urgent advise needed: EX1 / DSR-570 intercu

[Piotr Wozniacki]

Guys, I usually shoot live music with several EX cameras, so have long-established workflow for that. But this Sunday something completely different will be happening - a concert that we just couldn't give up recording - and all our EX cameras are out and busy elsewhere. So, it will be recorded in SD, using several DSR-570's plus 2x EX (including my EX1).

As to SD, I have no problem - will probably be recording HD to the SxS and 50 Mbps 422 IMX to my nanoFlash. The other EX camera I'll probably have to downrez in post.

But not knowing anything about the DSR-570, I have no idea how to set up their paint parameters (color matrices, gamma curves, etc), so that our EXs look at least similar to them. Please advise!

Piotr

[Bob Hart]

I can't help with setting up the DSR-570. My following comments, please treat with some suspicion as I am far from practiced in this business. Please defer to the opinions of those better than me who reply subsequently.

My personal preference, if the DSR-570 is a SD camera, would be to use the EX cameras for the wide shots where detail may matter more and the DSR-570 cameras for the close-ups and tighter frames, where any lesser resolving ability will be masked by subjects being larger in the frame and by backgrounds being likely to be in softer focus.

For white balance, my personal preference would be to use the fixed options each camera offers for whatever lighting source you are using and not manually whitebalance each camera.

I would also prefer to set the EX cameras to record their normal 1920 x 1080 but others may challenge me on this.

Another personal preference would be, if you have one, to take a Chroma Dumonde chart or similar colour chart and shoot vision of this chart with all cameras. In post you might then be able to more easily adjust the colour rendition of all cameras.

If you are unable to frame the chart tightly with all cameras, zoom in as close as you can on the chart. In post, you should be able to crop the frame or zoom in digitally so that the chart is framed. This may still give you an adequate RGB waveform display to enable matching of all cameras color rendition.

[Alister Chapman]

Your going to have a hard time getting an HD camera, with all it's detail settings etc tailored to HD to match an SD camera with the typical "chunky" detail correction circuits used in the SD world. Turn off the 570's detail circuits and the pictures will look soft, SD needs detail correction by the spade full to look sharp. The two cameras have very different looks. Colours should be similar, but gamma and detail will be quite different. You are going to have to dumb down the EX's to match the 570's. These are some EX settings that I've used in the past to attempt to get close, it won't look nice in HD but the SD will be closer to a 570. Also take great care over your downconversion, the usual issues with reversed field order (Pal DVCAM has the opposite field order to HD) and uneven line dropping to get from 540 lines to 288 per field apply, issues you won't see on an LCD but anyone with a CRT TV will see in great detail.

Matrix – Cinema, Level +15

Detail Level +20, Detail Frequency -45, White Limit +35, Black limit +45

Knee, Manual, Level 90, Slope 0.

Gamma Standard 2, Gamma Level +5

Black Gamma -10

Black Level -10

[Piotr Wozniacki]

Thanks guys!

Alister, do I understand you right that the PP settings you posted are good when recording HD to SxS (of course, there is no SD option on the EX1 !!!), while feeding the nanoFlash with SD-SDI? The SDI signal would of course be set to SD and squeezed (as we're aiming at 16:9 SD)...

The reason I'm not 100% clear on your advise is that your Detail setting is as high as +20, even though for good down-conversion to SD, I believe the sharpness of the original HD picture should be lowered (here is an excerpt from your own article on the matter):

"So what can be done?

Well the best way to improve the SD down conversion is to soften the HD image before it is down converted to prevent this single pixel light to dark switch from happening"

Please confirm, or elaborate if I misunderstood :)

Piotr

[Piotr Wozniacki]

Originally Posted by Bob Hart

My personal preference, if the DSR-570 is a SD camera, would be to use the EX cameras for the wide shots where detail may matter more and the DSR-570 cameras for the close-ups and tighter frames, where any lesser resolving ability will be masked by subjects being larger in the frame and by backgrounds being likely to be in softer focus.

.

So basically, your suggestion is to shoot with an HD project/delivery in mind, Bob?

[Tim Polster]

I agree with Bob but I don't think this means to deliver in HD. I think it just means the EX might hold more detail in SD than the DSR-570, so use the EXs for the wider shots.

But, with all of the HD downconversion issues, the DSR-570 might still have the best image when viewing the final SD DVD.

Piotr, do you have access to the DSR-570s before the concert? Can you get them a day ahead and setup a monitor to match the looks of all the cameras? that is your best bet if possible.

[Piotr Wozniacki]

Thanks Tim, but unfortunately I'll only have a couple of hours before the actual performance to set everything up :) This sounds crazy, I know - but that particular event has taken us by surprise (we were not planning it at all)...

[Alister Chapman]

The detail is deliberately set very high and with the frequency set at -45 to generate very thick, pronounced detail correction edges so that when downconverted these remain several pixels wide and won't twitter or aliase. You need to do this as you don't have any control over the built in downconverter.

You either do that or turn detail off and apply sharpening in post, but it's hard to simulate detail correction in post.

Frankly, I would not use the EX1's built in downconverter, it's not very good. I would downconvert in post where you will have much more control. Ideally you still need to add a 2 or 3 pixel radius blur prior to down convert to reduce the aliasing, but even then the large, hard edge correction helps because it eliminates most of the one or two pixel wide edges that will twitter.

In perfect workflow I would shoot HD with the EX1 with detail off. Then blur the HD with a 3 pixel radius blur, then downconvert to SD, then add detail correction to the SD to bring the edge contrast up to match the DSR570's

[Piotr Wozniacki]

Originally Posted by Alister Chapman

In perfect workflow I would shoot HD with the EX1 with detail off. Then blur the HD with a 3 pixel radius blur, then downconvert to SD, then add detail correction to the SD to bring the edge contrast up to match the DSR570's

Thanks Alister; I guess I'll do just that - shoot with detail off, and have my nanoFlash record the SD version for me. Then in post, I'll downconvert the HD version (following your proposed workflow), and compare results with the IMX nanoFlash file; for the final edit I'll use the better-looking of the two.

Just an additional question: when shooting HD with Detail off rather than at +20, are your other setting values (frequency, white/black limits) still valid for your "to-be-down-converted" PP scenario?

Piotr

[Alister Chapman]

With detail off you won't have the ability to change the other detail settings. I'd still use:

Matrix – Cinema, Level +15

Knee, Manual, Level 90, Slope 0.

Gamma Standard 2, Gamma Level +5

Black Gamma -10

Black Level -10

[Piotr Wozniacki]

Originally Posted by Alister Chapman

With detail off you won't have the ability to change the other detail settings.

Oops.. of course you're right; I forgot about this obvious fact.

As to the color/gamma settings, I'll give them a try.

Thanks again!

Piotr

[Piotr Wozniacki]

Dear Alister,

One more question, if you will. I've downloaded the 570 user manual, and can see the "paint setting" are very basic for this camera. However, what is possible to tweak is the black compressing/stretching.

Hence the question: when you recommend the Black gamma and Black level to be at -10 in my EX1, do you mean it will match the factory black setting on the 570, or stretched, or compressed?

Thanks,

Piotr

PS OK gentlemen - my nanoFlash has just died on me, so no SD recording :-( .I'll have to be recording 1080p as usually, and - since the number of DSR-570s imposes SD (PAL) delivery of the project - will be downconverting in post.

Apart from gamma, color etc one thing also bits me: should I record progressive as usually, or switch to 50i? Will downconversion to SD in Vegas e easier if I have 1080/50i rather than 25p as my masters?

Piotr