DSLR Horizontal Banding...?

[AndyUK]

Again... VERY impressive...! This has certainly opened up my eyes for using the Mak180 on some of those smaller DSO's

[Psychobilly]

They were taken in 10-13 Sept 2009... would have to check the exif data from the subs which i havent got access to at the moment... more subs would help but this wa s in an attempt to get the fainter out halo which is "out of reach" for a DSLR..... i will post that pic later...

From this older thread...

http://stargazerslounge.com/imaging-deep-sky/87410-m27-night-2-data-keeps-building.html

Billy...

[AndyUK]

Thanks again - It certainly shows what's feasible...! That must have been a little soul-destyroying to lose 2 hrs of 15min subs though... And I can't imagine what it would be like to lose one or more 30+ mins, but the results would certainly be worth it!

[Starflyer]

Hi all,

I've just caught up on this thread. I don't get any banding with an Exif temp below ~16C, it gets worse with higher temps and is awful above 22C.

It's interesting (but logical) that a higher ISO gives higher temperatures, I may try my next run at ISO 100 or 200. It's a common misconception that higher ISO settings make your camera more sensitive. A higher ISO just tells the electronics in your camera to multiply the signal that the sensor has recorded by x.

At ISO 100 the same amount of photons are collected by the sensor but with lower heat-generated noise. The image will need to be stretched more to bring out the signal that is there, but the noise should be less as the sensor is running cooler and the noise that is present isn't being multiplied by the electronics only by your controlled stretching.

This makes sense to me, just shout if I'm completely wrong but like Peter I don't like to get caught up in the maths of noise

[GrantEb]

Hi again all

@Starflyer. I have to disagree with you. I know the whole ISO thing has been going around for ages, but IMHO

When imaging with a DSLR, the noise generated by the CMOS/CCD, is due to current flow within it, which in turn generates heat.

If the ISO, was just the electronics telling the chip to multiply, surely this would be done with software? (Either before JPG output, or on your PC). A RAW file should contain everything, at whatever ISO setting.

In my limited experience, I have tried various ISO settings, and found ISO800 with the 1000d to be good, in low temps. I have tried lower ISO's on the same subjects, with much less capture, even when stretched in post processing. I think ISO1600, is a waste of time, unless very short exposures, or very low OAT.

I'm not saying this is gospel, and am open to discussion, and don't meen to be offending.

Cheers

[Starflyer]

No offence taken, quite the opposite... I don't think anyone has the definitive answer as far as DSLRs go. I think experimentation is the only way to find out. I know Psychobilly has spent a lot of time evaluating his cameras, perhaps he can chip in with his experience here.

AFAIK DSLR RAW files are manipulated quite heavily by the camera firmware prior to being made available for download, so you are correct that the image is boosted by software rather than electronics as I said in my last post. The RAW file is not what the sensor has 'seen' though.

If it was purely current running through the sensor that heated it, what do you think accounted for the 3C increase in temp that you saw at ISO 1600 compared to 800? Unless a higher current is used at higher ISO settings?

When I get time I'll run my 450 in the cooler box at different ISOs and see what the Exif temps do.

[GrantEb]

Thanks for the info about the RAW files.

IMO, the chip definately uses more power at higher ISO settings. This is where the higher temps come from, in turn producing more noise.

I know a lot say, the same amount of photons have hit the chip, but if you go back to film, it depends WHAT film is receiving the photons. If you use higher ISO film, the same amount of photons hit the film, but it is more sensitive.

So IMO, a higher ISO set chip, should be more "receiving" to photons, but at a loss to higher temps, resulting in more noise. Bring the temp down, at the same ISO, and you are winning.

Cheers

[MikeWilson]

Does all this talk about noise make anyone feel like giving film a go?

[dph1nm]

Take care when making deductions based on the EXIF temperature data. I have certianly seen it claimed that this does NOT measure the chip temperature on the 1000D, but is elsewhere buried in the camera body (to the extent that Peltier cooling the chip is not reflected in the EXIF temperature!).

NigelM

[dph1nm]

IMO, the chip definately uses more power at higher ISO settings. This is where the higher temps come from, in turn producing more noise.

Whereas this seems plausible, I have seen no evidence that higher ISO equates to higher dark current. Looking at this extensive analysis of the 300D

http://www.saratogaskies.com/articles/noise/index.html

the dark current (in electrons - or in this case ISO-normalised units) is much the same at any ISO.

BTW, I always use ISO1600 on my 1000D.

NigelM