New HD100 Clips 9-5-05
 

Friday night I dropped off into downtown L.A. on my way home from work...the light was getting low and nice and warm, and there was just enough atmosphere (ok ok, smog) to make things interesting but not make things ugly.

For those just tuning in, you must right click and "Save As" to your drive before you can play these. Otherwise your browser will try to open it as alphabet soup. I recommend VLC for both PC and Mac to play these correctly.

http://www.hdvinfo.net/media/nweaver/Downtown4.m2t
http://www.hdvinfo.net/media/nweaver/Downtown7.m2t
http://www.hdvinfo.net/media/nweaver/Downtown8.m2t
http://www.hdvinfo.net/media/nweaver/Downtown9.m2t

Lesson learned on this jaunt out with the camera? "Infinity" on many lenses is never quite where the focus ring stops turning...and you need to remember this doubly so when on the long end of the 16x wide open. In other words, you HAVE to take a second out and use Focus Assist even in situations where you think "Ah, just putting focus here under this circumstance will be ok". Nuh uh.

Nice...
 

Hi Nate,

I am very impressed with these bits of footage.

Is this literally out the box?

Waiting on mine at present and these images have made me even more impatient!!!

Glenn

Hey Nate,
Footage looks great. I gotta say the HVX is gonna have to produce some amazing images to top this cam from the few tests I've seen.

Thanks,
Eric James
htt://www.expertmagic.com/

Yeah. I have a few setting tweaks going on (low knee point, -2 detail, etc), but there's nothing going on that you couldn't reproduce yourself.

Somebody will note the super warm tone in the clips...I white balanced to white in the shade because the nice tones of the low sun weren't coming through, and all the numerous shadows were going blue.

Nate, this looks great. On a screen this size, it looks incredibly like 35mm film IMO. It doesn't have one single trait of video. The gamma specially looks extremely film like. Was it all shot in 24p HD?
Now that you have had the camera for a while, have you experienced any dropouts ? I'm wondering if a FS4 is really a must. I haven't seen many complaining about dropouts. The ProHD codec seems to be more reliable than the sony HDV one.
The more stuff I see from this camera, the more I want one. Yep, Eric is right. The HVX200 will have to be a really awesome camera to top this one off. The HD100 is sure a winner. It has great potential. Specially if you rent a mini35/cooke set up to REPLACE the stock lens. You just can't do that in progressive HD with anything less than a 60k Varicam.

Nate,

This was shot with the lens that comes with the camera, right?

Yup, it's 24. Note I took very special care in NOT having any large part of the frame going over 100IRE when exposing...blowouts are the surest way I know to give away DV.

So far I've shot about an hour's worth on my cam, but with LOTS of shuttling around and reviewing footage. No dropouts so far on Sony Premium DV tape.

Stock 16x.

That's very good news. The HD100 may put an end to the whole HDV dropout myth.

I deleted an email from ZGC by mistake about an upcoming show.

Nate,

I believe you did the test with Barry right.

What about the adapter that JVC says you can use other lens with it.

Do you know anything about that?

This was suppose to be part of this show.

Myth?

The very first shot I ever took on an HD100 was *riddled* with dropouts. Brand new tape, too. Unwrapped it, put it in the camera, shot, and this is what I got:

http://www.24puser.com/hd100/30p-ms-off-dropouts.m2t
http://www.24puser.com/hd100/30p-ms-on-dropouts.m2t

Now, the only things I can say are a) JVC specifically warns not to shoot anything valuable on the first 2 or 3 minutes of the tape, and b) I don't know what type of tape had been used in the camera before I got to it (it was a camera on display at WEVA).

But if you think dropouts are a "myth"... you do so at your own peril. This was enough of an experience for me to decide that I absolutely wouldn't ever shoot anything of value on HDV without using at least the DR-HD100 hard disk recorder as a backup. Others' standards may vary, others may find the risk acceptable. I sure don't.

But dropouts are no myth... Sony's even come out with an official statement acknowledging the problem.
http://www.sonydigital-link.com/dime..._tape.asp?l=en

I don't know that, Barry...

I can imagine the mixed lubricant used on a diplay model would make lots of dropouts possible....

But the MPEG2 is actually more robust against dropouts then DV - and with high quality tape, I have never experienced much (in DV - so I expect even less -second to none- in HDV with high quality new tape and always the same tape)...

Artefacts, that's no myth, although the JVC seems to do EXTREMELY well - as far as I've seen for now, and I guess that won't change anymore...

It seems that what SONY is talking about are dropouts caused by dirty heads. That would actually be in line with the fact that the camera you used was a display WEVA unit. Used, abused, different brands of tape, shuttling, etc.

Oh yeah - and always SHORT tapes. The plastic base is somewhat thicker there. Perhaps 30' tapes?

Wow
 

I have to admit, these are the kinds of clips I was expecting to see. My mind is at ease, because some of the earlier clips that I had seen pre-nate era :) were not very good at all. However in the hands of an expert....

And yes I second Michael Maier, It does look very 35mm like, if sometimes better, NO GRAIN !! super clean and crisp.

Speaking of which, I was watching Dark waters some weeks ago, and saw the trailer for Wolf Creek, and the image just blew me away, I had a hunch that it might be HD, it was! very clean

Pre-Nate, indeed -- Nate, those clips were extremely clean. Probably some of the best HDV footage I've seen. Very impressive, and really worked well with the stock lens too! If you remember -- were these mainly shot wide-angle?

I tend to stay wide unless I have a reason to punch in. The bus-stop shot was full wide. The mega-tilt to the street below was full wide, the generic street scene with cars driving away was prob 40mm, Disney hall shot prob started at about 60mm and went to full wide.

I just watched the clips again, and the more I see them, the more I like them. Great job Nate.

As you say yourself Barry, and as already put by others, you don't know what tapes have been used in the camera before you shot. It doesn't matter you used a new tape. I'm 100% sure several different brands were used and the heads got guped up with different tapes lubricants. Do the same with a DVX100, XL2 or PD170 and tell me they won't behave the same way, maybe even worse. You just can't compare a trade show unit, which has been used and abused, to a normal running unit. I rather go by Nate's experience, who I know is using the camera the way it should be used in normal shooting conditions, than go by some bad moments had in a trade show with a public unit. But that's just me.

I say myth, because I remember in the early days of DV, all the pros, and also sony, in a desperate effort to protect their higher end line, were saying DV wasn't good enough because it had lots of dropouts and etc. Even Panasonic was preaching how plain DV wasn’t good enough, because it wasn’t as robust as DVCPRO etc. Today, DV is used for professional work in large scale. Where did the DV dropout myth go?

I hear the HDV codec is less prone to dropouts than DV. The thing is when it happens, it's uglier, because of GOP. But it hasn't stopped DV and it will sure not stop HDV, specially ProHD, which has been proving to be a better solution. How many professionals out there shoot DV with a direct to disk solution? I would say nowhere close to 30%. If HDV is safer, I'm sure even more people will take the "risk". As many put it, just use high grade tape rather than the 5 tapes for a dollar deals one finds around.

Bottom line is, the HD100 is a winner IMO and as already pointed out, the HVX200 will sure have a run for it's money. Specially that it can't shoot HD for under 6k. I think they both have desirable features need in the field. One will just have to pick the ones he needs most. I'm actually really surprised with the HD100. I didn't expect it would be that good, based of the JVC HD1. But what I have seen from it, proves me so wrong.

The mini35 test, had the excuse of had used a $60,000 lens, which is beyond reach of the large majority of users for this camera. But Nate’s test is just straight out the box with the stock lens. It looks great! I know Nate knows what he’s doing, but it just proves the camera is a winner

Nate, was the white clip on 100 or 108?

108, with a knee of 80.

Aha, no trust in the auto-knee? or just because of the danger with the reflecting surfaces?

Gain on 0 dB?

Kinda. At least in manual I have an idea of what I THINK it should be doing...flattening out the top of the curve. Auto? Who knows what that means. Also for consistency's sake.

Yes, 0db gain.

I think that it is the best choice for the shot with the bus. I might have used the auto-knee with the big tilt-shot, but it looks great like this also...

Too bad you can't choose auto-knee with a button, instead of with a manu-setting...

I agree. Which is why I said what I said. But the point was not whether the camera had been used properly, etc., but to point out that HDV is still susceptible, and the only "myth" going on is that ProHD is "more resistant" to dropouts than DV. That's a silly (and even outrageous) claim. The only way to improve dropout "resistance" is by using a better tape. Dropouts are not dependent on the format, or on any shuffling of tracks, etc. Dropouts happen because there's a fault with the tape -- either a speck of dust, or something flakes off the tape, etc. The only thing ProHD could try to do would be to minimize the impact of a dropout, but they cannot in any way minimize the occurrence of dropouts.

And as the tape I showed demonstrated, ProHD does indeed appear to handle dropouts differently. On the Sony, each of those dropouts would have resulted in a half-second freeze-up. On the JVC, it just scragged portions of the picture.

When did that become a "myth", or a "desperate effort"? It's the truth. DV is a dropout-prone format. I get an average of three glitches per tape. It's not an overwhelming amount, but it's irritating and I can well understand why the news business is now 70% DVCPRO (and not DV). For average users that may not be much, but for pros who are shooting news (inherently a one-take business) dropouts are extremely irritating. Which is why Sony and Panasonic both immediately developed alternate formats to reduce dropout occurrence.


The "myth" I can't speak for, but the dropout reality is with us every bit as much as it always was. Moreso as a camera ages, too (my brand-new camera rarely had a dropout in the early days, but showed more as the hours racked up on the heads). Just shot a short film in the desert last February, and the additional factor of wind & desert meant that we got an average of two or three dropouts PER MINUTE. (never shoot in the desert without some sort of rain cover or something!) Every take we'd have to stop, rewind, pand play back, just to make sure there were no dropouts. Also, I used to shoot a 28-minute local television show in three 9-minute takes, and I could count on having to fix at least a couple of dropouts in every episode. Usually little tiny "sparkly" dropouts, usually not the big ugly "venetian blind" dropouts.

Dropouts are a reality. Their frequency can be mitigated somewhat by using the same brand of tape, high-quality tape, cleaning the heads, etc. But they do happen. It's the nature of the format (and it's a primary reason DVCAM and DVCPRO were developed). And dropouts affect HDV a lot more significantly than they do DV.

Ah, now there's the myth! That's patently impossible. Dropouts happen because of the tape, not the codec that's written on the tape.

What JVC has done is scramble the data around on the tape, so they can minimize the impact of a dropout AS COMPARED TO SONY's FORMAT. A Sony dropout seizes the entire GOP. A JVC dropout scrags a portion of the picture (and occasionally seizes a whole GOP, I've seen that too). But in both cases the dropouts are far more destructive to the picture than they are in DV. Look at the clips I posted again -- every dropout is a significant scrag that lasts for multiple frames. That's the nature of having a GOP-based codec -- if something gets scragged, it's scragged for the duration of the GOP and there's *no way* around that. Sony drops the whole GOP, JVC just lets the scrag play through, but it happens. With DV, that dropout may have glitched a little bit of one frame. With HDV, it's going to affect an entire GOP, each and every time. Dropouts are always uglier on HDV than they would have been on DV.

Don't know. But I do know that more of us are doing so. I have DV Rack on every shoot I do now. Obviously not possible for ENG work, but for anything where there's a "video village", I shoot on DV Rack every time. And I won't be caught without the FireStore on the HD100 either. Others may have different levels of tolerance, but one thing is inescapable: users *will* get hit with dropouts. Some will find the risk acceptable. I'm tired of it, I don't find it acceptable.

Anyone using cheap tapes, especially with HDV, is just asking for it. There's a reason Sony introduced $25 "HDV" tapes! If you're going to shoot on tape, the only thing you can do to minimize dropouts is to use better tape.

Look, all I'm saying is that dropouts do happen. Pretending they don't serves no valid purpose. They do happen, if you're recording to tape they will happen, and that doesn't mean the camera won't succeed, it just means that it's a risk-management game you have to play. I would think that anyone who values their footage (and, frankly, if you don't value your footage, what are you doing shooting on high-def?) would want to be aware of what the risks are, and what the options are for minimizing that risk.

For some people, the cost savings of using a $3 tape will be worth whatever dropouts they get (that ain't me, but I know that a lot of people will choose that option). For some, using a $5 tape will provide all the margin of safety they think they'll need (and I wish them well). For others, using the $18-$25 HDV tape will be what they think they need to do. For guys like me, with zero tolerance for scragged footage, that means a backup recording mechanism (HDV Rack or DR-HD100). Pick your level of risk and your strategy of management and go forward. But do so with your eyes open, recognizing the very real issue of dropouts and their more-serious impact on HDV footage. Don't cling to the false notion that dropouts are a "myth", or you'll be hating life when you get hit with 'em.

Dropouts can happen and it is in no way related to the format, but the tape (basicly). Very true.

Some things need to be said.

1) from personal experience: experiencing 3 dropouts per tape is an awful lot. Since the days I use an XL1 I never had more than 1 dropout per 3 or (perhaps) 2 tapes... (that's a factor 6 to 9 less!)

2) It is no myth that HDV is more resistant to dropouts than DV. Dropouts are caused by bad patches of tape, or dust or worn heads - not related to the format. BUT, HDV writes the image data in different places on the tape, making a difference whether or not the dropout will affect the image or can still be corrected by the error-correction. The dropouts might be there, but not visible to the way it is written to tape and then corrected if needed.
It is a reality that HDV is somewhat robust against *small* dropouts. Mind you, a big dropout will eat a big enough patch on the tape to make it impossible to compensate, so HDV isn't any more robust angainst serious dropouts then DV - just the very small ones.

3) I can't back Barry more on this: The ProHD format can't avoid any more dropouts than sony's standard HDV, but it handles them better apparently - and that is what it is all about, sometimes...

That may cause dropouts too... (especially in the desert...)

I don't know how ProHD recording works, but if it's 19Mbps it potentially has an extra 6Mbps they can use for error correction to fix dropouts. So in theory it could be significantly better than 1080i at handling dropouts.

Wow. I'm upset if I get a single noticeable glitch on a DV tape... the only time I usually see them is for a minute or two after changing tape brands.

Nope. That area is unused. They have plans to someday add uncompressed audio to that section, but right now it's not used at all.

Me too. But then again, I've met people who claim they don't get any dropouts, and I can point them out to them. They just don't see the little one-frame sparklies. I see 'em all the time and it drives me batty. I see on on broadcast television shows too, on small cable networks like TLC and HGTV.

Hey Chris, this thread should probably be split out -- it's taken a significant turn away from Nate's excellent clips and we've drifted off onto a dropout tangent... probably should be reorganized to keep Nate's original thread on-topic.

Well, those little dropouts bother everyone, I guess... I can't imagine anyone not noticing them.

On the other hand: if it is a way of shooting, it's not that important. Normally you end up using roughly 5-10% of shot footage (events). You might get lucky...

We rarely get dropouts on DV because we only use the suggested tapes per camera. Panasonic get's fed Panasonic. Sony get's fed Sony and JVC get's fed JVC. A drop out or block noise is pretty rare in these parts.

This brings up something else about the companion deck to the HD100. According to the literature. The deck has error corrrection. Does anyone have the companion deck?

i have access to a JVC BR-HD50. how can i check the error correction?

mark.

For what it's worth, I've shot several weddings now on Sony FX1 and Z1 cameras and have yet to experience the sort of blocky dropouts I'm used to in DV or as shown in the sample clips posted in this thread. I have had a few incidents where the video dimmed noticeably for a moment in a way which may be due to some kind of dropout, but no blatant digital artifacting.

I'm still waiting for Firestore to release their HDV firmware for the FS-4, so I can use that as an alternative recording solution.

Anyone notice on the mega tilt-down (Downtown 7) at the end, looking down at the road, the upper part of the screen has shifted a little to the magenta and the bottom has shifted a little to the green.

I've seen this problem with my old GY-DV 500. I sent still frames of this phenomenon to my JVC tech and he couldn't see it. Maybe they've grown accustomed to seeing it, who knows.

I saw this as well... well maybe not the green shift at the bottom. I thought the top may have been the lense flaring a bit from the sun, but definetly noticed it at the end of the tilt.

Can anyone else comment on why this may be happening?

Small white details show the CA...if something is on ONE side of the focus point, it will have green edging, if it's on the other, it'll have magenta.

I suspect all the white speckles in the street are inside the focus point the lens was set to. Even though they're technically in focus (because I was probably shooting a F8), each little speckle likely has a green edge.

Honestly, I never saw it until mentioned here, and I had to watch it 4 times to see it.

I don't mean to sound like a Fuji apologist, but if I have to watch something 4 times to see the problem, I'm not going to sweat it.

Thanks for the better look at the stock lens - it's sharp and the CA are not nearly as bad as in other samples, so maybe it's something that can be worked around, esp. if shooting for SD display on smaller screens.

Personally, I don't think this affects the HVX200 much - the HD100 is demonstrating that 24p and progressive scan count for a whole bunch is delivering a nice image. The Z1, especially those shooting cineform mode or deinterlacing in post looks weak compared to what these shots look like.

thanks for the clips!

cant see clips
 

I really wanna see these clips you posted Nate, but I cant seem to down load them successfully. I am on a mac with no right click but when i 'control' click i get an option to download but it comes out as jibberish like you said it would.
What am i doing wrong? The files, when downloaded, have a .txt ending added to the filename......??????????????????????????????????????

Craig