Which 300mm f/2.8 manual lens

[3dpan]

I'm looking for a 300mm f/2.8 manual lens to use on m4/3 for astrophotography. Auto focus has no use on stars. Oly has a ZD 300mm but the price is still prohibitive. There are other reasonably priced options such as Tamron, Tokina, Sigma, all of which I know very little about. The main requirements are sharpness, lack of CA, lack of coma, lack of dollars.
Does anyone have any ideas ?

[Daniel Bradley]

I would go with a good APO triplet and a field flattener if needed. I don't think you will find the same kind of resolution in retrofocus DSLR lenses. A friend of mine has a "good" Canon 300/4 and it weighs in at under 0.7 Strehl. He also has a really good copy of the new 100-400 II zoom, and at 400mm it manages just above 0.8 Strehl. I have compared a Sigma 150-600 at 600mm (actually only 560mm) and my good Canon 400/5.6 with a 1.4x TC (sharper than the Sigma at 560mm), and they can't even come close to my 600/6.3 triplet. It is 0.97 Strehl.

[3dpan]

Thanks Daniel, that's given me something to think about. BTW, are those Polystrehl numbers you are quoting ?
I already have a Vixen VC200L 'scope, but even with a focal reducer the focal length is too long for my main interest - nebulae, in or about the Milky Way.
I have found that 300mm is long enough for all except perhaps the Orion Nebula, the Horsehead Nebula, and the Running Man.
Down here in the Southern Hemisphere we get a very good view of the Milky Way all year round (weather conditions permitting), so I have not yet progressed to hunting distant galaxies.
Cheers,
Alec

[Daniel Bradley]

Hi Alec,
Nobody here talks about Polystrehl, just Strehl. The difference is that in measuring Polystrehl the lens/scope focuses on green. Exactly how, I do not know. There are different ways of doing it and some actually fudge the numbers for sales purposes. My telescope expert friend measures Strehl without any tricks, and if he comes up with 0.97 you know it is good. I had an 8 inch Newton that measured 0.985. That is really getting up there, when you consider that a theoretically perfect lens would be 1.0, or 100%. I only know that no astro photographer in his right mind () would touch anything under say, 0.95 Strehl, and that even the cheap Skywatcher APO doublets can hit that. The camera lens makers though deliberately do NOT publish such information as it looks so bad on paper and in the end is irrelevant. The fact is that SLR lenses have to meet such different requirements that they are corrected very differently than a scope. Leitz did for a while with their Telyt lenses, but soon stopped when people turned up their noses at numbers in the 0.70s. I tested my triplet against a Canon 500/4 and it blew it away in terms of pure resolution. BUT.... although my scope is REALLY sharp, backgrounds (bokeh) look terrible! Who cares about bokeh if you are shooting stars?
It seems though, that about 400mm is as short as anyone makes in scopes. Things like the Skywatcher Esprit or the various 420mm scopes around.
So even my friend has to resort to SLR lenses when he wants a wider field of view, hence the Canon 300/4 and now the even better 100-400 II. Here is a little article he wrote on his testing:
http://interferometrie.blogspot.co.at/
July 5th.
I have borrowed his 100-400 II and it is really good. My new 400/5.6 is a tad sharper center frame, but CA becomes a real problem as you move away from the center. There the zoom does really well. Good, you are not going to get f/2.8 at 300mm, probably closer to f/4.
Cheers
Dan

[3dpan]

Thanks Dan. That's a really comprehensive reply.
I try to get a f/2.8 lens so that the stars are a bit easier to focus on. For shooting I'm quite happy to stop down to f/4. even f/5.6 to minimise CA, and maximise sharpness. The new generation of cameras allow higher ISO settings to compensate. I have a Mamiya 300mm f/2.8 APO lens, but being designed for medium format film means the resolution is not up to 4/3 standards. I was hoping a "35mm film standard" lens might improve on the Mamiya resolution but still not cost a small fortune. I'm leaning towards a Tokina 300mm f/2.8 AT-X SD, but I don't know if I should be looking for a later model (auto) lens.
Cheers,
Alec

[Daniel Bradley]

The Tokina seems like a pretty good lens and used are in about the same price class used as the Canon 300/4. You also get resolution advantages with bigger front elements. A 2.8 stopped down to 4 will resolve more than a 4 wide open. Bigger the better! :-)
I know the Canon would also autofocus with a Metabones adapter, no idea about the Tokina, but if you get a Canon mount you could give it a try if you know anyone with a Metabones. The Tokina is not listed under the compatible lenses on the Metabones site, but that does not mean it would not work.
Cheers
Dan

[3dpan]

The Tokina AT-X 300 was originally only a manual lens, available in a variety of mounts, including Olympus OM. Later models included auto focus, but only in Canon, Nikon, and Minolta. (check this). It seems (but again I'm not sure), that the auto models also included improved ED glass.
I was only thinking of using the 300/2.8 for astro so the auto focus is not needed.
I have a 70-300 ZD that I have used in auto for birding etc on my E-620, and I'm OK with that combo, as long as the birds are not too far away.
Cheers,
Alec

[Daniel Bradley]

Alec,
Even if you don't need AF, you would be better off with the newer lenses if they do in fact have ED glass. CA is a PITA on stars! Check out all the nice little Italian flags away from Polaris.
https://c1.staticflickr.com/1/767/31...cb003772_o.jpg
As I don't have any kind of tracking mount I can only do star tests on Polaris, even though there is some elongation away from the NCP. Here is the same thing with my new 400/5.6
https://c1.staticflickr.com/3/2864/3...3da74079_b.jpg
To be fair, this shot also had the remove CA in Lightroom activated. It helps, but it does not replace a lens that has less CA. Still haven't had a clear sky to try it with the E-M1 II.

[3dpan]

Dan,
I've been reading a few reviews on the Tokina 300/2.8, and while they all say wonderful sharpness, some of them admit to purple fringing of highlights, and that's on the later autofocus lenses too. Some say a better choice is the Sigma 300/2.8. So I started reading reviews of that too. And I'm not really any wiser, just older.
Something else that has just come up is a seller on our local internet auction site, offering an E-M5 II kit and an E-M1 I kit. If they don't sell as kits he's thinking of splitting them and selling the bits. I originally asked him about the E-M5 II body only, but he said why not the E-M1 body. So I don't know why not. They both seem to have almost identical specs, and may cost much the same. One thing in favor of the E-M1 is that the price is already depreciated. I prefer the body style of the M1.
Over the past few weeks I've been trying various lenses untracked on the Milky Way. Either the Italian flags (your words), or elongated stars have been a killer. My present lenses are too slow, and my camera (E-620) ISO is too low.
So I resurrected my tracking mount which I have never used, and set it up in the back yard. Took two days to align roughly with the South Celestial pole (compass and clinometer), find sigma octans (that cluster is so dim it needs a really dark sky), and then read the manual, before I actually got it tracking. But what a difference. Use 800 ISO, stop the lens (Tokina AT-X 90/2.5) down to sharpen it, increase the exposure to collect more stars.




Wonderful circular dots.

[tomsi42]

When you are using the E-620, are you using RAW or JPEG fileformat? I find that shooting RAW with my E-30 (same ISO performance as the E-620) and using the latest Lightroom version, I can use ISO 1600.

[3dpan]

I'm using RAW. I had been using 1600 ISO but just my impression it was still noisy, even after stacking a dozen or so (untracked) images. I'm stacking in DeepSkyStacker.
Cheers,

[tomsi42]

I'm using RAW. I had been using 1600 ISO but just my impression it was still noisy, even after stacking a dozen or so (untracked) images. I'm stacking in DeepSkyStacker.
Cheers,

OK. Then it's time to upgrade

[Daniel Bradley]

Southern Celestial Pole.... what's that?

Completely forgot that you are Down Under.....

It is amazing how quickly you can get elongated stars without a tracking mount. My tests on Polaris are only 15 to 20 seconds, ans stars within a fraction of an arch second off NCP are already noticeably elongated.

For fun, here is a shot with my 600mm scope. It was windy and not super clear, (almost never is here) and there seems to be a little smearing. 20 seconds at ISO 800. No CA was removed.... none to remove! Notice it clearly resolves Polaris Ab. Can't do that with my Canon...



click picture to enlarge
Hoping for a clear night sometime. I want to try the Mark II and the scope, as well as the Canon with a 2x TC. It works pretty well on the E-M1:



click

[3dpan]

Dan, That's a seriously impressive shot of Polaris and Polaris ab. And the detail on the moon is "out of this world" (pun intended). Was the moon shot with the 600mm scope too ?
I wonder if a High Resolution shot of the moon would be successful, with the E-M1 II ?
Another question, it seems that the E-M5 II has the anti-aliassing filter removed, compared with the E-M1 I. Is this the same filter that gives all the problems with photographing nebulae ? ie cuts out all the reds. Does this mean there is no need to modify the E-M5 II for best nebulae photos ?

[Daniel Bradley]

It was my understanding that the E-M1 also has no AA filter, but I may be wrong. As pixels get smaller, the need for an AA filter decreases. The 7000 series Nikons have all done without it very nicely, and the E-M1 has even a slightly smaller pixel pitch than the 7100.
The moon shot is a stack of the two best of ten shots (air was not so good) with the 400/5.6+2x Extender III on the E-M1. I somehow doubt that a high res shot would be any advantage over stacking on something so far away and lacking in real fine detail, but I intend to give it a try, IF I ever see the moon around here again....
Here is a shot of the moon with the scope and an EC-14 on the E-M1.
click

As I remember the air was a bit better and this is a stack of four out of ten.


EDIT: No AA filter on the E-M1.

[3dpan]

Imaging-resource have a series of comparison reviews of the E-M1 and E-M5, Mark I and Mark II. They state a few times that the Mark II variants of both models do not have the AA filter, whereas the Mark I does. I did find another confirmation that the E-M5 II does not have an AA filter on one of the DPReview forums. That's as far as I got with that research.
And the infra-red response (or Ha response) is a separate filter anyway.
Your 4-stack moon shot is impressive, especially considering that to the naked eye the moon is just a small white blob in the sky.
For a High Resolution shot of the moon you may need a tracking mount, or maybe the exposure times are short enough that you don't. Can only find out by trying.
My longest lens is a 500mm Zeiss mirror lens.

[Daniel Bradley]

I think Hi-Res would be difficult without a tracking mount as there would be much too much movement within the second or so it needs to take the shots, not to mention changes in the air. I think stacking would be the better way to go, but in a way, the Hi-Res thing is nothing other than a stack. Interesting concept though, and worth a try for sure.
Everything I have found on the subject in both English and German indicates no AA filter on the E-M1. When the Canon 7D2 came out, I met up with a guy and did some testing with the same lens on the Canon and the E-M1, both having nearly the same pixel pitch. The Oly shots were noticeably sharper. Canon has had a history of being over generous with their AA filters, and they put a nice strong one on the 7D2. I can spot 7D2 files a mile away as they look very pleasing, but lack the crispness one gets from a body without an AA filter. In fact, the AA filter was one key thing that kept me from buying a 7D2.

[3dpan]

And I've just realised the AA filter is not the same as the IR filter, two separate pieces of glass. And it's the IR filter I want to remove for astro pics. So I'm back to square one. Interesting discussion though, and I learnt some more about the High Resolution mode.
I want an E-M5 II if just for that feature.
I'm sure a fast lens, 1600 ISO and 8 seconds will give me some detailed Hi Res Milky Way shots, or maybe not. I would need an f/1.2 lens I think. And I don't have one, or enough money to buy one. Ho hum, back to the drawing board.

[Daniel Bradley]

Ah, now that makes sense.
Most people here use modified Canons (IR filter removed) or naturally, dedicated cameras for astro work. The E-M5 II seems to have an IR filter, but it is very weak. No idea about the E-M1s.
Usually the IR "hot mirror" and an AA filter are all in the same glass block put in front of the sensor, so if you take it out on something like a Canon DSLR, not only will the sensor be more IR sensitive but sharper to boot. The down side is that the sensor becomes more sensitive to sunlight and can more easilly be ruined by accident.
1600 ISO is pretty noisy on our little Olys. Why not simply use a much lower ISO, longer exposure, and your tracking mount. True, such things may be a pain to lug around, but...
Speaking of IR filters, that is what I put on the eyepiece of my E-M1 nearly two years ago to protect the EVF, which has been replaced three times. Not had any problems since. So simple to do. I wrote to Olympus and asked them whether they had solved the EVF burn issue with the Mark II and got this reply,

"Thank you for your e-mail concerning the E-M1 Mark II.

Unfortunately the problem with the EVF was not solved. You might experience the same thing with the E-M1 Mark II.

If you have any further inquiries, we will be happy to address them.

We wish you a pleasant day."

Can you believe it??!! If I could "fix" it, why couldn't they? Needless to say, I have the IR cut filter on my Mark II and it is going to stay there!

[3dpan]

I have communicated with Olympus (Japan) in the past and found them infuriatingly polite and negative in the same sentence. Reminds me of our politicians.
I found a firm in the US that seems very knowledgeable about modifying DSLR's, including Olympus,
http://www.spencerscamera.com/astro-conversions.cfm
They will sell you a new, modified camera, or modify yours if you post it to them.
I was thinking of the High Resolution mode, which has a maximum exposure time of 8 seconds, and max ISO of 1600. With those constraints it seems I would need about f/1.4 for star shots, even wider maybe.
I have a long term goal to make a high res panorama of the Milky Way, preferably 360 degrees, and not just a few frames with a fisheye. Should take a year or so, stitching pics as the months pass. Maybe something like a 20mm lens. May have to save up for a Voigtlander.
Cheers,
Alec

[Daniel Bradley]

Ah, I understand. Sounds like an ambitious project. You must have lots of clear skies...

[Daniel Bradley]

Well, I tried it. Here is a High-Res shot of the moon early this eavening:


Tried to sharpen it and give it some contrast. Click picture for larger.

And here is a stack of 18 shots taken a few seconds later, stacked with Registax:



E-M1 Mark II with Canon 400/5.6 + 2x Extender III, electronic shutter.

[3dpan]

Current forecast here is 5 days of rain, but I'm sure the moon is up there.
Looks like Hi Res is no match for stacking, and that's effectively 800mm f/11.
The detail in your pic is something to aim for, a very high standard to match. Wow.
Cheers,

[Daniel Bradley]

Thanks, Alec,
Tonight and tomorrow night would be my favorite times for moon shots, sort of 3/4 full. Needless to say it clouded up last night before I had a chance to try a High-Res shot on Polaris and we are back to monochrome. Got to be Obama's fault....