D800 IQ is far from MF

falconeye · 02-09-2012, 02:16 AM

Originally posted by Class A

These are approximate figures as I'm disregarding the difference in aspect rations between FF and 645(D).

There seems to be consensus to compute crop factor based on the image circle, i.e., the sensor diagonal.

Crop factors with respect to 35mm full frame then are:
K-5: 1.52 (1/0.66)
D800: 1
645D: 0.79 (1/1.27)

But it is correct, a surface-based crop factor is required to compare pixel pitch. The pixel pitch difference between a D800 (5µm) and a 645D (6µm) is small though. Moreover the D800, D7000 and K-5 all share the same pixel pitch. So, I don't quite get it why everybody expects the D800 to outresolve lenses. E.g., the K-5 cerntainly doesn't outresolve the FA31Ltd. and that is a FF lens actually.

westmill · 02-09-2012, 04:10 AM

Originally posted by falconeye

There seems to be consensus to compute crop factor based on the image circle, i.e., the sensor diagonal.

Crop factors with respect to 35mm full frame then are:
K-5: 1.52 (1/0.66)
D800: 1
645D: 0.79 (1/1.27)

But it is correct, a surface-based crop factor is required to compare pixel pitch. The pixel pitch difference between a D800 (5µm) and a 645D (6µm) is small though. Moreover the D800, D7000 and K-5 all share the same pixel pitch. So, I don't quite get it why everybody expects the D800 to outresolve lenses. E.g., the K-5 cerntainly doesn't outresolve the FA31Ltd. and that is a FF lens actually.

The 31mm 1.8 is most certainly a corker and one of the best there is. Even so... At F4 which is where it performs at its best... It actualy surpasses
the resolving power of the K10 in the middle of the frame. Everywhere else it does not ! Fitted to 16 million pixel K5 the camera outperforms the lens
even in the centre at F4. Basicly... even the 10 million pixel K10 outresolves it at everywhere except F4 in the centre of the frame !
Pentax SMC-FA 31mm f/1.8 AL Limited - Review / Test Report
its not exactly difficult to see !

westmill · 02-09-2012, 04:18 AM

Originally posted by jsherman999

Let's see a much larger version of your avatar, and then all is forgiven

.

Yes, it does seem rather popular lol, it was deemed to be the seventh most veiwed picture of all time on one of the model sites.
The girl is actualy my best freind too..... say hello to Kelly lol

falconeye · 02-09-2012, 08:07 AM

Originally posted by westmill

Basicly... even the 10 million pixel K10 outresolves it at everywhere except F4 in the centre of the frame !
Pentax SMC-FA 31mm f/1.8 AL Limited - Review / Test Report
its not exactly difficult to see !

Borders and corners are difficult, esp. for wider lenses, agreed.

btw, the FA31 becomes better at f5.6 and f8 at the borders and corners. f4 is only the sweet spot for the center.

But this is where full frame shines (contrary to common belief). The equivalent for a FA31 at f/5.6 on full frame would be 50mm f/8 and there the corner performance is typically great. Moreover, part of the lens softness in corners can be corrected using DxO.

But anyway, center performance matters to me. I shoot enough sunjects where the center region contains the focus (both technically and composition wise).

ogl · 02-19-2012, 07:00 AM

Originally posted by westmill

the full pic.....

Compare with my K200d - it's just 10 MP, of course...
http://fotkidepo.ru/photo/132601/35882x2DrjaPV0i/8oQCmknK30/738066.jpg

and with K-5...
http://fotkidepo.ru/photo/132601/35882x2DrjaPV0i/8oQCmknK30/738085.jpg

bossa · 02-20-2012, 12:06 AM

People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways.

tuco · 02-20-2012, 10:25 AM

Originally posted by bossa

People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways.

That is assuming the lens on the APS-C has an image circle for FF. When the image circle just fits an APS-C, you are back to using wider lenses for equivalent focal length.

westmill · 02-20-2012, 10:31 AM

Originally posted by bossa

People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways.

That is very true.... Apsc gets to use the sweetspot of a lense centre designed for full frame.
A lot of lenses are DX too though so obviously you lose that edge in that instance.
Dont forget though full frame is larger than APSC so therefore needs less magnification so full frame genraly has the advantage when producing
really big prints, just as medium format has the advantage over full frame.
I like and enjoy the bennafits of four thirds system. I manage quite happilly without shallow depth of full frame and I produce too few really huge prints to miss that slight advantage of full frame. APSC is a nice middling compromise.

bossa · 02-20-2012, 11:25 PM

Originally posted by tuco

That is assuming the lens on the APS-C has an image circle for FF. When the image circle just fits an APS-C, you are back to using wider lenses for equivalent focal length.

Good point.

pinholecam · 02-21-2012, 01:44 AM

1st time I've found time looking at the Nikon samples, since I am not prepared to bust the bank for one.

I've not looked at all of them, but

1. Img_02 (the Japanese traditional building) - the DR seems good (areas under sun and black area under roofing ). However, edges are pretty bad (maybe the lens performance). The foreground is underwhelming too. The grass/moss details seems to me messy and 'made up'.

2. Img_03 (Modern Geisha) - The face does not seem the sharpest part (focus off?). The sleeve in front seems sharper, but then all the cloth textures seem 'messy' and "made up" again. I just can't see that so said resolving power making out the texture properly.

Doesn't anyone do proof reading/viewing when releasing these official samples?
Or do they need to be resized/sharpened to make the best of the file.
ATM, to me its just a more MP version of what a K5 can give.

02-21-2012, 01:44 AM	#55
pinholecam Pentaxian Join Date: Oct 2009 Location: North Photos: Gallery Posts: 4,709	1st time I've found time looking at the Nikon samples, since I am not prepared to bust the bank for one. I've not looked at all of them, but 1. Img_02 (the Japanese traditional building) - the DR seems good (areas under sun and black area under roofing ). However, edges are pretty bad (maybe the lens performance). The foreground is underwhelming too. The grass/moss details seems to me messy and 'made up'. 2. Img_03 (Modern Geisha) - The face does not seem the sharpest part (focus off?). The sleeve in front seems sharper, but then all the cloth textures seem 'messy' and "made up" again. I just can't see that so said resolving power making out the texture properly. Doesn't anyone do proof reading/viewing when releasing these official samples? Or do they need to be resized/sharpened to make the best of the file. ATM, to me its just a more MP version of what a K5 can give. Last edited by pinholecam; 02-21-2012 at 01:57 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Nikon D800 images now available	einstrigger	Canon, Nikon, Sony, and Other Camera Brands	33	02-07-2012 12:52 AM
Pentax K? and Nikon D800	Don Boyd	Pentax DSLR Discussion	35	01-23-2012 12:44 PM
Nikon D800 FF due soon?	joe.penn	Canon, Nikon, Sony, and Other Camera Brands	8	12-26-2011 11:53 AM
Ladies and Gentlemen, say hello to the D800!	JohnBee	Canon, Nikon, Sony, and Other Camera Brands	22	11-22-2011 06:05 PM
Nikon D800	Netsoft	Photographic Industry and Professionals	7	11-20-2011 03:26 PM

02-09-2012, 02:16 AM	#46
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871	Originally posted by Class A These are approximate figures as I'm disregarding the difference in aspect rations between FF and 645(D). There seems to be consensus to compute crop factor based on the image circle, i.e., the sensor diagonal. Crop factors with respect to 35mm full frame then are: K-5: 1.52 (1/0.66) D800: 1 645D: 0.79 (1/1.27) But it is correct, a surface-based crop factor is required to compare pixel pitch. The pixel pitch difference between a D800 (5µm) and a 645D (6µm) is small though. Moreover the D800, D7000 and K-5 all share the same pixel pitch. So, I don't quite get it why everybody expects the D800 to outresolve lenses. E.g., the K-5 cerntainly doesn't outresolve the FA31Ltd. and that is a FF lens actually.

02-09-2012, 04:10 AM	#47
westmill Veteran Member Join Date: Jan 2012 Location: Stoke on Trent Photos: Gallery \| Albums Posts: 1,146	Originally posted by falconeye There seems to be consensus to compute crop factor based on the image circle, i.e., the sensor diagonal. Crop factors with respect to 35mm full frame then are: K-5: 1.52 (1/0.66) D800: 1 645D: 0.79 (1/1.27) But it is correct, a surface-based crop factor is required to compare pixel pitch. The pixel pitch difference between a D800 (5µm) and a 645D (6µm) is small though. Moreover the D800, D7000 and K-5 all share the same pixel pitch. So, I don't quite get it why everybody expects the D800 to outresolve lenses. E.g., the K-5 cerntainly doesn't outresolve the FA31Ltd. and that is a FF lens actually. The 31mm 1.8 is most certainly a corker and one of the best there is. Even so... At F4 which is where it performs at its best... It actualy surpasses the resolving power of the K10 in the middle of the frame. Everywhere else it does not ! Fitted to 16 million pixel K5 the camera outperforms the lens even in the centre at F4. Basicly... even the 10 million pixel K10 outresolves it at everywhere except F4 in the centre of the frame ! Pentax SMC-FA 31mm f/1.8 AL Limited - Review / Test Report its not exactly difficult to see !

02-09-2012, 04:18 AM	#48
westmill Veteran Member Join Date: Jan 2012 Location: Stoke on Trent Photos: Gallery \| Albums Posts: 1,146	Originally posted by jsherman999 Let's see a much larger version of your avatar, and then all is forgiven . Yes, it does seem rather popular lol, it was deemed to be the seventh most veiwed picture of all time on one of the model sites. The girl is actualy my best freind too..... say hello to Kelly lol

02-09-2012, 08:07 AM	#49
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871	Originally posted by westmill Basicly... even the 10 million pixel K10 outresolves it at everywhere except F4 in the centre of the frame ! Pentax SMC-FA 31mm f/1.8 AL Limited - Review / Test Report its not exactly difficult to see ! Borders and corners are difficult, esp. for wider lenses, agreed. btw, the FA31 becomes better at f5.6 and f8 at the borders and corners. f4 is only the sweet spot for the center. But this is where full frame shines (contrary to common belief). The equivalent for a FA31 at f/5.6 on full frame would be 50mm f/8 and there the corner performance is typically great. Moreover, part of the lens softness in corners can be corrected using DxO. But anyway, center performance matters to me. I shoot enough sunjects where the center region contains the focus (both technically and composition wise).

02-19-2012, 07:00 AM	#50
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382 Original Poster	Originally posted by westmill the full pic..... Compare with my K200d - it's just 10 MP, of course... http://fotkidepo.ru/photo/132601/35882x2DrjaPV0i/8oQCmknK30/738066.jpg and with K-5... http://fotkidepo.ru/photo/132601/35882x2DrjaPV0i/8oQCmknK30/738085.jpg

02-20-2012, 12:06 AM	#51
bossa Veteran Member Join Date: Jul 2011 Location: Adelaide Posts: 4,546	People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways.

02-20-2012, 10:25 AM	#52
tuco Pentaxian Join Date: Jul 2009 Location: Pugetopolis Posts: 11,030	Originally posted by bossa People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways. That is assuming the lens on the APS-C has an image circle for FF. When the image circle just fits an APS-C, you are back to using wider lenses for equivalent focal length.

02-20-2012, 10:31 AM	#53
westmill Veteran Member Join Date: Jan 2012 Location: Stoke on Trent Photos: Gallery \| Albums Posts: 1,146	Originally posted by bossa People seem to be forgetting that a lens used on a K-5 has that 'extra' magnification to bring it up to the same print/screen size as a FF image and some of the limitations of a lens are going to show up early just for that reason alone. The edges on a FF are more likely to show softness as it's a much larger image circle and it's using way more of the lens than a APS-C crop. It seems to me that the FF is then disadvantaged when printed/viewed at the same size as APS-C because everything get's included in the image.. even the softer outer edges of the lens (unlike the aps-c). Both formats seem to win out in different ways. That is very true.... Apsc gets to use the sweetspot of a lense centre designed for full frame. A lot of lenses are DX too though so obviously you lose that edge in that instance. Dont forget though full frame is larger than APSC so therefore needs less magnification so full frame genraly has the advantage when producing really big prints, just as medium format has the advantage over full frame. I like and enjoy the bennafits of four thirds system. I manage quite happilly without shallow depth of full frame and I produce too few really huge prints to miss that slight advantage of full frame. APSC is a nice middling compromise.

02-20-2012, 11:25 PM	#54
bossa Veteran Member Join Date: Jul 2011 Location: Adelaide Posts: 4,546	Originally posted by tuco That is assuming the lens on the APS-C has an image circle for FF. When the image circle just fits an APS-C, you are back to using wider lenses for equivalent focal length. Good point.