Discussion of dv.com results

Lenilenapi said:
First- how did they record the uncompressed images from the Sony, JVC & Canon. Canon I know has SDI, but I thought the others had to go HDV.
Click to expand...
The analog component outputs of all these cameras are uncompressed. The Sony, HVX, and HD100 were connected to an AJA analog-component-to-HDSDI converter, and that HDSDI output was fed to the Mac workstation.

Also is this a viable way of capturing from these cameras in the field?
Click to expand...
No, not really. It was probably $20,000 worth of Mac, probably more, running a Fibrechannel RAID. Not exactly protable.

On the Canon - how easy is it to record SDI, I was under the impression that it is expensive.
Click to expand...
If you want to record uncompressed, it takes about $20,000 worth of computer hardware to do it. If you don't mind compressing the uncompressed output you can use something like the CineForm Wafian -- I think that's $10k or $15K.

I thought we were going to get a comparison of RECORDED images on this. Did we or didn't we?
Click to expand...
Again, I'm not sure. I would assume that Adam included the HDV footage in his comparisons, but I don't know that it was made clear one way or other in the article.

If not , its just the begining of real comparisons, almost more interesting from an engineer's perspective than a shooters.
Click to expand...
That's definitely what this test was. We intended to do both, but we started with the engineering and ran out of time.
 
One other thing... the HVX does not LOOK like it is low resolution and in fact it looks great and only loses percieved sharpness when you A/B it will a higher rez camera. Dunno what is going on in the Panny process but as Barry seems to indicate the percieved resolution, while less than the XLH, looks like more than 540. Maybe they are doing something similar to the Canon? Dunno. The Canon in 24F is the same way....sharper to the eye than the rez chart says.

As far as the XLH color, it is better than the XL2 in the cine settings but is FLAT FLAT FLAT out of the box. It can be tweaked and IMHO looks very film like.

Again, both are great... the XL2 creamed the DVX in 16:9 resolution but it didnt stop some great stuff from being made with it...



ash =o)
 
mochouinard said:
Is barry Hiding ?? :) I want his input on the Res, is it true, or a focusing problem or something else ?
Click to expand...
Why would I hide? I was out picking up the new Mac, I got back and logged on and here's this thread going on -- sorry if I can't answer everyone's post within five seconds, but sometimes I do have to step away! :)

The resolution question is puzzling, which is why Adam and I both want to see a second unit tested. We've seen better results from Kaku shooting a home-printed chart. Neither Adam nor I were convinced that what we were seeing was "right", so we are both wanting to see a second unit.

And, in subjective numbers, there's room for interpretation -- at what point do you draw the line? Adam calls the HVX at 540, but I can show you charts that clearly show it resolving something down to 630, and perhaps as deep as 700. That's where the whole MTF question comes in.
 
Barry,

I'm really surprised that the XLH1 wasn't set up at all... In the end, for me, it's sort of a moot point, because right now I couldn't record it in a format other than HDV, and I don't want to do that. But still....

Matt
 
Congratulations on the Mac. Once you get through your adjustment curve, I'm sure you'll be thrilled with it.

Matt
 
mochouinard said:
What I mean by that is does Barry really think that 540 is it res, or he also have reservation about the outcome. If he tell me, this is as good as it get, than I will know.
Click to expand...
Nobody felt that it was "as good as it gets" -- we thought it looked curious and incorrect. I expect more from it than that, and I want to see a second unit tested to know for sure.
 
PappasArts said:
...it improves the XLH1. H1 out of the box is like a f900 out of the box, if you have ever seen a stock F900, it doesn't do justice to what a fully tweaked F900 can look like. Same goes for the H1.
Click to expand...
What's this? A new advertisement method? :laugh: I know that I haven't yet XL-H1 hands-on approach but I already worked with a F900 and regarding Wilt's comment and Canon's pics (yours including -- the best that web can provide) and footage, I can say that statement is mere marketing. And I and many others, we are tired of propaganda rather...

:lipsrseal
 
Emanuel,

I can't speak for anyone else, but I certainly don't engage in propaganda. I can't speak for the XLH1 but assuming it's like the XL2, Pappas and Ash are right. The camera ships flat. Did you see the lists of settings that they never adjusted? Certainly you can understand that setting up a camera properly will make a difference in, not resolution, but rather how the image (color, noise, etc.) looks?

Matt
 
I think he only meant that, like an F900, the XLH must be tweaked to get the best image. I have only used an F900 a couple times and I can tell you that in the flat settings everything has sort of a flat pastel like look... XLH is similar in THAT regard.


ash =o)
 
Barry_Green said:
I had originally said that E-E (uncompressed recording) would be of much less interest than recorded format, since the users of these products are going to be using them to actually record footage. I know Adam was very interested in reviewing the live footage. We did have the ability to record uncompressed and did record uncompressed. I don't know how much of the HDV and DVCPRO-HD-recorded footage Adam used in his article. I sent him the footage, I will assume that he used the recorded HDV/DV100 footage in reaching his conclusions.
Click to expand...

I just want to clear this up a bit, since it keeps coming up. According to Wilt:


"- Numbers are for 10-bit uncompressed recordings at 1920x1080 resolution. Video recorded to the camera's native recording format may show lower horizontal resolution numbers due to subsampling: HDCAM, HDV, and 50i DVCPROHD top out around 810 TVL/ph; 60i DVCPROHD goes to 720 TVl/ph....

"- Numbers are for 10-bit uncompressed recordings at 1280x720 resolution. Video recorded to the camera's native recording format may show lower horizontal resolution numbers due to subsampling: HDV can capture up to 720 TVl/ph; 720p DVCPROHD goes to 540 TVl/ph....

"I urge you to look at the source materials rather than taking my numbers as gospel; not everyone agrees with my methodology for determining resolution. I've made two DV clips (assuming that everyone has a computer that can play back DV using QuickTime Player or other media programs) that contain clips from all the 1080 cams and all the 720 cams side by side. I've grabbed resolution trumpets from the uncompressed 10-bit clips and reproduced them 1:1 in the DV clips (field order "None", so there's no interlace-induced blurring; the High Quality bit is set for clean desktop playback); the detail in these DV clips is equal to the originals..."

So, everything seems to be based on the uncompressed footage, with no mention of having viewed the HDV footage. The only mention of camera-recorded image quality is:

"P2 playback of 1080i from the Panasonic matched what we saw in E-E mode, but 720p playback was a bit softer and yet (to me) more pleasing, as the 960-pixel sampling of 720p DVCPROHD stepped on some of the near-Nyquist aliasing of the camera's sensors whereas 1080i DVCPROHD's 1280-pixel sampling let it pass. At 960 pixels per scanline, limiting resolution is 540 TVl/ph; at 1280, it's 720 TVl/ph."

So, as far as I can tell, we've seen no results on the HDV codec at all...

-Barry
 
MattC said:
They didn't do anything! They didn't adjust gain, they didn't adjust knee compression, they didn't adjust corring, they didn't adjust the blacks, they didn't utilize either noise reduction setting, they didn't use either of the cine gamma settings. And I haven't even touched on the more complex color options.....

They did BUPKIS!

LMFAO!!!!
Click to expand...
That's not true Matt. We did go into the menus, we did experiment with the different gammas, we did do some black stretch and things like that. We didn't know about the "NR1" and "NR2" options until after the testing was complete, but the manual said that those don't affect the live image anyway, they affect the recorded MPEG2 footage.

We did what we could, with the collective brainpower that was there. What we lacked was the experience of an owner/shooter who had already conducted his own testing and knew what the results were, so we had to fumble through it without benefit of that experience. But we did go into the menus and did attempt to make it look its best based on general cameraman experience and finding elements in the menus and experimenting with them and reviewing the results on the waveform and the monitor.

What Adam is acknowledging is that we didn't know enough about the Canon to truly deliver the best results. But you're making it sound like we pulled it out, stuck it on a tripod and said "well, there it is." That's not accurate at all.
 
MattC said:
Emanuel,

I can't speak for anyone else, but I certainly don't engage in propaganda. I can't speak for the XLH1 but assuming it's like the XL2, Pappas and Ash are right. The camera ships flat. Did you see the lists of settings that they never adjusted? Certainly you can understand that setting up a camera properly will make a difference in, not resolution, but rather how the image (color, noise, etc.) looks?

Matt
Click to expand...

Agreed.

All this talk about the Canon being flat. Nobody adjusted the colors. I'm sure this has been improved over the XL2. But we've yet to see it in action because nobody's given us the grabs or footage to examine. I'm not doubting the HVX is going to be better right out of the box, said that before.

As far as MOTION is concerned, I'd love to see some more footage that deals with it. Barlow's footage was good and it had motion in it, and personally I didn't see anything wrong with it.
 
OK! Thx for the enlightenment, if you say so. Because is you that are saying...the source is very important and you are a reliable one.

EDIT -- Well...this it came fast today! I just speak for the first but it's true that it can be about any of you. Well, I'm just saying what I'm saying 'cause this bilateral struggle is not democratic :) -- I mean, fair without the other two contenders...and sometimes it's only a question of which brand work harder or not...that is which brand has the best marketing department...but we are here 'cause the motion pictures, at least, I think so...
 
Kholi said:
Hey, Matt--

I can't remember the company, but Barry posted a thread regarding a program that's supposed to be in development for the Mac systems that's like DVrack.
Click to expand...
That would be redrock micro's "Revolution" program.
 
Barry_Green said:
That's not true Matt. We did go into the menus, we did experiment with the different gammas, we did do some black stretch and things like that. We didn't know about the "NR1" and "NR2" options until after the testing was complete, but the manual said that those don't affect the live image anyway, they affect the recorded MPEG2 footage.

We did what we could, with the collective brainpower that was there. What we lacked was the experience of an owner/shooter who had already conducted his own testing and knew what the results were, so we had to fumble through it without benefit of that experience. But we did go into the menus and did attempt to make it look its best based on general cameraman experience and finding elements in the menus and experimenting with them and reviewing the results on the waveform and the monitor.

What Adam is acknowledging is that we didn't know enough about the Canon to truly deliver the best results. But you're making it sound like we pulled it out, stuck it on a tripod and said "well, there it is." That's not accurate at all.
Click to expand...

That's what I thought I read. You guys did try your best with the camera but you were all admitting, even Rawls, that you aren't professionals at all.

Those who think that's a cop out are wrong. That's just stating "We knew nothing about this camera but we gave it a shot."

HVX looks great still. The color issues are what attracted me to the DVX and ultimately this forum.
 
Barry_Green said:
That would be redrock micro's "Revolution" program.
Click to expand...

Thanks, Barry!

That's it. People should start checking that out no matter what camera you get.

DVrack is an awesome program. If the Redrock guys do one up just as well, you've got one powerful tool.
 
So it sounds like the resolution of the HVX200 is still up in the air (until another test can be conducted). Can someone else with the HVX200 make that happen? Perhaps someone with testing knowledge could instruct the HVX200 owner on how to properly perform the test. If the results are similar, we can assume that the first test was accurate.

As many of you said, resolution isn't everything so how about this...

One thing you guys did at the test was to shoot the girl sitting at the table with every camera. No other clips on the internet show each one of these cameras shooting the same exact scene (using the same lighting, etc...). It would be awesome if we could get a short clip from each camera of the girl sitting at the table. This would be a better way to judge how we feel about the look of the camera (not a perfect test, but something new to examine).

It would be great if the clips were available without knowing which camera they came from (sort of a coke vs pepsi test). In a later post, someone could reveal which camera shot each clip. Anyone up for doing that? The results could be interesting.
 
Sorry Barry, I came to that from this paragraph by Adam:

Here's where we missed an opportunity: the Canon XL H1 has a variety of menu settings we failed to exploit. Its knee was set to "middle" whereas "low" might have bought us a smidgen more highlight detail. The Canon also has two noise reduction options (NR1, which appears to perform recursive or "3D" noise reduction, and NR2, which is said to be like "applying the skin detail function over the entire picture") as well as coring (which reduces enhancement noise in the shadows), all of which we neglected to engage, so the Canon's image was a tad noisier than it otherwise might have been. The Canon was used with normal gamma and no black stretch.
Click to expand...

It just seems like none of the settings which could have had a positive affect on the image were utilized (along with killing the sharpnes, dropping the master pedestal way down, etc.).
 
You must log in or register to reply here.
Back
Top