Advice Please On Lunar / Planetary Imaging Cameras?

[Garethr]

James,

Your right it was the older neximage

I think my rant was about the quality of these cheaper cams and what they produce as an end result. A lot of hard work to get anything near to a good image. The zwo seem to be able to produce a workable image without too much mucking about

Gareth

[jambouk]

I have to say, I'd think twice about buying a ZWO again.

I've had two which just stopped working. No idea why. They are both in China currently being looked at. When they worked they worked well, but like other USB imaging devices mine have always had a tendancy to freeze up, and drop out. I'm not saying they are worse than other devices, just that they are not perfect.

James

[JamesF]

As chance would have it last night I was diddling about trying to make oacapture work with a Neximage Burst.  Given only the choice of Neximage cameras and going by the specs, that's probably the one I'd say was most likely to be suitable for planetary imaging.  I believe internally this is an Imaging Source DFK42 (or a DMK42 for the mono version).

However, it appears that the data is transmitted from the camera in YUY2 (or UYUV) format, which is lossy, and that the frames are interlaced.  I'd suggest that this is not ideal for a planetary imaging camera.  I'd be inclined to reject it as an option on those counts.  I have no idea if the Neximage 5 is the same, but I'd not be entirely surprised if it were.

James

[Scorpius]

As chance would have it last night I was diddling about trying to make oacapture work with a Neximage Burst.  Given only the choice of Neximage cameras and going by the specs, that's probably the one I'd say was most likely to be suitable for planetary imaging.  I believe internally this is an Imaging Source DFK42 (or a DMK42 for the mono version).

However, it appears that the data is transmitted from the camera in YUY2 (or UYUV) format, which is lossy, and that the frames are interlaced.  I'd suggest that this is not ideal for a planetary imaging camera.  I'd be inclined to reject it as an option on those counts.  I have no idea if the Neximage 5 is the same, but I'd not be entirely surprised if it were.

James

James - Obviously the Neximage 5 isn't at the top of most folk's list as a planetary webcam. However, to say it's not ideal for that application would seem to be a stretch since that's what it's specifically designed to do. http://www.celestron.com/browse-shop/astronomy/astroimaging-cameras/neximage-5-solar-system-imager-(5mp)

A new planetary camera like the ZWO isn't really in my budget right now so once the planets are in a better position, and I get some extra time, I'll have another go just to see what else I can do with the Neximage 5. I've always been a believer in the philsophy that the "proof is in the pudding" so to speak and Mars is next on my list. Therefore the ongoing saga of the poor ole Neximage 5 will surely rear its ugly head once more... 

[michael.h.f.wilkinson]

James - Obviously the Neximage 5 isn't at the top of most folk's list as a planetary webcam. However, to say it's not ideal for that application would seem to be a stretch since that's what it's specifically designed to do. http://www.celestron.com/browse-shop/astronomy/astroimaging-cameras/neximage-5-solar-system-imager-(5mp)

A new planetary camera like the ZWO isn't really in my budget right now so once the planets are in a better position, and I get some extra time, I'll have another go just to see what else I can do with the Neximage 5. I've always been a believer in the philsophy that the "proof is in the pudding" so to speak and Mars is next on my list. Therefore the ongoing saga of the poor ole Neximage 5 will surely rear its ugly head once more... 

It might be designed as a cheap planetary camera, but if it ups the transfer speed by including lossy compression, and uses interlacing as James says (and he knows his stuff) then the camera is less than ideal. No two ways about it. I would personally rather go for a second-hand ASI120 (MM or MC). Quite a few pop up due to the newer ASI120MM-S (or MC-S),  ASI174(MM and MC) and ASI224MC coming out.

[JamesF]

James - Obviously the Neximage 5 isn't at the top of most folk's list as a planetary webcam. However, to say it's not ideal for that application would seem to be a stretch since that's what it's specifically designed to do. http://www.celestron.com/browse-shop/astronomy/astroimaging-cameras/neximage-5-solar-system-imager-(5mp)

I don't wish to be uncharitable to Celestron as there are some pretty good guys there and they've been very helpful when I was working to get the Skyris cameras supported on OSX and Linux.  This is purely my interpretation and therefore may be entirely wrong

Celestron have a different business model from (say) ZWO, QHY, Atik etc. in terms of their cameras in that they bought in the electronics wholesale.  The guts of many (if not all) of their cameras come from Imaging Source.  They still even use Imaging Source's USB vendor ID and product IDs.  I think the Skyris range evolved out of TIS's own range of astronomy cameras, but I'd guess Celestron decided they needed a different range to offer to people who couldn't justify the cost of the Skyris cameras and "went shopping" amongst TIS's large range of existing products (they do a lot of kit designed for industrial/machine vision and microscopy) to find something suitable.  What I think they came up with is this:

The "new" Neximage (not the old SPC900-alike) is, I think one of these

I'm fairly sure the Neximage 5 is a DFK72

And the Neximage Burst are this one and this one (colour and mono respectively)

I've even heard that some of them can be flashed with TIS firmware updates.

From the specs for the TIS camera the Neximage looks like it only does compressed video formats which is not the best of news for a dedicated imaging camera.  When the SPC900 was the best thing there was going for anyone who didn't have £300 to blow it was bearable in a camera that wasn't ever designed to be an astro camera anyhow.  If it were the only game in town at a price I could afford that doesn't mean I wouldn't buy it though

I've never touched a Neximage 5, so I have no idea how it delivers data.

The Burst C I've already covered.  The mono model presumably won't do the same compression as the colour model because there's no chrominance data to start with, but I'd not be surprised if it still does the interlacing.

Looking at the range I've always thought that the Burst models might be better suited to planetary imaging than the others in the range -- higher frame rates, larger pixels, greater sensitivity, whereas the Neximage 5 might be best used for lunar and solar imaging where the lower sensitivity isn't such an issue and the small pixel size suits a focal ratio that most people will be able to reach easily with pretty much any OTA.

In the light of the fact that the data from the C has undergone lossy compression and is interlaced I'm not convinced my initial opinion was correct.

If it were my money paying for a camera for my use I don't think I would personally buy any one of the three on the basis of what I know.  If I couldn't afford anything else new I'd probably sit and watch the used market for a while to see what came up.  In no way does that mean someone else shouldn't buy one or that they won't get images out of one that they're happy with.  We all have different budgets, expectations and supplies of patience.  I'd far rather see someone getting a kick out of imaging with a camera I wouldn't recommend than not imaging at all on the basis of my opinion.  Go to it, and post your images

James

[JB80]

Just thought I'd add a couple of pics that I have taken with my ZWO ASI120m and 6" f8 reflector.

Definitely for solar system imaging it's a great camera, sure it is a bit out of date with the newer models coming out but with the results below who needs 'em.

This is a mosaic to create the full image.

And this is roughly(I probably cropped it) what a single panel will be like.

Something else worth mentioning is I'm now at the point where to get more out of my set up I am changing more of my hardware like uprgrading barlows, focusers and such.

I'm not convinced that upgrading cameras will see a big improvement with the 6" scope although I'll happily test one out to see.

I also have a DMK41 which is a great camera too but I don't think it is as good on lunar as the ASI120 and the frame rate too slow for planetary although it does a surprisingly good job.

[JamesF]

As an addendum, I'm having an absolute nightmare trying to unpick the frame data from the Burst C.  Each frame does look like one field of an interlaced image and the configuration data from the camera appears to be consistent with that, but something isn't right -- perhaps it's not interlaced or perhaps it's not YUY2.  I've got to the point where I'm wondering if the camera hardware doesn't actually return the correct information and the Windows drivers don't care because they know what they're expecting anyhow.

Or, and let's face it, this is the far more likely solution, my code is just broken in some way that I have yet to discover

James

[Garethr]

Jb80

I rest my case

Great images mate

Gareth.

Ps sorry about the rant it was late, I was tired ect but still from the heart

[CSM]

I have both the SPC900NC and the ZWO MC120-S and although the latter should be better (have not had chance to use it in hi-speed mode) I think it comes down to the seeing conditions as to the results you get.  And use AutoStakkert!2 initially, then Registax for the waveletting.

[JamesF]

I wanted to revisit this  thread just to clear up some information on the Neximage 5 and Neximage Burst cameras which I now believe is incorrect so if anyone comes across this thread in the future it may at least avoid pointing them in the wrong direction.

I've spent the last couple of weeks up to my elbows in the gory mess that is the Neximage Burst firmware.  It's not been pretty and I have come close to throwing things on occasion.  There's no public documentation that I'm aware of for either the Neximage 5 or Neximage Burst.  Neither is there for the DFK42 and DFK72 on which they are based.  Right now it appears that whilst the Celestron and Imaging Source cameras are largely the same thing, the Celestron cameras have slightly different firmware which I believe reduces the functionality available on the camera.  With the Neximage 5 at least it appears to be possible to flash the camera with the DFK72 EUVC firmware and restore the missing functionality.  It's a strange arrangement though.

In terms of working out what it's possible to do with the camera, TIS appear to have followed the UVC (USB Video Camera) spec. as far as it suited them, but then diverged from it when it didn't provide the functionality they needed.  As the UVC spec. has evolved that's meant that the TIS firmware may not have functionality present where it would be expected in a UVC camera, or it may have different functionality (and it has some extra features too).

Unless you know where the two diverge (which is hard without documentation it's very tricky to interpret what's going on.  One of the places it all goes horribly wrong is the specification of the supported frame formats.  Read in a UVC-compliant fashion this data says that colour cameras have YUY2 frames and they're half the size you'd expect from the resolution of the camera, making it look like they're interlaced.  However, TIS appear to have decided that perhaps because there was no way at the time within the UVC spec to say something like "this is a raw colour frame with a GRBG colour mask" they would instead tag the frame as YUY2 and add some extra control functionality to be able to query what the colour mask ought to be.  I guess when you see the camera claim to be YUY2 you're just supposed to know that it's actually raw colour.  I have no idea why it says the frame is half the size I know it to be though.  There's some other exceptionally strange stuff the Burst does too, such as claiming that the largest block of data it will attempt to transfer from the camera is about 5MB when the full frame is only 1.2MB long.  That's another mystery I've not got to the bottom of yet.

Anyhow, now I actually have got to the point where I have the Burst delivering images and I can display them I can confirm that it delivers raw colour frames to the user and they're not interlaced.  As TIS do appear to have a moderately consistent approach to their cameras, I'd assume the Neximage 5 is exactly the same.

I'm hoping now that I've broken the back of this work and I won't have to come back and correct this again

James