If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
Hasselblad
Alan Browne wrote:
If the highlight problem of digital sensors could be solved, then 16 bits would be a benefit, however as the lowest order bit would end up higher in the overall range v the present case. (eg: do any current digital sensors retain detail beyond 2.5 stops above middle grey?). Most of them go slightly above that. Nothing else would make sense! Consider that if we (some would say correctly) define middle grey as 12.5% reflectance... then 25% reflectance is 1 stop brighter, and 50% is 2 stops, and 100% is exactly 3 stops above middle gray. 100% is as bright as it gets, and there is nothing to save above that. Now, we all know that various camera, lightmeters, and people define middle grey as anything from 12% to 18%, and all that does is change how far down from 100% we put middle grey. 18% would make it 2.5 stops below 100%. If this isn't sinking in, let me be blunt: there is no such thing as "the highlight problem of digital sensors". And with most cameras you can adjust where it thinks middle grey is on a per shot basis with something called Exposure Compensation. That is exactly what it does. On at least some Nikon cameras there is also a menu option to apply a bias "permanently". Same effect. A 16-bit data path does not affect any of the above. It would allow recording greater dynamic range (if the sensor and ADC noise levels permit). -- Floyd L. Davidson http://www.apaflo.com/ Ukpeagvik (Barrow, Alaska) |
#12
|
|||
|
|||
Hasselblad
In article , Alan Browne
says... The FinePix S3 solved this by pairing each sensor site with a low sensitivity sensor for the highlights. Very clever. But couldn't keep up with the resolution race at a time when conventional sensors improved just enough. They could increase dynamic range by having some sort of vertical capacitor cell in each pixel. Small pixel area, but if you make the cell deep enough (in the 3rd dimension), the well capacity should be big and you get a large dynamic range. This vertical capacitor principle is used with DRAM cells. -- Alfred Molon Olympus E-series DSLRs and micro 4/3 forum at http://tech.groups.yahoo.com/group/MyOlympus/ http://myolympus.org/ photo sharing site |
#13
|
|||
|
|||
Hasselblad
On 2014.11.09, 13:01 , Alfred Molon wrote:
In article , Alan Browne says... The FinePix S3 solved this by pairing each sensor site with a low sensitivity sensor for the highlights. Very clever. But couldn't keep up with the resolution race at a time when conventional sensors improved just enough. They could increase dynamic range by having some sort of vertical capacitor cell in each pixel. Small pixel area, but if you make the cell deep enough (in the 3rd dimension), the well capacity should be big and you get a large dynamic range. This vertical capacitor principle is used with DRAM cells. Interesting - but if it were that simple, then even a modest increase (double the depth) would add a stop to the DR. Maybe doubling the depth is not a 'modest' change, however. Would also cause bucket brigade issues that would slow down the shooting rate. I'd happily give up frame rate for highlight detail. -- Among Broad Outlines, conception is far more pleasurable than “carrying [the children] to fruition.” Sadly, “there’s a high infant mortality rate among Broad Outlines—they often fall prey to Nonstarters.” "Bestiary of Intelligence Writing" - CIA |
#14
|
|||
|
|||
Hasselblad
On 10/11/2014 12:48 p.m., Alan Browne wrote:
On 2014.11.09, 13:01 , Alfred Molon wrote: In article , Alan Browne says... The FinePix S3 solved this by pairing each sensor site with a low sensitivity sensor for the highlights. Very clever. But couldn't keep up with the resolution race at a time when conventional sensors improved just enough. They could increase dynamic range by having some sort of vertical capacitor cell in each pixel. Small pixel area, but if you make the cell deep enough (in the 3rd dimension), the well capacity should be big and you get a large dynamic range. This vertical capacitor principle is used with DRAM cells. Interesting - but if it were that simple, then even a modest increase (double the depth) would add a stop to the DR. Maybe doubling the depth is not a 'modest' change, however. Would also cause bucket brigade issues that would slow down the shooting rate. I'd happily give up frame rate for highlight detail. Nikon kind of achieved this with the D810 (compared to D800). Full well capacity is nearly doubled, hence base ISO reduced from 100 to 64. But there isn't a corresponding increase in DR, a fraction of a stop, not almost a stop. QE has reduced, with corresponding slight decrease in high ISO performance. The differences are trivial - so trivial that I can't see any difference between D810 and D800E raw files. |
#15
|
|||
|
|||
Hasselblad
On 10/11/2014 3:29 a.m., Alan Browne wrote:
The FinePix S3 solved this by pairing each sensor site with a low sensitivity sensor for the highlights. Very clever. But couldn't keep up with the resolution race at a time when conventional sensors improved just enough. Here's the D200, s5 Pro, and D300 dynamic range compared. The S5 (based on D200 body) had a clear and significant advantage over the "normal" CCD architecture in the D200. But the D300 killed the S5 in resolution, dynamic range (through the entire ISO range), and general camera ergonomics. http://i.imgur.com/9AjQjTX.png |
#16
|
|||
|
|||
Hasselblad
On 2014.11.11, 03:32 , Me wrote:
On 10/11/2014 12:48 p.m., Alan Browne wrote: On 2014.11.09, 13:01 , Alfred Molon wrote: In article , Alan Browne says... The FinePix S3 solved this by pairing each sensor site with a low sensitivity sensor for the highlights. Very clever. But couldn't keep up with the resolution race at a time when conventional sensors improved just enough. They could increase dynamic range by having some sort of vertical capacitor cell in each pixel. Small pixel area, but if you make the cell deep enough (in the 3rd dimension), the well capacity should be big and you get a large dynamic range. This vertical capacitor principle is used with DRAM cells. Interesting - but if it were that simple, then even a modest increase (double the depth) would add a stop to the DR. Maybe doubling the depth is not a 'modest' change, however. Would also cause bucket brigade issues that would slow down the shooting rate. I'd happily give up frame rate for highlight detail. Nikon kind of achieved this with the D810 (compared to D800). Full well capacity is nearly doubled, hence base ISO reduced from 100 to 64. But there isn't a corresponding increase in DR, a fraction of a stop, not almost a stop. QE has reduced, with corresponding slight decrease in high ISO performance. The differences are trivial - so trivial that I can't see any difference between D810 and D800E raw files. If that's the case then there would seem to be more issues regarding DR than just well depth. -- Among Broad Outlines, conception is far more pleasurable than “carrying [the children] to fruition.” Sadly, “there’s a high infant mortality rate among Broad Outlines—they often fall prey to Nonstarters.” "Bestiary of Intelligence Writing" - CIA |
#17
|
|||
|
|||
Hasselblad
On 2014-11-08 10:15:23 +0000, Alfred Molon said:
I read the blog of a pro the other day, where this photographer was claiming that Hasselblad cameras are so much better (he didn't explain why). Is that true and if so, what makes Hasselblad cameras better? Is it just the higher pixel count due to the medium format sensor or are the sensors better (lower noise, higher dynamic range etc), the AF, the JPEG engine etc? If you are shooting in the studio or relatively static scenes off a tripod, they are good. Careful comparison of test images shows that they are not really superior even at ISO 100 when compared to the best and latest Canon and Nikon full frame SLRs. If you look at how they perform at high-ISO speeds the Hasselblad loses big time. It's terrible. Their imagers are about eight generations behind Canon and Nikon's performance at ISO 800 and above. If you're like me and you enjoy natural light photography there is simply no place in your bag for a Hasselblad. At high ISO my Canon S90 (a sub-compact camera) gives better performance than the Hasselblad. This is really sad. The Hassy is good if you run a photo studio or a especially wedding operation where you use artificial lighting and want to be able to trump all the guests who brought equally capable, but vastly less expensive cameras. So in that sense it's a professional tool because it allows you to maintain the image of superiority over all of the enthusiasts and amateurs. After all, THEIR camera didn't cost $45,000 now did it? 16 bits of color per pixel also sounds good until you realize that there is no digital photo printing technology that is even remotely capable of such a wide gamut. Even 10 bits per pixel is overkill for the very best printers. |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
FS- TON OF Hasselblad | Scungili | General Equipment For Sale | 3 | July 15th 04 12:52 AM |
HASSELBLAD PME 5 | Magister | Medium Format Photography Equipment | 3 | June 23rd 04 07:37 PM |
HASSELBLAD 553 EXL KIT | cohenandy | Medium Format Equipment For Sale | 0 | February 24th 04 02:37 PM |
FS: Hasselblad 553 ELX | Tom Miller | General Equipment For Sale | 0 | July 30th 03 07:08 PM |