[Pentaxor]

some real world samples from the NX30 to those who would want to know what to expect from the lens.


[

[Laurentiu Cristofor]

also, a word of caution. the 2 other previous tests made the FA35 look poor.

Thanks for the links.

The NX30 looks excellent, there's no doubt about it. I liked the result in the first test, with the apple label crops. I don't expect the FA35 to be better than what you're getting from the NX30, but in some shots it looks particularly bad. I saw that you mentioned using a hood for some tests only - were most of these done without a hood?

Anyway, it's nice to see such output from the NX system. Samsung is putting out some nice stuff. Did you also try their NX20 lens? When you can, please share more results from your NX in the non-Pentax cameras forum.

[Pentaxor]

Thanks for the links.

The NX30 looks excellent, there's no doubt about it. I liked the result in the first test, with the apple label crops. I don't expect the FA35 to be better than what you're getting from the NX30, but in some shots it looks particularly bad. I saw that you mentioned using a hood for some tests only - were most of these done without a hood?

Anyway, it's nice to see such output from the NX system. Samsung is putting out some nice stuff. Did you also try their NX20 lens? When you can, please share more results from your NX in the non-Pentax cameras forum.

to be honest, some FA35 results did look bad that made me think I had a bad copy. but having done a lot of re-tests suggests that it was not. it just that it had it's time in the limelight pass already. some FA35 tests were done without a hood and surely it would had helped. what is interesting is that the NX30 doesn't have one nor do I feel it even needs one. I'm not sure if I would get a hood for it.

unfortunately, I haven't tested their NX20 which was a mistake and I should have atleast tried it while I was at the store. probably because I was not interested on the 20mm focal length. although I might try their NX16 which I'm carefully considering as a portable UW lens. I would share my findings by then. but now, time for me to put some of my old lenses up for sale.

[ManuH]

Something doesn't look right in these tests. It almost look like if the Samsung 30mm f/2 simply got more sharpening applied. Does Samsung apply software correction (like distortion) on their lenses? From Photozone: "The vignetting characteristic of the Samsung lens is a bit ... odd. Basically there's no significant vignetting to speak of. Regarding the basically non-existing increase at f/2 we suspect an auto-correction within the camera". The Photozone results also suggest that we shouldn't see that much differences between the two lenses.

It would be interesting to compare the results with the FA 35mm f/2 mounted on a K-7.

[ManuH]

OK out of curiosity I just tried my "theory".

I sharpened the FA 35mm pictures and got almost exactly the same sharpness as the NX 30 results. Looking closer, at the flower picture for example, on the NX30 we can see there is more sharpening as there is a fine white line around the flower. For some reason it seems to me that the camera is applying more sharpening to its native lenses.

[Pentaxor]

Something doesn't look right in these tests. It almost look like if the Samsung 30mm f/2 simply got more sharpening applied. Does Samsung apply software correction (like distortion) on their lenses? From Photozone: "The vignetting characteristic of the Samsung lens is a bit ... odd. Basically there's no significant vignetting to speak of. Regarding the basically non-existing increase at f/2 we suspect an auto-correction within the camera".

interesting theory you got there. I'm not sure how Samsung would had done such sharpening by electronically relay IQ rendering thru software/firmware. what I only do know is that Pentax for example has able to optimize their DA LTD lenses for APS-C but not thru firmware but optical modification. although I'm not dismissing the possibility of image post-processing application on the Samsung lenses, particularly distortion but not so much on sharpening and vignetting, there is no way of telling it. regardless of which, we still get excellent results.

The Photozone results also suggest that we shouldn't see that much differences between the two lenses.

maybe I missed out on something, but I haven't read somewhere that Photozone had suggested such statement.

It would be interesting to compare the results with the FA 35mm f/2 mounted on a K-7.

see part I and part deux comparison threads, the FA35 were shot with a K-7. part III FA35 images were shot with the NX100 thru an adapter.

[Pentaxor]

OK out of curiosity I just tried my "theory".

I sharpened the FA 35mm pictures and got almost exactly the same sharpness as the NX 30 results. Looking closer, at the flower picture for example, on the NX30 we can see there is more sharpening as there is a fine white line around the flower. For some reason it seems to me that the camera is applying more sharpening to its native lenses.

come to think of it, if that were case, then such sharpening would be done on other NX lenses as well and would result to similar IQ performance. I didn't see such sharpening happen or being applied on the NX20-50 kitlens. although it performed very good, it wasn't near the performance of the NX30.

[ManuH]

but not so much on sharpening and vignetting, there is no way of telling it. regardless of which, we still get excellent results.

Vignetting is obviously corrected (as per Photozone results) and is done by others (Panasonic on m4/3?). Sharpening is a bit weird unless Samsung has a sharpening algorithm that can be optimized for a lens?

maybe I missed out on something, but I haven't read somewhere that Photozone had suggested such statement.

I just compared both graphs, tested on K10D and NX10 (I know I know results shouldn't be cross comparable). But both lenses follow a comparable curve, quite sharp from f/2, extremely sharp stopped down.

see part I and part deux comparison threads, the FA35 were shot with a K-7. part III FA35 images were shot with the NX100 thru an adapter.

I missed that one, thanks. But if the NX is sharpening its results, the problem is the same. Although on that particular sample, I couldn't match the NX with sharpening. But the FA35 looks really blurry and not up to its reputation (sharp at all apertures).

LR could also be the culprit, applying different type of demosaicing. That's why some reviewers prefer to use dcraw for its neutrality.

Interestingly Lenstip also tested both lenses. They noted that the NX10 sensor reached 1-3 lpmm more than the K-7 / K20D. On their tests the NX30 scored 36 and the FA 35mm 30 in the center, on the edges they're equal. That's a small difference that shouldn't be that visible IMHO.

BTW Lenstip used dcraw, they could detect the difference in vignetting that Klaus couldn't.

[borno]

Looks to me like the samsung lens is as sharp in the corner wide open as the Pentax film lens is at the center. I'm real happy with mine. (obligatory cat shot)

[Pentaxor]

Interestingly Lenstip also tested both lenses. They noted that the NX10 sensor reached 1-3 lpmm more than the K-7 / K20D. On their tests the NX30 scored 36 and the FA 35mm 30 in the center, on the edges they're equal. That's a small difference that shouldn't be that visible IMHO.

BTW Lenstip used dcraw, they could detect the difference in vignetting that Klaus couldn't.

it was reiterated further however that the difference in sensor performance is not that huge. besides, the test was made on different systems (indicative of the first 2 threads that I posted). lenstip uses a different resolution measurement. one could easily interpret the a 5 lpmm MTF50 resolution gap to be not that significantly far or is it? mind you that the interpretation of any lens below 30 lpmm is considered to be underperforming. and further as stated "we must stress here that the MTF50 values reaching 30-35 lpmm, like those of the Pentax at the frame edge, are good and allow to generate sharp photos. Perhaps those photos won’t be outstandingly sharp but they will be sharp enough for most uses.", where the FA35 barely reached 30lpmm but still quite good to be honest. so a 4-6 lpmm advantage could also mean a lot to such interpretation. overall however, it looks like the NX30 is the better performer. to be honest, it is good to see some comparative numbers as well for scientific purposes. although I must say that the images looks to be more of a better comparison personally speaking if we would like to see how a lens does in real life.

[ManuH]

one could easily interpret the a 5 lpmm MTF50 resolution gap to be not that significantly far or is it?

MTF50 is supposed to be close to what a human eye perceive as sharp. But for the fun, I sharpened the FA35 pictures similar to the NX30:

NX30


FA35



NX30


FA35



NX30


FA35



Now they are much closer don't they?

[Pentaxor]

they do look a bit more closer now in terms of sharpness. although it is still significantly noticeable the elaborate differences in saturation, contrast and detail between the two lenses. if one was to copy the exact rendering of one lens, it would take a bit more careful pp work.

[ManuH]

they do look a bit more closer now in terms of sharpness. although it is still significantly noticeable the elaborate differences in saturation, contrast and detail between the two lenses. if one was to copy the exact rendering of one lens, it would take a bit more careful pp work.

That's a valid conclusion if you're going to use the NX system: there's no point to use a FA35 on this setup as the NX30 performs as good if not better. And certainly better out of the box. Personally I prefer the 30mm focal length, it's a perfect match as a "normal" on APS-C. The Samsung NX10 appeals me very much, its form factor is ideal for a mirrorless camera IMHO. But when I tried it I was underwhelmed by the loud shutter and the low-res EVF. Also MF doesn't seem to be that easy. I have a Samsung EX1, it performs well, good hardware but GUI is a let down. It seems that Samsung still has to mature a bit on that front.

But from your tests, we cannot conclude the NX30 is better than the FA35 optically. Yes saturation and contrast are not the same, but I'm pretty sure it's a matter of profile applied. I cannot see a big difference in details rendered, at least with these subjects.

[audiobomber]

I've followed Pentaxor's three threads. I also read MightyMike's thread on DPR that attempted to equalize normal lens MTF performance using Lenstip test results. The FA 35mm is a fine lens, there's no doubt about it. I have to conclude that the Samsung 30mm is a finer lens. Unfortunately, it won't fit on a Pentax camera and I have no interest in a Samsung camera, so the NX30 performance is not terribly significant to me.

[Pentaxor]

Personally I prefer the 30mm focal length, it's a perfect match as a "normal" on APS-C.

this would be the most ideal reason for having the lens, the IQ performance made it much more valuable IMO. indeed it is almost perfect focal length.

The Samsung NX10 appeals me very much, its form factor is ideal for a mirrorless camera IMHO. But when I tried it I was underwhelmed by the loud shutter and the low-res EVF.

true, the form factor is really very nice and I like it. it is not necessarily for pocket cam enthusiasts, but more of a compact dlsr alternative. the loud shutter does indeed may need some improvement although it's not that bad nor far behind the K-5/K-7's shutter. and far from the clunky noise of other brand dslrs make. although to be fair, it's AF is more of a silent whirl compared to the zipping noise that a Pentax screwdrive make. so as far as shooting sound goes, they have some compromises. EVF on the otherhand is not that bad but can be improved.

But from your tests, we cannot conclude the NX30 is better than the FA35 optically. Yes saturation and contrast are not the same, but I'm pretty sure it's a matter of profile applied. I cannot see a big difference in details rendered, at least with these subjects.

as I mentioned, one can choose to accept or reject such conclusion. I'm not sure what you meant by profile applied if it has to do with software and not optical rendering, but if we have that kind of thinking, does that make it different from comparing a DA LTD over an FA LTD comparison? like contrast and colour is due to profile applied and not the lens' own optical characteristic? besides, it would be illogical to impose different degree of profile application on such lenses if that was possible. for what I know and what I saw is that the NX20-50 and NX30 don't share the same characteristic that would suggest such profile application.

one may have doubts maybe due to Samsung being a new market player and hasn't established a reputation on glasses, especially if that would make Sony look stupid for not constructing a lens on their own without having to have someone specifically make one for them. although come to think of it, Samsung does have some of Pentax's glass blueprints where they could always make a reference of whenever they want to construct a new lens design. so they are not totally inexperienced. remember that they did produce the sensor for Pentax before, so why not lenses?

it's a matter of budget research, manufacturing and dedication that would enable a company to produce a lens. the important thing is that Samsung is able to optimize lens rendering that one camera manufacturer can only hope for. now who wouldn't be happy with that? just imagine if such 30mm lens existed for Pentax as a cheap alternative instead of the DAL35. I'm sure no one would even discuss about profile application now, would we?