C22: HDR / panos
Hi,
When C22 was announced, I was quite enthusiastic about the 2 main features of this release: HDR merging and panorama stitching. However, after having used C22 for some time now, I’m a bit frustrated (so say it politely)…. .
The stitching is far from perfect as I see too many times ugly patterns in the pano, that cannot be found in the individual RAW files or areas that do not fit together even if the overlap is approx 30% as recommended; if I use the same images with other software (Affinity Photo, Autopano Pro, et al.) to stitch a pano I don’t have any problems whatsoever (of course I don’t have a dng, but the stitching is perfect….)
In HDR the lack of a de-ghosting functionality creates strange patterns when there is movement such as clouds, and what is even more weird is that the newly created HDR .dng file show some distortion as well as chromatic aberration that do not exist in the 3 source files. All HDR images were taken on a tripod without any (physical) movement of / touching the camera between the shots.
I don’t know about the experience of other users, but I think these 2 areas of the software need a major re-work.
Best,
Mark
p.s.
I don't like buying "banana-ware" and yes, I know that a .0 version of a software might not be perfect but this version was rushed out to customers without being tested properly. And the software crashed quite often.
I use Capture One since version 8 but I have never seen such a bad software upgrade from them....otherwise I love the software and never want to go back to Adobe LR.
-
This lack of performances for these two new functionalities has been reported many times here and there. We just have to wait for Capture One to improve them in such a way that they could compete with basic (and free) HDR or Pano stitching softwares.
1 -
Performance - at least to me - is related to speed. The new functionalities do not work at all, because the results are unusable. I basically paid for these 2 functionalities and expect that they work when the new software version was launched - HDR and panos were the 2 major selling points!
And, yes, I know that there is nothing else we can do but wait until this is fixed.
1 -
I've only stitched several panos and haven't tried HDR yet. That said, so far most of my panos, including one multi-row, have come out very well. I couldn't figure out why C1 didn't do well on the 1-2 that didn't work. Overall, though, I've gotten much better results from this first C1 pano iteration than I did for the first several from PS years ago.
I suspect that problems may have to do with details in the areas needing blending. I'd suggest that you file a support request and include at least JPEGs of the frames you're trying to blend to help their engineers improve their algorithms.
0 -
I reported HDR pano problems during Beta testing and compared C1 performance to Photomatix, PTgui, etc. C1 HDR performance was far below the performance of the other apps
I have used many stitching programs over the years and while the C1 version works, there are two problems: None of the other apps required adjusting light fall-off before stitching (not a show stopper). However, the absence of being able to manually add control points can be a problem. I have not experienced this problem because I normally overlap my shots between 40 to 45%.
I will continue to use Photomatix for merger and stitch in Affinity until they get it fixed. Workflow is cumbersome but I believe that image quality is more important than easy workflow.
0 -
CAPTURE ONE - WE MAKE YOU USE YOUR IMAGINATION
pay in advance - get a working software later (maybe)
Overall, though, I've gotten much better results from this first C1 pano iteration than I did for the first several from PS years ago.
this explains a lot so c1 never tried to mach state of the art pano/hdr software of today or what user expected but instead what adobe accomplished 10 years ago.
-1 -
Today, S/W is delivered on a schedule regardless of maturity; this philosophy is not limited to C1.
Back in the days of Hollerith cards (I am really old), I led teams that developed S/W, Beta tested S/W, and debugged major S/W products, and over the years, I learned that most problems with complex packages comes from leaving insufficient time for trouble shooting. When I was doing this kind of work, you started Beta testing only when you thought that the product was working well and you want to see if a larger user base would be able to "break it". The purpose of Beta is confirmation that the program is working well not to debug it.
Version 22 entered Beta testing with obvious flaws and leaving two weeks between Beta versions was not adequate to implement any real changes. So the product manager had no real choice (if he did not want to slip the release date) but to release it anyway in order to maintain the schedule.
Their aspirations got ahead of their ability to deliver!
0 -
i understand this but I still think this is totally wrong. everything now follows a date marketing and management has determinate and the consumer pay the price. but not everyone in the software world operates this way as there are still many companies which values their customer, adobe is a very good example for this. in my view c1 is a rotten company the spirit which made it successful many years ago is gone. this job hopping CEO now braggs 1/3 of the employees are hired last year, well everybody familiar with corporates knows this is not something you want to communicate but explains why things are so bad.
0 -
You are correct, not every S/W company is operating this way. But it is not only C1 that is doing this, the MBA philosophy is pervasive.
BTW, it used to be an auto industry standard to sell cars and hope that the buyers would not keep coming back with problems. The Japanese car makers put an end to that way of doing business
0 -
The difference between a manager and an administrator is that a manager provides value added technical guidance based on experience while an administrator makes decisions based on an algorithm. Master of Business Administration (MBA) sys it all.
1 -
for a company which labels itself the "leader" in imaging software this is a total disgrace. you do not have to be a fortuneteller to understand that many user will skip the 22 update because of this and the insane upgrade prices.
"We are witnessing the emergence of a multi-device, collaborative photography workflow, including Cloud, Mobile, Social and Collaboration platforms, and we understand"
seems the exploited interns have the say now....
0 -
Anyway, it seems to me photo software companies are struggling to innovate in order to keep their customers, but is there anything more to go through in that field ? Beyond the HDR/Pano functionalities, which are a real disaster with the released 22 version, the only way for such companies to survive seems to be the annual or monthly subscription....
0 -
even with a subscription everybody except maybe adobe needs to innovate to keep their customer and to accomplish this you need a management which understands the industry but also inspired and talented developer both seem missing now. phase one started as a very innovative company in the 90ies now they are the opposite copying old adobe ideas with lot of marketing BS will not even persuade gen z photographer.
0 -
I agree with Robert.
Most of the options one can add to photo-processing software will have existed in some other application for a decade or more. Thus it is quite difficult to come up with anything that will prove to be impressive and "new".
And of course whilst some new features may be impressive to those who do not use them regularly or rigorously, those who already use such applications a lot will likely find the newcomer wanting.
Only new tech from the camera manufacturers pushes the need for technical innovation in the guts of the system. But that is so often buried deep and not visible other than by its absence when a manufacturer comes up with yet another way of compressing RAW files or shifting pixels without any guarantee they will continue that experiment into the future. (Even if they survive the mobile phone onslaught that has been changing their market.)
The one thing that is a big "buzz", currently, for younger generations is the concept of Artificial Intelligence in the form of "machine learning". A danger that comes with heading down that route is the presumption a starting set of data sources from which the system might learn will be adequate to provide a starting point that the target users find acceptable. Acceptable enough to want to participate in the process and provide further examples that can be used to enhance the results.
The theory is perfectly fine when fully defined for a subject and applied to machine learning when managing readily measured processes. A factory environment for example where machine can be monitored automatically and tolerances compared in order to spot potential failures by using the feedback from actual failures as information that can be used to adjust the process
That is a world away from the approach that hands-on photographers are likely to be using when they create their art.
In the "filter" driven AI world of mobile device "snap and share" where volume of output often out-demands quality of image in order to achieve 5 minutes of fame, the huge volume of pre-processed images available probably helps the AI analysis whilst lowering the bar for what is required to reach general acceptance. General acceptance being, I would guess, 80 to 90% of the people who might use the "special" technology in the first place. That group may itself be limited to not more than 10 to 20 % of all people who regularly capture images via any sort of device. I suspect my estimated figures are probably optimistic.
It seems unlikely that any advanced functionality is truly "mass market" and much more likely that it is niche activity.
That leaves the development of the technology at risk of failing to reach its marketing objectives (the reason for its existence unless the development is undertaken primarily because the developers are enthusiasts who want to do it) by failing to meet the expectations of those who are already invested in applications that address the same need yet still not being attractive enough at the price point to move occasional users away from their lo-cost or "free" solutions.
0 -
I agree with Stanley's comments about modern software development strategies and beta testing.
Once the industry was able to use the Internet as a "Just in Time" delivery method (or more accurately a "sometimes not quite in time or a bit too soon" method) all the old time controls started to disappear.
In the Desktop world that was driven harder and faster by the proliferation of lower-cost components and, especially in Windows world, extreme flexibility to perform upgrades and create configurations of variable standards that are impossible to fully test before releasing software.
Long ago, in the corporate software world, it became apparent that the most successful software companies - especially those that appeared from nowhere and replaced the longstanding incumbents - were often selling a combination of vapourware and older technology dressed up in new finery.
Moreover, the buyers did not really care. If one delivered perfect software (in terms of no known bugs after extensive testing prior to release) in a controlled package they were not impressed. It meant nothing to them.
In effect it was simply one less argument they could put forward as justification for a discount when discussing upgrades or annual maintenance contracts.
Beta testing should be little more than ensuring that the installation package works; that no conflicts with other installed software on user systems are discovered; and that the functionality works as intended at the point of release in a way that the users understand (mostly) and do not find common, consistent and complete show-stopper results.
We might like it to be more than that but given the timing issues related to OS version releases and so on, that would probably mean, inter alia, a later release date (too late for the purpose demanded by users) and the risk that the entire thing would always be obsolete due to lateness of delivery.
And that's just the technical consideration. There would also be commercial factors and those would include the perceived need to keep up with whatever moves "competitors" in the market were making.
0 -
Permanently deleted user
you do not have to be a fortuneteller to understand that many user will skip the 22 update because of this and the insane upgrade prices.
Or maybe many users will simply have tested v22 properly before buying it - or not.
I don't have any problems whatsoever with 22. Why? Because I trialled it first and decided to save my money by not upgrading.
Simple.
In fact I'm still on v20, which supports my cameras - right up to my Canon R5 - extremely well, and because gimmicky pixel-editor functionality like Pano stitching and HDR have no place in a Raw converter, regardless of (supposedly) "added value" bells and whistles apparently being an inexorable trend.
Result - Capture One is just fine with me...
And it beyond tedious to me that the same people who whine about the lack of this functionality or that feature in a piece of software, are so often the same people who also whine about features being added "too quickly"...
this explains a lot so c1 never tried to mach state of the art pano/hdr software of today or what user expected but instead what adobe accomplished 10 years ago
Not REMOTELY the point being made, which is that it's absolutely normal for new features to start out rough-and-ready, and improve over iterations of the software.
It's called "The Real World".
The take-away point is this: if you buy the software and THEN decide it's not fit for purpose, that's on you, not the software.
0 -
the perception that a company is innovative is more or equal important for its market position than the real value of an innovation. I also see much room for creative new concepts and ideas and btw. video software is in some regard far more advanced compared to imaging software !
Beta testing should be little more than ensuring that the installation package works; that no conflicts with other installed software on user systems are discovered;
very true but unfortunately not the case with c1.
0 -
Permanently deleted user
very true but unfortunately not the case with c1.
You have NO IDEA whether that's true or not. A sample size of one - you - is not statistically relevant, no matter how important and representative you believe your opinions to be.
0 -
Not REMOTELY the point being made, which is that it's absolutely normal for new features to start out rough-and-ready, and improve over iterations of the software.
It's called "The Real World".
maybe in the amateurish world of image editing but nobody would risk using a half baked software for professional video or print production especially when the tech is more then 25 years old !!
ps.: as affinity is mentioned here as an alternative, this people got it right on the first try, pano, HDR and focus stacking work so it seem doable.
0 -
For those of you who stayed behind with v21 (I wish I had), I wrote a work flow tutorial on how to go from C1, merge in Photomatix, stitch in Affinity and then go back to C1. Not as easy as pushing one button, but gets the job done and creates a good image.
I have not beta tested it although a friend used and he said that he had not encountered any real problems. The file is 2.4 MB. If there is any interest, I'll try to post it.
0 -
I was a software developer for 35 years and I know the requirements of a software release only too well. Today there are different models for developing successful software, but a critical issue is always the release date. Normally it is not just the software developers who test, but an independent quality assurance system supported by modular test routines. Beta testers are optional, but do not have the opportunity to test in depth.
What C1 has delivered in the past since the C20 is simply presumptuous towards customers. To want to sell functions such as the new export workflow as a great feature, but then to row back again - an absolute nogo. Functions that nobody needs to be touted as groundbreaking - arrogant. For HDR and Pano there is simply special software that can do this well - why is the program inflated here just to attract a few customers who, however, look disappointingly at the result after the first attempts and still have to switch to the special software.
C22 does not need a Pano / HDR, it needs a "metal" implementation (Intel), it needs a future-oriented DAM with Ki-based elements and it needs absolute stability in the individual workflows. This is where C1 should invest and provide real manpower. C1 basically has excellent functions, but they suffer from the fact that the rest of the structure is sloppily programmed - the result is reflected here in the forum. (IMHO)
2 -
Helmut,
AMEN!!!! You are right on.
1 -
Yep !!
0 -
Helmut hit it for me. I'd love to see C1 put some serious effort into their DAM, as well as their Print dialog. I find their entire printing approach sadly lacking—why can't we save a comprehensive preset (= printer + ICC profile + output size + borders)? And why do we have to set printer and page layout parameters every time Capture One launches? (it can't even remember the last printer used) I know these are shortcomings that Capture One have been aware of for years.
However it strikes me that the new company is mostly about doing 2 things: attracting new users (LR users?) and wrestling with whether to match/beat LR's features or come up with innovative new ones (the former seems to be the current approach).
Innovation's hard, and having to release products on a management-set timeline (i.e. "scheduled innovation") doesn't work very well. Apple is the first company I'm aware of that institutionalized annual "major" software version releases a couple of decades ago. I remember when one of their V.P.'s stood up at a conference and proudly proclaimed that new OS versions, each with at least 100 "new features" would be released annually. Not the best approach for customers, and it must drive the Engineering and Marketing departments crazy to figure out how to name and market those "major new features".
I get the feeling that C1 21-22, while introducing some useful features, may be more about bringing newer engineers up to speed than really refining the software. Also, I wonder how much customer research they do when coming up with and prioritizing new version requirements. From recent releases, I get the feeling that they're doing a lot of back-pedaling rather than forward thinking.
Now, though, many software companies have adopted annual "major" release cycles. Sad, but seems to be the reality. At least C1 has retained their dual perpetual/subscription strategy, which I hope continues.
0 -
+1 Helmut
C22 is a sad step backward...
0 -
+1 for Helmut.
I'd like to add better soft-proofing for those using a commercial printing service.
I like to have dng pano functionalitity in C1 though. HDR will become less important in general, imo, as camera technology will advance, but their is no technical camera solution for pano stitching (aside the old panorama cameras, which we will not see again).
0 -
BeO,
it's true, that new cameras have much better dynamic range which makes HDR less important; on the other hand - at least in my case - I have thousands of images taken with cameras with non-so-great DR which I would like to (re)process in C1.
0 -
True. And cameras still have a way to go until e.g. high-quality interior architecture images which include windows can be made with a single shot (at least if you want to show what's outside). I was thinking in rather longer terms.
0 -
@mhsch8
"it's true, that new cameras have much better dynamic range which makes HDR less important; on the other hand - at least in my case - I have thousands of images taken with cameras with non-so-great DR which I would like to (re)process in C1."
I suspect you may be in a very small minority with little pull for a future customer base in business developement terms.
For clarity, I probably fall into the same category although with less volume of interest in re-processing old images with potential Pano or HDR opportunities. I mighthave a few I could revisit - but only for my personal interest.
Both of my daughters bought quite competent DSLR's a little less than a decade ago. I'm not sure if they still have them. If they do they don't use them - but they do use their phones to take images all the time.
My wife used to shoot 35mm film quite a lot with a decent quality Olympus bridge camera. Once the kids had matured to their teens (and the camera film load mecahnism failed) she lost interest in anything obtaining else ... until the smart phone arrived in her life. Now the darned thing never seem to leave her hand and stills and video are shared worldwide at the press of a touch "button".
Lacking inlimited space on her phone she "manages" the videos and images via Dropbox and a couple of SD cards for archival purposes. For "Live" stuff on the phone (and tablet) the "Gallery" is sufficient.
I have never heard her say she needed a DAM cataloguing system. Whether she knows what such a thing might offer I have no idea. I doubt it and I doubt she would be interested to discover what it might offer.
I think she may be quite typical of the mass market.
If so, the potential for attracting users to software by offering something that the most interested and knowledgeable potential customers probably think should already exist within the product would not seem to offer much marketing incentive. As we all know "Change", whether real or not, generates interest if it can be opresented as something "new"; but only if it can be presented as something "new" - and preferably with references to things of "the future". Things like AI.
Put A.I. into something of interest primarily to traditionalists and one might run the risk of being too intrusive - especially if doing so relieves the traditionalist of his or her duties.
Out of pure interest I would love to see some numbers to support the development decisions being made. I would be very surprised if they are ever discussed in the public domain.
0 -
Out of pure interest I would love to see some numbers to support the development decisions being made.
not sure I would want to know this as it would further degrade my view of this company.
considering "creative" is the most used word in advertising and social media by this company it clearly does not apply to them. when you add a HDR tool years to late one might think they had time enough to rethink and develop a new and better approach to HDR or better an option for exposure blending taking into account the much larger DR of todays cameras but no the even failed to copy adobe, what looser.
1
Post is closed for comments.
Comments
29 comments