PHD versus MaximDL?

[ncjunk]

That's interesting.

I like the fact that you don't have to calibrate with the same setup.

If you get a computer crash for some reason, it's a pain having to recalibrate, and it also takes up valuable time at the start of the session.

I reckon that I'd still calibrate when changing target, but if you're going for the same object several nights running, this would be great.

I only recalibrate if i change the orientation of the camera i.e. if i have taken it off to clean. Otherwise i only do it once and leave it. i Definately dont bother recalibrating for anither target..i am not saying this is the correct way

[afm038]

As Dave has already said you do not need to calibrate if you haven't changed things from session to session. I think that you do however need to either enter manually or have Maxim to be able to read the mounts declination position (this is my setup) so it can automatically compensate for different tracking rates at different positions in the sky. You can also use the auto flip function to take into account the meridian flip.

ATB, Andy

[buzz]

I too had issues with PHD random behaviour but I think I tracked it down.

a) I had hot pixels on my Lodestar and I didn't do darks. I'm pretty sure it locked onto a hot pixel on more than one occasion.

I started using an off axis guider and put the mirror too far into the edge of field. My stars were more like pale streaks and this too seemed to cause issues.

c) I didn't always re-calibrate PHD when I moved in DEC -

d) When I switched scopes, some of the calibration routines were a little suspect (in the cloudy nights article - it specifies a cal routine over 5% of the imaging area.)

e) I had play in the mount axis which would cause twitchy behaviour as the scope went past the vertical.

I'm now routinely achieving 0.2 arc seconds rms error with an EQ6 mount.

lastly - a thought starter - I lined up my camera with the scope axis - but if I turn it 45 degrees.... it should give 40% more resolution...I think it is worth a back to back test....

[chrisshillito]

lastly - a thought starter - I lined up my camera with the scope axis - but if I turn it 45 degrees.... it should give 40% more resolution...I think it is worth a back to back test....

If you're going to do that make sure you're guide camera has square pixels. PHD only measures star movememt in terms of pixels and has no knowledge of the pixel dimensons. As a result it will not measure the camera tilt angle correctly if the pixels are not square and this could result in a level of correction being applied to the wrong axis.

Chris.

[Jessun]

PHD is great as it's free but I can't get my head around that it looses a star and then does nothing.... While you're asleep...

As a backup routine it should really be able to just pick another one. Why don't these programs guide on multiple stars anyways. The computer is at that point doing pretty much nothing else but waiting for a new sub through the USB port, so number crunching should be easy enough to do for a whole bunch of stars.

I never used Maxim, but that feature alone could make it a deal breaker. Could save an unknown but huge amount of dosh on AP absolute encoders...

[Astrosurf]

I'm glad this thread is of interest, even if I don't understand all of it!

[nightvision]

I am really new to all this and just starting guiding and AP. I have been trying many of the free guiding programs and it seems there is more choice than going from free to very expensive.

I did some initial testing just with a long exposure webcam connected to a 50mm guide scope. I found PHD to be the most flaky and inconsistent; I don't think it gets on with webcams. Metaguide seemed to be the most stable and sophisticated and several others where very reliable but lacked features (features I don't yet know the importance of).

I now have a guide camera (AG130) and on some brief testing PHD seems slightly less unstable (than with a webcam) but not as reliable as several other packages. I am just building a test rig to do some proper testing and will post findings in a new thread.