Jump to content




The above adverts really do help Fotozones. Please click on them if they are relevant to you. Not seeing them? Just exclude Fotozones from your ad blocker. Thanks!


Photo
- - - - -

Fujirumors: Fuji X-Trans support in Capture One Pro not far away


  • Please log in to reply
20 replies to this topic

#1 Lars Hansen

Lars Hansen

    Advanced Member

  • Life Member
  • 647 posts
  • LocationDenmark
  • Edit my pics?:Ask Me

Posted 07 December 2012 - 21:33

http://www.fujirumor...-series-support

Interesting to see if they can surpass the free software Fred Nirque has been writing about :-)

Edited by Lars Hansen, 07 December 2012 - 21:35 .


#2 aerobat

aerobat

    Advanced Member

  • Life Member
  • 550 posts
  • LocationSwitzerland

Posted 07 December 2012 - 22:04

Hi Lars,

This sounds like xmas and easter in one go. These superb cameras deserve an adequate RAW converter. Phase One is certainly one of the very capable companies to do it.

Regards Daniel

#3 schwett

schwett

    Advanced Member

  • Life Member
  • 1,227 posts
  • LocationSan Francisco
  • Edit my pics?:Ask Me

Posted 07 December 2012 - 22:20

as noted, the question isn't whether it will be supported.... it's "supported" in adobe camera raw - and many other converters. the output looks like junk.

nobody has really demonstrated that the x-trans sensor CAN be properly demosaiced without introducing artifacts far, far worse than the problem it was intended to cure (moire). i enjoy my x-e1 but it would be a far better camera with the same basic technology but a bayer array over the sensor.
http://photo.sfmthd.org/ [under construction]

#4 Alan7140

Alan7140

    Advanced Member

  • Life Member
  • 5,623 posts
  • LocationTasmania, Australia
  • Edit my pics?:Ask Me

Posted 07 December 2012 - 23:17

http://www.fujirumor...-series-support

Interesting to see if they can surpass the free software Fred Nirque has been writing about :-)


The more I use Helicon Filter, and the more obscure things I discover and press into use, the more I think that to justify the price C1Pro will have to be a bloody revelation rather than just another X-Trans processor to justify that extravagant price tag. Fact is that for all but the most idiotic exposure stuff-ups, Helicon does a simply extraordinary job of initial processing of the X-Pro1's RAF files, and has a very comprehensive set of additional controls to adjust those stuff-ups, even to the ability to modify the basic processing in DCRaw via a GUI interface to the command line. Even ACR doesn't have a facility like that (if it did all the current drama would be a moot point, as individual users would have a way to opt out of Adobe's currently non-optional Van Gogh painterly processing filter, which is still there in the latest ACR 7.3 RC).

Plus it only costs $75, with a 30-day unrestricted demo available, which anybody thinking of buying C1Pro just for the X-Trans raw processing would be nuts not to try first. You could then buy nearly two thirds of a Samyang 8mm fisheye with the money saved.

#5 schwett

schwett

    Advanced Member

  • Life Member
  • 1,227 posts
  • LocationSan Francisco
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 00:32

....if it did all the current drama would be a moot point, as individual users would have a way to opt out of Adobe's currently non-optional Van Gogh painterly processing filter, which is still there in the latest ACR 7.3 RC....


only if there actually was a tool available that could easily plug into it!
http://photo.sfmthd.org/ [under construction]

#6 bjornthun

bjornthun

    Advanced Member

  • Life Member
  • 2,963 posts
  • LocationFredrikstad, Norway
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 01:41

If they have a proper de-mosaic algorithm for the Fuji X-trans sensor, you could have a winner. C1, in my opinion ,does a better job with Leica M8 files and with files shot with PC lenses than does LR. C1 allows you to tell where the optical axis of a PC lens is, i.e. How much you have shifted the lens. On M8 files there are fewer artifacts than with LR, remember M8 has no AA filter.
Bjørn T

#7 Lars Hansen

Lars Hansen

    Advanced Member

  • Life Member
  • 647 posts
  • LocationDenmark
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 12:24

This sounds like xmas and easter in one go. These superb cameras deserve an adequate RAW converter. Phase One is certainly one of the very capable companies to do it.


Daniel - I agree. I already have C1 and if they have succeeded to create a good converter it would certainly remove one of my last excuses for sitting on the fence with regard to buying the X-E1.

The more I use Helicon Filter, and the more obscure things I discover and press into use, the more I think that to justify the price C1Pro will have to be a bloody revelation rather than just another X-Trans processor to justify that extravagant price tag.
......
Plus it only costs $75, with a 30-day unrestricted demo available, which anybody thinking of buying C1Pro just for the X-Trans raw processing would be nuts not to try first. You could then buy nearly two thirds of a Samyang 8mm fisheye with the money saved.


Fred - as mentioned above I already have C1 and bought it with 50% discount - ended up around the same price tag as Adobes LightRoom I think. An C1 upgrade is priced approx. the same as the Helicon Filter so I just hope C1 will be at the same level at Helicon etc.


Bjørn and Schwett - I'm not a techie but realise that this sensor has it's pro's and con's. For me it's a question of having a good RAW support before buying. I'm not going to print big and I simply like the camera and it's interface plus the lenses Fuji made.

Edited by Lars Hansen, 08 December 2012 - 12:36 .


#8 Andrea B.

Andrea B.

    Lost in the Desert

  • Administrators
  • 11,096 posts
  • LocationNuovo Jersey
  • Edit my pics?:Yes

Posted 08 December 2012 - 16:48

Lloyd Chambers has been writing about the "fractal" artifacts of XPro and XE1 files. I don't subscribe currently to his in depth analyses, but his calling attention to it on his blog pages and showing examples there might help kick Fuji into action. Not holding my breath on that however.

Diglloyd Link: http://diglloyd.com/...-artifacts.html

Andrea B.
UltravioletPhotography.com


#9 Andrea B.

Andrea B.

    Lost in the Desert

  • Administrators
  • 11,096 posts
  • LocationNuovo Jersey
  • Edit my pics?:Yes

Posted 08 December 2012 - 16:59

There is currently a 20% off Christmas sale for various Helicon products & bundles.
Look for the little red Santa tag in the upper left if the sale offer doesnt come up initially.
http://www.heliconso...m/purchase.html

Andrea B.
UltravioletPhotography.com


#10 Lars Hansen

Lars Hansen

    Advanced Member

  • Life Member
  • 647 posts
  • LocationDenmark
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 19:10

Lloyd Chambers has been writing about the "fractal" artifacts of XPro and XE1 files.


That looks nasty. As I understand Adobe is working on improving their X-Trans converter - or do you know whether the results shown are from their "improved" version?

Hopefully Phase One and/or Adobe will be able to improve this to an extent that will at least reduce these artefacts to a level that is only of academic interest. I'm trying to get rid of my (bad) habit of pixel peeping, but this stuff probably impact IQ above pixel level.

#11 schwett

schwett

    Advanced Member

  • Life Member
  • 1,227 posts
  • LocationSan Francisco
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 21:53

it certainly does affect image quality when printed at moderate sizes or even when cropped and shared online for large displays. if you downsample about 50% it more or less goes away. the images are unusable for any critical purposes unless they're downsampled.

the fact that it appears in JPGs and in ACR output (to varying degrees) is the main reason i'm not fully satisfied with the camera. the dcraw results looks promising which is the *only* reason the camera isn't on ebay!!
http://photo.sfmthd.org/ [under construction]

#12 Alan7140

Alan7140

    Advanced Member

  • Life Member
  • 5,623 posts
  • LocationTasmania, Australia
  • Edit my pics?:Ask Me

Posted 08 December 2012 - 23:46

Thanks for the link, Andrea - I was still running the demo Helicon Filter, but now own the perpetual licence as it qualified for the Christmas discount. A once-off $60 for this program is a bargain (or maybe an indication of what such programs which run a GUI on a GNU free command line program should cost: PTGUI and AutoPano Giga should take note...).

As for the Diglloyd article, it really only illustrates what people have been bitching about for months now, nothing new there at all. ACR, whether in Bridge or in LR, and SilkyPix, are crap in processing X-Trans files. And yes, pixel-peeping reveals the problem starkly, and equally, yes, as schwett points out, it does have an effect on "normal" sized prints or screen-views. However, not really enough was made of the colour smearing that goes on as well as the fractal-type effect that seems to preoccupy discussion - as far as affecting the visual appearance of regular sized prints this has a far greater influence than the painterly fractals effect.

Interesting was the feedback comment published in that article where the writer blames Fuji for ACR's crap processing. Extraordinary transfer of blame there. Not to absolve Fuji for releasing hardware with inadequate software support (you'd think they would have learned that lesson by now), however Adobe is just as culpable for releasing something inadequate, which ACR most certainly is.

Edited by Fred Nirque, 08 December 2012 - 23:56 .


#13 schwett

schwett

    Advanced Member

  • Life Member
  • 1,227 posts
  • LocationSan Francisco
  • Edit my pics?:Ask Me

Posted 09 December 2012 - 06:36

if the effect was absent from the jpgs, I think the blame would fall squarely on adobe. But it's not, which suggests that a more optimal demosaicing routine is beyond the reach even of fuji's own engineers with all the time eh ad to develop the algorithms during the development of the camera. this suggests a flaw in the basic premise of the product, or some I solved problems which accompanied the product to market for all comers. we may yet see it fixed - by adobe or fuji or whoever, but one should really not expect adobe to reinvent their entire image processing strategy for a small handful of cameras. the majority of ACR users don't have an x-trans... But given the poor detail in the jpgs, the majority of x-trans owners need a raw converter!

I really wish they had just used a bayer array and left off the AA filter.
http://photo.sfmthd.org/ [under construction]

#14 Alan7140

Alan7140

    Advanced Member

  • Life Member
  • 5,623 posts
  • LocationTasmania, Australia
  • Edit my pics?:Ask Me

Posted 09 December 2012 - 07:11

There is a better raw converter than the lot of them, based on DCRaw, and I'm using it. The more I get used to Helicon Filter, the more it pleases me, particularly having found the way to do batch processing.

Some will gripe about the zippering on very hi-con edges, but that's literally 1x1 pixel wide, which is really of no consequence in the actual real-life use of photographs. The noise reduction in Helicon is excellent as well, far better than in either ACR, Photoshop or DFine 2.0, for that matter. Plus there are still things I'm discovering. The only thing I'm not overly happy with thus far is the way it deals with blown highlights, but I'm willing to blame user-error for this as I really haven't worked through everything yet.

Just because both Fuji and at their behest SilkyPix in concert couldn't come up with a decent processing algorithm doesn't mean it can't be done. Dave Coffin didn't have to rewrite his DCRaw program in its entirety virtually proving that Adobe would not have to do a bare-bones rewrite to get a passable result, so it would rather appear that Adobe simply stuffed it up royally and is now ignoring the problem, hoping it will go away.

Fine by me, it's just another step closer to the day I can turn my back on Adobe and use better tools for my purposes than the now overly arrogant and self-assured Adobe can provide. I've already been using 3rd party programs for panos and stacking, and with only minor improvements and a bit more experience with use I can see that Helicon Filter might well be my raw processor of choice for my NEF files as well. The thought that I also now have a license in perpetuity for the paltry sum of $60 further pleases me. No more forced, expensive, price-gouging upgrades of the entire Photoshop program just to have a new model camera supported (we pay 50% more for Adobe products for no other reason than we are not the USA, something which our government is actually holding an inquiry on right now, with Adobe so far being the main company mentioned by name in this). Helicon may not be as quick off the mark in camera support, but being quick and shit doesn't cut the mustard, either.

As I have an X-Pro1 which I enjoy using immensely, and I now have a way of processing the files to a better than passable standard, I'll be damned if being ignored by a corporation such as Adobe will force me away from the X.

As an aside, the promised delivery of 14/2.8 lenses by Fuji did not happen (surprised? - not really), which led me down the path of substitution, (I need a wide angle other than the fisheye, the 16-35 AFS is a pig to use on the Fuji), and for versatility parting with only $400 gets me a Samyang 14/2.8 which will work on both the D3s & the Fuji, and is chipped for the Nikon as well. Problem solved, at the expense of Fuji not selling a $900 lens to me. I'm getting beyond fed-up with manufacturers promising things and not delivering without explanation, and thinking that this is somehow OK. At some stage customers will gain the upper hand again, just as happened in the 1970's & '80's.

#15 aerobat

aerobat

    Advanced Member

  • Life Member
  • 550 posts
  • LocationSwitzerland

Posted 09 December 2012 - 08:59

Hi Fred, as soon as Helicon Filter supports the X-E1 I'll sure give it a try. The price for the perpetual license is certainly very reasonable.

Regards, Daniel

#16 aerobat

aerobat

    Advanced Member

  • Life Member
  • 550 posts
  • LocationSwitzerland

Posted 23 December 2012 - 22:20

Latest news from the X-Trans RAW converter front:

http://www.fujirumor...ial-unlocked-2/

I finally feel xmas coming :good:

Daniel

#17 Lars Hansen

Lars Hansen

    Advanced Member

  • Life Member
  • 647 posts
  • LocationDenmark
  • Edit my pics?:Ask Me

Posted 27 December 2012 - 13:41

Latest news from the X-Trans RAW converter front:

http://www.fujirumor...ial-unlocked-2/

I finally feel xmas coming :good:

Daniel


And here is some pixel peeping for the Christmas holiday - Fuji RAW conversions made with C1 beta compared to Lightroom:
http://www.seriousco...om-4-2-a-15229/

Edited by Lars Hansen, 27 December 2012 - 13:41 .


#18 Alan7140

Alan7140

    Advanced Member

  • Life Member
  • 5,623 posts
  • LocationTasmania, Australia
  • Edit my pics?:Ask Me

Posted 27 December 2012 - 21:48

In all the discussion I forgot to mention that Helicon Filter has a plug-in to Photoshop (and I presume LR) as well - in other words you can use its image modifiers with any image you are working on in Ps at any time, and without question "sharpen fine details" in the sharpening choices produce the best, most artifact-free digital sharpening I have yet seen. With the X-Pro1 files it sharpens very subtly and in higher ISO shots produces the most pleasing film-like sharp "grain", which pleases me greatly.

As for the samples of C1 - while obviously better than ACR (which really would want to be the case), the comments so far as being a bit better than SilkyPix makes me think that perhaps all is not as wonderful as it should be. DCRAW/(Helicon Filter) has flaws at edges with zippering and spurious pixels in high-contrast areas, but these microscopic drawbacks aside (really only visible at pixel-peeping magnification), the image quality otherwise is hugely better than SilkyPix, and of course leaving ACR completely out of contention.

I was hoping that reports on C1 would universally report things like hugely better, but the absence of such hoped for unanimous praise seems a bit worrying to me. I've even seen comment that though it's better, it's still not doing full credit to the X-Trans files.

The biggest point being made seems to be that it doesn't show the ACR painterly effect, as if this was some sort of a triumph - the ACR mess was of Adobe's own invention, not any flaw in the X-Trans files, and should not be assumed as being something that happens in processing X-Trans raws by default as seems to be the case. Being equal or slightly better than SilkyPix is certainly not something I'd be jumping up and down with joy over. The ideal solution would be DCRAW image quality without the zippered edges and spurious pixels, not just something slightly better than the mushy SilkyPix rendition.

So lots of grains of salt here at the moment - the fanboy comments of some C1 junkies are a bit over the top as well - they don't gel with comments of others who obviously don't have such C1 stars in their eyes. Time will tell when some properly processed images comparing various developers using head-to-head examples of the same files processed are posted.

#19 schwett

schwett

    Advanced Member

  • Life Member
  • 1,227 posts
  • LocationSan Francisco
  • Edit my pics?:Ask Me

Posted 27 December 2012 - 23:55

...The biggest point being made seems to be that it doesn't show the ACR painterly effect, as if this was some sort of a triumph - the ACR mess was of Adobe's own invention, not any flaw in the X-Trans files, and should not be assumed as being something that happens in processing X-Trans raws by default as seems to be the case. Being equal or slightly better than SilkyPix is certainly not something I'd be jumping up and down with joy over. The ideal solution would be DCRAW image quality without the zippered edges and spurious pixels, not just something slightly better than the mushy SilkyPix rendition.


i have to disagree here fred. the mess is not of adobe's invention. fuji invented the x-trans sensor layout, and with it created a massive challenge in demosaicing the files. their own JPGs show the results of that challenge. adobe's result is even poorer in some ways, and silky and all the rest have made their own tradeoffs in solving a problem which appears to be exponentionally more difficult than demosaicing a bayer-matrix raw file.

this is why i haven't been holding out much hope for C1. when so many converters (which produce good results with other types of files!) have failed (dcraw, gc, fuji jpg, silky, acr...) it's pretty obvious that it's a tough problem.

i've finally been able to try dcraw with the x-e1 files. completely not convinced.
http://photo.sfmthd.org/ [under construction]

#20 Alan7140

Alan7140

    Advanced Member

  • Life Member
  • 5,623 posts
  • LocationTasmania, Australia
  • Edit my pics?:Ask Me

Posted 28 December 2012 - 00:56

The 'watercolur' mess is most certainly of Adobe's invention.

Short-cutting by attempting to use a modified Bayer demosaicing with gobs of blur, spreading, contrast-adding and sharpening does not a proper solution make. If you can't do it properly, best not to do it at all, or do it in a time-frame which allows proper development, as so many of the other processors appear to have chosen to do.

DCRAW is not an X-Trans demosaicer. It also uses the Bayer model as a base for its algorithm, just that it doesn't destroy the whole image by bleeding and blurring to kill the spurious pixels and zippered edges visible at 100% as was Adobe's solution. It's those artifacts visible in DCRAW that Adobe tried to kill - if you process the same file in both & pixel-peep, you can see how the Adobe blur-smudge-sharpen has hidden the zippering and spurious pixels, unfortunately the fact that this then stuffs up the image at print size seems to have missed Adobe's attention entirely. Same goes for SilkyPix - you can see the underlying faults in a head-to-head comparison and how the excessive blurring has been used to kill them.

At some stage someone will actually do the math needed to get the X-Trans algorithm right, whereupon this will all be just a bad memory. By the sounds of initial reports, C1 appears to have taken the modification of the conventional algorithm route, but applied modifiers with far more care than Adobe's ham-fisted effort and Silky Pix's blurry solution. As soon as the C1 release is posted I'll obviously be downloading it and directly comparing it against Helicon Filter's DCRAW interpretation, but from the initial underwhelming reports I'm not exactly holding my breath.

Getting away from pixel peeping at 100%, at my normal printing resolution of around 34% the DCRAW problems combined with Helicon Filter's additional image modifiers are all but invisible, and in actual printed form the images are stunning, as there is no colour loss as with ACR & SilkyPix, and the pixel-sized defects become smaller than the inkjet micro-blobs that make them up, in other words, effectively invisible. It has taken an awful lot of time to work out how to use Helicon properly as the effects of similarly named controls are in some cases entirely different to those in ACR, which is the program I know backwards.

So for the moment I'm happy to choose invisible artifacts which have a minimal effect on the final use of the image as opposed to a solution which both kills colour and detail in the final use image. In an ideal world it would be preferable of course to have a perfect image at any viewing size, but as my final use is unsampled printing at either 288 or 360ppi on the Epson 7800, what I see at 96ppi on a 30" screen is of little relevance.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users



Receive a FREE CAMERA BAG from Think Tank Photo


An appeal to all Fotozones visitors: please help me to keep this site going by starting your gear purchases using any one of the affiliate links shown below:

Amazon.com | Amazon.co.uk | Amazon.de | Adorama.com | thinkTank Photo | DigitalRev.com | OWC | B&H or Donate via PayPal

Starting your shopping here doesn't cost you anything more, but by using the links above (or any others found on the site) you are advising the affiliate that you support this website. This results in a small commission that helps with the running costs. If your preferred outlet isn't among those listed above you can also support the site by making a donation of any amount via PayPal (no PayPal account required). Any donation will be most appreciated.