CONTAX CLUB Carl Zeiss 討論區

香港互聯網康泰時攝影會
現在的時間是 週四 28 3月 24 16:39:40

所有顯示的時間為 UTC + 8 小時




發表新文章 回覆主題  [ 103 篇文章 ]  前往頁數 上一頁  1 ... 34567  下一頁
發表人 內容
 文章主題 :
文章發表於 : 週一 04 9月 06 23:42:02 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
I was looking at the white sign "Quali Walk" "6=>" on the lamp post, at the top left corner of your pictures. Clearly, both Canon 17-40 and 24-70 were sharper than VS24-85 at the top left corner, so I wondered if that was what you were focusing on... and asked you that question.

Also, you would notice that the lamp post was suddenly very sharp for the two Canons. The rest of the frame was less sharp. I was not sure if that was indeed the focus plane, or whether it was because of field curvature. (I thought the lamp post on the left was indeed the focus plane, judging from the scrub in the center.)

To sum up, from the two Canon pictures, what you focused on was less sharp than what you don't focused on! :? Strange.


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 00:23:50 
理事
理事

註冊時間: 週六 08 11月 03 07:56:21
文章: 5762
cyberstudio 寫:
I was looking at the white sign "Quali Walk" "6=>" on the lamp post, at the top left corner of your pictures. Clearly, both Canon 17-40 and 24-70 were sharper than VS24-85 at the top left corner, so I wondered if that was what you were focusing on... and asked you that question.

Also, you would notice that the lamp post was suddenly very sharp for the two Canons. The rest of the frame was less sharp. I was not sure if that was indeed the focus plane, or whether it was because of field curvature. (I thought the lamp post on the left was indeed the focus plane, judging from the scrub in the center.)

To sum up, from the two Canon pictures, what you focused on was less sharp than what you don't focused on! :? Strange.


First of all thanks for reading the pictures in detail!

As a long-term manual focus user, sometimes I would use hyper-focal distance (or DOF) for shooting landscape or street snap shoot, especially for a rather wide angle lens (say 24mm this time) and moderately small aperture (such as f/8 or f/11), while the result on film usually won't let me down..........

Considering the setting for the test this time, i.e. 24mm and at f/8, would the DOF for all these 3 lenses still not enough to cover all the ranges from near and far as in the picture? (even a mis-focus did happen on the lamp post on the left hand side) I had been trying to figure out the answer also before I decided to post these pictures on the forum.............at lease what I can see are the lamp posts far away on the winding walkpath tend to be in focus in 3 of the pictures.

I add another 100% crop from the centre here:

Canon 17-40L

圖檔

Contax 24-85 T*

圖檔

Canon 24-70L

圖檔

Should it be the problem because I had used ISO 1600? or should I say the AF system is not 100% trustful? That I dunno also..... :roll:

_________________
圖檔
Wah
理事 - Board of Members
www.flickr.com/dicksonlau


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 03:02:19 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
Hi wah,

The concepts of hyperfocal and DOF are based on a circle of confusion - a measure of what we consider as sharp. Usually, the CoC is chosen to be 0.03mm for 135 film and 0.025mm for APS. The EOS 5D has a pixel pitch of 8.2 micron, or 0.0082mm. So, according to DOF and hyperfocal theories, a blur of 3.7 pixels is still considered "sharp". Clearly, this is inadequate for critical lens testing.

I found that even stopping down to f/8, I can still very easily find the focus plane by comparing relative sharpness. There must be a point in the picture where it is the sharpest. If that point is where you really intend to focus on, great. If not, it becomes an AF accuracy test instead of a lens test.


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 08:03:45 
理事
理事

註冊時間: 週六 08 11月 03 07:56:21
文章: 5762
Hello cyberstudio,

Thanks for the technical information! I suppose I should stick with manual focus for the test next time for a better control of the result, or probably better with a tripod because I had to do it in rush this time.

In other way round, did I make a very good situational simulation for the shooting habit of the AF camera users? I probably didn't notice any problem from the viewfinder even (may be my eye-sight is not that good......... :cry: ) when the camera told me that it is in focus for where I intended to........ :roll:

_________________
圖檔
Wah
理事 - Board of Members
www.flickr.com/dicksonlau


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 15:49:38 
CONTAX CLUB 會員
CONTAX CLUB 會員

註冊時間: 週六 08 1月 05 22:54:51
文章: 2268
來自: Hong Kong
In the conversion description, it computed that the difference between EF and N mounts is 0.65mm and two copper plates 0.55mm and 0.11mm which are used. Then :

1. although the two plates add up to be closest to 0.65mm, there is still a 0.03m difference. This 0.03mm is about 5% and 5% error on focusing can be a lot, especially at 85mm focucal length. If there is a 0.14m plate and it can be used to replace the 0.11mm plate, I think the focusing accuracy can be largely improved.

2. i am not sure if copper is a good material here as copper has very high expansivity. The thickness can vary a lot when temperature changes say from 0C to 30C. This further affect the focusing accuracy. Copper is also a soft material which can easiy lose its shape. Or is the material actually a copper alloy?

Nevertheless, it is still an exciting trial! Excellent!

_________________
Contax : RTS, RTSII(50Y), RTSIII, Aria D, Aria(70Y), N1, ND, T3D(70Y), G2D
Nikon : FM3A, FM3AB
Leica : M9, X1
Fuji : X100
OM : EP2


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 16:45:44 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
The shims are not an end, it is a means. Zeiss publishes technical data of the Vario Sonnar 24-85 on its web site. The backfocus distance is 42.7mm at the wide end and 64.1mm at the tele end. After the lens has been manufactured, careful measurements are made, and enough shims are added to the lens to ensure the backfocus distance is as designed (42.7mm at wide end). The purpose of the shims is to control the backfocus distance. If the shims are off by 0.03mm, the backfocus distance will be off by only 0.07% (42.7mm vs 42.73mm).

Think of this analogy: I needed to build a building which is 427m high, but I was cheap, I built it 420.8m high and put a 6.2m antenna on top. The antenna maker made a 5% error, and gave me a 6.5m antenna. Now my building is 427.3m tall - but that is only a small difference from my original design.

Of course, being the kind of perfectionists we are, an error is an error. It becomes meaningless as we head down to the micron range, as the metal expansivity you mentioned kick in, and also metal compression as you tighten the screws (metal is malleable), not to mention that well-calibrated equipment is required to make a measurement that small. Therefore, many critical dimensions of a camera, such as film-to-flange distance, is toleranced to +/- 0.02mm or +/- 0.03mm.


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 16:55:12 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
Ah forgot to mention the backfocus distance is defined to be the distance from the rear element to the film plane.


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 21:08:44 
討論區會員

註冊時間: 週日 09 11月 03 02:22:35
文章: 286
來自: HK
wah 寫:
In other way round, did I make a very good situational simulation for the shooting habit of the AF camera users? I probably didn't notice any problem from the viewfinder even


Wah, 100% agree.

For me, it's a great user test. It's a test done in a hurry, in a open area, hand-held, ISO 1600....Yet, the result is the most detailed piece of reference material available in the web so far.

Thank you for your hard work! :D

However, it's better you go to a lab to do the lens test next time! (though I am not interested in those figures.....NEVER....)

I am just hoping to see another detailed, scientific test from somebody else...... but who else? :evil:

_________________
http://www.brianmok.com/


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 23:13:42 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
Of course, wah's contribution cannot be overemphasized. He is the first person in the world to get this new product installed.

Sorry if I sounded like I was criticizing his test. That was not the intention. I was only trying to interpret his test results, given that, as Brian puts it, that is the most detailed piece of reference material available on the web so far. I don't recall seeing any other person doing any A/B comparison on the net.

wah, the 300D samples on page 2 all look very good. Are they all taken with AF?


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 23:40:35 
理事
理事

註冊時間: 週六 08 11月 03 07:56:21
文章: 5762
cyberstudio 寫:
Of course, wah's contribution cannot be overemphasized. He is the first person in the world to get this new product installed.

Sorry if I sounded like I was criticizing his test. That was not the intention. I was only trying to interpret his test results, given that, as Brian puts it, that is the most detailed piece of reference material available on the web so far. I don't recall seeing any other person doing any A/B comparison on the net.

wah, the 300D samples on page 2 all look very good. Are they all taken with AF?


Yup..........and concerning the AF speed I suppose 300D is still reasonable good regardness of its "age" (300D was released in 2003!)

And somehow............I can see even 5D sometimes cannot gain absolute advantages regarding picture quality...........or I should say does not sound very reasonable for the price difference especially with the current price of a 2nd hand 300D (which may be HKD$2000 something nowadays :wink: )

_________________
圖檔
Wah
理事 - Board of Members
www.flickr.com/dicksonlau


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週二 05 9月 06 23:57:14 
討論區會員

註冊時間: 週三 24 12月 03 17:09:08
文章: 112
來自: Canada
I particularly like your "Perrier". Zoomed to the max, wide open, close focusing distance - the conditions are as tough as it gets for a zoom lens. The result, however, is surprisingly good! At 100% I can read the label very clearly, even some of the fine print. And, the barcode on the bottle - I can probably use a barcode scanner to scan your photo instead of the real bottle.

Your photos of Tish (on 5D) on page 5 are also great. I can very clearly see the background reflected from her pupil. (I am peeping at the EXIF and these are AF, too.)

Thank you for sharing with us these pictures!


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週三 06 9月 06 01:04:53 
會計
會計

註冊時間: 週日 09 11月 03 00:26:48
文章: 6252
來自: Hong Kong
看了這裡幾位的討論,令小弟佩服大家深入而認真的態度,也令自己非常汗顏,因為我係一個「差不多先生」 :oops: ,明知「差之毫釐,謬之千里」係玩改鏡的大敵,但又沒有鑽研其中奧秘,所以每次玩都必需請教樺少同埋揾師傅操刀。今次樺少能自己動手,又做埋測試,好難得。希望大家繼續俾意見,令有興趣玩的朋友有所参考。 :D

講番測試,好期待有更多的範例圖片,試不同的機款、設定、題材,鏡頭。反正 N 系還有好幾款鏡,尤其係 VS17-35/2.8 及 MS100/2.8 最引起我的興趣!有時我認為測試方法還是其次,因為影響相片效果的要點有一百幾十,又無人能比實驗室更 'scientific',玩法又非出自原廠之筆,此 'ZEF'發明經已令人叫絕!每個人的使用習慣不同,故在任何條件下測試也具參考價值。我想,最重要是有玩 ZEF 的興致與能力吧,有人能為 N 鏡另尋途徑真係功得無量! 8)

_________________
會計
* * * * * *
鯨魚愛Zeiss
圖檔


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週三 06 9月 06 11:44:53 
討論區新手

註冊時間: 週三 30 8月 06 15:32:37
文章: 10
來自: Taipei
I have purchase one adaptor for makro-sonnar 100/2.8.
I will test 1DsII+EF100/2.8 macro and 1DsII+makro-sonnar 100/2.8.
Please wait.


回頂端
 個人資料  
 
 文章主題 : 唔好意思
文章發表於 : 週四 07 9月 06 21:50:23 
討論區會員

註冊時間: 週一 19 12月 05 11:56:17
文章: 194
唔好意思, 因為部機之前影開人, set左+0.3step, 原裝canon鏡因為測光準確, 所以d相over左, d色淡左


回頂端
 個人資料  
 
 文章主題 :
文章發表於 : 週日 08 10月 06 01:56:59 
理事
理事

註冊時間: 週六 08 11月 03 07:56:21
文章: 5762
jason.lau 寫:
Wah dada, can NAM-1 be used on your adapter :?:


I didn't test.........and not able to test coz I dun have the NAM-1

but as per suggestion from Conurus the firmware version would not match and the 645 lenses should not be able to function.......

_________________
圖檔
Wah
理事 - Board of Members
www.flickr.com/dicksonlau


回頂端
 個人資料  
 
顯示文章 :  排序  
發表新文章 回覆主題  [ 103 篇文章 ]  前往頁數 上一頁  1 ... 34567  下一頁

所有顯示的時間為 UTC + 8 小時


誰在線上

正在瀏覽這個版面的使用者:沒有註冊會員 和 12 位訪客


不能 在這個版面發表主題
不能 在這個版面回覆主題
不能 在這個版面編輯您的文章
不能 在這個版面刪除您的文章

搜尋:
前往 :  
cron
Powered by phpBB® Forum Software © phpBB Group
正體中文語系由 竹貓星球 維護製作