For denizenx: Test Shots from Vivitar 19-25mm f3.5-4.5

[Tweek]

These test shots are requested by Ser Wei, but may be useful to those who are interested to find out more about this lens, which I think is only a small handful as most are looking out for the Sigma, Canon and Nikon actually.

The shots are taken by EOS D30, at 3MP Fine, straight from the camera (unless otherwise stated).

Firstly, a comparison between wide-open at F3.5 and stopping down to F8. Ser Wei, this is what I meant by very soft at wide open and a marked improvement at F8, tho still a little too soft but I can live with it.

Full size crop from original image

Next, an outdoor scene that looks like this originally:

USM applied

Here all full size crops of pics taken at F3.5 and F8 respectively. One noticeable problem is that at wide open, the scene is consistently underexposed.

Lastly, the aliasing problem I was telling Ser Wei about. This sample is cropped from the fence-like structures on the building. You can see the red/green streaks due to aliasing. I've had similar streaks appearing on closely-packed windows on a CBD building.

I must say that despite all these drawbacks, the lens is still quite a good bargain, and it will give me sharp enough pictures when printed at 4R, which is what I do usually. I don't need the Ls and the EXs (yet) to satisfy my needs as an amateur, hehehehe, despite what Simon always tells me.

 

[Knighthunter]

I think that phenomenon is called chromatic abberation

 

[Tweek]

Originally posted by Knighthunter
I think phenomenon is called chromatic abberation

CA doesn't look like this, CA is blooming over edges of high contrast, and it doesn't happen in streaks. Aliasing happens to areas where resolution is very fine, too fine for the lens to resolve (above Nyquist Frequency). This phenomenon featured in my post only happens to areas where resolution is very fine or details are very packed.

 

[Knighthunter]

Originally posted by Tweek
Nyquist Frequency

:what: wow I never tought that Nyquist Frequency also generated by lens.....I never think CCD alogarithm (FFT) related to the lens. If this statement is true this will only happen on DSLR, it won't appears on the film since no alogarithm involved in 35mm (Only CA may present in 35mm).
I am not sure, do you have any reading source for this? Kindly enlighten me.

 

[Tweek]

You can refer to dpreview's reviews of various cameras. One of the tests is the resolution test, and Phil Askey will explain a bit on aliasing and Nyquist Frequency. I hope my understanding on this topic is accurate though.