Approximate ASA of the A1U/HC1

The quick answer to that is "it depends what you set the gain to", but I doubt that's what you want to hear. It's not as frivolous an answer as it may at first sound. On a video camera, adding gain has the effect of increasing the effective ISO rating and degrading the image via noise.

If all cameras had the same S/N level at 0dB gain, it would make sense to compare ISO ratings at 0dB in each case to get a measure of their relative sensitivities. Unfortunately their 0dB S/N figures vary model to model, and one camera may look poor with only 6dB in, another may look better with 12dB.

There's a lot to be said for how digital still cameras manage this. You can dial their ISO rating to whatever you want, low settings obviously being cleaner than high. To compare two cameras you'd then set both to the same settings, compare images of the same scene and the cleanest pictures indicate the most sensitive camera, though they are both set to the same ISO.

ummmm............I hope that's not what I think it is.............

:eek:

Nope. It's not my version of Saving Ryan's Privates.

:D

Good point. I was hoping that someone might have done a couple of tests with the camera such as with it's 0dB gain and perhaps with a 2 or 3dB, as anything higher would likely be too noisy with this camera.

My informal tests give about 200ASA. But, as Infocus says, it all depends on the gain you set it to and the noise level you're prepared to accept. The Sony HDWs are all about 320ASA and have pixels 5micorns apart, while the A1/HC1 has pixels 2.5 microns apart (and therefore 1/4 the area, two stops "slower") but the camera has at least 6dB more gain (and therefore noise) than the HDWs.

This is a wide ball-park figure, and I wouldn't go to court with it.

Speaking of gain, the A1 starts adding gain at an exposure setting of 19 at 3dB and then goes up in increments of 3dB after that.

Thanks so much again you guys!

The A1 exposure control is really subtle. It starts at lens wide open/+15dB gain, steps down 3dB at a time until it gets to 0dB, then stops down half a stop at a time until it gets to F/4, then leaves it at F/4 and pulls in a 3-stop neutral in half stops until it's fully in, and only then does it start stopping down smaller than F/4. I found this out the obvious way, I recorded a few seconds at each "exposure" position, and then played back with the metadata showing, there was a 3-stop gap at F/4. So I peered into the lens with a little torch and saw the filter moving across the iris. It's only a single filter, it moves across gradually, Nice touch.

I know the ND is slowly crossing the light path Alan, and that of course it's way out of focus inside the zoom lens, but does it come up from the bottom of the camera and ND the 'sky' first? For lots of footage you must be shooting through the cut edge of the tiny filter. Sounds horrible in theory, but is fine in practice.

Your exposure tests were obviously done in the auto mode. So did you lock down the shutter speed to do this, or does the A1 cling to 1/50th sec even in the auto mode? The PDX10 didn't - it raised the shutter speed to alarming (and CCD smear) levels in the auto exposure mode.

tom.

No, done in Manual, so that I could say into the mic exactly what condition I'd set it to. It hangs on to 1/50 until it's run out of everything else.

It moves the ND in from one side. There's no problem with this, and it becomes obvious when you think of how the lens works:.......

The front part of the lens focuses the image such that the light is traveling in parallel lines at it passes through the iris, the back part of the lens converges this onto the image plane. Another way to look at it is to consider that every pixel travels through every part of the iris plane, the lens spreads each pixel out to cover the entire diameter of the (fully open) iris. The iris, when stopped down, does not affect the image in any way other than reducing the amount of light for each and every pixel. So, the shape of the iris aperture is irrelevant (you'll see all sort of odd shapes of iris in cameras, from two-blade curves right up to 12 or 16 blade straight, it matters not, except that you see the shape of the iris on extremely defocused highlights (e.g. car headlamps). So, the ND is moved across the light path in or very near the iris, so it affects the entire image, irrespective of how much of the iris aperture it covers.

Funny, I was going to mention the ND filters coming into play at F/4 myself. I tried to put a little table up showing the exposures settings, F stops and gain, but when it didn't work when posted, I forget to mention the ND blades... Interesting little innovation, no?

Not really an innovation, I'd be surprised if other cameras didn't do it as well. It shows that Sony understand the optics well enough.

The auto ND in the light path has been around for years. My Canon 310 Super-8 utilised the same idea back in the early 70s.

Quite so :) The problem we have here is that people don't seem to understand that it needs only one thick neutral, gradually slid in place, it doesn't need lots of thin ones.

'it needs only one thick neutral, gradually slid in place, it doesn't need lots of thin ones' - quite so Alan, but Sony's PDX10 (TRV950) did indeed use three little slivers of ND as far as I could see. The tiny 1"/5 chips allowed f/4.5 to be used, then turning the 'iris' wheel resulted in stacking the filters one on top of another.

Agreed, and the A1 does it all in one. It takes time for good ideas to get accepted :)

Incidentally, I doubt they "stacked" as such, because that would change the length of the optical path. I suspect that each filter is a complete disc, but with the neutral occupying about 180 degrees. That makes the glass thickness constant. It's far less important to do it this way in very small cameras though.