[yusuf]

It's often stated here that Pentax 50 f1.7 lenses are more sharper than 50 f1.4. So I decided to do a quick real-life comparison along with two other non-Pentax lenses - at wide open and at f2.

IMO, A50/1.4 @f1.4 is as sharp as M50/1.7 @f1.7 at the same focus point, any difference is just marginal. However, both the lenses renders quite differently & it's quite evident at f2 where A50/1.4 blurs background more compared to M50/1.7 @f2. M50/f1.7 also have slightly better corners at wide open.

All photos on FF Sony A7. Please share your opinion and your pick among #1 to #4

Wide Open

#1 A50 @f1.4


#2 M50/1.7 @f1.7


#3 @f1.4


#4 @f1.8



@ f2
#1


#2


#3


#4

---

Last edited by yusuf; 03-06-2015 at 09:41 AM.

[jrpower10]

Sorry, I'm probably being a bit thick but I'm not following. The #1 photos are the A50 1.4 and the #2 photos are the M50 1.7 correct? and what are #'s 3 & 4?

Wide open hte A50 1.4 is going to have shallower DOF than the M50 1.7 so yes, the OOF areas will look different. I'd be curious to see the A50 1.4 topped down to 1.7 compared to hte M50 1.7 wide open (or at the same aperture). Guess I'm wondering if using a wider aperture lens and stopping down a little gives better results than starting with a lens with a narrower max aperture. At f2 I think I prefer #2 (the M50 1.7 I presume).

[ChristianRock]

The first aperture click down on M 50 1.7 isn't f2. From my non-scientific calculations based on shutter speed required for the images to look exactly the same, I think it's an f/2.4. So yeah, it's going to blur the background less than the 50 1.4 at f2. Will probably look a bit sharper as well.

But I'd say it's a bit hard to look for differences in the examples posted. First because these differences probably aren't going to show unless you show a crop. Second because every channel, except for the red channel, looks a bit underexposed. The camera probably did that so the red channel wouldn't blow up.

[yusuf]

The first aperture click down on M 50 1.7 isn't f2. From my non-scientific calculations based on shutter speed required for the images to look exactly the same, I think it's an f/2.4. So yeah, it's going to blur the background less than the 50 1.4 at f2. Will probably look a bit sharper as well.

That's a good information. The point is f.17 lenses are not sharper than f1.4 as it's often claimed.

...I'd be curious to see the A50 1.4 topped down to 1.7

Unfortunately, A50 has no stop between f1.4 and f2 but you can refer f2 pictures (with ChristianRock's clarification above)

[stevebrot]

differences probably aren't going to show unless you show a crop.

Yep...even back in the film days, the evaluation was done using a resolution target and a direct examination of the negative using a microscope. Taking the vintage view of things, the sharpness claims regarding the Pentax 50s (f/1.7 vs f/1.4) go back to published magazine tests from the 1980s and has been confirmed with other tests since. At f/2, the slower 50/1.7 is sharper than the 50/1.4 and continues to lead until f/5.6 and narrower where results are comparable. At least that is how I remember it.

As for the series of photos, I see 8 reasonably sharp photos of a low-contrast subject representative of what I would expect of wide aperture shots from about 90% of fast 50s made in the last 50 years.

In regards to the first click on the 50/1.7, the increment is 1/2 stop and the value is nominally f/2. Whether it is a true f/2 is a matter of conjecture. Response from the in-camera meter is not a good indicator of lens aperture since multiple factors (vignette and bokeh) affect the TTL metered exposure. I suppose I could go outside and test my M 50/1.7 against the blank sky, ;)

Edit: Silly me, I had forgotten that the Pentax convention for the dot or unmarked stop is one full stop narrower than the maximum aperture rather than the 1/2 stop interval for the rest of the ring. I only remembered when I looked in the FA series manual.


Steve

[ChristianRock]

That's a good information. The point is f.17 lenses are not sharper than f1.4 as it's often claimed.

Well... you have made a conclusion based on the copies that you have... so while it might be true for those two copies, one would have to test more copies to say that's a definite.

You wouldn't be alone in saying it, though. Yoshihito Takinami's site for example has tests claming the same - that basically all the 50 1.4 and 50 1.7 versions have the same resolution wide open. The 55s and the 50 1.2s are a bit behind in his tests.

[geomez]

I had read that the 1.4 was softer up to f2 but then came out ahead over the 1.7.

[stevebrot]

FWIW, I have purchased both Pentax-M and ST 50/1.4 lenses and took both back because I was disappointed by their performance relative to my slower 50mm lenses.


Steve

[stevebrot]

I suppose I could go outside and test my M 50/1.7 against the blank sky,

Ok...I was able to confirm that the click stop is indeed at f/2 on both the M 50/1.7 and A 50/1.7. At least that is how it registers on my cameras having the non-crippled K mount. As for how the M50/1.7 meters at f/2 vs f/2.8 on a Pentax dSLR, that is a whole 'nuther matter. Stop-down metering with the crippled mount is not particularly linear, not even on the current model cameras.

Edit: I am not quite so sure after spending some more time with the camera. I did the look-see on my Ricoh XR-2s which has an analog needle for the shutter speed read-out. The difference between f/1.7 and the unmarked stop is indeed a little more than 1/2 stop, but the difference between there and f/2.8 is also about 1/2 stop. I would love to compare to the KX, but that camera is in the shop.

That being said, I now tend to agree with ChristianRock. The reason is a paragraph in the lens manual for the FA series lenses. In that manual, it says that the dot position (equivalent to unmarked stop) is one stop narrower than the maximum aperture. For f/1.7, that would be f/2.4.


Steve

[troika]

Thanks for doing this, I enjoy these tests. I did a very similar one with a K55/1.8 and a K50/1.4 https://www.pentaxforums.com/forums/10-pentax-slr-lens-discussion/229434-shoo...k55-1-8-a.html

I loved both lenses and was surprised at little actual difference there was between the two, except potentially subjective rendering difference. What kicked it for me was 8 blades vs 6. In middle apertures with a busy and well seperated background, I dramatically preferred the 1.4 lens for the geometric bokah. It might be interesting to see you test that with these two lenses and see what happens.

[timo]

I have had a variety of M and FA 50s over the years, and my impression has been that the 1.4 is just as sharp as the 1.7 at equivalent wide settings, but has lower contrast wide open; and that stopped down the corners of the 1.4 are very slightly better. But the differences are tiny, and I am not going to waste more of my life doing more comparisons! I'm sure some of the comments on the 1.4's alleged softness wide open are down to the difficulty of getting the focus spot-on with such a shallow DOF.

[pinholecam]

I have had a variety of M and FA 50s over the years, and my impression has been that the 1.4 is just as sharp as the 1.7 at equivalent wide settings, but has lower contrast wide open; and that stopped down the corners of the 1.4 are very slightly better. But the differences are tiny, and I am not going to waste more of my life doing more comparisons! I'm sure some of the comments on the 1.4's alleged softness wide open are down to the difficulty of getting the focus spot-on with such a shallow DOF.

+1 to this.

I too have owned all the m50's and A50's.
And the FA50/1.4
No real difference in sharpness between them at the respective apertures.

The very hexagonal oof highlights between f2.8-f5.6 on the f1.7 and f2 50s drives me nuts though.
Other than that, the 50/1.7 is easily on par with anything out there.
Even a Rollei 50/1.8 I compared it against.
Don't even want to mention the RF Cron 50 DR which did not appeal to me at all.

[Not a Number]

Edit: I am not quite so sure after spending some more time with the camera. I did the look-see on my Ricoh XR-2s which has an analog needle for the shutter speed read-out. The difference between f/1.7 and the unmarked stop is indeed a little more than 1/2 stop, but the difference between there and f/2.8 is also about 1/2 stop. I would love to compare to the KX, but that camera is in the shop.

Every test I did with my SMC Pentax-A 50mm f1.7 indicates that it is less than 1 full stop from f2.8 to the unmarked click stop between 2.8 and 1.7. I checked it on a K1000, Program Plus and K10D. I even did Lowell Gouge's histogram measurement procedure and it still came out as less than 1 stop. I even posted the histograms and numbers and everybody kept saying I was wrong just because some Pentax literature says it is one stop. Well may be it is my lens that is off even though others have reported similar findings. It's not like the Pentax literature has always been correct either or there may be exceptions.

Perhaps somebody with an optical test bench can measure these lenses accurately.

I just had a thought that I can photograph the iris and we can measure the pupil to derive the f-stop mathematically.

[monochrome]

That's a good information. The point is f.17 lenses are not sharper than f1.4 as it's often claimed.

This is what I've learned over the years - actual professionals who've been active since the 60's might amplify this summary.

The general historical explanation of the difference in design goals between the 55/1.8's and 50/1.7's, and the 50/1.4's, related to their different intended purposes.

As a general rule the 50/1.4's were thought to be designed intentionally with softer edges and a sharp center area in order to create a pleasing 3D effect for a centered subject. That was a desirable artistic, aesthetically preferable goal in the 60's - subjectively, a 'good' subject image.

The 55/1.8's and 50/1.7's were designed for 'flat work' or 'copy work', i.e. they were intentionally designed to be uniformly sharp corner to corner. At the time the printing industry required an actual photographic negative of a 'paste up' from a typesetting machine. A camera was mounted to a Copipod or camera stand (a device to hold a camera precisely vertical) so a flat sheet could be photographed on a photo table. A printing plate was made from the negative and mounted to the printing drum, etc. There was specialized equipment for this, but an SLR and accessories with a 'flat' lens could be used in small print shops.

Apparently, 1.7's and especially f/2.0's were marketed as 'consumer' lenses and often included as a kit lens - partially to allow increased margins on the less common 50/1.4's. A 50/2.0 of course starts nearer the naturally sharper part of the aperture range, so . . . . .

Today, with digital image viewing, our aesthetic tastes have evolved to value images that are sharp across the frame. When we compare legacy 1.2's and 1.4's to legacy 1.7's and 1.8's we are expressing our current aesthetics and forgetting the original design goals.

[audiobomber]

I have had a variety of M and FA 50s over the years, and my impression has been that the 1.4 is just as sharp as the 1.7 at equivalent wide settings, but has lower contrast wide open; and that stopped down the corners of the 1.4 are very slightly better. But the differences are tiny, and I am not going to waste more of my life doing more comparisons! I'm sure some of the comments on the 1.4's alleged softness wide open are down to the difficulty of getting the focus spot-on with such a shallow DOF.

Another problem is that the FA 50 1.4 front element is not recessed at all, making it more subject to veiling flare. That lens definitely needs a lens hood, at all times, and can handle a deep one on APS-C. Here's the hood I use:



And here are some gratuitous images @ f/1.4: