Menu
What's new
The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

New firmware is here

V

vjbelle

Well-known member
ecsh said:
Update is done, but the screen does not come back on, for me to hit the finish button on the screen. Anyone else?
Click to expand...
I didn't hit the finish button on the screen.... That was done on my computer. The process was a little clumsy and I had to exit the program once and restart the update but the software knew I had gone through all of the previous steps and started the update.

Victor
 
nostatic

nostatic

New member
The first one (in the original post) seems to have more definition - transitions from shadow to book and the thin lighter shaft in the upper left.

 
V

vjbelle

Well-known member
The examples show subtle differences at best. I hope I don't have to use that ISO for anything serious..... at least in its current state.

Victor
 
V

Viramati

Member
Not that I use jpegs but I believe I now see much better high iso files with a lot less aggressive noise reduction etc
 
tashley

tashley

Subscriber Member
More opinions please? And don't forget:

1) we're looking really for noise performance not definition because, thought the shots were both carefully manually focussed on the same spot, sublet focus differences might give the impression of more detail.

2) this files was effectively pushed two stops: in other words it 'should' have been shot at ISO800, in which case it would have looked a whole lot better. This despite the fact that it had +2/3rds stop comp applied. The A7R meters extremely protectively of highlights, which is all well and good but it does mean that one can end up giving the sensor less light than one might with another camera.
 
nostatic

nostatic

New member
Well, #1 seems to have less or more controlled noise throughout the frame. And I'm sticking with my definition as I see it in other parts of the image :p

For me iso 3200 is a "good lighting" night. I'm routinely into 6400 and 12800 so if they've manage to improve those files without excessive NR I'm happy.
 
tashley

tashley

Subscriber Member
Well the bad news folks is that I agree, #1 look overall less noisy and the noise is more even. And it's the old firmware :-(
 
Godfrey

Godfrey

Well-known member
I'm sorry, but to me these cameras see in the dark anyway. Have any of you seen ISO 3200 film lately? ;-)

The quality of noise on a +2/3 EV exposure, pushed another two stops, at ISO 3200 isn't really of much use in evaluating performance to me. I'm more interested in what the best they can do with correct exposure might be.

Firmware download has completed. No time to work on installation just at the moment, but I'm ready.

G
 
tashley

tashley

Subscriber Member
Godfrey said:
I'm sorry, but to me these cameras see in the dark anyway. Have any of you seen ISO 3200 film lately? ;-)

The quality of noise on a +2/3 EV exposure, pushed another two stops, at ISO 3200 isn't really of much use in evaluating performance to me. I'm more interested in what the best they can do with correct exposure might be.

Firmware download has completed. No time to work on installation just at the moment, but I'm ready.

G
Click to expand...
I think the problem is, there is no 'correct' exposure. Think about it, this frame is already given + 2/3rds over the 'as metered' (i.e. 'Correct for JPEG') exposure. Full ETTR with the A7r can be three stops, even more, over but to me it seems that the highlights suffer well before they blow.

The crops I show aren't, therefore, boosted two stops in LR to correct them, but to bring out the noise for a comparison...
 
Godfrey

Godfrey

Well-known member
tashley said:
I think the problem is, there is no 'correct' exposure. Think about it, this frame is already given + 2/3rds over the 'as metered' (i.e. 'Correct for JPEG') exposure. Full ETTR with the A7r can be three stops, even more, over but to me it seems that the highlights suffer well before they blow.

The crops I show aren't, therefore, boosted two stops in LR to correct them, but to bring out the noise for a comparison...
Click to expand...
Well, we will disagree. To me, correct exposure is the point at which the interesting highlight bits in a photo do not 'suffer' from saturation/compression while as much data at the dark end of the DR is collected. If the subject DR fits wholly within the DR of the sensor, great, otherwise you decide what you lose and where.

Artificially pushing the gain up to reveal a noise characteristic is interesting, yes, but it's the noise characteristic of a severely bent up dynamic range. While it can be curious and even interesting to work in that characteristic from time to time for aesthetic reasons, I don't really consider it indicative of the sensor output in the 'correct' exposure range it was designed for.

Just a difference of opinion...

G
 
tashley

tashley

Subscriber Member
Hmmm, like a terrier with trouser leg in his teeth I'll have another go: it certainly isn't 'fair' or indicative of a sensor's performance to push it around like this if you want to see how noisy it is: but I do think that if you want to see whether a sensor has more or less noise with a different firmware version, it is useful. Sort of like making an X-Ray of an oil painting to see if it's a palimpsest! The X-Ray isn't there to show you how beautiful the painting is.
 
You must log in or register to reply here.
Top