Reading mode: Light Dark

Does the Alpha 700 really do ISO 12,800?

Go to last post
D

David Kilpatrick

Veteran Member
Messages
5,438
Reaction score
8
Location
Kelso/Scotland, UK
This is a follow-up to the pBase gallery of D300/A700 images:

http://photoclubalpha.com/2007/11/06/does-the-alpha-700-really-do-iso-12800/

It is slightly technical in the closing section, but in the key to it is that my Sony Alpha 700 and Nikon D300 files were not differently exposed in terms of real densities, and the Alpha 700 was giving 'half' the exposure of the D300.

For tungsten and floolight shots, you should probably compare the Alpha 700 at ISO 3200 with the D300 at ISO 6400, because then the exposures will be roughly the same.

Sample variations and operator error excepted!

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
David Kilpatrick said:
It is slightly technical in the closing section, but in the key to it
is that my Sony Alpha 700 and Nikon D300 files were not differently
exposed in terms of real densities, and the Alpha 700 was giving
'half' the exposure of the D300.

For tungsten and floolight shots, you should probably compare the
Alpha 700 at ISO 3200 with the D300 at ISO 6400, because then the
exposures will be roughly the same.
Click to expand...
That's going to complicate using the D300 for those of us who still possess exposure meters and know how to use them. I don't use mine that much, but it is my final authority.

Interesting complication in trying to have standardized "tests" for all cameras. Even using a different ISO is only an approximation. It's value will depend highly on the particular color of the subject. Or in the case of most photos, colors.

It boils down to each camera is different and you have to learn them individually to get the best out of them. What works for one may have no bearing on what works on another.

I do hope you find the time to run the comparison in the same color space. I would like to dig through it all, but don't want to have to play with color space. So, I've held off the download ordeal for now.

Walt
 
Hi Dave,

Interesting thing with the Light meter and the ISO "boost". So, I decided to go back to your PBASE photos and check out the EXIF info. For some reason, the NIKON EXIF info - ISO setting is absent in all of the ISO 6400 NIKON shots? They are there for NIKON ISO3200 shots and I assume lower ISO shots as well...

Can you check the EXIF info on the actual files and see if they're missing as well, I'm thinking maybe it's a glitch with PBASE, but the ISO 6400 EXIF info shows up for the A700, so not sure if this is a PBASE error or not... ?

--
-Alex

From the minds of Minolta to the imagination of Sony, Alpha, like no other.

http://www.pbase.com/lonewolf69
 
lonewolf69 said:
Hi Dave,

Interesting thing with the Light meter and the ISO "boost". So, I
decided to go back to your PBASE photos and check out the EXIF info.
For some reason, the NIKON EXIF info - ISO setting is absent in all
of the ISO 6400 NIKON shots? They are there for NIKON ISO3200 shots
and I assume lower ISO shots as well...

Can you check the EXIF info on the actual files and see if they're
missing as well, I'm thinking maybe it's a glitch with PBASE, but the
ISO 6400 EXIF info shows up for the A700, so not sure if this is a
PBASE error or not... ?
Click to expand...
Check the ISO 100 ones as well - no ISO 100 shown for Nikon. That's because they refuse to allocate a real ISO to these settings outside 200-3200. They state the camera has a range of 200-3200, and extension beyong this to 'equivalents' of 100 to 6400.

It is excluded from the EXIF data at both ends of the scale. The Alpha 700 reports it, but Sony do say that 100 and 6400 are an extended mode as well.

I have no idea why Nikon does this as it prevents you telling whether LO 0.3, 0.5, 0.7 or 1 or HI the same has been used.

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
I guess I didn't get why 14 bit would change the channel boost since it doesn't change DR just the slicing of the DR... but the rest is excellent.. May explain why your shots are off one EV and it looks better than some other D300 shots I have seen.. an anolomoly that might vanish with RAW. If I got the jest of the posting.

------------
Ken - KM 5D (A700 Joy)
http://www.cascadephotoworks.com
 
David,

When I look at those two pictures, just in terms of actual noise, I see more noise in the A700 image, but I see that it has been mostly averaged out in the Nikon image, together with a loss of resolution. For instance, in the feathers there are the typical chroma noise streaks which are largely gone in the Nikon image, but along with a reduction in detail in the frogged threads around the eye holes, and loss of some feather details. So, is it the discrepancy in apparent ISO that is causing this, or is Sony actually doing less on chip NR then Nikon at this level?

--
Only my opinion. It's worth what you paid for it. Your mileage may vary! ;-}
http://www.dougwigton.com/
 
Very interesting David. I just commented in the thread where someone PPed your tests to make the A700 look more like the D300 and I noticed the biggest difference in the blue threads. Specifically, I cannot see much if any detail loss in the A700 as compared to the D300. In some cases (wood grain) I actually see more with the A700, things like the fiber detail look very similar. This could be my eyes, as I know every other part of me is aging, but I did see that at 3200, the blue yarn threads do largely disappear in the A700 sample. Is this tungsten WB blue boost you speak of connected with this blue channel loss of detail too? If not, is it possible that Sony is being more agressive with the blue NR to smooth skys and the like?

Is it possibe I cannot see anymore and have no idea what I am talking about? hehe
 
very interesting insights David - many thanks
--
Bernard

AS/SSS rocks!
lens reviews and more on dyxum.com!
 
Alpha Doug said:
David,

When I look at those two pictures, just in terms of actual noise, I
see more noise in the A700 image, but I see that it has been mostly
averaged out in the Nikon image, together with a loss of resolution.
For instance, in the feathers there are the typical chroma noise
streaks which are largely gone in the Nikon image, but along with a
reduction in detail in the frogged threads around the eye holes, and
loss of some feather details. So, is it the discrepancy in apparent
ISO that is causing this, or is Sony actually doing less on chip NR
then Nikon at this level?
Click to expand...
Maybe Nikon is doing more JPEG conversion NR and less on-chip. Since Sony has three stages of NR - on-chip, raw NR and JPEG High ISO NR - it's complicated. We know that on-chip NR can't be turned off, but we do not know whether firmware v2 altered raw NR, or just high ISO NR.

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
HUM469 said:
Very interesting David. I just commented in the thread where someone
PPed your tests to make the A700 look more like the D300 and I
noticed the biggest difference in the blue threads. Specifically, I
cannot see much if any detail loss in the A700 as compared to the
D300. In some cases (wood grain) I actually see more with the A700,
things like the fiber detail look very similar. This could be my
eyes, as I know every other part of me is aging, but I did see that
at 3200, the blue yarn threads do largely disappear in the A700
sample. Is this tungsten WB blue boost you speak of connected with
this blue channel loss of detail too? If not, is it possible that
Sony is being more agressive with the blue NR to smooth skys and the
like?
Click to expand...
Not a bad guess. Targeting blue and red for chroma NR and making the green channels do more luminance work is a good strategy. I have not really examined the blue sky thing as our blue skies are not very blue right now :-)

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
It would be interesting then to compare RAW images with similar adjustments from Sony and Nikon then, since that would give a better picture of NR done on chip vs. in JPEG conversion in the camera. I still see better details, especially around the eye holes in the frogged threads in the A700 images than the Nikon images, but I see more noise in the Sony images than in the Nikon. There definitely is more NR going on somewhere in the Nikon chain than on the Sony.

--
Only my opinion. It's worth what you paid for it. Your mileage may vary! ;-}
http://www.dougwigton.com/
 
we shuld compare the sony 3200 to the Nikon 6400. That will probably even the scales a bit - the Sony 3200 does not seems worse than the same-exposure Nikon 6400. Also, it will be definitely an advantage for the alpha on really low-light shots, where the Nikon will just need too long an exposure.
 
I am waiting for reviews before I believe nikon released a pro camera that is one stop worse in all iso's than the sony. Sorry, somethings wrong here, I don't believe that. If thats true they won't sell and nikon knows it. Something else is up in this workflow.
 
sandy b said:
I am waiting for reviews before I believe nikon released a pro camera
that is one stop worse in all iso's than the sony. Sorry, somethings
wrong here, I don't believe that. If thats true they won't sell and
nikon knows it. Something else is up in this workflow.
Click to expand...
I have done controlled tests now and the Nikon sensitivity is either 1/2-2/3rds lower than the Alpha, or four of their most expensive lenses have completely miscalibrated aperture settings.

The 'one stop worse' is in terms of actual TTL metered exposures - the Nikon tends to produce a lighter result. It can be calibrated by the user to match any meter, and if I was using this body, I would be programming in a -2/3 calibration for exposure permanently.

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
David Kilpatrick said:
I have done controlled tests now and the Nikon sensitivity is either
1/2-2/3rds lower than the Alpha, or four of their most expensive
lenses have completely mis-calibrated aperture settings.

The 'one stop worse' is in terms of actual TTL metered exposures -
the Nikon tends to produce a lighter result. It can be calibrated by
the user to match any meter, and if I was using this body, I would be
programming in a -2/3 calibration for exposure permanently.
Click to expand...
David does your recommendation apply to RAW captures or only Jpegs ? Many have observed that slight over-exposure at high ISOs often results in less noise when any over-exposure is corrected in RAW conversion.

Keith-C
 
KEITH-C said:
David does your recommendation apply to RAW captures or only Jpegs ?
Many have observed that slight over-exposure at high ISOs often
results in less noise when any over-exposure is corrected in RAW
conversion.
Click to expand...
I have shot a load of 14 and 12 bit RAWs but I don't yet have a proper processor to handle them. Unfortunately, ACR4.2 mis-exposes A700 raws by +1 stop relative to JPEG densities which is very odd. If it does the same with D300 files they will need -2 correction in the ACR controls.

Some but not all of the Nikon low noise may be down to allowing fuller exposure and a sensitivity on the low side in true ISO terms.

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
OMG, the poor A700 is not only noisy and unsharp but also badly calibrated. Sensitivity is perfectly and objectively measurable, what a shame that Sony can't do that...
 
MCsaba said:
OMG, the poor A700 is not only noisy and unsharp but also badly
calibrated. Sensitivity is perfectly and objectively measurable, what
a shame that Sony can't do that...
Click to expand...
?????

You may want to read the article carefully again

--
Bernard

AS/SSS rocks!
lens reviews and more on dyxum.com!
 
Done:
"In Nikon’s exposure and sensitivity terms the Alpha runs to 12,800".

But I'm still not sure. Is there a sensitivity "in Nikon terms" ????? That don't make sense to me. ISO is an international standard. At least it was in the past. Anyway, the conclusion is that A700 is doing 12800 instead of 6400.
ab012_KM7D said:
You may want to read the article carefully again
Click to expand...
 
MCsaba said:
Done:
"In Nikon’s exposure and sensitivity terms the Alpha runs to 12,800".

But I'm still not sure. Is there a sensitivity "in Nikon terms" ?????
That don't make sense to me. ISO is an international standard. At
least it was in the past. Anyway, the conclusion is that A700 is
doing 12800 instead of 6400.
Click to expand...
I should have said it the other way round - in Sony's terms, the Nikon is only doing 3200 at the 6400 setting.

Digital sensitvity can not be expressed in ISO. That's why Nikon won't even use an ISO rating for 100 and 6400. At the best, digital sensitivity is variable and subject dependent. It changes with AWB apart from anything else. The ISO figures are just an approximation, to aid when comparing with film, or using a lightmeter.

David

--
Photoclubalpha - the afterlife of the Minolta Club, post APOcalypse...
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top