Just for New Users.. A700 High ISO NR Why Adobe Tests are not valid.

[Ehrik]

Jeez, I just spent the last day on another more "professional" forum,
and this RAW NR was a topic of conversation. The difference was,
they were talking about the difference in NR between Nikon and Canon.
They all acknowledge that both cameras do NR on RAW

I think they confuse NR in the raw converter they are using with NR
on the
raw files. A number of people have analysed Nikon and Canon raws
(prior to
conversion) and found no NR smearing on them.
...

Nope, definitely not. It was all about RAW NR in camera.

Yes, but they judged it by looking at the output from their raw converters,
didn't they? They didn't analyse the raw data directly using mathematical
methods.

Just my two oere
Erik from Sweden

 

[WaltKnapp]

The question is - why if the files should be so similar is one
algorithm used for Sony, and another for Nikon? Are the raw files
somehow instructing ACR to vary its presets? Or are they genuinely
different?

Of course the algorithm is different. Otherwise there would be no need to issue new raw converters for each model of camera.

ACR reads the data in the RAW file as to camera type, then adobe has a set of presets for each model of camera it uses. Clearly they are using quite different presets on the two brands.

What you might try is hack into the RAW files and change the camera type data. Give Adobe an a700 file that is marked Nikon D300 internally and vice versa. The Canon folks did some interesting experiments along these lines when ACR was doing the same as the a700 to their files. You won't get ideal processing as the cameras really are different, but you'll get a better idea what's going on.

Walt

 

[MrScary]

--Well KEN---Thanks for poping in n showing us how Bibble works on the a700 RAW...Looks GOOD to me...You just might be proveing to me that the a700 is not as bad as I thought....How do the images look from iso200 to iso800...Are the far better than the KM5D & 7d.....
Thanks Ken5D..

MrScary (DennisR)
Swansea, Wales. UK

http://www.pbase.com/dennisr
http://community.webshots.com/user/mrscarecrow
http://www.russ4to.KMD5DImages.photoshare.co.nz

 

[b shaw]

Ken,

These were taken with firmware v.02 and with DRO + 3. I think it might be instructive to re-shoot the same still life with v.03 and (as we've learned since your original shoot), turn off DRO for the shot.

Obviously, do this only if you want, since you are doing us a favor for posting these in the first place.

--
b shaw

http://bshaws.blogspot.com/

 

[David Kilpatrick]

Ken,

These were taken with firmware v.02 and with DRO + 3. I think it
might be instructive to re-shoot the same still life with v.03 and
(as we've learned since your original shoot), turn off DRO for the
shot.

Obviously, do this only if you want, since you are doing us a favor
for posting these in the first place.

I don't think Ken needs to as v3 does not change that aspect of the files - at least officially. DRO may reduce exposure, and they do look a touch dark.

It might be instructive to have some different coloured objects in a test. For many reasons. We need to try some colour-specific test targets.

David

--
Publishing & Editing Photoworld (photoclubalpha.com) and Master Photo Digital
Currently writing for f2 and the BJP

 

[b shaw]

David,

Sorry, I should have been more complete. This particular image was taken at ISO 3200. And, as you know, according to Sony the upgrade "Reduced noise for pictures taken at ISO 3200"

Then, it would be instructive to see how would the various converters handle the noise with the new firmware.

--
b shaw

http://bshaws.blogspot.com/

 

[lonewolf69]

Dear ALL,

So then Bibble is the pen-ultimate solution then to the HIGH ISO problems that we've been seeing? Sorry for beating a dead horse but everyone especially Barry and Bill are avoiding answering this question?

I've posted and posted many times requesting if the END results of the Bibble is satisfactory and if NOT and IF they still have a beef with the NR to explain in detail. I will boldly assume then Bibble is the HOLY GRAIL for processing HIGH ISO (1600 & up) A700 RAW files.

Unless someone wants to still "fight the fight" about NR and HGH ISO A700 RAW files....

Again, thank you VERY VERY much Ken!

Bookmarking this thread right now!!!
--
-Alex

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

http://www.pbase.com/lonewolf69

 

[WaltKnapp]

Ken, I am also trying to become enlightened on this subject since I
shoot RAW almost exclusively on my A100 and an considering upgrading
to the A700. You stated that you have heard second hand reports that
Adobe has not spent much time with the A700 RAW. Have you spoken with
anyone at Adobe to verify that they are applying NR to the A700 RAW
even when the NR slider is set to zero? What I don't understand is
why of all the cameras that use RAW, only the A700 seems to have this
problem with the Adobe RAW conversion. Why would they single out the
A700? If they indeed are doing uncalled for NR, then they should be
made aware of the fact that it may well cost them a good number of
sales.

Last fall the Canon folks were up in arms over Adobe doing the same thing to Canon Raws, particularily the 40D. They were getting a somewhat milder form of what Adobe does to a700 RAW. In fact they and Adobe forums traced this in some form to all Bayer pattern sensors operating at ISO 1600 or above. There was extensive discussion in both Canon and Adobe forums. They understood the exact problem far better than anything I've seen discussed in this forum. Adobe appears to have fixed the problem with Canon, but not a700. They did not stop doing NR ahead of user settings, just toned it down until the complaints got low enough. So, you see, Adobe did not single out the a700, they more like ignored it and did not care to deal with the problem.

And I did point this all out last fall. Since Sony's own raw converter was designed by or in cooperation with Adobe it's also suspect. Many others use the same core software components as Adobe.

We really don't have a raw converter that's specifically designed for a700 RAW files. And until we do what we can get is speculation. Bibble works well, but it's approach is generic rather than a700 specific.

I would say that at this point probably many of the tests that have been done on the a700 may be just as suspect as Adobe images.

Sony did, to Ken, verify what I'd worked out, that the RAW files really are what comes out the output lines of the sensor. No more talk of the Bionz cooking the raw files in Raw output please. It does a pretty good job when it does cook the raw as part of processing the in camera jpegs.

I believe the definition of RAW is what comes out the output of the A/D lines stored in a file. And that appears to be exactly what Sony puts out.

The a700 is a nice camera, and for those that try poduces some good images. But there is a lot to learn.

Walt

 

[Chadd Thompson]

Really software can't have an effect in how it demosiaces and process
it.. tell me why they same file with both NR and sharpening zeroed
has a finer noise pattern in Bibble then Adobe.. Magic? or did Sony
put a bug in the RAW to make Adobe smooth out the detail..

Kiklop, I'm with Ken on this one, because the Bibble results are
simply beautiful...? Maybe it's like Ken says it's magic?
Bottomline, IF Bibble can do this, Adobe should be able to as well,
but since it's NOT happening.

One might draw the conclusion that Adobe doesn't care, or that they
simply did not put enough effort to optimize the A700 Raw conversion.
What's your take, and please do reply with your view of the Bibble
files? Please??

Thank you,

--
-Alex

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

http://www.pbase.com/lonewolf69

--
Thanks,
Chadd
http://www.pbase.com/chadd

 

[lonewolf69]

Chadd,

???

--
-Alex

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

http://www.pbase.com/lonewolf69

 

[lonewolf69]

RJSanchez wrote:
Last fall the Canon folks were up in arms over Adobe doing the same
thing to Canon Raws, particularily the 40D. They were getting a
somewhat milder form of what Adobe does to a700 RAW. In fact they
and Adobe forums traced this in some form to all Bayer pattern
sensors operating at ISO 1600 or above. There was extensive
discussion in both Canon and Adobe forums. They understood the exact
problem far better than anything I've seen discussed in this forum.
Adobe appears to have fixed the problem with Canon, but not a700.
They did not stop doing NR ahead of user settings, just toned it down
until the complaints got low enough. So, you see, Adobe did not
single out the a700, they more like ignored it and did not care to
deal with the problem.

And I did point this all out last fall. Since Sony's own raw
converter was designed by or in cooperation with Adobe it's also
suspect. Many others use the same core software components as Adobe.

We really don't have a raw converter that's specifically designed for
a700 RAW files. And until we do what we can get is speculation.
Bibble works well, but it's approach is generic rather than a700
specific.

I would say that at this point probably many of the tests that have
been done on the a700 may be just as suspect as Adobe images.

Walt, RJSanchez,

There are quite a few comments by some of the A700 owners here who were STUNNED at the JPG output from the A700, and were less than enthused by the RAW files, to the point that they were comtemplating switching their workflow to be JPG only (some PROBABLY already have...), might this revelation have an impact on those users now??!?? I remember people saying stuff like:

"...Wow! JPGs are better than RAW..."
"...no longer need to shoot RAW anymore..."

This is getting very interesting indeed...

--
-Alex

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

http://www.pbase.com/lonewolf69

 

[Ken_5D]

Really software can't have an effect in how it demosiaces and process
it.. tell me why they same file with both NR and sharpening zeroed
has a finer noise pattern in Bibble then Adobe.. Magic? or did Sony
put a bug in the RAW to make Adobe smooth out the detail..

Kiklop, I'm with Ken on this one, because the Bibble results are
simply beautiful...? Maybe it's like Ken says it's magic?
Bottomline, IF Bibble can do this, Adobe should be able to as well,
but since it's NOT happening.

Part of the problem is both Sony and Adobe decided that there should be a bit more NR below the users control so they collide badly. I am not saying that the A700 doesn't have some extra NR at ISO 1600 even in Low mode. It does, enough number tests show the noise curve changes. etc.

But Sony gets blamed a lot because of the output choices of the RAW converter.

If adobe is tweaking out put by camera.. they should just back off totally on the NR for Sony.

The one thing that did come out of Phil's response to Barry's reaction to my first posts was that even the D300 gets better results (more native noise and less NR out of Bibble)

I am all for and advocated to Mark, like a calmer versions of Barry for choice at PMA..

But the same people need to be making the same stink about too much NR on the Abobe boards since I don't think they give a darn about us or even PMA. Adn frankly there were moments where while Mark was doing his job as a representative of the Alpha line, I think he gets it.. and I hope can get the message to teh design team.

Of course their next problem is how to give choice without reviewers assuming all thier tests should be with most of Sony's innovation turned off.

One might draw the conclusion that Adobe doesn't care, or that they
simply did not put enough effort to optimize the A700 Raw conversion.
What's your take, and please do reply with your view of the Bibble
files? Please??

Thank you,

--
-Alex

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

http://www.pbase.com/lonewolf69

--
------------
Ken - Happy A700 Owner
http://www.cascadephotoworks.com

 

[b shaw]

No fight, just note, there are a few Converter algorithms available for use.

Frank Holub has been developing a free RAW conversion tool, RAWHide, with emphasis on support for the A-mount. Here is what he says from his site:

"NO single routine (not even RAWHide's) can perfectly interpolate this
data every time. All Routines have trade-offs. "

http://www.my-spot.com/RHC/RHC_Demosiac.htm

The other component to regard, separate from raw conversion, is Noise Reduction. The very popular tool, Noise Ninja, does an outstanding job. A partial version is included in Bibble. I use a different, equally outstanding tool called NeatImage Pro, and I have been able to get equally outstanding results.

Yesterday, I compared the two, hands-on, and I can get about the same results with either of the tools and each tool has strengths unique to itself. So, one of those tools (or any of the other competent tools), would be useful. If you go with Bibble, consider going with NoiseNinja since it's integrated.

By the way, if you consider Bibble the pen-ultimate, which tool is the antipenultimate and which is the ultimate tool? ;-)

--
b shaw

http://bshaws.blogspot.com/

 

[Ken_5D]

--Well KEN---Thanks for poping in n showing us how Bibble works on
the a700 RAW...Looks GOOD to me...You just might be proveing to me
that the a700 is not as bad as I thought....How do the images look
from iso200 to iso800...Are the far better than the KM5D & 7d.....
Thanks Ken5D..

MrScary (DennisR)
Swansea, Wales. UK

ISo 1250 and down.. Adobe is good enough if you are careful of the detail slider in the sharpening section. I can still start the water color effect. bibble still gets more detail so if you are going to print larger than A3+ Bibble beats LR 1.3.1

1600 and above I right now often use ACR 3.4 or Raw Therapee... I am thinking about going back and buying Bibble.. If I were shooting alot of high ISO it would be a very strong need David K has some issues with aritifacts in some situations he is hoping Bibble will clean up.. I really appreciate his pushing them on this.

A700 gets you more keepers because of more DR, better focus, even in low light and faster response time.

At lower ISO there is a bit more noise in the sky... but easily fixed.. but it is there.

Also DRO will bring up noise where the shadows fall at the low end of the DR in jpgs. The give back is Sony has put alot of the DR in the highlight end so it is much harder to blow out the highlights with the A700 than the KM 5D.

At higher ISO... even with adobe 6MP scaled images from the A700 are slightly better than the KM 5d (so should also be for the 7D) . But because the rest of the camera handels so much better in low light.. it means the A700 is a better low ISO camera in my opinion.

There are some minor noise costs for the 12 MP that may be because Sony is backing off the NR on low ISO more. But as I said.. It takes a minor tweak of the simplest NR to make the skies clean again. The jpgs will be great because the Bionz is a really great image processor.

http://www.pbase.com/dennisr
http://community.webshots.com/user/mrscarecrow
http://www.russ4to.KMD5DImages.photoshare.co.nz

--
------------
Ken - Happy A700 Owner
http://www.cascadephotoworks.com

 

[Ken_5D]

Ken,

These were taken with firmware v.02 and with DRO + 3. I think it
might be instructive to re-shoot the same still life with v.03 and
(as we've learned since your original shoot), turn off DRO for the
shot.

Obviously, do this only if you want, since you are doing us a favor
for posting these in the first place.

Thanks.. right now my trial of Bibble is out and I am on the fence only because I don't shoot a lot of high ISO right now and am getting by with LR for ease of use.

V3 for all my reading was only a locking up type of Bug fix all the image stuff was in V2 but the way they wrote the notes to include the V2 it sometimes reads like there is more IQ stuff in V3.. anyone know different speak up?

These are from RAW so DRO can affect exposure about 1/3 EV.. as it makes room for DRO.. But this shot is actually accurate for what the eye would see when shot.. The office is very dark at night.. I have 150 watts ceiling directed and 60 watts under a stained glass shade. I was actually looking to shoot to see noise in bad lighting vs brighter "low light" like sports etc)

--
b shaw

http://bshaws.blogspot.com/

--
------------
Ken - Happy A700 Owner
http://www.cascadephotoworks.com

 

[b shaw]

I don't have LR ... so, if you want to send me some ARW files, I'll convert.

Moreover, I'll convert with about 5 other RAW converters (from RAWHide) and dcraw as integrated with RAWTherapee.

Now that I re-read Sony's site, the high ISO improvement was supposed to be between v1 and v2. However, it would still be interesting to see with v3

--
b shaw

http://bshaws.blogspot.com/

 

[b shaw]

Thats what I get trying to type and read the sony support site, while watching a big sporting event in the US. Got it. v02 had the noise improvement.

--
b shaw

http://bshaws.blogspot.com/

 

[Ken_5D]

I don't have LR ... so, if you want to send me some ARW files, I'll
convert.

Moreover, I'll convert with about 5 other RAW converters (from
RAWHide) and dcraw as integrated with RAWTherapee.

Now that I re-read Sony's site, the high ISO improvement was supposed
to be between v1 and v2. However, it would still be interesting to
see with v3

What I would like to do is take some of David's great studio tests and do some comparisons... he had both D300 and A700 files.

One think I need to try to do is normalize WB Bibble seems to come in warmer than other converters.

I am not going to have time to shoot more tests this week.. Work that pays for this will be pretty heavy.. I will keep this in mind or if David wants to let us try some of his great RAW files.

--
b shaw

http://bshaws.blogspot.com/

--
------------
Ken - Happy A700 Owner
http://www.cascadephotoworks.com

 

[b shaw]

... I'm traveling for work this week, but the offer stands to help in anyway I can.

--
b shaw

http://bshaws.blogspot.com/

 

[David Kilpatrick]

I have already had issues with exceeded file space the raw files for these are huge - approaching 20MB each.

A full set of raws will be 100-140MB per camera.

I'll look info putting some on my own machine rather than our externally hosted websites, and using a download caching alias to shift the load onto nyud.net

David

--
Publishing & Editing Photoworld (photoclubalpha.com) and Master Photo Digital
Currently writing for f2 and the BJP