Pages: 1 [2] 3 4 ... 8   Go Down

Author Topic: Why are the tests of IQ250 against CCD-backs taking so long?  (Read 35291 times)

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
customizing a column-parallel ADC CMOS sensor: ADC sweep rate and CFAs
« Reply #20 on: February 06, 2014, 07:33:26 pm »

Is there is fixed data path here or can you bypass the on-chip gain and/or ADCs and roll your own?
I am fairly sure that the Sony EXMOR sensors (and all sensors that do column-parallel ADC) always do ADC at the bottom of each column of pixels, and are only wired to output a digital signal from there.

However, the ADC process can be adjusted: reducing the sweep rate of the sawtooth voltage input used with ADCs can increase accuracy (maybe adding a bit or two) in exchange for a lower frame rate.
That and CFA design seem to be the main sensor customizations available. (Demosaicing algorithms and such are also relevant to in-camera JPEGs.)
Logged

synn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1235
    • My fine art portfolio
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #21 on: February 06, 2014, 08:47:15 pm »

Doug,

Thank you for all the hard work you're putting into the tests.

If it is at all possible, please do a test comparing the 250 to the 140 and the 260 (And perhaps the Leaf equivalents too) for portraiture at base ISO and under strobes.

There's certainly no rush, but it would be nice if this could be done at some point.

Thank you!
Logged
my portfolio: www.sandeepmurali.com

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: customizing a column-parallel ADC CMOS sensor: ADC sweep rate and CFAs
« Reply #22 on: February 06, 2014, 09:09:13 pm »

top shoulder adjust.

E.

I am fairly sure that the Sony EXMOR sensors (and all sensors that do column-parallel ADC) always do ADC at the bottom of each column of pixels, and are only wired to output a digital signal from there.

However, the ADC process can be adjusted: reducing the sweep rate of the sawtooth voltage input used with ADCs can increase accuracy (maybe adding a bit or two) in exchange for a lower frame rate.
That and CFA design seem to be the main sensor customizations available. (Demosaicing algorithms and such are also relevant to in-camera JPEGs.)

Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Ken R

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 849
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #23 on: February 06, 2014, 09:55:10 pm »

I'm insulted by this. I've got six years of history here being very direct about both the pros and cons of anything we sell. To imply it's taking "so long" because there is some conspiracy to keep the results from you vastly underestimates the amount of work we put in to providing these tests.

I spent 7 hours at the Morgan Library on Monday capturing every combination of:
IQ250, IQ260, IQ280
32HR, 32XL, 40HR, 60XL
0/15/30mm rise
0/15/-15mm shift

I also added an ISO sweep test, and tested the 47XL and 28LS with the IQ250.

With LCCs that 271 images.

It took most of three days (90% done at this point) to backup, organize, apply LCC's, and adjust those files. The organization is especially important because we intend the raws to be able to be processed by people not present at the test, who will therefore need to know exactly what each file represents.

We'll be making all these files available to our clients, and most of them available to anyone (there are some bandwidth concerns in sharing the entire inventory to the entire world). Both the ones that show the IQ250 shining (most of them) and the ones that show the IQ250 really struggling (like the 35XL with 15mm movement).

Keep in mind that in that time since the launch we've also
- changed 13 time zones, and did a 16 hour flight
- written a lengthy article that will be published on LL tomorrow, based on extensive interviews and conversation with the R+D team
- written an extensive FAQ on the IQ250 which had information no one else published
- tested the new version of C1 and Capture Pilot
- posted answers to dozens of questions asked here on the forum
- captured and published a video of the IQ250 live view
- rushed TIFFs of the raw files we had to client's that have time sensitive requests which could not wait for the new version of C1 (e.g. one is planning a project for next month and needs to know if an IQ250 or IQ260 will be better for his project)
- done all the normal office work (ya know, like supporting our existing customers - our #1 priority, whose questions and issues don't suddenly stop because there is a new product to test)

If anything, I do accept blame for overestimating the speed at which we'd get the tests out. For that I do apologize.

If I sound a bit annoyed at the accusation, I am. I've basically not slept a full night since the launch, doing everything possible to get good, relevant, real world testing done, while also taking care of our existing customers (as a dealer we are very hands on with our client's and any issues/questions they have). I do this, despite knowing that as soon as I post the test the comments will be filled with posters telling me I'm dumb because I didn't do XYZ ("you should have done 17.5mm of shift, not 15mm blah blah"), because I know there are many here who have a genuine curiosity, and don't have access to rent/evaluate a back and therefore rely on such testing to help them make very large financial decisions.

Anyway, back to work...

I got tired just by reading this!

Thanks for all the work. The initial tests you guys sent me look really promising and really confirm your initial thoughts on the back. It is a great back to use on an SLR in any condition. Anything that the D800E can do (in regards to sensor/image quality) the IQ250 can do better basically. The fact that the back is usable on tech cameras its just a bonus.
Logged

RobertJ

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 706
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #24 on: February 06, 2014, 10:08:06 pm »

But... is it better than a Hasselblad Lunar?
Logged

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #25 on: February 06, 2014, 10:17:25 pm »

Doug,

And now you will have to test the Pentax, and then the Hassy ;)

Edmund
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11288
    • Echophoto
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #26 on: February 06, 2014, 10:43:22 pm »

Hi,

I agree with Doug.

Phase One has published some images early shot by real photographers and that got them some bad critics.

Doing any real test/demo is some real effort. Not least, you need to put the test in a contest. Also, having raw images is nice, but pretty meaningless until the camera is supported by raw converters.

Doug's article had a bit of bad timing with LuLa as the owners/editors were off to Antartica, but it's nice to see it published tomorrow.

Best regards
Erik


I'm insulted by this. I've got six years of history here being very direct about both the pros and cons of anything we sell. To imply it's taking "so long" because there is some conspiracy to keep the results from you vastly underestimates the amount of work we put in to providing these tests.

I spent 7 hours at the Morgan Library on Monday capturing every combination of:
IQ250, IQ260, IQ280
32HR, 32XL, 40HR, 60XL
0/15/30mm rise
0/15/-15mm shift

I also added an ISO sweep test, and tested the 47XL and 28LS with the IQ250.

With LCCs that 271 images.

It took most of three days (90% done at this point) to backup, organize, apply LCC's, and adjust those files. The organization is especially important because we intend the raws to be able to be processed by people not present at the test, who will therefore need to know exactly what each file represents.

We'll be making all these files available to our clients, and most of them available to anyone (there are some bandwidth concerns in sharing the entire inventory to the entire world). Both the ones that show the IQ250 shining (most of them) and the ones that show the IQ250 really struggling (like the 35XL with 15mm movement).

Keep in mind that in that time since the launch we've also
- changed 13 time zones, and did a 16 hour flight
- written a lengthy article that will be published on LL tomorrow, based on extensive interviews and conversation with the R+D team
- written an extensive FAQ on the IQ250 which had information no one else published
- tested the new version of C1 and Capture Pilot
- posted answers to dozens of questions asked here on the forum
- captured and published a video of the IQ250 live view
- rushed TIFFs of the raw files we had to client's that have time sensitive requests which could not wait for the new version of C1 (e.g. one is planning a project for next month and needs to know if an IQ250 or IQ260 will be better for his project)
- done all the normal office work (ya know, like supporting our existing customers - our #1 priority, whose questions and issues don't suddenly stop because there is a new product to test)

If anything, I do accept blame for overestimating the speed at which we'd get the tests out. For that I do apologize.

If I sound a bit annoyed at the accusation, I am. I've basically not slept a full night since the launch, doing everything possible to get good, relevant, real world testing done, while also taking care of our existing customers (as a dealer we are very hands on with our client's and any issues/questions they have). I do this, despite knowing that as soon as I post the test the comments will be filled with posters telling me I'm dumb because I didn't do XYZ ("you should have done 17.5mm of shift, not 15mm blah blah"), because I know there are many here who have a genuine curiosity, and don't have access to rent/evaluate a back and therefore rely on such testing to help them make very large financial decisions.

Anyway, back to work...
Logged
Erik Kaffehr
 

Steve Hendrix

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1582
    • http://www.captureintegration.com/
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #27 on: February 06, 2014, 11:40:09 pm »

Proper testing takes much longer than many assume. Certainly more than a few hours!  ;D

It's difficult, especially when comparing products and results, to ensure that apples are compared to apples, and that the results are conclusive and repeatable.

It takes a lot of time - and this is on top of everything else - the world doesn't stop just for the tests. They often require significant before and after hours time (more than usual), to make sure {existing as well as prospective} clients still get taken care of. Last week, we had 2 employees spend the night in the Atlanta office to complete some of the initial testing* It did not help that an un-prepared city had an ice storm and 20 minute commutes turned into 6 hour drives....  :(

Testing (and continued testing) will be forthcoming from "sellers"....

I don't think anyone is afraid to state the results from the IQ250 compared to CCD sensors - I think it is already indicated that the results will not be as bad as we feared, but not as good as we hoped (when it comes to technical camera usage with shifts, etc).


Steve Hendrix
Capture Integration
Logged
Steve Hendrix • 404-543-8475 www.captureintegration.com (e-mail Me)
Phase One | Leaf | Leica | Alpa | Cambo | Sinar | Arca Swiss

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11288
    • Echophoto
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #28 on: February 06, 2014, 11:46:07 pm »

Hi,

I guess that is often the case. Miracles are long between.

Best regards
Erik



I think it is already indicated that the results will not be as bad as we feared, but not as good as we hoped (when it comes to technical camera usage with shifts, etc).


Steve Hendrix
Capture Integration
« Last Edit: February 06, 2014, 11:55:02 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

buckshot

  • Guest
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #29 on: February 07, 2014, 08:01:04 am »

I'm sure Phase did testing on a tech camera during development - or, at least, you would hope they did - this thing costs $35k after all. Why the results aren't made available at launch is odd.

Imagine BMW bringing a new car to market (costs about the same - yes, still hard to get your head around), and then after it goes on sale they declare 'no - we don't know how it drives in the wet - someone had better do some testing and find out'. Doh !
Logged

pedro39photo

  • Full Member
  • ***
  • Offline Offline
  • Posts: 111
    • PedroNunesPhoto
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #30 on: February 07, 2014, 09:23:58 am »

By the way, here is the challenging, real-world, but beautiful location, in which we tested the back. More details tomorrow:

(IQ250, 32HR, 8mm rise)

Amazing work Doug, perfect and beautiful place to do the quality test.
Thanks for the effort.

Pedro 
Logged

Paul2660

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3978
    • Photos of Arkansas
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #31 on: February 07, 2014, 09:46:20 am »

I'm sure Phase did testing on a tech camera during development - or, at least, you would hope they did - this thing costs $35k after all. Why the results aren't made available at launch is odd.

Imagine BMW bringing a new car to market (costs about the same - yes, still hard to get your head around), and then after it goes on sale they declare 'no - we don't know how it drives in the wet - someone had better do some testing and find out'. Doh !

I had some of the same thoughts, but then gave it a bit more thought. 

If you look at the previous big Phase One rollout from last year, the IQ260, I believe in the featured photographers area, Phase had at least one who was a tech camera user.  So far on the Phase Site, all the featured users for the 250 seem to be action/wedding shooters.  Phase is a small company, privately owned, so total resources are not published.  It's also clear to me now that Phase had been working on this camera for quite a while as they are shipping it at announcement, something they never have done before that I can remember. 

Yes they had plenty of time to set up a test with anyone familiar with the workings of a tech camera, either for landscape, architecture, or interiors etc.  So far nothing on their site has shown any of this style of shooting, i.e non DF+ body.  I don't know if this means Phase One did not have time, or their direction was for a different photographic market.  They also may have interpreted the results from the P30+ which was not a good back for tech camera use, as possible reasons not to lead with a tech camera.  However with the DF+, Phase One lenses, etc. they have much more control over the end results as they have a greater familiarity with these Phase One branded products.

It's also interesting to note that both DT and CI have been working to get the samples posted from various tech cameras/lens combinations and even Alpa added some material of their own.  The fact that Phase One did not lead with at least one landscape tech shooter from the day of the announcement tells me that Phase focused on other photographic areas for this back in their own testing.   Phase One has a very through website and they do seem to keep it updated with new products, demos, video etc.   

The results seem to show that for wides, the 250 may not be the best possible solution but if you are in the 55mm range or larger it very well might work great.   

I appreciate the fact that my dealer, Digital Transitions, was able to work with me for 3 days last week when my 260 came back from Phase One for a top cover fix  with a definite problem at base ISO 50.  This involved going through over 6 GB of sample raws from me, and then forwarding the images on to Phase One.  This getting done so fast, along with the terrible weather that has plagued both my state and NY was greatly appreciated considering the duress that they are under to get a new product out to the market while they have the advantage of being the only product with this capability.   It's a terrible feeling when your back comes back with a problem that wasn't there before which makes it non-usable at base iso.

I can appreciate the time it takes to get a test done, published, and put into a format that others can interpret it.

Paul C
Logged
Paul Caldwell
Little Rock, Arkansas U.S.
www.photosofarkansas.com

Gandalf

  • Full Member
  • ***
  • Offline Offline
  • Posts: 112
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #32 on: February 07, 2014, 01:25:35 pm »

Doug,

Do you know if any tech cams will be coming on tour with the IQ250 demos? I know I won't really be able to test much in a studio, but I would love to see them together when Allison/Michelle come to Denver.

Logged

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4169
    • http://www.doug-peterson.com
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #33 on: February 07, 2014, 01:28:06 pm »

Do you know if any tech cams will be coming on tour with the IQ250 demos? I know I won't really be able to test much in a studio, but I would love to see them together when Allison/Michelle come to Denver.

If you contact Michelle directly she should be able to arrange the tech camera body/lens of your choice.

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4169
    • http://www.doug-peterson.com
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #34 on: February 07, 2014, 05:19:39 pm »

Good news is that we got the article up (and all the various small typos and issues changing from final draft to the blog platform and formatting and the LL team being swamped during and after their trip to Antarctica).

IQ250, CMOS Fully Realized

Bad news is I'm still organizing and uploading sample files and writing up our findings. Great news is I think they were be very useful in evaluating the real-world usable image circle for each of the tested lenses for each of the tested backs.

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4169
    • http://www.doug-peterson.com
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #35 on: February 07, 2014, 05:26:29 pm »

Was all tech camera testing indoors? 

I did a 32HR vs 35XL of the empire state building in late afternoon light, with full range of movements with the 250 and the 260. I screwed up the 35XL shots (badly misfocused - it's hard to do tests when it's freeeeezing cold outside! You miss stupid things – mad at myself about that) so I won't be posting them as part of our testing but glad to share with you - especially the 32HR shots (but also the misfocused 35XL shots I guess if useful) if relevant to you. Probably not for a few days though - still lots to do on the main vein of testing/writeup.

Christoph C. Feldhaim

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2509
  • There is no rule! No - wait ...
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #36 on: February 07, 2014, 05:51:46 pm »

Good news is that we got the article up (and all the various small typos and issues changing from final draft to the blog platform and formatting and the LL team being swamped during and after their trip to Antarctica).

IQ250, CMOS Fully Realized

Bad news is I'm still organizing and uploading sample files and writing up our findings. Great news is I think they were be very useful in evaluating the real-world usable image circle for each of the tested lenses for each of the tested backs.

After your writing about "taming CMOS color" by the P1 guys I'm quite curious about CCD / CMOS comparisons.
It seems its still not clear for everyone if it is just the CFAs or something else.
From my limited understanding of physics and electronics I couldn't see why there should be a difference,
but I have no real MFD experience (only a short weekend with a loaner) and all the pros say CCD is color magic ....
What I would really like to see is a sunset/sunrise comparison and candle lights / fire comparison, I would even like to see it compared to film.
Cheers
~Chris

Steve Hendrix

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1582
    • http://www.captureintegration.com/
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #37 on: February 07, 2014, 06:48:18 pm »

Anyone interested in doing some in person testing is invited to our Grand Opening next Thursday at our new England office:

https://captureintegration.com/ci-new-england-grand-opening-party-213/

In addition to much of the CI team being on hand, representatives from these companies will be on hand:

Phase One - Murray Elliot
Mamiya Leaf - Ziv Argov -
Hasselblad - Andrew Trumback -
Alpa - Andre Olgani
Broncolor - Colin King
ProFoto - Bill Gratton
Eizo - Jeff Deeken

It's a big day!

Tons of gear from the above companies, including (2) IQ250 units for hands on testing.


Steve Hendrix
Capture Integration

Logged
Steve Hendrix • 404-543-8475 www.captureintegration.com (e-mail Me)
Phase One | Leaf | Leica | Alpa | Cambo | Sinar | Arca Swiss

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #38 on: February 07, 2014, 06:57:10 pm »

Good news is that we got the article up (and all the various small typos and issues changing from final draft to the blog platform and formatting and the LL team being swamped during and after their trip to Antarctica).

IQ250, CMOS Fully Realized

Bad news is I'm still organizing and uploading sample files and writing up our findings. Great news is I think they were be very useful in evaluating the real-world usable image circle for each of the tested lenses for each of the tested backs.

Ever so beautifully written - such wasted eloquence - you should have been a copywriter Doug.

I think my monitor is really broken; it shows these wonderful images from your new  Phase, and yet they don't really make me want one - in fact my monitor is so bad it even shows the lady of the lake a bit out of focus. I must really get that stupid thing fixed.

Quote
The little isle is all inrail'd
With a rose-fence, and overtrail'd
With roses: by the marge unhail'd
The shallop flitteth silken sail'd,
       Skimming down to Camelot.
A pearl garland winds her head:
She leaneth on a velvet bed,
Full royally apparelled,
       The Lady of Shalott.
Edmund
« Last Edit: February 07, 2014, 10:37:28 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Fine_Art

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1172
Re: Why are the tests of IQ250 against CCD-backs taking so long?
« Reply #39 on: February 07, 2014, 07:49:09 pm »

It must be contagious, she looks blurry on my screen too. You might want to sub a different shot for the article.

Looking forward to some large sample shots. Not that I would buy one, just curiosity.

Logged
Pages: 1 [2] 3 4 ... 8   Go Up