again, this points to what I said before: part of the chroma subcarrier is decoded as a luma signal. The reason for this however I don't know. Is it because the Marvel (or RR-G) does a poor job of separating the two? Or is it just simply normal? In fact, the luma and the chroma signals can never be 100% separated because they are so tightly intermixed.
Michka

[Prospero]

I tried some LD caps last night with the composite conector and didn't have any trouble. I do however, have some pretty expensive video cables. My LD player is the LDW1, so it doesn't have a svideo out.

[fluggo99]

Thought I might add something; I was watching cable TV just now (Gundam Wing :P), and noticed the same checkerboard pattern-- on my TV. Could it perhaps be in the signal? Could your new $$$ VCR be filtering that kinda stuff out? If so, the problem would be present/absent on both composite and S-video, right?

[Dr Mordrid]

Wellll.....a lot of cable is now MPEG-2 encoded so who knows what kind of artifacting could be introduced in the decoding process by the box?

I guess we'll have to update our weird-s**t lists ;-))

Dr. Mordrid

[fluggo99]

Well, I have analog cable TV, no box, but of course that doesn't mean MPEG-2 wasn't used in the distribution phase.

[Phire]

Although the patterns have dissapeared a new problem has arisen, So it's out of one problem into a new one every time almost. My new problem is that I can almost never make a video without a corrupt frame getting in now, short captures wont get it but long captures have a 90% chance of getting in a corrupt frame, it's extremely annoying and turns my video black and white, is there anything I can do to fix this or does anyone know what exactly causes the corrupt frame?

[AVATAR-X]

Out of curiosity, do you get a corrupt frame no matter what the input source? I ask because at one point I figured out that the "Video Stabilizer" feature in my VCR seemed to cause intermitant corrupt/snowy frames while capturing with my Marvel G200.

I'm not sure if this is your problem, but maybe this will help someone out.

[egov]

Ok, after weeks of misery we found the cause of the zigzag patterns -- power supply! Matrox G200 and G400 is very sensitive to the noise from the power supply. I tried my G200 TV on few different supplies with no improvement. Here's my original message:
http://forum.matrox.com/mgaforum/For...ML/002509.html
If you find a sollution, please, tell me.

[Jkun]

If you're sure it's the PS that's the cause, get an Enermax powersupply.

If these don't fix the problem, it's either not the PS or the G200 card is broken. Enermax is the Rolls Royce of powersupplies.

C'T regularly tests powersupplies, the Enermax gave the cleanest power of 'em all.

J-kun