4-cam comparison -- initial response

[Mr. E. Trane]

Hello,

Let me add my thanks to Barry and everyone for the test. I can’t wait for the final results.

Now… Don’t we kind of owe it to these guys to let them finish the comparison?

Barry noted over and over that this was only a LIVE footage comparison. Personally I think the recorded format’s gunna make a HUGE difference for a lot of people who may not what to be tethered to a laptop every time they want to shoot something. And then again maybe not, but the point is the jury’s still kind of out for now, eh?

 

[ullanta]

Your logic is going to fail here, because you are locking onto your values as the standard. IF lowest price for a camera to shoot HD is the key parameter, no question Sony wins. Is that the only possible reason for picking a camera? NO!!!

Nah, I think Sanyo has 'em all beat. And it fits in yer pocket, too!

 

[Barry_Green]

My only big surprise and hiccup is that if the jvc is so close in image to the hvx then isnt that a great arguement for HDV?

No, because we didn't even look at playback yet. This was all live camera head stuff. Remember that I said we haven't looked at codec performance at all, and in the report I said that could change the results a little, or even a lot -- it all remains to be seen.

 

[Barry_Green]

Live outputs are not exactly much use to most of the people on this board

I agree, I think the playback tests will be more important since we all have to record on something at some point.

As a matter of interest, was the live output from the HVX (when you compare resolution) in 720p or 1080p?

We did both.

Also I have heard that the camera is more noisy in 1080 than 720. Does this seem to be the case to you?

No, although we didn't do a side-by-side looking specifically for that, I don't think it looked any different.

If it all originates from a 1080p capture it seems odd if there is more noise in 1080.

Well, I guess there could be a reason, if the image is getting scaled down to 720 then some noise might be averaged out... I'll have to test that notion side-by-side.

The sensitivity is a big suprise. It goes to show that all the talk of a native 1440x1080 block being far less sensitive are not enitrely true.

Yep -- two big surprises there. I thought the HVX might be as many as two stops faster, and I thought it would hold a significant edge in dynamic range. Instead it was a little under 1 stop faster, with directly comparable dynamic range.

Oh and whatever results you manage to get; could you please make some available to PC users too, even if they are stills?

Stills would be about as useful as video, since we really didn't shoot much of anything moving. Well, there was a little tiny bit of motion on the res charts so we could discern the true resolution past the aliasing, but that's not much motion, so a still would probably serve as well.

 

[Barry_Green]

It may be a bit early, but can you give un as idea to what lattitue the HVX was handling?

Jay made a scene that had about 12 stops spread throughout it, and he's going to make an overlay that shows what stop was on each section. Then we'll post some stills and you can see what stops got resolved where.

Im a bit confused with you saying the HVX was 2/3 stop faster than any other camera, but you say the JVC handled highlights better - or is this a totally different thing?

Different thing entirely. When looking at the same scene, the HVX would be at maybe f/5.6, the HD100 at maybe f/4. So the HVX could get comparable brightness in half the light.

Highlight handling is unrelated to sensitivity; the HD100 had a nicer rolloff and the ability to retain some detail at the higher end before blowing out to white.

 

[Ernest_Acosta]

I think Adam Wilt's testing will clear a lot of things up. I've been checking his site but nothing yet. The one thing that maybe should have been put on the HVX would have been a 4:4:4 uncompressed output. Huge files but it may have separated it from the other cameras, albeit the variable framerate does put it in a league of its own. May be a hint to what should be in an HVX200A. I think the rule of thumb is to shoot 720p because on another forum someone claimed that the resolution of the HVX is a little more than 720p and the reason that there is more noise in the 1080i stream is that the chip size cannot resolve it and that it is upres from 720 to 1080.

 

[joelnet]

Well, I guess there could be a reason, if the image is getting scaled down to 720 then some noise might be averaged out... I'll have to test that notion side-by-side.

That's a really good point Barry. If noise is the same size when captured at 1080 as when captured at 720 then shrinking a 1080 image to 720 should shrink the noise, right?

 

[Cees Mutsaers]

Good point Brian, suddenly there is not talk anymore of the motion artifacts of the HDV format when moving the cam or when there is a lot ofmovement in the shot? Shouldn't the DVCpro codec be superior in this area as steated before in many threads????

Forgive my question, but shouldn't the HVX be noticeably sharper than any of the other cameras with pans/zooms/dolly/jib shots? Due to the DVCPROHD vs HDV compression.
Obviously, I have found with static shots the HDV looks great but once you hand hold or start moving the camera it looks very soft...
Any feedback or comments?

Brian Broz

 

[holyzoo]

Also Steve Jobs did not mention HVX at MacWorld San Francisco at all. It was surprise compared with the last year's Sony FX1 presentation. He talked about Sony's 16 hundred HDR-HC1.

What the hell does this have to do with the 4-cam comparison?? Start a new thread and spare us the embedded large images as well.

 

[pastywhiteboy]

To me, true progressive, flexibility (many modes of operation), VariCam, NO Dropouts, DVCPRO50 support, and a NON-MPEG compression are key points. Am I willing to spend 2x to 3x over the Sony Camera for a system (Camera + P2 Cards) that meets my needs and wishes? YES!!!

Bob, are you crazy?! The Sony workflow is cleeeaaarly better than the HVX's. I mean, all you have to do is fill up a tape with a bunch of takes that you don't need, then load them up in real time, then convert them to an intermediary format to edit the super easy to work with mpeg-2, then edit, render, then re-render for approximately 6 days using magic bullet to achieve decent 24p. Clearly that is the best workflow for everyone!

And since Sanjin is an owner of the FX1, and compared tests with secondhand footage from 3 other cams, his conclusion of the FX1 being the best value is ever so apparently accurate.

Ok, so that was a ridiculous amount of sarcasm. But seriously, I too own a couple FX1's, but I'm all in favor of simplifying my workflow, working with better color sampling, an easier to work with codec, and let's not forget the indispensable benefit of no more tape errors. The FX1 may be the best value for a consumer looking to do HD, but as a professional, every shoot I do with the FX1 honestly makes me nervous.

-pastywhiteboy

 

[jbeale]

Sony FX1 edge sharpening

Sony FX1 edge sharpening

I just did a quick-and-dirty test of my Sony FX1 resolution yesterday.

http://www.bealecorner.com/fx1/res1/index.html

Like you, I was amazed by just how out-of-focus the image appears at "0" sharpening. It really looks misfocused, but it was at the exact same focus point as the image that looks crisp at sharp+14. It's as if the sharpening control was creating resolution from nothing.

Seems like you can trade off native light sensitivity for native noise performance to some degree, just by boosting the gain in post. (At least if you're not just shooting raw footage- pretty much everything I do has some post color correction.)

 

[sanjin]

You do not need Magic Bullet, what you need is Shake and FCP 5 to achieve any task in postproduction with FX1 footage, second hand or professional. And for all people that expected that HVX, HD100 and XLH1 will solve their HD frustration I am giving them once more to download my 4 cam test footage that was already downloaded more than 12.000 times. Now it is again on my home server that is a bit slower than my official pro web server.

For all that could not believe that 3K doing the same job as 11K I wiil now start again my 4 cam footage in 720p for download, 41.7 MB, compression H264, (15 users simultaneously) at

ftp://84.112.86.213/Library/FTPServer/FTPRoot/Movies/4CamTest_H264.mov

Download is open for the next 18 hours.


SJUSER

 

[Isaac_Brody]

4 cam test footage that was already downloaded more than 12.000 times. Now it is again on my home server that is a bit slower than my official pro web server.

For all that could not believe that 3K doing the same job as 11K I wiil now start again my 4 cam footage in 720p for download (15 users simultaneously) at

ftp://84.112.86.213/Library/FTPServer/FTPRoot/Movies/4CamTest_H264.mov

Download is open for the next 18 hours.


SJUSER

You mean that footage that was taken from a bunch of different sources none of which you shot personally and none of which were shot at the same time? No thanks. :huh:

 

[sanjin]

Isaac Brody plays a clever boy!!! I did FX1E footage and all others were chosen by "A CHANCE OPERATION" if you would know what it is. To get more about it look for John Cage writings.

SJUSER

 

[Isaac_Brody]

A chance operation? You mean you chanced to download the footage and host it together under the guise of being your own comparison shoot? I chanced to post this reply.

 

[pastywhiteboy]

You do not need Magic Bullet, what you need is Shake and FCP 5 to achieve any task in postproduction with FX1 footage, second hand or professional. And for all people that expected that HVX, HD100 and XLH1 will solve their HD frustration I am giving them once more to download my 4 cam test footage that was already downloaded more than 12.000 times. Now it is again on my home server that is a bit slower than my official pro web server.

For all that could not believe that 3K doing the same job as 11K I wiil now start again my 4 cam footage in 720p for download, 41.7 MB, compression H264, (15 users simultaneously) at

ftp://84.112.86.213/Library/FTPServer/FTPRoot/Movies/4CamTest_H264.mov

Download is open for the next 18 hours.


SJUSER

You're missing the point, Sanjin. The point is that a solid state format, whether you choose to use HD or not, will improve your workflow significantly and a level of reliability that tape will never be able to achieve. No one is doubting that the FX1 has a good price, but whether it's the best value for professional use?

 

[xray]

Bob, are you crazy?! The Sony workflow is cleeeaaarly better than the HVX's. I mean, all you have to do is fill up a tape with a bunch of takes that you don't need, then load them up in real time, then convert them to an intermediary format to edit the super easy to work with mpeg-2, then edit, render, then re-render for approximately 6 days using magic bullet to achieve decent 24p. Clearly that is the best workflow for everyone!

Zenn.....it takes time to ...breath......think...dream and to make choices...But IT based flow is way faster but do you always need that? For ENG work it does but for Indyfilm? You can give more time to writing...With tape you always have something in the back. But seriously, each workflow has its advantages, so has tape and so has P2 (with more than 4GB/card). Not digitizing is a great time saver. But always a backup tape can be too!

 

[The Machinist]

A chance operation? You mean you chanced to download the footage and host it together under the guise of being your own comparison shoot? I chanced to post this reply.

I chanced to laugh at Brody.

The irreverent Boston wit i miss so much.

 

[Malcolm Wright]

And I chance to wallow in the hilarity of it all.

Seriously SJ, you seem incapable of factoring in anything but price and the results of preliminary image comparisons which don't take into account the adulteration of the raw data after recording compression.

Like everyone else has said, there are SO many other factors to take in to account that your appraisal can only seem comical to us. A belly laugh doesn't necessarily mean to be mean... it just happens when something is genuinely funny.

Barry and all those who participated: thanks for doing all this! It seems we all agree on one thing here at least - that you folks are doing an awesome and appreciated job of helping us make informed decisions.

I'll chime in at this stage with so many others to say it seems work flow and bells and whistles seem to be the differentiating factors here, and for my purposes, the HVX remains far ahead in this respect. It would still do so if it had faired somewhat less well in the image comparison tests: the other factors are so compelling for my needs. The fact that it delivers comparable image quality, and in some respects superior image quality relative to its competitors is all bonus

But I await the next phase of conclusions from you folks!

Thanks again!

M.

 

[Emanuel]

Well...peace to all! :kiss: