Will EX-1 be leaps and bounds better than HVX?

[harddrive]

The EX1.

It uses lots of small pixels to create a 4:4:4 image, ........... Seems like a lot of wasted tightly crammed (small) pixels to me.

I suspect one answer might be oversampled originals then tend to look very, very good at the output resolution, for a similar reason that even on a final VHS, it's pretty easy to tell the difference between a master on 35mm film and one on a cheaper camera - even though both are far better than the VHS itself.

Doing it this way also means a straightforward approach to processing, which can likely transform into low noise figures etc. Pixel shifting has a theoretical advantage in that it allows bigger pixels (=more light collection) and an output with greater luminance than chrominance resolution. So yes, perhaps better suited in theory if you intend recording at less than 4:4:4 anyway. But it also involves more processing, hence noise etc, which may well counteract the potential benefits.

 

[mikkowilson]

Yeah, it's a very fine line both ways ... keep in mind that the biggest culpit for noise is too small pixels that as a recult don't have enough sensitity to give out a clean signal.

Again, it's a limitation of camera and overall sensor size. It's the age old balance between resolution and dynamic range & sensitivity.

That's one of the reasons/benefits of the EX1 having 1/2" sensors as opposed to 1/3" on the competition like the HVX .. they have more surface area to play with to allow them to fit all those extra pixels into without loosing huge amounts of sensitivy over 1/3" cameras.

And that's all part of not just the technical, but marketing side of the features: Sony can proudly (and quite correctly) advertise that their new camera has both larger chips AND more pixels! The fact that the larger chips are a necessity for the greater pixel count just happens to be a convenient technical restriction.


Of course there are other variables in there too, like sensor technology and depelopment. (CMOS performs differently to CCD, and todays chips wil outperform the ships from 5 years ago..) And that's why it's really hard to compare cameras on paper (or web page), and why the final definative results only become evident in use in the field. And even they vary for each shooter, time, place...



- Mikko

 

[Stevet]

The EX1.

It uses lots of small pixels to create a 4:4:4 image, but then the highest format you can get out of the camera for recording is 4:2:2 (and that only over HD-SDI), which has only 50% of the color information intact. Seems like a lot of wasted tightly crammed (small) pixels to me.


- Mikko

Oh plesase....
LOL

 

[harddrive]

The fact that the larger chips are a necessity for the greater pixel count just happens to be a convenient technical restriction.

I don't see the bigger chips as a technical restriction, quite the opposite. Bigger chips normally means better all round for quality, and is a big reason why 2/3" cameras are generally preferred to 1/3"; DOF being another big reason.

Problem has been bigger chips normally means bigger lenses, bigger batteries, etc etc, and bigger normally means much heavier and more expensive. THE big rabbit that Sony have pulled out of the hat is having those 1/2" chips in a package size and price normally the preserve of 1/3", and the significance of that can't be underestimated.

Second big innovation is keeping a fixed lens (for size, cost, etc reasons) but making it truly manual! Why has the past choice either been fixed servo or manual interchangeable?

Those two aspects alone put all the other pros and cons in the shade for me.

 

[amguy]

The EX1.

It uses lots of small pixels to create a 4:4:4 image, but then the highest format you can get out of the camera for recording is 4:2:2 (and that only over HD-SDI), which has only 50% of the color information intact. Seems like a lot of wasted tightly crammed (small) pixels to me.


- Mikko

Now I'm not the brightest bulb on the tree, but... I believe one of the greatest challenges for manufacturers right now is compression. I think most of the newer cameras have 4:4:4 at the head, but the stream would be so large, the average user would not have the equipment to edit it.

Add to that the fact that broadcast bandwidth is still an issue, and it is even more complicated.

In the broadcast world this is an "inexpensive" camera. Sony, like many other manufacturers has to choose where to compromise, that will have the least affect on the intended user group they are targeting.

Much discussion about long GOP vs. Intraframe vs. wavelet, and what is best. Myself, and a whole host of other people, try to "dumb down" the function of compression in order to get our minds around it. The truth is that there is a lot more going on with the "new" implementations of long GOP, than the original. Due mostly to the on-board processing right at the chipset. I don't fully understand it, but it is "better".

I am not certain, but I believe colorspace is affected by compression. Or that colorspace has an effect on the size of the stream?

Thomson/Grass Valley has their new Infinity Series camera. Go take a look at it and tell me if you see anything similar to the EX. Granted it is a different beast in a different price point, but CMOS chips, albeit 2/3", flash recording, etc. There are at least 3 different compression options. Why? Compromise. Each different scheme affects the output in some way.

Panasonic paved the way with P2. When they released the HVX200 it was for a specific group. It doesn't have HD-SDI. Why not? It uses "smaller" chips to create the 720p image. How come? Compromise for price point. It has mulitple ways to capture/output video. Why? Not everyone can handle, or wants the highest 100mbps stream for editing, nor does every single project demand the highest level of quality. Sometimes light and quick is better.

So as long as we examine the "low end" of the spectrum, we will always be picking out and examing manufacturer compromise, and how we will work with it.

At least that is what I have come to realize, differing opinions may vary. :thumbup:

 

[Morpheus_23]

The EX-1 is heaps better than the HVX, well on paper at least.
[/qoute]

Some will believe this to be true.

The EX-1 has a bigger lens, it shoots 1920 x 1080, whereas the DVX uses 960x540 to achieve 720p, plus the EX-1 features a fully manual moving lens.
[/qoute]

Is this what makes the EX1 "heaps" better? If your buying based on resolution as the only criteria for your decision, then the EX1 is your choice. But if you are using a feature list from each camera (all cameras), at least for me, the EX1 still has some limiting factors.

The thing about video cameras is that they are like computers. I could buy a camera tomorrow that shoots 3000p, and then six months later, something different replaces it. It's almost an impossible rabbit to chase. The best that you can do is focus on storytelling. As long as it's a 3CCD HD camera, you should be okay.

I agree. No doubt I will end up with the EX1 in addition to the HVX.

 

[Barry_Green]

As long as it's a 3CCD HD camera, you should be okay.

Any of the 3-CCD HD cameras can deliver fine results, but they're certainly not necessary to be "okay" -- a 3-CCD standard-def camera's films have won the best cinematography award at Sundance three times (PD150 and DVX) and have been nominated for Oscars for Best Documentary twice (DVX 2006 & 2007).

You don't need HD. And you don't need three chips; the SI-2K and the Red and the HV20 owners would all argue against needing those.

Buy (or rent) something that does the job, but don't get too hung up on the gear because it is *not* what's going to make or break your project.

With that said, "better is better" and we all always want better. And as someone said earlier, there is no global "better". There's only "better for a specific purpose."

Okay, so let me try saying it a different way so I can avoid the whole "it's not the camera, it's the person behind the camera" platitude. Let's say it like this: If you think a better camera is going to be what makes your project better, stop right now. Because it won't. It's like what I used to say a couple of years ago when people would ask "should I shoot on film or video" -- the very fact that you'd be asking the question answers itself: shoot video. When you're ready for film, when the project will benefit from film, etc., you'll know it and you won't even have to ask. Until then, save your money.

 

[mikkowilson]

Oh plesase....
LOL

Please what?

I don't see the bigger chips as a technical restriction, quite the opposite. Bigger chips normally means better all round for quality, and is a big reason why 2/3" cameras are generally preferred to 1/3"; DOF being another big reason.

Problem has been bigger chips normally means bigger lenses, bigger batteries, etc etc, and bigger normally means much heavier and more expensive. THE big rabbit that Sony have pulled out of the hat is having those 1/2" chips in a package size and price normally the preserve of 1/3", and the significance of that can't be underestimated.

I never said that it was a bad thing - quite the contrary, I'll take a larger sensor over a smaller one any day (with all other thigns beeing equal).

I was just commenting how clever Sony was with beeing able to build a product that's also marketable beacuse of the way it's made.
Perhaps "technical requirement" is a better term than "restriction" for larger chips when you increase pixel count.

3 Full HD sensors is a huge (also marketable) featrure of this camera.
1/2" sensors is also an equally huge feature.
A fully manual proper Fujinon Lens also also a big feaure.

My point is the convinience of the fact that these all depend on each other to actually work succesfully.



Oh, one more side note, regarding your rabbit: As the feaures increase, so to the requirements, even with this camera. For example the EX1 requires 12v batteries (not 7.2) - meaning that it has a whole new battery type.
Also that nice lens also now has a brand new type of lens control port that is as of yet unsupported by anyway (even Sony themselves couldn't tell me the exaclt plan for it!) - so no more low cost LANC (or tradiational professional either for that matter!) on this camera.
As the move forwards, they do have to replace technology that has been around for years, and that does mean some increses in costs across the board too.

It's a new class of camera, it changes pretty much everything about these cameras. With both it's benefits and problems too.


- Mikko

 

[harddrive]

Much discussion about long GOP vs. Intraframe vs. wavelet, and what is best.

The obvious answer is wavelet, but things are never simple. (And to date I believe wavelet codecs all are I-frame.) You may ask why everybody doesn't use it then? It tends to require a lot of grunt for processing, which in turn can mean higher power etc etc. The big plus point is that when it gets challenged it gets soft, when most other codecs get challenged they get blocky. That's also very good in a broadcast chain, and cascaded compression.

And what is meant by best? Is it quality for a given bitrate, consistency, or ease of editing? You tend to have to trade one off for another, so which do you consider the best compromise? Because no matter how all the manufacturers squeal that theirs is the best compromise, they all are a trade off.

I am not certain, but I believe colorspace is affected by compression. Or that colorspace has an effect on the size of the stream?

Colourspace throws colour information away, making use of the physiology of the eye - more sensitive to luminance detail than colour. Exactly as with all analogue systems. That reduces the stream size, so *for a given desired output stream size* reduces the level of compression then required.

Thomson/Grass Valley has their new Infinity Series camera. Go take a look at it and tell me if you see anything similar to the EX. Granted it is a different beast in a different price point, but CMOS chips, albeit 2/3", flash recording, etc. There are at least 3 different compression options. Why? Compromise. Each different scheme affects the output in some way.

Infinity has been through a big redesign as they decided to use CMOS instead of CCDs. It's taken a year, but first reports now are very impressive, and apparently it is starting to ship.

Most interesting aspects to me are the choice of media and the (wavelet) JPEG2000 recording option. It's like the EX with 3x1920x1080 chips, but is a full size shouldermount. Unlike the EX it offers a choice of recording options - SD/HD, and different codecs, which is something I don't like about the EX. Some users may want to go tapeless before going HD - Infinity allows that. And the codec choice means integrating it into any larger system is easier than if they'd said "JPEG2000, take it or leave it". But two cameras at different prices aimed at different markets.

 

[amguy]

So based on information in this thread, can the original question be answered?

Are these 2 cameras similar enough to compare?

Both record to flash.
Is that where the similarities end?

2 different codecs.
1 has tape, the other doesn't
1 is CCD, the other CMOS
1 is 4:2:2 color space, the other 4:2:0
1 is SD/HD, the other only HD
1 has HD-SDI, the other doesn't
1 is 1/3" sensors, the other 1/2"
1 is 13x zoom, the other is 14x
1 is .14 lux min, the other is 3 lux min.
1 has 3.5" LCD w/ 900K res, the other 3.5" w/ less res

I think there is a huge difference in these cameras. I also believe that although the EX1 has the Cinealta badge, it is going to be better suited or geared toward broadcast. The Long GOP just might limit it's wide adoption by Indie filmmakers, because from what I understand, the Long GOP will "break" easier.

Indie filmmakers do a lot more in post to footage than ENG does. ENG is fast, quick turn, get it out. Clear, crisp, "leave the color alone" kind of shooting.

Filmmakers tend to add color effects and other enhancements to achieve the look they are going for.

I think, perhaps, that the DVX will continue to outpace everything else as a dual role camera. Maybe the EX1 would have greater limitations in a "film" style of shooting. Maybe we'll learn something new.

One thing for sure. About another week until people begin to get their new EX1's. It will certainly be fun to see what people turn out initially.

 

[Tzedekh]

Let's say it like this: If you think a better camera is going to be what makes your project better, stop right now. Because it won't. It's like what I used to say a couple of years ago when people would ask "should I shoot on film or video" -- the very fact that you'd be asking the question answers itself: shoot video. When you're ready for film, when the project will benefit from film, etc., you'll know it and you won't even have to ask. Until then, save your money.

Actually, you just said a better camera can make your project better. Or let's put it this way: if the script, the acting, and the direction are good and the DP competent, and the project doesn't call for a Super8 or home-video aesthetic (a la Oliver Stone), higher-quality origination generally improves the end product and the chances of getting a distribution deal. But you're right: higher definition and greater dynamic range won't save an otherwise poorly made film and may only magnify its shortcomings. In other words, make sure the cart and the horse are in the right order.

 

[harddrive]

The Long GOP just might limit it's wide adoption by Indie filmmakers, because from what I understand, the Long GOP will "break" easier.

Then again, it does have the option of HD-SDI out. 10 bit, true 1920x1080 resolution, 4:2:2 uncompressed, able to be recorded to whatever is desired. I'm not so keen on tethered recording options, but Indie filmmakers seem much less bothered by that.

 

[mikkowilson]

... as long as they arn't wanting to shoot offspeed, then that too is a very viable option. The HD-SDI output is another high end feature.

- Mikko

 

[karapetkov]

Oh, my head hurts... too much noisy rolling wobble ...

Maybe I should just focus on polishing my script. I've read more tutorials on that.

I liked the "choose your poison" thing. There will always be something better and there will always be defects and flaws... and let's not forget that sometimes defects can be turned into effects...

A good story doesn't become obsolete so quickly and sometimes they're timeless [I don't pretend to have one though].

And maybe it's best to leave the companies chase each other in improving their pixels and rolling shutters.

"Somebody has to die"... so I choose my poison... she's a brunette .

P.S. "Somebody always does"...

 

[moviemaker999]

Heres one thing that I don't understand about the Ex1 and sony. It seems that in the demo videos of the sony ex1, sony is constantly showing that the SxS cards kick P2. Now the sony cards are like 100mbs and Im not sure what p2 is (something close). Why wouldn't they give you a higher quality compression option if the SxS cards are so much better. Why limit it to 35mbs HQ if tthey can reach 100mbs. I understand that the express cards are the future,but they are making way too big a deal about this. The difference is marginal. I just wish that being free of xdcam disks they would offer a higher quality recording option, even if it would mean less recording time. I guess they are just keeping it for the ex2. And by the way I own a dvx and have stayed out of the Hd world and I think the ex might be my next camera.

 

[harddrive]

It seems that in the demo videos of the sony ex1, sony is constantly showing that the SxS cards kick P2. Now the sony cards are like 100mbs and Im not sure what p2 is (something close). Why wouldn't they give you a higher quality compression option if the SxS cards are so much better. Why limit it to 35mbs HQ if tthey can reach 100mbs.

Don't confuse BITS with BYTES. SxS is capable of a quoted 100M*BYTES*s which is 800 M*bits*s, the comparable figures for P2 are 80MBs (640Mbs) - so in practice they are very comparable.

But as to the question you posed, then yes indeed, exactly what I said in the last paragraph here - http://www.dvxuser.com/V6/showpost.php?p=1097685&postcount=90 . My own suspicion is that they didn't want to do anything that would cause a headache as far as marketing XDCAM disc cameras is concerned.

As far as P2 v SxS is concerned, then speaking as someone with no investment in either system to date, all the technical pros do seem to be with SxS. When I heard this put directly to a Panasonic rep at a seminar, he didn't dispute it, but did point out in P2s favour that it was an established product, was available then, whereas SxS was still 6 months away at the time. True enough, but it does raise longer term question marks: SxS has the edge in max speed (about 20%), price (about 10-15% street price), but most importantly is based on current technology and is natively supported in most modern laptops. The question is whether commercial considerations permit Panasonic to use it in future.

 

[Barry_Green]

all the technical pros do seem to be with SxS.

Really? Read the manual -- they say that if you eject a card while it's being written to, or read from, or if the power goes out while a card is being accessed, you can lose *everything* on that card. Sorry, that's just profoundly unacceptable. At least with the P2 cards, the worst you'll ever lose is two seconds.

 

[mikkowilson]

[...]if the power goes out while a card is being accessed, you can lose *everything* on that card.

Ouch...


- Mikko

 

[Stevet]

Ejecting the memory is not a smart thing during sessions, but it could accidently happen.
Well, for that matter, accidently dropping the camera could also happen.
Fortunately, I have not had this happen, but have heard stories from others.

 

[Noel Evans]

Really? Read the manual -- they say that if you eject a card while it's being written to, or read from, or if the power goes out while a card is being accessed, you can lose *everything* on that card. Sorry, that's just profoundly unacceptable. At least with the P2 cards, the worst you'll ever lose is two seconds.

I agree with you Barry its a big RISK and probably should have been addressed, though for me this would hardly be a deal killer if I was considering this cam insofar as - this has never happened to me recording to p2. One thing Panasonic did right on the HVX and HPX was cover the p2 cards to ensure they couldnt accidently be popped out whilst shooting - Sony seems to employ something similar.


What worries me is the phenomenon of rolling shutter wobble. Ive seen it on a HV20 myself and some footage posted here. Can these companies physically employ a better or improved system to remove it? Sorry I am just not fully understanding it. I mean what is the likelyhood that it may effect someone while usint the cam for proffessional purposes?

Then Im on the next question. I saw mention that RED was rolling shutter, can then RED be effected by the same thing? And going back, have they employed a better system to prevent it? My point is RED (and maybe this is best suited to another thread ) is going to be used on higher $ productions where a rolling shutter wobble event could cost much more to reshoot. Surely its been addressed here by Sony and RED right?