Just curious to know if anyone is using an A7S for imaging - not the other A7 variants. On paper its ISO performance in low light should slay anything from Canon/Nikon but paper specs aren't worth much IMHO...
From the data flowing around the net the A7s and Canon 6D should be very close.
Cooled however could be different story - lets see what happens in the coming months!
In practical usage yes but in straight compare each other at high ISO I think the Sony is the king even better than the Nikon D4s and DF.
But probably for norma astro use ISO6400 is probably as far as you would go in which case not a lot of gain? If you imaged at ISO12800 or 25600 the differences would show as I believe the Sony is good up to 25600.
Also the full well depth is massive in the Sony - 157,000 electrons compared to 9 micron CCDs like the KAI11002 and KAF16803 which are around 105,000 that is huge. That may be the biggest gain as stars now should retain colour rather than the often seen white DSLR stars with little colour.
I read the pixel size of the Sony is 8.54 microns. That's similar to the 11002 and 16803 at 9 microns. CMOS have 40% of the surrounding area taken up with individual pixel circuitry and ccd does not have that.
Greg.
Last edited by gregbradley; 09-12-2014 at 11:42 AM.
Don't worry John, I'll try to break the Internet with my findings!
UPS delivery note says that it's due Thursday, sounds too good to be true. Might even make it in time for Heathcote on Saturday?
I have seen tests showing that the 6d has better noise control.
From what I have seen across the net, 12800 will be the practical limit of the Sony, but maybe even higher for NB, but that's still pretty fast. Until I get it cooled by JTW Astronomy that is, who will be releasing an ultra cooled version of the Sony in the future.
Its base ISO is 3700. That's SIX times faster than I use my 40D at. My 10 minute subs at heathcote will take just over a minute. My first test will be for 60 shots at that setting.
The beauty of the JTW deal is that it has a 30 day no questions asked return policy. So I'm only risking the price of postage if it turns out to be a dud.
Sounds exciting. Now don't be naughty and get lazy and do 20 subs at 30 seconds ISO12800 and think that will do. You still need total exposure time to get the details and depth. On ED180 you've got both aperture, speed and a very sensitive camera but it doesn't let you off the hook and do 3 minute total exposure times!
Remember ISO is merely amplified signal. You still need to get the signal.
I have question that may not be able to answer as yet,
With a 450d at iso 1600 a 45 sec image still needs ~45 sec to process on the camera and then to download. I use the EOS utility to make sequences of 10 images, with delays to allow for onboard processing.
how much time is sent on onboard processing in the A7??
Alan, it sounds like you have long exposure noise reduction ON. That means your camera will take a 45 sec dark pic for every 45 sec light pic you take. You're better off deactivating it and taking a few separate darks.
I just tested the Sony, it can take a 10 sec pic with virtually no delay between shots.
I'm going to our dark site tonight, glorious blue skies out, I can't understate how excited I am to give this baby first light!
Here's a shaky HANDHELD quickie I took last night with a 200mm lens, at 1/4 of a second, at 12800 ISO. Very pleased. not with the image, terrible, but the noise, sensitivity at stupid exposure times and star colours impress me. At 1/2 a second the light pollution starts to be exposed! So I could try 1200 subs for an hour?
I've had the camera for a couple of months. It's high ISO performance is indeed impressive as others have mentioned. However, the dark current noise is quite high, much higher than 6D at equivalent ISO/shutter. I've had a lot of trouble removing the hot pixels in DSS
Attached image straight out of camera ... ISO 12800, f/2.8, 8 s, 24mm
I've had the camera for a couple of months. It's high ISO performance is indeed impressive as others have mentioned. However, the dark current noise is quite high, much higher than 6D at equivalent ISO/shutter. I've had a lot of trouble removing the hot pixels in DSS
Attached image straight out of camera ... ISO 12800, f/2.8, 8 s, 24mm
Is there a way to map the hot pixels? I remember with Nikon's there was a way to get the camera to update its hot pixel map.
Worth a post in a Sony Group somewhere.
Is that a hot pixel on the shore and it looks a bit orange?
Greg.
Is there a way to map the hot pixels? I remember with Nikon's there was a way to get the camera to update its hot pixel map.
Worth a post in a Sony Group somewhere.
Is that a hot pixel on the shore and it looks a bit orange?
Greg.
Not that I've found Greg. The pixel noise is most noticeable in electronic shutter mode. First curtain and mechanical are much better ... but still worse than 6D. I think the orange light is a distant light on the horizon.
I've read electronic shutter can cause some issues with certain scenes like fluro lighting. No doubt the electronic process causes some artifact somehow.
With such short exposures wouldn't it be easier to simply use long exposure noise reduction?
Or take 30 darks and make a master and bias and use a master adaptive dark?
I'm interested as I was thinking about perhaps getting one at some stage. But then the A7r is no slouch and it seemed to me unless you are taking night video then ISO12800 has little advantage over 30 seconds/ISO6400 which is already extremely bright or even better 90 seconds/ISO1600 with a Polarie.
By the way I saw a nice little DSLR tracker at Bintel the other day. It seems to guide in both axes. That could be good for a 200mm Milky Way shot for 30 minutes or so.
The 6D does seem very clean. I wonder how Canon are doing it? Perhaps one reason is dark subtract occurring on every long exposure/high ISO or some sort of very effective hot pixel subtract. Its hard to believe the basic sensor is that noise free as its the same architecture they have been using now for quite some time. The main difference between a Sony CMOS full frame and a Canon is the Canon has the A/D converter circuitry off chip and a separate piece of silicone. Sony has theirs onchip and around every pixel. No doubt there are advantages to both ways with Sony's producing lower noise due to less noise picked up from the travel off chip. The Canon though may lead to less heat.
I think you're right Greg, for regular wide field nightscapes there is not much to be gained over 6D. My main reason for getting it is high iso video, plus the electronic shutter for higher mag zero shake timelapse.
Here's an example of through the scope video - ISS/Moon transit. I pushed the cam to the limit ... 1/2500, iso 6400, 60 fps, f/10, 2000 mm. Shot in APS-C mode @ 1920x1080 video which is sharper than full frame, almost as sharp as 1:1 pixel.
don't forget sony's s log2 for extreme dynamic range in video particularly would be super handy for lunar eclipses and events like the Saturn moon / occultations.