How useful really are bias frames?

[lukebl]

... just wondered, as I can't tell ANY difference between my CCD images taken including bias frames and those taken without. Not even a subtle difference. So why bother?

Darks and flats are obvious. Am I missing something here? Perhaps my images are so bad that they mask any potential benefit if bias frames!

[RikM]

With my DSLR, I found that if I stacked in flats without first subtracting bias I got odd blochy gradients. It could have been a coincidence or maybe just the wrong settings, but a master bias is the easiest thing to take and update every few months so I include it by default.

[lukebl]

With my DSLR, I found that if I stacked in flats without first subtracting bias I got odd blochy gradients. It could have been a coincidence or maybe just the wrong settings, but a master bias is the easiest thing to take and update every few months so I include it by default.

As I understand it, bias frames aren't needed anyway with DSLRs due to the their mechanics, so your bias frames shouldn't have any effect. My problem is with my SXVF-H9, as conventional wisdom has it that bias frames are required for CCDs. I can't, however, see the benefits.

[robbieince]

If you are taking darks at or around the same temperature and exposure as your lights then you don't need bias frames. Your darks contain bias signal already. If you do both you are subtracting bias twice.

Regards

Rob

[ollypenrice]

I don't think conventional wisdom says take bias for CCD. The concensus on here is that the bias signal is contained in the dark frame and you don't want to remove it twice. I certainly never do.

Like many others I do make a master bias but use it only as a 'dark for flats.' This saves making dedicated darks for flats and according to the authoritative Narrowbandpaul will give an effectively identical result.

There may be some mysterious scaling of darks made possible by including bias but AstroArt 5 has removed the box in which you were invited to place bias frames in AstroArt 4. I'm not competent to comment further on that.

Olly

[lukebl]

If you are taking darks at or around the same temperature and exposure as your lights then you don't need bias frames. Your darks contain bias signal already. If you do both you are subtracting bias twice.

That's my point. Why do people take bias frames at all, then? They seem completely unneccesary, but people seem to do them all the same!

I'm only saying that bias frames have no impact on my images, so I don't see the point. Perhaps if my images were better, or with a better sensor, then I might see a difference?

[r3i]

I could be wrong on this but I think if you're using flats but not darks the bias frames come into play. You have to subtract the bias from both your lights and flats prior to dividing the lights by the flats:

(lights - bias) / (flats - bias)

It's claimed that cameras like the QHY8 don't need darks so this scenario would apply to them, but images taken with my QHY8 certainly benefit from the subtraction of dark frames.

[Jessun]

This old thread:

http://stargazerslounge.com/discussions-cameras/167498-bad-column.html

deals with a problem that a BIAS solved quite nicely. I don't wish to take darks so I apply the BIAS to the LIGHTS and the FLATS like r3i said.

So I'm stuck with them...

[Ant]

I would much rather have a camera (like the H9) that doesn't require darks but does benefit from Bias.

Taking Bias frames are a hell of a lot easier than Darks. Bias frames are so quick and easy to take I take them every time...

Cheers

Ant

[lukebl]

I would much rather have a camera (like the H9) that doesn't require darks but does benefit from Bias.

Taking Bias frames are a hell of a lot easier than Darks. Bias frames are so quick and easy to take I take them every time...

Thanks a lot gents. I have always taken darks with my H9, but if I can take bias frames instead then that would certainly save time.

[RikM]

I use DSS for stacking and it can't extract the bias info from the darks to apply to the flats so I need a separate set of flat darks or bias for that. Bias are easier to take and seem to work fine so that's what I do.

[steppenwolf]

I have the same CCD as you and I always take BIAS frames but never DARKS and the benefit is very very subtle but it is there as the LH edge and top 20% of my sensor shows a very subtle difference in sensitivity. I also have the SXVF-M25C and here the BIAS frame makes a significant contribution as the much larger pixel count results in some issues with column defects making an appearance but the BIAS resolves these very well indeed - again, I don't take DARKS with the M25C either.

[ollypenrice]

Do bias need to be renewed every time? I've read that they can be affected by the cables etc but then another guy who understands electronics said no, it's all in- camera.

I can see that on a chip like the 285 bias ought to do nicely instead of darks. I never took darks for mine.

Olly

[Psychobilly]

I tend use bias and no darks with my later canon DSLR cams... My moutn is slightly misaligned so i get a small dift oevr a few subs and coupled with an Outlier stacking method it seems to work...

Although this information is all historic as I haven't done any long duration imaging for a almost 2 years

Peter...

[RobH]

You need to subtract bias from your flats in order for them to apply properly (if you're not using flat darks).

If you're using darks, then bias subtracting your lights isn't required.

I shoot a set of 100 bias subs which are then stacked as a master bias and used for 6 months or so as I believe that your sensor response changes slowly over time.

Rob

[lukebl]

Thanks again all. So, how do I subtract bias from the flats, or stack 100 bias subs, with just DSS, AstroArt3 and Photoshop at my disposal?

[RikM]

If you load a single light frame of anything, and then load your bias frames and tell DSS to go, it will combine the bias frames into a master bias (master offset.TIFF) which you can file away for later. It normally saves automatically into the folder the bias frames came from.

When you come to process your actual image, load your lights, darks, flats and then load your master bias and DSS recognises it as a master and applies it accordingly. DSS seems to go the long way round for calibration but you end up in the right place if you loads the right sets of frames.

[ollypenrice]

To subtract bias from flats just tell whatever software that the flats are images and the bias are darks. This will make a master flat which will simplify life till you next move the camera. I try hard to leave all 'as is' at the scope so my flats last for several images. A square chip makes this easier!

Olly

[Shibby]

Sorry to bump an old thread, but I have found that my bias frames make the whole image darker (I'm using flats but not darks). I'm just wondering if this is normal or is there something wrong with my master bias?

[themos]

"Making the image darker" is entirely a function of your renderer. Most packages give you sliders to let you control which ADU value corresponds to monitor-max-dark and which to monitor-max-bright.

[Russe]

In sharpcap, how do you take BIAS frames with an SPC880? Shoot an avi and process the single pics?

[Russe]

Has anyone here managed to take FLATS with an LX modded webcam? How would you do it - which programme to use I mean...

[Freddie]

As per answer to your other post, whatever you use to take your lights.

[Russe]

DSS doesn't allow to take single pics less than a second long...

[Russe]

Sharpcap I meant...