Let me throw my technician hat on for a moment and explain what's happening, given the description of the technician's actions that Panos described. The firmware updates are most likely adding more 'calibration points' than were originally available. This makes it possible for the technician to use the scope and monitor to calibrate both sides independently at more points in the response curve so that their outputs match up in all modes under all conditions. BTW, the high tech term for a calibration point is 'fudge factor'. :-)

This is how you overcome the inherent flaw or limitation of the original design. Had the same scenario with a new to market machine in our wafer fab a few years ago. The factory had to revise the software to allow for more calibration points in the magnet response curve so that what was actually happening at the physical level would jibe with what the front panel monitor displayed. Without those extra points, the machine couldn't accurately auto tune itself at various points in the AMU curve.

-gb-

Greg - you are spot on. I had another visit from the JVC engineer this morning to fix my SD timecode problem and I specifically asked him to confirm whether firmware updates were responsible for improving the removal of the split screen (specifically because of this thread).

Unprompted he informed that what the firmware updates were doing was adding more calibration points for the camera (originally they only had calibration points at 0dB and +18dB), now they have calibration points at +12dB and he thinks more may be added in future.

So mate - you are officially a smart cookie.

I'm still not sure why anyone needs to deal with "imply" when JVC has made it very clear that those units that have "Inspected by JVC USA" stickers have the latest of everything. Whatever SSE these units might have they are not "defective." (Which does not mean that an occasional unit might be defective and need to be replaced for all sorts of reasons.



Steve - mate you're missing the whole point of the firmware. It doesn't matter if all the USA units have the latest firmware and "Inspected by JVC USA" stickers, because it is the automated calibration of the camera at the factory which is failing, not the firmware. And the fact is having the latest firmware DOES allow the camera to be better calibrated (see Panos' report), and it seems the best calibration is done by hand by a technician. Don't believe me - ask JVC that specific question - are there calibration points in the firmware for minimising the split screen and can a technician adjust them? Better still ask if you could see it being done (preferably on your camera). If they are inspecting every unit ask what percentage of cameras they are rejecting because of excessive split screen?

On & off topic - I was frustrated last evening, as I was trying to resolve an image problem while using my mini35 with my HD100. Turns out I had some micro dust on my rear glass element on the mine35 adaptor. I had compressed air and Qtips for dinner last night - my favorite.

So I was very happy as I previewed my final footage on a 34" widescreen WEGA to see that the specks/blotches had disappeared - a nice clean image - then I noticed the split effect. This one was a subtle/slight discolored right half, not matching the left. Been a while since I've seen it.

I was shooting in my living room, with a good amount of light, plus a lowel omni bounced into an umbrella, with a Canon f/2.8 70-200 zoom, through the mini35, into the HD100, using variable test gains 0, 3 or 6db, 1/60th shutter. I would adjust the back iris on the mini35 to keep the video legal (zebras set to report at 100%).

I believe the last setting where I saw the SSE was at 6db, mini 35 back iris around 1.5-2, with that Canon lens at 1/60th. SSE may have gone away at 1/30th.

Image looked nice - I was surprised to see the effect - never seen it with that much available light. The lens and adaptor certainly hinder lightflow (f/2.8 plus lose another stop or stop & 1/2 with the adaptor), then back iris, then gain, etc.

Just a quick test case, and maybe not the most ideal way to engineer the shoot, but it wasn't an extreme environment either. I recently had good luck with the stock fuji lens testing at a wedding reception at 9db, 1/30, with about 30-50 (variable) watts of diffused on-camera chimera mini-light - no SSE in that footage. Footage was dark, but certainly usable. I can't see using more than 9db of gain and still being happy with it.

I've been staying out of these discussions for a while (the posts seem to give me optimism and then concern about every 5 minutes) - just thought I'd post this for anyone using the mini - perhaps something to watch out for. I'm still holding out faith (blind or not) for even subtle fixes. Still a very big fan of the camera and I've had very good experiences with JVC for years - just not completely sure I trust the cam (or myself with it) in every scenario yet. That's why I test I guess.

Even if I thought the JVC high definition camera was a paperweight it is still better than nothing...

Tommy, I haven't followed this whole thread since it became a whole nonsense debate war with kilometric posts. But if anybody called the HD100 a paperweight, that is the dumbest, most ignorant and unfortunate remark I ever heard about the camera. One should ask this person what he’s shooting with. But chances are, after that dumb remark, he wouldn't get good results even with a Viper, because his ignorant remark clearly shows the error is between the viewfinder and the floor, i.e. operator error. His remark is something only somebody who has no idea of what to do with a camera would say. So just ignore such person.

But in the end, you have to allow for the possibility that the design will be updated to help reduce or eliminate this SSE issue without fanfare or public acknowledgement of any kind.

-gb-

They are called ECOs and are always occuring. But, after you have said that -- so what? We'll never know when of if and so such changes have no relevance to the here and now.

I can work now with a camera that won't appeat for 3 months. I'm simply saying, this thread is no irrelevant. We know the answer.

Steve - mate you're missing the whole point of the firmware. It doesn't matter if all the USA units have the latest firmware and "Inspected by JVC USA" stickers, because it is the automated calibration of the camera at the factory which is failing, not the firmware. And the fact is having the latest firmware DOES allow the camera to be better calibrated (see Panos' report), and it seems the best calibration is done by hand by a technician.

I said exactly that several days ago. The firmware does not FIX SSE it enables better calibration. Since by now all PAL units should have this firmware and all USA always had the firmware -- I see no reason to talk about it.

If you want to start another Thread on CALIBRATION that makes sense to me.

But, other than when the PAL owners take their init in for the firmware upgrade -- I know of no way to have an NTSC unit re-calibrated -- unless the unit is obviously horrible.

Thus the focus should be on calibration not firmware.

Can you pay a local JVC tek to calibrate your unit?

Should JVC offer a a VIP calibration for $300?

Should JVC offer a DELUXE version that includes a VIP calibration for $500 more?

These are all good questions. But debating future potential firmware is a waste of time and confusing to every new person who reads this thread. If you bought an NTSC unit you have the firmware and your unit was calibrated.

I can't see using more than 9db of gain and still being happy with it.

Because noise can be encoded into blocks, it looks worse than DV noise. For this reason, +6dB may well be the safest limit outdoors where there is black night. In short, adding gain to black night is a very bad idea!

On the other hand, indoors with moderate light, +12dB seems OK. In short, adding gain to get a better indoor image seems to work fine!

Likewise, if you do not have an acceptable minimal level of light, adding +18dB will be of no value because you'll not have much color and the color will be wrong!

In short, I agree with you: 9dB +/-3dB seems the limit.

NIGHT = 0dB to +6dB
INDOORS = 0dB to +12dB

I said exactly that several days ago. The firmware does not FIX SSE it enables better calibration. Since by now all PAL units should have this firmware and all USA always had the firmware -- I see no reason to talk about it.

If you want to start another Thread on CALIBRATION that makes sense to me..

Steve it was you who said that Greg Boston failed to "grok" what was going on yet my information is that his explanation is spot on. I thought you didin't believe that firmware could be "improved" to allow better calibration.

But, other than when the PAL owners take their init in for the firmware upgrade -- I know of no way to have an NTSC unit re-calibrated -- unless the unit is obviously horrible.

Of course NTSC units can be re-calibrated in exactly the same way as PAL units can. And why should a unit have to be "horrible" to demand re-calibration - surely it simply has to be out of spec.

Thus the focus should be on calibration not firmware.

Can you pay a local JVC tek to calibrate your unit?

Should JVC offer a a VIP calibration for $300?

Should JVC offer a DELUXE version that includes a VIP calibration for $500 more?

These are all good questions. But debating future potential firmware is a waste of time and confusing to every new person who reads this thread. If you bought an NTSC unit you have the firmware and your unit was calibrated.

If every unit being delivered had identical split screen you might have a point, but that is simply not the case. Are you saying that every HD100 being delivered has been optimally calibrated - because as a user that is what you'd expect. I promise you that that is not the case. If you'd read my previous post then you'd know the culprit is the automatic calibration at the factory, which is failing on a percentage of units. Why should any user pay extra to have their camera calibrated correctly?

I agree that there is little point in debating future firmware fixes at this point - it is far more likely that JVC will bring out the HD100UA and never admit there was anything wrong with the 1000's of units in the field (see Panasonic for details on how this is done)

Steve it was you who said that Greg Boston failed to "grok" what was going on yet my information is that his explanation is spot on. I thought you didin't believe that firmware could be "improved" to allow better calibration.

I said exactly the opposite -- I said because of his background he did grok calibration.

And why should a unit have to be "horrible" to demand re-calibration - surely it simply has to be out of spec.

Since we don't know the sprc, all we can tell JVC is "it looks horrible!"

Are you saying that every HD100 being delivered has been optimally calibrated - because as a user that is what you'd expect. I promise you that that is not the case.

I would love it to be the case, but of course, they are not perfectly calibrated.

If you'd read my previous post then you'd know the culprit is the automatic calibration at the factory, which is failing on a percentage of units. Why should any user pay extra to have their camera calibrated correctly?

Because mass production doesn't yield perfection, you can argue it should, but it doesn't. When I buy a $4ooo HDTV I still expect to spend $300 to have it ISF calibrated.

I agree that there is little point in debating future firmware fixes at this point - it is far more likely that JVC will bring out the HD100UA and never admit there was anything wrong with the 1000's of units in the field (see Panasonic for details on how this is done)

Sony and Panasonic both do "A" versions -- nothing new here. This is how business is done. You seek a solution that gets you what you want. What's a few hundred dollars?

It looks like Greg was spot on about the calibration and firmware. I have not been here in some time because of the above noise - too bad that's not fixable in firmware either :)

There was some earlier discussion though that heat and progressive scan were the issues - perhaps explaining why it's shown up in unexpected situations. I'm pretty sure I'm not getting an HD100 but the HVX is still on my HD shopping list. Has anyone heard recently that the calibration can drift if temps go up?

It looks like Greg was spot on about the calibration and firmware. I have not been here in some time because of the above noise - too bad that's not fixable in firmware either :)

There was some earlier discussion though that heat and progressive scan were the issues - perhaps explaining why it's shown up in unexpected situations. I'm pretty sure I'm not getting an HD100 but the HVX is still on my HD shopping list. Has anyone heard recently that the calibration can drift if temps go up?

Sorry Stephen that's my fault, I brought up heat in another thread - it was merely speculation on the part of the JVC engineer that I was talking to that the automated calibration might not be working properly on some units because of temperature, not so much field performance.

Thread closed -- Some Brand X vs. Brand Y platform bashing removed -- some all caps shouting converted -- thread re-opened.

Sorry Stephen that's my fault, I brought up heat in another thread - it was merely speculation on the part of the JVC engineer that I was talking to that the automated calibration might not be working properly on some units because of temperature, not so much field performance.

Okay thanks - so many posts on SSE, hard to keep track.

Adrian, I use this camera for weddings, if you want to check the picture quality, here is the links. Take about 20 mins. to download with hi-speed.
I very happy with this cameara, I ready to buy one more.
I don't have a splitscreen effect at all at any situation.
I tried other HDV camera but I sent it back, and I ended up with this one.
I love it.
I hope these little footages help for your decision.

http://www.megaupload.com/?d=7D4VC6EV

http://www.megaupload.com/?d=2F2ZFXKG

Laszlo

Sorry Stephen that's my fault, I brought up heat in another thread - it was merely speculation on the part of the JVC engineer that I was talking to that the automated calibration might not be working properly on some units because of temperature, not so much field performance.

You are correct about heat! JVC now recommends a 5 minute warm-up.

By the way I wasn't shouting -- only differentiating my answers to your points as quickly as possible!!!

Anyway, I think it's now clear that:

1) There may have been, many months ago firmware specifically to decrease SSE -- or firmware that modified any of the functions that could affect SSE. It seems likely that was done before any units shipped to PAL areas.

2) The factory calibration was discovered, via these PAL models in the field, to not yield acceptable camcorders. Or, the PAL region managers said to hell with SSE, we want the units because IBC is coming. Who knows the truth?

3) The USA said "no way" and waited for firmware that improved calibration. And, it looks like they may re-calibrate ALL or a statistical SAMPLE of the incoming units. Who knows the truth?

4) This firmware can be installed on the PAL units and they will be recalibrated.

5) Calibration is never a BINARY process. At it's simplest it means running X tests and checking if the camera meets Y specs. At it's best, a skilled tek can take hours and beat these specs -- perhaps by a significant amount. Perhaps even totally eliminating SSE.

6) Since time is money -- I'm simply suggesting that JVC offer a "Premium Tune-up" that optimizes all aspects of the camcorder. In reality, you can bet that if CNN buys a hundred of these you can bet JVC will provide them with full information and CNN will calibrate every one!

In fact, any big name renter or buyer will be able to calibrate their units. Think of the Panavision mods to the CineAlta. Hell, someone here could start a business buying HD100's and adding all the film options and calibrating them. Then sell for $10K.

Just got off the phone with a JVC factory technician, and he basically confirmed everything Steve's been saying.

There is split-screen in every camera manufactured, and it's not going away.

The confusion between US and non-US cameras has to do with the delay between US and non-US units. non-US territories received their product long before the US did. Non-US territories are now receiving the same level of firmware that the US is using. The version that I have (version 1.17) is the latest, and that's what everyone else is getting too.

The splitscreen is not going away. It will be with the HD100 forever. It cannot be "fixed", it can only be minimized by not shooting in low light circumstances, etc.

He did say that units that haven't "warmed up" yet are more prone to showing the effect, so after it's been on for five minutes it should be more accurately self-calibrated than it is straight out of the box.

He said that you can also sometimes notice the autocalibration in effect when changing gain levels -- sometimes you boost the gain and see the split-screen but it goes away after a couple of seconds. That's the calibration at work.

But there are scenarios where you're going to see split-screen. It is the nature of the beast. And it will not go away. Those who are claiming their cameras don't have it, probably don't know where to look for it, but he says they all have it and it's the price they (and you) have to pay to get a native-pixel 1/3" HD imager at this price point.

But there are scenarios where you're going to see split-screen. It is the nature of the beast. And it will not go away. Those who are claiming their cameras don't have it, probably don't know where to look for it, but he says they all have it and it's the price they (and you) have to pay to get a native-pixel 1/3" HD imager at this price point.

Thanks Barry - any thoughts of if this is infact true - that all native pixel 1/3" HD progressive imagers will have this effect? Sure throwing some CPU and DSP power at the issue would deal with it?

In addition to Barry's post. I was told by JVC that the issue is the circuit does not update fast enough to make split screen not noticeable. I believe this information (along with Barry's post) is not the entire story. I do believe the camera can be calibrated to different tolerance levels (as witnessed by another user in Jakarta) by a technician that can take the time with the camera.

I was told 9db out of the box and I'll hold them to it. Later down the line, if the camera goes "out of tune", I'd be willing to pay a tech to get it back into spec and chalk it up to the price of business. The camera is most certainly worth 5-6K.

Yet, the HD100 is selling like hot cakes. Go figure.
I don't think so...not as far as I was told by a JVC dealer here in BC.

I don't think so...not as far as I was told by a JVC dealer here in BC.

I wonder why I was in line for almost 2 months before I got mine then.

I wonder why I was in line for almost 2 months before I got mine then.
Early adopters. Simple. My dealer has had their in-house/demo unit for weeks now and everyone is waiting to see what's going to happen.

Early adopters. Simple. My dealer has had their in-house/demo unit for weeks now and everyone is waiting to see what's going to happen.

Yet, people reporting to have just bought the camera are popping up all over the place online. This is just the online community. If you count all the other people who don’t post it (which is the majority), then you have no telling.
Waiting you say? You do realize that the huge majority has no idea of the so-called SSE, right? Just the online community knows. But anyways, I still don't understand all the fuzz about it. I have been shooting with the HD100 for over a week now and have never seen SSE in my normal footage, night or day.

So, anybody knows how many have been actulay sold?



Adrian


____________________________________________________
Al zodape de naco que me saca los mensajes se puede ir a la reconcha de la lora.

Camera manufacturers seldom release that information, sorry.

Who cares how many are sold. I know 90% of those who have one are very satisfied with the camera. Barry Green is correct, in those cases that I have seen SSE have been when the camera was cold, after a few minutes of warmup it has dissapeared.

A friend of mine who is a experienced DP visited me to check out my camera and was intrested to see the infamous SSE in person. We hooked it up to a projector and we could see it when we zoomed in to a dark area on a white wall, but after a couple of minutes we could no longer see it, not even with +18db!

I guess Im one of the lucky owners as I rarely see the SSE, I have tried to force it by all means and I have seen it about 3-4 times since I recieved it. I might have been able to avoid those situations as well if I had let the camera get warm up!

The bad thing about the camera is that its not so light sensitive and is a bit noisy, but tell me one HDV camera that is light sensitive (compared to DV cams)?!

Well, I just got mine last Monday, and the ONLY time I have noticed an almost invisible SS is when I had it under extreme low light, 18dB gain, and using the macro focus feature. I am SO pleased with the outcome of the test footage quality I have been shooting!

I have an almost done deal with a producer/director for shooting a feature for TV, with a possible 35mm transfer for Latin America distribution. He is so pleased with the footage I have been showing him, shot by others and by myself, that he does not want to go to 35mm acquisition for his upcoming projects.

By the way, I just did an acquitition via Firewire to my laptop in real time using Premiere Pro and CineForm. I just connected the camera to the laptop and activated the Premiere's "F5". Without even hitting the Record button on the camera, I clicked Record in the Premiere capturing window and beautifully captured a 24fps HDV footage. So, I hope that this can be used in lue of the Firestore as direct-to-edit capturing workflow, except that it will be in my hard disk already without the need to transfer from the Firestore to the computer!

Luis

By the way, I just did an acquitition via Firewire to my laptop in real time using Premiere Pro and CineForm. I just connected the camera to the laptop and activated the Premiere's "F5". Without even hitting the Record button on the camera, I clicked Record in the Premiere capturing window and beautifully captured a 24fps HDV footage. So, I hope that this can be used in lue of the Firestore as direct-to-edit capturing workflow, except that it will be in my hard disk already without the need to transfer from the Firestore to the computer!

Luis

Congratulations for the camera and thx for the above info. Good luck with your project!

My here Malaysia have, Now in my hand JVC GY-HD101E are have SSE.
So will inform the service by next week.....