Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Jul 24, 2003, 1:37 AM   #21
Member
 
Join Date: Oct 2002
Posts: 31
Default

Quote:
Originally Posted by WmAx
I referenced both references.
Where? You mentioned both Dpreview and Pop Photo but the numbers don't mach (you said Fuji got 1500 and 7i got 1400).

Quote:
"Mistake number two. Most digicams seem to have a sweet spot somewhere between f/4 and f/5.6."

Very true, considering far edge sharpness and CA behaviour. But, the diffraction limit lowers as a smaller aperature is used. The measurement references are near the center of the lenses in these tests.
I was talking about effective sharpness, measured from photos. That automatically takes all factors into account.

Quote:
"This sounds interesting. I though the cells in our retina are not in any particular pattern. I take that this was published in some respectable medical journal. Can you tell me the name of the author?"

When did I say this? I did not. I specified adaptive conditioning to an orientation. IT is a physchological effect, and has been documented that the human percepton has difficulty registering an object not in the standard orientation it is used too. This can be readapted to another orientation, at least temporarily, with repeated exposure to the new orientation. It has nothing to do with the eye structure.
HAH! I knew it! There is no such article. That web site is indeed dealing with pattern recognition. It means they are studying how humans (or specialized programs) can identify objects. Of course I'll have problems identifying my friends, if the photo is shown me upside down - I'm not used to see them that way. It does NOT mean the resolving power of my eyes drops 90 % when the image is upside down.

So, why you mentioned an unrelated article? It has nothing to do with SCCD. It does not proof SCCD is a good idea. It does not proof most real world features are horizontal or vertical. 90 angle was mentioned, but only due to the way the experiment was carried out "because at 90 the room features visible from the familiar viewing direction would disappear from view".

I'm asking again: where's your proof?

Carry out to following test: draw two adjacent, thin lines near each other. Rotate the paper slowly. If your theory is correct, you should be able to see there's two, separate lines instead of just one only when the lines are horizontal or vertical.

BTW, I've now seen how credible you are and I'm not going to continue this. You can have the last word.

Marko
MarkoB is offline   Reply With Quote
Old Jul 24, 2003, 6:32 AM   #22
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

Well everone already knows the differences in resolutions and lenses between the 3 cameras, let me add what else the D7hi brings to the table since we are bringing up Photoshop:

1. It's the only one in the group that can output RAW file (beside TIFF, and light compression JPEG).
2. The picture can be mapped to AdobeRGB (in addition to 2 other sRGB colorspaces, one of which is the Vivid which gives the highly saturated look that some one mentioned).
3. This camera has a TTL-wireless flash system found only on much more expensive dSLR systems (also including different ratio control)
4. The built-in flash has programmable power so you can use the camera to wirelessly trigger studio slaves as well.
5. A real-time histogram superimposed in the viewfinder to verify exposure.
6. Grid and Scale that can be turn on @ leisure rather than changing viewfinder screen.
7. Filter effect wheel (that's right electronic filters even in B & W)

The fact is this camera might cost more than others, but the D7hi is probably the closest thing to a dSLR this side of $1000 from the manual zoom to the way the camera operates (or in Photoshop workflow). In fact when you put the D7hi on manual it actually mimics an SLR in the stop-down mode, ie when the aperture or shutter is changed the camera EVF viewfinder brighten or darken accordingly by EV. This WYSIWYG mode alone is worth all the price differences since it allows even a tyro to shoot in manual just by looking in the EVF (augmented by the real-time histogram)! ie What You See Is What You Get in the final pictures on the flash cards... Imagine doing this feat @ night!

BTW the D7i is $200 cheaper than the D7hi without the high speed buffer and sync terminal, but you can still use the built-in flash for studio shoot. You have to like the silver color instead of black though! :lol:


FYI, here the new Fuji: http://forum.digitalkamera.de/dcboar..._id=5121&page=
... and they talk about an impending Minolta D8i in august too! 8) 8) 8)
NHL is offline   Reply With Quote
Old Jul 24, 2003, 10:09 AM   #23
Senior Member
 
Join Date: Mar 2003
Posts: 205
Default

"Where? You mentioned both Dpreview and Pop Photo but the numbers don't mach (you said Fuji got 1500 and 7i got 1400). "

Incorrect. I stated imaging-resource.com and dpreveview.com as references, because they provide their test results for download. I referenced popphoto only for their color tests only, as popphoto is the only online reveiew site i have noticed to give objective color analysis.

I mentioned imaging-resource a few times, but here are the links to the captured charts at imagng-resource, for your reference:

http://www.imaging-resource.com/PRODS/F602/F62RESLF.HTM

http://www.imaging-resource.com/PRODS/D7I/D7IRESWLF.HTM

http://www.imaging-resource.com/PROD...7HIRESWLXF.HTM

"I was talking about effective sharpness, measured from photos. That automatically takes all factors into account. "

Measured sharpness? Do you mean measured resolution? I am not understanding of your reference to measured sharpness. Sharpness is an effect of accutance(low level contrast elements), and you can vary the 'sharpness' in any one image and create different versions of it that are percieved very differently buy still have the same actual amount of resolution information present.

"HAH! I knew it! There is no such article. That web site is indeed dealing with pattern recognition. It means they are studying how humans (or specialized programs) can identify objects."

Yes, it is indeed demonstrating adapted preference of orientation, just as told you. Ha? Ok, well it's just what i said it was.

"Carry out to following test: draw two adjacent, thin lines near each other. Rotate the paper slowly. If your theory is correct, you should be able to see there's two, separate lines instead of just one only when the lines are horizontal or vertical."

Fuji published several references, but i have long lost those. THis paper I presented shows evidence that strongly suggest humans are sensitized to a particular orientation. In case you did not notice, that orientation is v/h. Almost everything(een in nature) is v/h oriented. I suggest you carry out a little experiment as well. Take a simply piece of graph paper. Look at it in standard v/h orientation. Now hold it at a 30 or 45 degree angle; you will have a much harder time focusing on the details of the scene. You can do with with any landscape as well: take a photo of a scene of trees, house, etc. and do the same. Your brain will have a very difficult time processing the scene, since it will very likely(dominantly) contain v/h data, and shifting the orienation to an obtuse orientation creates difficulties(until you have been able to readapt to the orientation, which will take some time). Under no normal circumstances would a human be adapted to obtuse angles for perception.

Here, my friend, is a 1:1 crop from a portion of scene, identical framing, using a SCCD(diagnol array) vs. normal oriented CCD of the same sensor count. You should have viewed this earlier when i referenced it. It is excellent example..... the v/h data is dominant in one section, and it shows drastic perceptibple improvement... while the otehr part of the capture on the right is of an object shifted 45 degrees.... even thogh the sccd is weakest at this angle, it's difference that is percieved as almost identical to the standard v/h sensor which is more efficiet at resolivng diagnol data than v/h. You would both to be instantly recogiziable of drastic differnes if diagnol and v/h data had the same weighting to us.

http://www.ddisoftware.com/reviews/s2/rescomp.jpg

"So, why you mentioned an unrelated article? It has nothing to do with SCCD. It does not proof SCCD is a good idea. It does not proof most real world features are horizontal or vertical."

FUji had a set of FFT analysis data performed on 'everyday' images, and demonstrated that by far the most dominant features in nature and artificial were v/h. I do not know where this data is now, but it's not worth spending hours to track it down for you. Besided, it rather obvious just looking outside taht everything is primarly v/h, due to effect of gravity on nature. Never noticed that palnts, grass, primary grains of tree bark....trees......almost everything is primarly dominant of lines in the v/h orientation or at slight angles to v/h where the SCCD is most efficient as opposed to 45 degree angular content.

"BTW, I've now seen how credible you are and I'm not going to continue this. You can have the last word. "

O.K., but you have incorrectly remembered the references i posed for the resolution charts and blamed me because the ones at places i never specified for those purposes is different.....uhm...yeah, iw ould expect that since i never quoted popphoto as a reference for juding extinction points of data.....popphoto does not even publish the chart so how would you ever come any conclusin at popphoto regarding this? Same goes for the journal of vision article, i never said it was a specific research project on the dirct validity of the SCCD, but you seem to think i said it was? It was exactly as advertised: study on the adapted orientation of humans recognition, and the results of readapting recognition.

-Chris
WmAx is offline   Reply With Quote
Old Jul 24, 2003, 10:24 AM   #24
Senior Member
 
Join Date: Mar 2003
Posts: 205
Default

NHL - I do like the idea of the live simulated EVF of the exposure, that definately would be nice feature to have. The RAW output would be great as well, so that especially white balance could be easily changed later and that disk space could be saved with the smaller raw files: i assume the Dh71 raw processing software allows WB application after shooting like most dslr raw conversion software?

-Chris
WmAx is offline   Reply With Quote
Old Jul 24, 2003, 11:30 AM   #25
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

Quote:
i assume the Dh71 raw processing software allows WB application after shooting like most dslr raw conversion software?
Yeap, just like any other work flow and just like them the quality of the result varies with the vendor's plug-in...

The other thing that I forgot to mention, and I don't know if it's the same on the other digicams here for that night shot: The D7's actually takes two long exposures (one with the shutter close) and subtracts out the dark frame which substantially removes any noise from the picture in the dark... It's all depends on if theses extra features are worth to the buyer or not, but all digicams nowday are more capable than most average shooter! Let's all improve on our shooting skill and stop beating on the cameras :lol: :lol: :lol:
NHL is offline   Reply With Quote
Old Jul 24, 2003, 12:00 PM   #26
Senior Member
 
Join Date: Nov 2002
Posts: 332
Default Message to NHL

"Let's all improve on our shooting skill and stop beating on the cameras."

NHL - You are as wise as you are knowledgable. [/quote]
fporch is offline   Reply With Quote
Old Jul 28, 2003, 11:26 AM   #27
E.T
Senior Member
 
E.T's Avatar
 
Join Date: Jul 2003
Posts: 921
Default

Quote:
Originally Posted by NHL
1. It's the only one in the group that can output RAW file (beside TIFF, and light compression JPEG).
2. The picture can be mapped to AdobeRGB (in addition to 2 other sRGB colorspaces, one of which is the Vivid which gives the highly saturated look that some one mentioned).
3. This camera has a TTL-wireless flash system found only on much more expensive dSLR systems (also including different ratio control)
4. The built-in flash has programmable power so you can use the camera to wirelessly trigger studio slaves as well.
5. A real-time histogram superimposed in the viewfinder to verify exposure.
6. Grid and Scale that can be turn on @ leisure rather than changing viewfinder screen.
7. Filter effect wheel (that's right electronic filters even in B & W)
You forgot fast to use manual/mechanical zoom (only digicam other than DSLR which have it), great wideangle (28mm) and settings can be done without using lot of menus.
E.T is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 12:47 PM.