Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Is Sony Really Alpha?

  • Please log in to reply
2186 replies to this topic

#2126 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 18 March 2021 - 07:38 PM

I'm sorry, but what are you looking at, because the computed MTF curve for the Canon 35mm f1.8 certainly is NOT better than the actually measured (with 10 copies) MTF from the Sony 20mm f1.8 by Roger. The 10mm san/tag lines from the Sony are substantially better than the Canon, they also show no separation out to the full frame corner while the Canon has substantial separation (hence, the awful aberrations you see in the example images as posted in the various threads linked to the Canon user reports). The 30mm tan/sag lines are sharper for the 20mm before falling off to similar levels, the 20mm lines are help show the lack of distortion and again have virtually no separation. 

 

FWIW there's really no point in comparing MTF charts from such different focal lengths, especially ones provided from a manufacturer which are computed and those from a private tester which are "real". Based on reports (and my own personal experience with the Sony 20mm) it's quite apparent that the Canon 35mm can't be used wide open unless you are ok with pretty bad coma and LoCA, the 20mm on the other hand can be used wide open with virtually no coma and LoCA. 

 

To the other comments regarding Sony lenses, I have used quite a few now for astro, the Zeiss 50mm f1.4ZA, the Voigtlander 50mm APO, Voigtland 65mm APO, Voigtlander 110mm APO, Voigtlander 21mm Nokton, Voigtlander 40mm Nokton, Sony 24mm GM, Sony 20mm G, Sigma Art 14mm (e-mount), Sigma Art 35mm, and a few others. I would say the ones that come closest to "perfect stars to the edge" are the Voigt 50mm APO (this is the best 50mm currently available except maybe the Leica APO) and the 20mm G. The Voigt 50mm APO is a work of art, one of the few lenses capable of resolving the high resolution A7RIV sensor, no coma, very very little LoCA. The 20mm G is probably the best lens in the 16 to 24mm range, I've also used the Nikon 20mm f1.8S which is very close, but the Nikon has worse distortion and a baked in lens profile that sucks. The Sony 20mm G has very little distortion, has slight amounts of coma, but no LoCA. The 24mm GM is a little bit sharper, but has slight coma and visible LoCA, it needs to be stopped down to f2.2/f2.5 IMO. The rest of the lenses that I've used have varying degrees of coma/LoCA or sharpness fall off. The Voigt 65mm APO is also fantastic, but quite a bit heftier than the 50mm and I think the 50mm is sharper. I've been using the 110mm APO on my ASI2600mc, I'm working on a high resolution mosaic of the MW this summer with that combo. Voigtlander is going to be releasing a 35mm APO which should also be fantastic, possibly the best 35mm available based on its MTF. I tend to mostly use focal lengths wider than 35mm for time lapse, so in those instances I'm doing untracked single exposures, focal lengths 35mm and longer I use for tracking mount based nightscapes and stuff of that nature. I've got quite a few time lapse videos with my A7s and 20mm G (plus a few other lenses) here:

https://www.youtube....grjX658pmmGWaUw

 

Lots of tracked nightscape shots with various lenses here:

https://www.flickr.c...s/75706432@N02/

 

 

Hmm which do you like better the 20 1.8 or voigtlander 50 apo? I know very different focal lengths but hard to choose which would be more versatile and both look good. 



#2127 erictheastrojunkie

erictheastrojunkie

    Apollo

  • *****
  • Posts: 1,017
  • Joined: 15 Jul 2016
  • Loc: Salt Lake City

Posted 19 March 2021 - 09:07 AM

Hmm which do you like better the 20 1.8 or voigtlander 50 apo? I know very different focal lengths but hard to choose which would be more versatile and both look good. 

Different tools for different purposes, but if I had to have only one it'd be the Voigtlander, it's the best 50mm available for astrophotography IMO. The only thing that is tough with it is the focus throw, it takes a very fine touch to get precise infinity focus, but it's a macro lens afterall and it has a long focus throw for all other distances. The 50mm is insanely sharp, coma free, mostly LoCA free, has round aperture at f2 and f2.8, is one of the highest resolving 50mm lenses available, is absolutely tiny/lightweight, etc etc. The 20mm is good, but there are a lot of offerings in the 20mm to 24mm range that are very good. I actually think I'm going to sell my Sony 20mm G, I have the Voigtlander 21mm f1.4 and while it's not coma free and has some LoCA, it's very sharp and I think I can get away with it for my purposes. Voigtlander is releasing a 35mm APO next month that will almost surely be the best available 35mm for astro (it's MTF chart, fwiw, is astounding), I'm tempted to pick that lens up this year. 



#2128 erictheastrojunkie

erictheastrojunkie

    Apollo

  • *****
  • Posts: 1,017
  • Joined: 15 Jul 2016
  • Loc: Salt Lake City

Posted 19 March 2021 - 09:26 AM

So you should check this analysis of a Sony lens:

 

https://www.kenrockw...es/50mm-f12.htm

 

The first paragraph above the MTF charts is quite telling.  Evidently they look so good because Sony didn’t actually MEASURE them, and they left out diffraction and flare.  They are “fairyland” MTF charts.

 

Probably true for the 20mm f/1.2 lens as well.

It needs to be said again, Ken Rockwell is a borderline worthless lens and equipment reviewer, he uses his website purely to drive money making traffic and his opinions are insanely subjective, not based on any kind of standardized testing procedure, and he is WILDLY biased (and has been for a very very long time) for certain manufacturers (unironically of which you are also wildly biased). Anyone using his website for referencing anything equipment related lacks any and all credibility, as far as I'm concerned the only reason I'd give Ken's website a visit is to find tables of physical characteristics of lenses (sizes, weights, etc). I also suggest people stay away from DXO for anything equipment testing related, they come up with their own methodology that is unpublished and generates random numbers that don't mean anything/aren't comparable. They create "scores" for different purposes, but without publicly stating their procedures and what goes into making those scores they are pointless. I also laugh at Ken's thoughts on the MTF released from Sony, of course they aren't generated from actual results, virtually EVERY manufacturer's MTF charts are CALCULATED. Very very few manufacturers publish MTF charts from multiple copy real world testing, Voigtlander being a recent exception with their newer APO lenses. Until people like Roger Cicala can get their hands on these lenses real-world MTF charts don't exist. And in any case, despite Ken's absurd thoughts on the matter, the chart is generated from optical bench calculations and not some fairyland world. This isn't some game of charades being pulled on us by Sony, all lens manufacturers do this, even Canon.  

 

For those looking for proper lens testing websites I suggest you stick with the following, all of which use non-biased, objective, standardized (and published) testing procedures:

Roger Cicala's blog at lensrentals.com, it is the single best lens testing website on the internet, Roger is the ONLY person doing multiple copy rigorous testing for MTF as well as curvature and distortion. 

the-digital-picture.com, uses a lot of Roger's hard core testing stuff in a more typical layout with additional testing and sample pictures.

lenstip.com, very useful for a wider variety of lenses in a typical layout with standardized testing of things that are also very useful for astro like coma and LoCA. 

 

A few others that I use if I can't find anything on the above websites:

opticallimits.com

cameralabs.com



#2129 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 25 March 2021 - 06:08 PM

Different tools for different purposes, but if I had to have only one it'd be the Voigtlander, it's the best 50mm available for astrophotography IMO. The only thing that is tough with it is the focus throw, it takes a very fine touch to get precise infinity focus, but it's a macro lens afterall and it has a long focus throw for all other distances. The 50mm is insanely sharp, coma free, mostly LoCA free, has round aperture at f2 and f2.8, is one of the highest resolving 50mm lenses available, is absolutely tiny/lightweight, etc etc. The 20mm is good, but there are a lot of offerings in the 20mm to 24mm range that are very good. I actually think I'm going to sell my Sony 20mm G, I have the Voigtlander 21mm f1.4 and while it's not coma free and has some LoCA, it's very sharp and I think I can get away with it for my purposes. Voigtlander is releasing a 35mm APO next month that will almost surely be the best available 35mm for astro (it's MTF chart, fwiw, is astounding), I'm tempted to pick that lens up this year. 

Im curious about the new sony 35 1.4 gm. The 50 1.2 im sure will be good but too expensive. 



#2130 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 01 April 2021 - 01:10 PM

In a full-spectrum mod, for that matter any mod, moving sensor forward is advisable to compensate for removed filter/s. Modern Sony has shims which can be removed/re-arranged (...while being cognizant of orthogonality) to move sensor forward a bit (...although, given the thin shims, limited move forward may not fully compensate for the removed filters). Regards

I'm planning to take a 2nd old a7 apart to convert to full spectrum.

 

I've read a lot about shimming the sensor but I can't find any specific guidance on this. Do you need to get additional shims? 

 

Or as you said removing the shims actually moves the sensor forward? I'm not familiar with how the assembly works and can't find info. I've heard .3mm shimming the sensor forward can compensate for the removed glass but I would like to know exactly how to do this before taking the camera apart. 


Edited by xonefs, 01 April 2021 - 01:10 PM.


#2131 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 11,894
  • Joined: 13 Jan 2012
  • Loc: USA

Posted 01 April 2021 - 02:53 PM

Following is what shims look like; you need to keep track of them since they mostly are NOT the same thickness and sometimes multiple at one mount point. Only way you can move sensor forward on Alphas is by removing and/or re-arranging shims but keeping track of orthogonality. Movement forward will be quite limited and may not fully compensate for the filters removed.

 

 

In short, don't read too much into it. Do the mod, keep things orthogonal and enjoy. Auto-focus may still be preserved; in the end auto-focus in NOT needed for astro. Regards

 

 

Note: a7S shown; mark shims before removing or re-arranging. I put dots on them in 1, 2, 3 order.

Attached Thumbnails

  • Shims.jpg

Edited by mmalik, 01 April 2021 - 02:55 PM.


#2132 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 01 April 2021 - 03:35 PM

Following is what shims look like; you need to keep track of them since they mostly are NOT the same thickness and sometimes multiple at one mount point. Only way you can move sensor forward on Alphas is by removing and/or re-arranging shims but keeping track of orthogonality. Movement forward will be quite limited and may not fully compensate for the filters removed.


In short, don't read too much into it. Do the mod, keep things orthogonal and enjoy. Auto-focus may still be preserved; in the end auto-focus in NOT needed for astro. Regards


Note: a7S shown; mark shims before removing or re-arranging. I put dots on them in 1, 2, 3 order.


Thank you. This is also for terrestrial which is why im more concerned about spacing. I’ll give it ago.

#2133 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 03 April 2021 - 01:15 AM

Following is what shims look like; you need to keep track of them since they mostly are NOT the same thickness and sometimes multiple at one mount point. Only way you can move sensor forward on Alphas is by removing and/or re-arranging shims but keeping track of orthogonality. Movement forward will be quite limited and may not fully compensate for the filters removed.

 

 

In short, don't read too much into it. Do the mod, keep things orthogonal and enjoy. Auto-focus may still be preserved; in the end auto-focus in NOT needed for astro. Regards

 

 

Note: a7S shown; mark shims before removing or re-arranging. I put dots on them in 1, 2, 3 order.

so these shims look like different thickness and one has two. if I just remove them all seems like there would be the possibility of introducing tilt, so how would you check that and keep it orthogonal while removing/rearranging?



#2134 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 11,894
  • Joined: 13 Jan 2012
  • Loc: USA

Posted 03 April 2021 - 01:19 AM

Measure thickness of all, and do your best to keep things orthogonal as you use either the thinnest at appropriate mount point or best case not use any. Regards


Edited by mmalik, 03 April 2021 - 01:20 AM.

  • xonefs likes this

#2135 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 03 April 2021 - 01:30 AM

Measure thickness of all, and do your best to keep things orthogonal as you use either the thinnest at appropriate mount point or best case not use any. Regards

Ok so I have .4mm, .25mm, and .3mm. There's no way to really rearrange and subtract .25 equally from everywhere.

 

So if I don't use any it seems like it could have one side lower by .15mm than it started. I'm not sure how bad that is.

 

Or if I should just keep the stock shims as is and deal with potentially losing infinity focus on some lenses without an internal filter

 

appreciate any guidance as I would like to reassemble soon 

 

edit: Ive found some plastic m42 shims and I cut one to size that is .08mm thick. I plan to replace the .40mm with that and leave the other two bare. hope that is good enough to stay mostly orthogonal that's the closest I can realistically get with what I have. 


Edited by xonefs, 03 April 2021 - 01:52 AM.


#2136 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,668
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 03 April 2021 - 02:18 AM

edit: Ive found some plastic m42 shims and I cut one to size that is .08mm thick. I plan to replace the .40mm with that and leave the other two bare.

That sounds great, as long as the sensor ends up orthogonal, which is not clear from the info you have provided.

 

Mark



#2137 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 11,894
  • Joined: 13 Jan 2012
  • Loc: USA

Posted 03 April 2021 - 03:03 AM

As mark pointed out, you need to provide the individual shim as well as the total for each mount point. If you have already figured it (orthogonally) out, you should be good. Regards



#2138 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,668
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 03 April 2021 - 03:26 AM

I think I've worked out what you mean.  Currently your 3 mounting points have single shims of .4mm, .25mm, and .3mm.  You intend to replace them with 0.08, 0.0, 0.0 respectively.  This will introduce .07mm of tilt i.e. 70 microns.  70 microns is a lot of tilt - especially if you use fast optics.

 

Mark


  • SandyHouTex and xonefs like this

#2139 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 03 April 2021 - 11:07 AM

I think I've worked out what you mean. Currently your 3 mounting points have single shims of .4mm, .25mm, and .3mm. You intend to replace them with 0.08, 0.0, 0.0 respectively. This will introduce .07mm of tilt i.e. 70 microns. 70 microns is a lot of tilt - especially if you use fast optics.

Mark

Yes that’s correct on the spacers and what I was going to try. Wouldn’t how much tilt depend on the mount point of where the spacers are? The posts are not right at the edge of the sensor but pretty far outside of the whole assembly, so I would think from looking at it that it would be like half that much actually across the sensor area diagonol. And I don’t know how precise it could have been in the first place or what is acceptable.

Is there any way to test for tilt after without very expensive software? And without using star fields?

Edited by xonefs, 03 April 2021 - 11:16 AM.


#2140 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,668
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 03 April 2021 - 04:20 PM

And I don’t know how precise it could have been in the first place or what is acceptable.

This teardown states that shims of 40, 30 and 25 microns were used:

https://www.lensrent...nys-new-camera/

 

If it is correct, it suggests that Sony aims for an accuracy of 5 microns or less.

 

Mark


Edited by sharkmelley, 03 April 2021 - 04:28 PM.


#2141 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 03 April 2021 - 09:29 PM

This teardown states that shims of 40, 30 and 25 microns were used:

https://www.lensrent...nys-new-camera/

 

If it is correct, it suggests that Sony aims for an accuracy of 5 microns or less.

 

Mark

Well it reaches focus with all lenses I've tested and I couldn't tell any tilt or variation in corners from some quick photo tests today with a 55mm f1.8 lens and 180mm 2.8 lenses wide open and  at different apertures . But it's cloudy out to try to test on stars at the moment. 

 

I guess ideally I would have used a .10 and .05 shim. 


Edited by xonefs, 03 April 2021 - 09:30 PM.

  • nofxrx likes this

#2142 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,668
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 04 April 2021 - 01:23 AM

Well it reaches focus with all lenses I've tested and I couldn't tell any tilt or variation in corners from some quick photo tests today with a 55mm f1.8 lens and 180mm 2.8 lenses wide open and  at different apertures . But it's cloudy out to try to test on stars at the moment. 

 

I guess ideally I would have used a .10 and .05 shim. 

I think you mean .15 and .05 ?

 

The big test will be a star field but your initial results sound hopeful for the lenses you are using.

 

Mark


Edited by sharkmelley, 04 April 2021 - 01:30 AM.


#2143 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 04 April 2021 - 11:38 AM

I think you mean .15 and .05 ?

 

The big test will be a star field but your initial results sound hopeful for the lenses you are using.

 

Mark

yes that's correct

 

I might also use this on a scope, but it would be hard to test on the redcat to know if the scope or the sensor is tilted since it already seems slightly off on my other full frame camera. 



#2144 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 05 April 2021 - 06:30 PM

Well, I’ve been spoiled with the A7RIV... I always thought the star-eater thing was overblown but it’s pretty horrible on the a7 for long exposures... so probably won’t use this much as another astro cam. Guess it might be ok for nightscapes but I was still hoping to track for a minute or two.

Id like another a7riv but too expensive and maybe even a used a7iii but used a7iii prices are still stupid high.

#2145 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 11,894
  • Joined: 13 Jan 2012
  • Loc: USA

Posted 05 April 2021 - 08:11 PM

The Alpha game...

 

 

 

Attached Thumbnails

  • Comparison_Dedicated++.jpg

  • t_image likes this

#2146 SandyHouTex

SandyHouTex

    Fly Me to the Moon

  • *****
  • Posts: 5,811
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 06 April 2021 - 08:48 AM

Warning on Sony firmware updates:

 

https://www.cloudyni...ce-and-for-all/

 

A7rIII now a paper weight.



#2147 nofxrx

nofxrx

    Vendor (HyperCams & Mods)

  • *****
  • Vendors
  • Posts: 5,697
  • Joined: 12 Jul 2005
  • Loc: Melbourne, Florida

Posted 06 April 2021 - 05:18 PM

The Alpha game...

Mike,

 No offense but these are not comparable other than the sensors used. Everything else about these cameras are different.

You are making the comparison that an a7R4 is comparable to the QHY600 which is absurd, IMHO.

Carry on...


  • xonefs likes this

#2148 t_image

t_image

    Gemini

  • -----
  • Posts: 3,499
  • Joined: 22 Jul 2015

Posted 06 April 2021 - 05:56 PM

Mike,

 No offense but these are not comparable other than the sensors used. Everything else about these cameras are different.

You are making the comparison that an a7R4 is comparable to the QHY600 which is absurd, IMHO.

Carry on...

No offense,

but this reply to Mike on a DSLR/gen cam forum doesn't make sense???????

What makes you think Mike is trying to mislead that the cameras are the SAME besides the sensor,

which IMO is more important than 'everything else' especially when there is a price difference (context of the DSLR forum section) $3k v $7k, not to mention ability to use as a normal camera.

Just because you don't find his sensor sharing devices chart helpful doesn't make it not helpful to others,

therefore your post violates your own requirement.

proceed forth....



#2149 xonefs

xonefs

    Mariner 2

  • *****
  • Posts: 295
  • Joined: 23 Aug 2020

Posted 06 April 2021 - 06:05 PM

Mike,

 No offense but these are not comparable other than the sensors used. Everything else about these cameras are different.

You are making the comparison that an a7R4 is comparable to the QHY600 which is absurd, IMHO.

Carry on...

The sensor is kind of the biggest component.. and comparing it to the color versions is not that far off. I shoot with a full spectrum a7riv sometimes and have mono zwo cams. Honestly results are really not far off shooting with the a7riv aside from it not playing nice with astro software. Dark current is so low cooling doesn't seem like it would really add much. I have a 2600mm now which is pretty close to an aps-c sized version of the 6200, which is the same sensor in the a7rIV as in the OSC version.  Even comparing mono to OSC which is not a fair comparison the results from the 2600mm with LRGB look really close to what I got a couple weeks previously shooting some galaxy targets with the A7RIV.

 

Mono astro cam vs OSC mirrorless will never be a fair comparison, but I would definitely think twice about spending the kind of money these APS-C and full frame OSC astro cams cost when the mirrorless versions are cheaper, more versatile, and close to the same performance (after modding, but I would want it modded anyway). 


Edited by xonefs, 06 April 2021 - 06:21 PM.


#2150 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 11,894
  • Joined: 13 Jan 2012
  • Loc: USA

Posted 07 April 2021 - 12:47 AM

You are making the comparison that an a7R4 is comparable to the QHY600 which is absurd, IMHO.

This is just a sensor level comparison; nothing more. I think folks understand that. Regards




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics