M33 - Process A & B

[centroid]

The same image of M33 taken last night, but one is processed with DDP filtering, and the other without. I've tried to keep the 'levels' of both images the same.

The difference is quite subtle, but which do you prefer, Process A or Process B ?

ED80 + f/6.3 FR - SXVF-H9C and 10x10 min subs, and guided with the 12" LX.

Although the total image time was 100 minutes, it was captured in 'stop & go' fashion, over about 4 hours, due to the swathes of cloud that was drifting down the East Coast.

Dave

[trevboyd]

I'm not an expert by any stretch of the imagination, but to my eye, A has a nicer core.

Trev

[Ant]

I prefer B.

Can't really explain why, just looks smoother to me.

Edit: Really like the images BTW, I'd be very happy with either - M33 has some fantastic detail in it!

Ant

[MartinB]

The bottom one seems to have a bit more stretch applied overall - it just looks a bit more "pushed". I prefer the first one - the detail appears finer.

Great M33 Dave.

[barkis]

Hi. Dave. Both are much of a muchness, and may I say also both are superb images. I have to err on the side of process one, simply beacause it appears sharper and clearer in the central regions of the galaxy. But this is purely nit picking by me. I would be ecstatic if I had produced either one of these images. Well done Dave, very nice indeed.

Ron.

[Cloudgazer]

Their both nice images Dave but if I had to choose I'd go for the first one.. It just looks a little bit sharper and has a bit more detail than the second one.

Scott

[centroid]

Its amazing how we all see things differently.

I did say that the difference was subtle.

Your are correct Martin the second one (Proc has a bit more stretch applied, as this is needed to recover the image after DDP filtering is applied.

DDP filtering works well for some DSO's, especially globular clusters, and some galaxy cores (e.g. M31), but as with all filtering, there's always some other impact upon the image. Generally, this is an overall 'dimming' of the image, which then requires some adjustment of the levels, i.e. 'stretching'.

With these two M33 images, I prefer some aspects of image A, and some of image B, but if I had to make a choice, I'd probably opt for the non-ddp version A, but its very marginal.

Thanks for your comments guys.

Dave

[legion48]

Both great images Dave, I can't see much difference with my old eyes.

[CELESCOPE]

two very nice images Dave, ok the first one has it for me, find the detail from centre out to be very good, the second one, looks a bit to processed or its taken on that smooth detail, so its the first one for ole Rog. had agreat night myself have to process them yet ,

Rog

[MartinB]

Dave, my understanding of DDP is that it is simply a curve pushing up the dark detail but keeping the bright stuff the same. It effetively incorporates a levels adjustment to the black point (setting the background level) and then you can manually set the point at which the curve starts to flatten out (the mid point adjustment). It traditionally incorporates sharpening which I like to turn off. I find I have to play around with the mid point quite a bit to get what I'm happy with.

I think it became popular for working with 16 bit FITS when PS could only use 8 bit images. Got the stretching done before transferring to minimise data loss. It stayed worthwhile with tiffs since, although they are 16 bit, they work in integer format rather than floating decimal point (less data loss if stretched prior to saving as a tiff). Now there are FITS plugins for PS the case for DDP is harder to make. I have played around ad nauseum comparing DDP with a completely manual stretch in PS. I find DDP nice and simple and generally does a good job. With lots of practice with levels and curves in PS I'm beginning to find I can pull more out than I can with DDP.

Isn't the ED80 great!!

[centroid]

Its a kind of combination of gamma stretching and unsharp masking. The gamma stretch function of DDP, uses a complex algorithm to compress the dynamic range of the image. Theoretically, this allows both the bright and feint areas of the image to be displayed at the same time, in a more balanced manner.

I find that it usually performs a nice 'balancing act' between an over bright 'core' and the fainter gas and dust lanes, but it doesn't suit all images.

When I've finished the initial stacking and processing in AA4, I save two copies of the image, one back to FITS format, and the other as a 16 bit TIFF. The TIFF being the one that has the 'post processing' work carried out on it.

The ED80, yep, I reckon, for the money, it has to be the best performing little scope there is.

Does seem a bit unusual, using a £4.2k 300mm SCT, to guide a £250 80mm refractor.

Dave

Dave

[markcol]

Super M33 images Dave, personally I prefer the first edition too!

Regards

Mark