Hey everyone, as a recent owner of the jvc, one of the main problems people told me about was the split screen effect in low-light conditions. I was told that with the introduction of the U(A) model, that problem was one of a few that was going to be addressed and fixed. However, I was just playing around with it in my house and my heart sank when I noticed that same split screen effect that I thought was taken care of was very visibly present on the viewfinder and tv. What do you think should be done about this or can there be something done. Thanks

No thier can't be anything done. I have it on mine to, but it only shows up at 18db so I have my pre set at 15db at its highest point haven't had a problem since. Its a native problem to the progressive ccds so you just have to work around it and what would you use 18db for anyway, I never even get close to that high of a gain setting in anything I record.

I would argue that if you are even close to 18db gain you are using the wrong camera.

I just got my hands on an "A" version and that was the first thing I tested. I tried my best to induce SSE at 0dB, but was impressed that I couldn't. I managed to do it at +12, but once again was very difficult and the extent of the SSE was night and day compared to the way it used to be.
My understanding is that the "fix" to this problem is proper calibration, which is dependent on a technician spending a couple hours on it.

What dB level were you using when you saw it.

Agreed, (maybe stay away from 18db) although I've never observed an SSE with my camera, I've heard that it only shows up with the right shooting/visual conditions present. It's not a problem that is there all of the time (like a plasma screen burn in).

The manual addresses the issue and I never heard anyone say from JVC that the SSE would be solved with the "A" version. The A upgrade is a firmware upgrade that adds battery monitoring (Anton Bauer) and NLE digitizing adjustments. See JVC website www.jvc.com .

However, if you're seeing the SSE below 18db you might call JVC. They've been very helpful and responsive to a lot of camera owners.

Don't let your heart sink.....

Whether you have an "A" model or pre "A" model the rules still apply. You'll have to white balance your camera and turn it on for a couple of minutes. Try the routine and then please report back the findings.

I have tried and tried to get SSE on my "A" model and it just is not there and frankly I'd forgotten about it. You can always get the camera calibrated again if need be.

thanks everyone for their feedback, I have found the SSE at 0db but sometimes it shows up and sometimes it doesnt. And it's all in the same lighting conditions. I will definitely try to white balance and leave the camera running for a few minutes and give a reply back. I bought 2 of these cameras so I will have to test the other one along with this one to see if there is any difference.

thanks everyone for their feedback, I have found the SSE at 0db but sometimes it shows up and sometimes it doesnt. And it's all in the same lighting conditions. I will definitely try to white balance and leave the camera running for a few minutes and give a reply back. I bought 2 of these cameras so I will have to test the other one along with this one to see if there is any difference.
Alan,

At 0db there is a problem. I would get it calibrated.

Definitely!

Ok, I will look into getting it calibrated. Is this process something i have to pay for or will jvc just do it for free if i give them a call. If I have to pay for it that pretty much sucks because I bought these new cameras expecting them to work right especially for their price.

Alan, I had the same problem as you. Just call the guys at the Cypress location and tell them about your split screen. It needs to be calibrated. The A upgrade has nothing to do with that and they don't even look into that issue when updating a camera. Mine is great now but it took about 9 days.
Mike

The SSE seems to be rather like the old tube video cameras, you needed to do a line up to get them into registration.

With the HD100 it seems best to test the camera for SSE when you purchase the camera and if it's not up to standard, get it back to JVC to re-calibration.

In the longer term, it'll be interesting to see if the calibration drifts with time. Chances are it will, and cameras will need to be sent in to be re-calibrated every so often. The important thing will be that JVC can manage a turnaround time appropriate for professional equipment just being calibrated.

thanks everyone again for their information. I guess I will call up jvc it's just kind of annoying because I sent back my first camera because there was something wrong with it and they sent me a new one and now I'm going to have to send this one back to get calibrated. Lol i wish there was an easier way to do things. You guys have been very helpful, it's nice to know there is a community out there for questions such as these.

Ok, I will look into getting it calibrated. Is this process something i have to pay for or will jvc just do it for free if i give them a call. If I have to pay for it that pretty much sucks because I bought these new cameras expecting them to work right especially for their price.

If it's less than 30 days old try using the JVC Perfect Experience guarantee. They'll ship you a new one and yours send your back in their packaging (I think) at their cost.

http://pro.jvc.com/prof/support/pepolicy.jsp

I had a very early camera and when I sent mine in for an update I also asked them to check calibration because I was seeing SSE fairly easily. To my surprise they replaced the motherboard without explanation. It's MUCH more difficult to stimulate the effect now.

I've had my A revision camera for about three weeks now and have not been able to produce the split-screen effect, even when intentionally shooting horribly lit rooms with 18db gain with no proper white balance and a cold camera. At this point I'm not worried about it at all.

I did, however, have one dead pixel in the upper-left corner which only showed up when gain was applied, and was only noticable on a high definition display. The "Pixel Mask" feature in the service menu solved this problem and I'm now a very happy camper.

We had our camera upgraded to "A" recently.

I will soon post a screenshot of SSE Horror that reared itself on a shoot outside last week. I don't believe the gain was on at all as it was full sunny/partly cloudy that day.

Does anyone know if the HD100 will show metadata on the LCD when tapes are played back? specifically the gain in db and the f/stop and shutter?

My GL1 was able to do that.

I'd like to know exactly what the shooter had it set for when it happened.

Adam Wilt mentioned the SSE could show itself if a lot of Green was in the shot. There was a rather large grassy field in 90% of this one that had a really bad SSE goin.

We had our camera upgraded to "A" recently.

I will soon post a screenshot of SSE Horror that reared itself on a shoot outside last week. I don't believe the gain was on at all as it was full sunny/partly cloudy that day.

Does anyone know if the HD100 will show metadata on the LCD when tapes are played back? specifically the gain in db and the f/stop and shutter?

My GL1 was able to do that.

I'd like to know exactly what the shooter had it set for when it happened.

Adam Wilt mentioned the SSE could show itself if a lot of Green was in the shot. There was a rather large grassy field in 90% of this one that had a really bad SSE goin.


Mark, it can happen at 0db and in bright conditions. For the record and for those who insist that's it's an 18db low light thing, it isn't. It can happen in bright sunshine.
However, as I've said before, the fact that it can happen and the likelihood that it will are tow different things.
JVC is very specific about how to prevent it and how to deal with it if it occurs: make sure that the camera is warmed up for 5 minutes before shooting, re-white balance, change you exposure.
Also the fact that Adam suggests that SSE could occur against a solid green background is definteily not to say that it's likely at all. It's just that with the right (wrong) combination of elements it might appear.
I'd be willing to bet that if you took your camera back to that location under the same circumstances that you couldn't reproduce the problem.
If it did occur again, then definitely send it into JVC for calibration.

I asked the shooter how long the camera had been on and he said at least an hour.

Well, I guess I'm lucky. I've owned my camera for over three months and have yet to see SSE.

"Its a native problem to the progressive ccds"

Not true at all. It is only JVCs implementation in this camera that cause this problem. There are many progressive cameras around and none of them except the JVC have this problem.

All,

This technology has been discussed at length on this forum, so I won't go into great detail, but here's a short explanation:

The GY-HD100U is a revolutional camera in that it offers full 1280 x 720 native resolution, with TRUE progressive scan. This is more pixels being scanned in one frame than ANY other HD camera in this price class, and even more than the Panasonic Varicam. This is done with CCD's that have two processors on board to divide the work load, and it's important that the two sides of the CCU be calibrated for the same output levels. Precise camibration on a camera that is mass produced can sometimes not come out as good as possible, so sometimes the calibration needs to be touched up.

As some have already said, SSE calibration is a free, warranty service. Just send it into JVC Pro Service. I recommend the Cypress service center for SSE calibration, although the other pro service centers can also do it.

Here is a link to our service center contacts:

http://pro.jvc.com/prof/support/index.jsp


And, keep in mind our "Perfect Experience" 30 day exchange policy. This can be implemented through you reseller or directly with JVC. Here is a link on this policy:

http://pro.jvc.com/prof/support/pepolicy.jsp

Regards,

Carl

You forgot that funny little phrase the FAQ, "At the required 74Mhz clock rate, the imager's power dissipation would be excessive and the chips would self-destruct."

Therefore the problem is related directly to the high pixel read out count not if the CCD is progresive or interlaced. Lower resolution progressive CCDs would not have this problem, neither would larger CCDs with the same pixel count. I dare say there are some ways to dissipate that heat without the need for dual processors. Perhaps a bigger body design with some decent heat piping.

This issue could have also been avoided by designing the camera as a minimum 1/2" block from the get go. So many advantages to being 1/2" its not funny:

Higher light sensivivity
Less noise (greater signal to noise ratio)
Larger surface area to disipate heat
Shared lens format to XDCAM HD (maybe only a different mount)
Existing 1/2" lens options
etc etc etc

The price difference in producing a 1/2" CCD over a 1/3" CCD would be negligable in terms of the system cost. Heck, a full Nikon DSLR has a 6MP 1.2" sensor and sells for US$650 retail even with a half decent lens. Large sensors clearly arn't expensive anymore.

You forgot that funny little phrase the FAQ, "At the required 74Mhz clock rate, the imager's power dissipation would be excessive and the chips would self-destruct."

Therefore the problem is related directly to the high pixel read out count not if the CCD is progresive or interlaced. Lower resolution progressive CCDs would not have this problem, neither would larger CCDs with the same pixel count. I dare say there are some ways to dissipate that heat without the need for dual processors. Perhaps a bigger body design with some decent heat piping.

This issue could have also been avoided by designing the camera as a minimum 1/2" block from the get go. So many advantages to being 1/2" its not funny:

Higher light sensivivity
Less noise (greater signal to noise ratio)
Larger surface area to disipate heat
Shared lens format to XDCAM HD (maybe only a different mount)
Existing 1/2" lens options
etc etc etc

The price difference in producing a 1/2" CCD over a 1/3" CCD would be negligable in terms of the system cost. Heck, a full Nikon DSLR has a 6MP 1.2" sensor and sells for US$650 retail even with a half decent lens. Large sensors clearly arn't expensive anymore.

Guy, thanks very much for your valuable input.

We wanted to give our customers as high a native pixel count as possible -a real HD native pixel count of 1280 x 720 - not a derived HD pixel count using some sort of scaling or spatial offset. Native 1280 x 720 progressive is a good thing, isn't it?

We also wanted to give our customers a light weight, compact shoulder mount design.

These two reasons dictated that the CCD size had to be 1/3"

To go with a 1/2" or 2/3" CCD would have meant much higher cost and a larger body.

From what many of our customers are telling us, and from the reaction we've had at NAB so far this year, our decision to stay with the small shoulder mount design, and to stay with the native 1280 x 720 progressive design seems to be a very popular decision.

Regards, Carl

Guy, thanks very much for your valuable input.

We wanted to give our customers as high a native pixel count as possible -a real HD native pixel count of 1280 x 720 - not a derived HD pixel count using some sort of scaling or spatial offset. Native 1280 x 720 progressive is a good thing, isn't it?
Absolutely, indeed, without a doubt!

We also wanted to give our customers a light weight, compact shoulder mount design.

These two reasons dictated that the CCD size had to be 1/3"

To go with a 1/2" or 2/3" CCD would have meant much higher cost and a larger body.

From what many of our customers are telling us, and from the reaction we've had at NAB so far this year, our decision to stay with the small shoulder mount design, and to stay with the native 1280 x 720 progressive design seems to be a very popular decision.

Regards, Carl
It takes all types, and I personally would much prefer a slightly larger and heavier camera to get 1/2 or 2/3" sensors. I am more than happy with the size of my DV500.

Specifically though:
To go with a 1/2" or 2/3" CCD would have meant much higher cost and a larger body.
I don't buy that. 5 years ago perhaps with the cost, but not anymore,now its just marketing to hold the price and margins of 2/3" cameras where they are. You do realise how small a 1/2 sensor is don't you? No bigger than the size of your thumbnail.

Guy you should do what the "RED" camera people are doing.

Make the camera you want and at the low cost your talking about.

I'll buy it!

Carl, if my camera is showing SSE with plenty of light and good exposure is it in need of Calibration?

It would be nice to know in concrete terms what would be considered needing calibration.

I'm damn lucky that the problem happened on a shot that wasn't important or I'd be in real trouble right now. I will be posting various screen shots from this shoot (good and bad) in a thread of my own shortly. I would like the thread to address the shortcomings and how to minimize them. I do love this camera and I'm willing to work with it.

[QUOTE=Carl Hicks]

And, keep in mind our "Perfect Experience" 30 day exchange policy. This can be implemented through you reseller or directly with JVC. Here is a link on this policy:

/QUOTE]


The Perfect Experience concept is another credit to JVC. They took care of us every inch of the way, and unfortunately that can be a rarity in 21st century business.

Great camera and great support, it definitely moved us from Canon to JVC.

Carl, if my camera is showing SSE with plenty of light and good exposure is it in need of Calibration?

It would be nice to know in concrete terms what would be considered needing calibration.

I'm damn lucky that the problem happened on a shot that wasn't important or I'd be in real trouble right now. I will be posting various screen shots from this shoot (good and bad) in a thread of my own shortly. I would like the thread to address the shortcomings and how to minimize them. I do love this camera and I'm willing to work with it.

Mark, if after warming up the camera for several minutes and doing a manual white balance, you are still seeing SSE at zero or low gain settings, then you probably need calibration. Send it into the Cypress, CA service center.

If you've had it for less than 30 days, you could also opt for an exchange.

Regards, Carl

This is more pixels being scanned in one frame than ANY other HD camera in this price class, and even more than the Panasonic Varicam.
That's not accurate; the VariCam CCDs are 1280x720 as well.

Mark, send the camera to Cypress as Carl suggests but also try to tell them the frame rate and the gain settings where yout hink this was encountered.

I would explain this aspect of the camera architecture differently than Carl but it takes me at least 5 minutes of talking and I have to be able to see your eyes so I can tell you understand.

This can't be typed because of length and because typing leaves the thread open to replies like are in this thread from people who don't understand.

For your information the information in this thread is insufficiently accurate.

I'm tired.

I don't buy that. 5 years ago perhaps with the cost, but not anymore,now its just marketing to hold the price and margins of 2/3" cameras where they are. You do realise how small a 1/2 sensor is don't you? No bigger than the size of your thumbnail.

I'd still have to imagine that one of the big manufacturers would have decided to bite the bullet by now and release a 1/2" camera for less than 10K if cost wasn't a major factor. Marketing only keeps things propped up for so long in an industry as competitive as this one.

The reason for this is because these companies do not want to kill their middle market. If they are selling 1/2" or 2/3" chips in 5K cameras why would you buy their 25k-50K models?

Just coming back from NAB it's very clear what is going on. JVC really understands where things are headed and they will have a head start. The most dissapointing showing was from Sony.

That's not accurate; the VariCam CCDs are 1280x720 as well.
In the final output yes, but not the CCD which is 960x720.

In the final output yes, but not the CCD which is 960x720.

This is my understanding.

This is why the JVC HD100 is unique amongst its competition.

The reason for this is because these companies do not want to kill their middle market. If they are selling 1/2" or 2/3" chips in 5K cameras why would you buy their 25k-50K models?
I agree completely. You know Sony have even said their HDV market is 1/3" only, XDCAM HD is 1/2" and HDCAM is 2/3", which makes no sense except for marketing. Screwing the customers is no way to earn respect.

On Varicam, I thought the CCDs were at least 1280x720 but recorded as DVCProHD 720p 960x720 4:2:2.

In the final output yes, but not the CCD which is 960x720.
Not quite. The CCDs are 1280x720. The HD-SDI output is 1280x720.

The recording format is 960x720.

Barry, why in the devil are you comparing a $65,000 camera that has 2/3 in chips to a 5K camera? Carl Hicks said the HD-100 has more pixels than any cam in it's price range and he is correct.

but then he went on to specifically add that it had more than even the varicam, using it by name. Hence the discussion is more than valid in my opinion.

If that's the case then what Carl said is still true. There is more recorded rez than the Varicam. Anyway, who gives a rats? If somebody was giving me a Varicam I'd take it...

"This is more pixels being scanned in one frame than ANY other HD camera in this price class, and even more than the Panasonic Varicam."

ie: The comparison was about how many pixels were being scanned by the CCD block, not how many pixels are recorded to tape.

"If somebody was giving me a Varicam I'd take it..."
Which just shows that pixel count on tape isn't everything.

That's not accurate; the VariCam CCDs are 1280x720 as well.

Barry,

I stand corrected. The Varicam is 1280 x 720 at the CCD. What I was thinking of is the fact the even though the Varicam is 1280x720 at the CCD's, it records only 960 x 720 pixels on the tape.

By comparison, the JVC ProHD cameras have 1280 x 720 pixel CCD's and they record 1280 x 720 pixels to the tape.

Thanks, Carl

960x720 @ 4:2:2 vs 1280x720 @4:2:0 using two very different codecs at very different bit rates.

My point only being that fact DVCProHD is not recording those lines doesn't make it a better or worse approach in itself. Each pixel of DVCProHD is less compressed on average.

Having said that there is no need to go into it any further, we all know the pro's and cons of HDV vs DVCProHD, its been done to death.

All HD-100's are capable of SSE. If I hook up my HD100 to my TV, turn all the lights off in my apartment, boost the gain to 18db and point the camera at plain white wall, you will see the line down the center.

The reason for this phenomenon is that the 2 DSP engines need a certain amount of light in order to calibrate with each other properly. If there isn’t enough light, they can’t sync together.

If you are shooting a dark area for an event, use a camera light. If you need a night shot, drop the gain down and add some light.

What some people on these forums don’t seem to understand is that if you are stressing your camera to the point of showing SSE, you will have so much gain noise and murky color, SSE will be the least of your problems. You are not using the camera properly.

18dB is ridiculous, I've never pushed a camera that far and actually put it to tape. I won't even go above 6dB with 2/3" cameras that aren't even as susceptible to gain noise as 1/3" stuff like the HD100. Maybe 12dB if we're in a situation where I just have to get the shot no matter what, but I'll try to add more light first if possible. By the time I'm pushing 18dB I'm telling whoever I'm shooting for there's simply not enough light to shoot, or I'm pulling out my handycam which has a "Night Shot" mode. For some reason Green Vision just ends up looking more acceptable.

I've used +18 db gain on Betacam shooting a fly on the wall documentary with the police. It's best only used when content is important and you don't want to create a distraction using a light in volatile situation. It actually looks worse in the V/F than on tape because of the peaking.

However, that is rare and + 9db gain on a Betacam (when required) covers 99% of documentary work, including under street lights. For drama etc you want much lower gain settings (if any).

I suspect that the above sensitivity might be the same as using +12db gain on the JVC using Standard gamma.

It's a lot better to light than just to put in gain. All cameras have limitations and you have to select the right camera for the job you're doing.

Guy - Read this and see if 4:2:2 and 4:2:0 with DVCPROHD and JVC Flavor of HDVB is really that different.

http://www.bluesky-web.com/HDVHVX.htm