Gain -3db vs 0db?

[Cal Bickford]

I've heard people say that a -3db gain setting on the A1 really represents the actual signal power and that 0db on A1 is really +3db (relative to other manufacturers) and this is one reason the A1 does well in low light. Does anybody know if this is true? When you set A1 to 0db gain is the camera actually amplifying the signal??? On a related note, does increasing gain raise the value of the gamma curve uniformaly across its entire range, or is it weighted towards the black end?

[Bill Pryor]

Zero db seems to be zero db to me, with no apparent gain. Minus 3 is -3.

[Nick Weeks]

I second Bill's response

[Tom Roper]

Well intuitively there's got to be gain at zero compared to minus 3 or you couldn't call it "gain".

Minus 3 *is* the true amplifier zero for the same reason you can't pull a sled by pushing on a rope.

When you set the camera to 0 db you ARE amplifying the gain on the Canon A1. People are probably saying it's non-amplified because they don't see noise. Fair enough.

But wait...I forget what they call it, but my Canon 5D dslr camera has some sort of a so called minus gain. No come to think about it, it's an ISO 50 setting as opposed to the default 100 ISO setting. You have to go deep into the menu settings to make it available. But the tradeoff at that lower (ISO 50) setting is that to get lower noise, you lose a bit of dynamic range. I wonder if similar logic applies to the A1?

[Nick Weeks]

Well, I see what you're saying Tom, about the gain, makes complete sense. But I've heard something about dropping the power to the CCDs below their "norm" to produce the -3db gain, but I could be wrong. Supposedly the 0db is the normal operating power for the CCDs and the db rating (+ or -) is relative to the power applied to the CCD at 0db of gain.

But compared to other manufactures might be a different issue. I've seen some posts about the 12db gain different from the Sony 12db gain (can't remember if it was better or worse), and I've also heard some people say the A1 is nosier at 0db than other cams. I personally think it has zero noise (or very very little) at 0db and -3db is just a darker picture similar to a weak ND filter. I guess its all in who manufactures the CCDs and what the power requirements of each are.

All I know is, the picture looks great at 0db of gain without noise, and I can open my aperture wider on -3db combined with ND and a polarizer for a really nice outdoor picture! I think this camera has great dynamic range at 0db and -3db.

[Richard Hunter]

Originally Posted by Tom Roper

Well intuitively there's got to be gain at zero compared to minus 3 or you couldn't call it "gain".

Minus 3 *is* the true amplifier zero for the same reason you can't pull a sled by pushing on a rope.

The analogue signal coming from the CCDs can be amplified (+ gain in dB) or attenuated (- gain in dB). Zero dB should represent unity gain, or x1, which is neither amplification nor attenuation.

Because the A1 seems to be a bit noisier than expected, some people have speculated that the default zero dB is actually amplifying the signal and that you have to go to -3dB on the A1 to get the true x1 setting. Of course there is no practical way to verify this without input from Canon.

Don't know about the pushing on a rope analogy. -3 can't be the true amplifier zero for such a reason. It would be easy to put heavier attenuation settings in the design, e.g. -6dB, -10dB, etc, if there was a good reason to do so. This would not change the reference level of 0dB which is (or should be) unity gain.

Richard

[Tom Roper]

Originally Posted by Richard Hunter

The analogue signal coming from the CCDs can be amplified (+ gain in dB) or attenuated (- gain in dB). Zero dB should represent unity gain, or x1, which is neither amplification nor attenuation.

Because the A1 seems to be a bit noisier than expected, some people have speculated that the default zero dB is actually amplifying the signal and that you have to go to -3dB on the A1 to get the true x1 setting. Of course there is no practical way to verify this without input from Canon.

Don't know about the pushing on a rope analogy. -3 can't be the true amplifier zero for such a reason. It would be easy to put heavier attenuation settings in the design, e.g. -6dB, -10dB, etc, if there was a good reason to do so. This would not change the reference level of 0dB which is (or should be) unity gain.

Richard

Thank you and you're exactly right about unity gain, amplification and attenuation. Thanks for the excellent explanation. As a board designer for many op-amp circuits, I am ashamed for the memory lapse. I'm scheduling an appointment with my physician.

And I was just so proud of the "pushing on a rope analogy" (sigh) :(

[Cal Bickford]

so if you guys aren't in need og gain do you usually leave it at 0 or -3?

[Richard Hunter]

Hi Cal. I usually use -3dB, unless I need more exposure or (occasionally) more depth of field.

Richard

[Mark Fry]

Because I sometimes need a bit of extra gain, I have set the three-possition gain switch to 0, +6 and +12, so that I don't have to go into the menu when I'm in a hurry. Switch stays at 0 for perhaps 90% of the time, though. If all gain settings were available on a dial, maybe I'd use -3 more often.