Less exposure/more detail with regular VCR than pro S-VHS VCR?
 

Hello everyone. I have made an interesting discovery. I found that videos that play back on my VCR/DVD combo unit (Panasonic DMR-EZ37V) actually show more detail in high-exposed areas than my Panasonic AG-1980. Less exposure is shown which actually reveal more detail. I was playing this VCR out of curiosity to compare the picture with my AG-1980. Does anyone know about this kind of thing? Even though the picture may have improved I would still prefer to use my S-VHS VCR because that's what many people suggest in video capturing. Is there a way I can fix the exposure on my Pro unit to match that of my combo unit, or even better?

I have included a couple of example pictures to show you what I mean.

Actual video files, unprocessed, I think would be preferable.
That being said I see more red color bleeding from the woman’s dress, down on the right side, in the Svhs jpeg.

There are trim-pots insinde the VCR that adjust the Y level, though I woudln't recommend messing with them unless you know the level is higher than it's supposed to be. You should be able to adjust the level on the capture card side (or tbc). The AG1980P does also have some image settings to adjust sharpness and noise reduction - probably someone that uses these can fill you in on how to set it.

My TBC (AVT-8710) won't correct the levels in a way that more details show up. Adjusting the levels on that only brightens/darkens the picture, nothing else. Same thing with levels in software (VirtualDub). The problem lies within the VCR (those trimpots you were talking about). Is it easy to access them so I wouldn't damage the VCR (AG-1980)?

What capture card are you using with the AG-1980? In VirtualDub you need to adjust brightness and contrast so the video levels fit within the blue area of the histogram. Does that fix the overexposure?

What i notice from these pictures, is the inperfection of the vertical lines, at one picture....
which means the video signal is too strong ?
I guess you should go for the combo, and move on, i've seen worse captures....

It looks like the TBC in the AG1980P is off at least.

You could also use the Intensity Shuttle with the Panasonic DMR-EZ37V because the Panasonic DMR-EZ37V has also component video output, which you also can switch to progressive output, check the manual for this.
This way you can set (the capture) video levels in the video setup driver settings, or in post with Davinci Resolve, works perfect.

(the DMR-EZ37V already gives a better and clean video signal output, it looks very much alike the DMR-ES35V)

btw. you did not tell which capture device you use, it's important :)

This will result in far less detail, not more, as it requires deinterlacing (and not a good method, either).

Why ? i only can find info, that component video output (also de-interlaced, with this EZ37V model available) is better then s-video or composite video, in anyway ...never worse.
de-interlacing will always give less detail, anyway ?
A side effeect of component progressive output is also, that it removes Macrovision (if present on tape) i noticed.

I am using an AIW 9800 Pro.

I have already tried adjusting the brightness and contrast, as well as putting it in range of the histogram, and that doesn't work.

I don't have that unit. Right now I'm just going to try to use the AG-1980P. It's highly suggested, and I already have it. Many people use it, and I don't think many people would if there is an exposure issue. I think there's some sort of problem with it. Maybe a caps issue? I am already planning on sending the unit to TGrant Photo so they can look at the caps, the unit already has other problems.

The AIW 9800 Pro has also component video input

Not impossible, I had issues with blown out video and other things on a Panasonic AG5700, replacing certain caps and adjusting some gain pots improved it (though haven't completely figured out other issues.) May even be more plausible on a 1980P due to the video path involving digitization (which could result in clipping and other issues if things are off with the VCR).

Why do you insist on component (Y+Cr+Cb), from an analog tape format that is Y+CrCb, and can use s-video (Y+CrCb). What is the purpose of it? You gain nothing. In fact, you limit your workflow options. As well as introduce potential crosstalk issues by switching carrier cables.

It makes no sense. :question: :screwy:

If you're grabbing the processed output from a DVD-Recorder it may give a marginal quality improvement over S-Video, especially for PAL as it would avoid doing vertical chroma smoothing done in PAL a second time by the capture card. It may evade macrovision, or may make it worse, as component doesn't support all forms of macrovision. Granted, DVD-recorders tend to screw up the image when seeing copy protection in any case.

The raw video on vhs (and video8,betamax etc) is Y/C and not component so any component output will have had to go through a chroma decoder, and most likely a digitization step.

I think I found out why the combo shot is less exposed than the SVHS one, and if so there is no problem at all with my pro unit. I had captured test shots of the same scene with both units. The thing is, when I advanced each test shot frame by frame to the subject presented in this post, I tried to match to the exact frame on both shots, but actually couldn't match them to a tee. It's a movement shot, so the exposure/details fluctuate as I move. Since both frames are not at the exact location of the shots, the exposure is a tad different between each frame. I just tested this out and found it to be true.

I found out that this is not the reason why the exposure is different for each of those frames. I've been doing more experimenting, and have test captured the same scene multiple times with only the AG-1980. At the same frame as above the exposure is different for each take. I think it's a caps issue after all.

As noted way back in post #2, we need short, unprocessed video samples attached.

As hodgey said, the AG-1980 line TBC appears to be turned off in the post #1 screenshot. Or it's not working properly.

Maybe. Did you give the capture card's AGC enough time to settle? If you just did FF/RW directly to the desired spot, AGC can cause differing levels. You need to seek to a bit before the comparison scene each time.

Unfortunately my combo unit just stopped working. I was able to get a raw avi from my AG-1980P of that scene but wasn't able to get a copy from my combo unit. The tape is now stuck in there and I can't get it out.:(

No I didn't. I did let it settle before the new capture I did today, but comparing that with the images I have from past captures with the AG-1980 the exposure level is still the same. I used the timing graph of VirtualDub to see when the tape had relaxed enough for a capture, capturing only when the lines all merged together. I hope I did it right.

It was turned off, I had discovered. The capture I did today was made when the TBC was turned on.

Here attached is the capture I did today. The painting shot I got the snapshots from is at the very end.

The sample does indicate clipping is happening before the capture card. What was the workflow used for the capture? If it was AG-1980 direct to capture card, then the VCR is clipping the bright parts of this tape internally.

I don't think the timing graph would really relate to the idea I was referring to. But the sample does include enough "pre-roll" before the painting shot, regardless.

The way I have things set up is: AG-1980 > Green AVT-8710 TBC > Capture Device. The camcorder wasn't a very high quality one. That's how it recorded the video. There are quite a few over exposed areas in the tape. I've tried to fix it by keeping within range in the histogram but that doesn't help.

The AVT-8710 may be what's clipping the overexposed areas. Need a VCR direct to capture card video sample.

My AG-1980 was sent to the shop (TGrant) so I only have my regular VCR to plug directly into the capture card. It's not the combo unit, that stopped working. It's a JVC regular unit from 2002. I did a test capture of that one shot you were talking about. It's still overexposed. Like I said, it was recorded that way so I don't think there is anything that can be done. I still have the original Hi8 tape, so I did a test capture from that too, directly from my CCD-TRV138 Hi8 Sony camcorder. It's not the same camcorder the video was shot from. There is no S-Video out on it so I only was able to connect via composite. As expected, still overexposure. Both clips are included right here.

If it helps to why it's overexposed, the video was shot with a Sharp branded camcorder from 2001.

Looking at vcr.avi and hi8.avi, there was too much dynamic range in the scene to properly expose both the bright sunlight building outside the window and the relatively dark indoor windowpane at the same time.

Also some interlace artifacts.....

AVI files, these ones, are supposed to look like that, I think. It's when they're converted to other formats when the artifacts disappear.

Comparison screenshot with waveform attached as ZIP (too wide for forum).

Some highlights are intact on the VHS copy while the Hi8 cam player is pumping up the levels of the original. But most of what you'd want to see in this shot was never recorded in the first place and isn't correctable, as you say.

The Hi8 cam player is also adding oversharpening artifacts like false darks around the bright windows.

That's weird. You would think this would be the other way around. I didn't think you would get more detail out of a copy than the original, and the copy on VHS was recorded in EP mode, which is lower quality than Hi8.

Maybe it's the low quality that is EP mode that tones down the false darks? I don't see how details could emerge on a copy. One thing I haven't yet mentioned to you is that the VHS copy was recently recorded from the Hi8 camera I currently have, recorded on the very day I uploaded those clips. The VHS clip was not recorded a long time ago from the SHARP camera I owned back then. The false darks were recorded onto the tape, but EP mode probably toned them down a bit.

I guess the original has some/most details, high and low, different things you try to do with it will make different things visible from that,
would be a simple conclusion of that, try different things again only from the orginal material.
But outside of that low - high range, you will not get any information, the scene in question is maybe not a good example because there is not a human face in it, to measure the skin tones, a scene with more color in it, would be better to use.
white is offcourse which has all colors.... but in this scene it is the most over exposed area, which maybe the camera tried to compensate also too much, or not in a good way, due to the quality of a consumer camera, compensating this will be very hard, if not impossible. so you can only improve on that in a limited range.

In that case, I don't understand it either. :hmm:

I also have an old VHS tape that this scene was transferred to back in the day from my old Sharp camcorder. Same thing.

Analog video recorders will clip the video signal at some set brightness level before encoding it to tape. The signal has to be clipped before being put through the FM-encoding process as one would otherwise end up with horizontal black streaks on playback at white/black transitions when the signal level goes too high. (E.g See this source)

Maybe that's what you're seeing on the recordings. If the recording is overexposed (whether due to bad manual setup or camera auto-exposure not being good enough) the brighter parts may end up being clipped. Same with digital stuff like miniDV for that matter, though in that case it's pretty easy to know if it's from the recording process since the video is all digital.

Are you saying that a VCR that records from a Hi8 camcorder adjusts the brightness levels so that they wouldn't seem so bright after recording? I do notice the levels are a bit better in the VHS version.

I did a little reading in the link you provided and it looks like a VCR can also adjust the dark levels too, so that they wouldn't look so bad, am I correct? This is something that VCRs do, adjust the bright and dark levels so that they become more balanced?

I guess AGC will have different "limits" on different devices, it's analog, not digital, digital is set to standard values, which most manufactors will value/follow.
Maybe you can use different analog passthroughs to get a better result, but the source material is all what you got, and no magic.
Check if you can do any adjustments in the devices settings/menus.