[interested_observer]

The new Pentax didn't fair so well against the Tamron - especially on the long end. One of the comments suggested that Tamron did all the glass manufacturing themselves, keeping the best for their build and providing less than perfect glass for the Pentax build. So, read the comments.

You Tube

You Tube

[Madaboutpix]

One of the comments suggested that Tamron did all the glass manufacturing themselves, keeping the best for their build and providing less than perfect glass for the Pentax build.

Frankly, that bit sounds like a conspiracy theory to me.

[Sandy Hancock]

Copy variation.

[Rondec]

I think Sandy is right. This is why it is incumbent on the lens purchasers to shoot with a new lens and see if there are issues. If a lens is soft or has decentering or whatever, you can send it back and get another copy.

There's no way that Tamron is sending worse quality lenses to Pentax (see comparative reviews of Pentax DFA 15-30/24-70 and Tamron versions).

[cport]

Hopefully.


Anyway, it shows at least the Ricoh's QC.

[bdery]

One of the comments suggested that Tamron did all the glass manufacturing themselves, keeping the best for their build and providing less than perfect glass for the Pentax build.

Because someone says it on the internet does not make it true...

I'm reviewing / testing the D FA 70-210 right now, and apart from bokeh which is a letdown, it's hard to find big flaws with that lens.

[normhead]

Hopefully.


Anyway, it shows at least the Ricoh's QC.

You really need to read the lens rentals.com page on sample variation. If you replace Ricoh's QC with standard camera industry QC then you're on to something. You can buy a $15,000 lens from canon, Nikon or Sony and have QC issues.

So, that begs the question, if people don't understand industry wide QC and base a whole report one copy of a lens, is it not incumbent upon them to pay at least lip service to sample variation. After all, many have sent a lens in to a repair depot and have it sent back as "up to spec." when it still performed poorly. Lensrentals.com points out in their article most facilities do not have the equi[ent to test for decentering or mis-aligned lens elements. Such equipment would be extremely expensive. The guy at lens rentals makes adjustments with shims using pretty much informed trial and error. He knows of no facility to send his lenses out where repairs can be done to his satisfaction

While it would be great if they only shipped lenses from the top half of the QC curve, that would double the cost of less as we'd still have to pay for the bottom half. It's sad.

On the positive note, my first DA 55-300 was broken in half my second one is much better right out of the chute. I paid for two lenses, but the one I have now, I'll never part with. I also suspect my 18-135 is from the high end of the curve, but I got lucky.

The sad thing is when people take your advice, look at your pictures, buy the lens and then don't get the results you did. It really is a crap shoot.

Looking at my own expereince, most of my lenses performance is acceptable. My first DA 55-300 was on the low end of the scale (but I was still happy using it, my new one is just better) and is the only one of my 10 or so Pentax lenses where a noticeable performance was likely to improve replacing the lens. My conclusion is industry QC keeps lenses affordable. Adding a whole new level of QC to make sure every lens is above the current median for sample variation would be very expensive and not necessarily all that worthwhile. Most image don't depend on good sharp lenses for their IQ. That being said, if you get an excellent lens at any focal length don't sell it. The thought that you can buy another that's just as good may not in fact be the case without two or three tries.

A lens comparison without reference to the lens rentals documented sample variation for that lens is probably not worth a whole lot. You need to know how 10 less tested out, not one. And you need to know where on the curve compared against other copies of the lens the lens tested out.

The biggest problem with these comparisons is that there is nothing in it for the camera companies to run the tests themselves. If people had a way fo selecting only the best performing lenses from a production run, who would buy the still within spec, but poorer performing lenses? You'd have to charge more for the high performing ones and less for the lower end of the spectrum. A totally unworkable arrangement.

Sample variation is just something you have to live with.

---

Last edited by normhead; 03-25-2020 at 06:40 AM.

[pres589]

I take the review statements from the Cameraville guy with large grains of salt. I do like that he shows images from his tests but sometimes they don't support his statements at all. I do like that he's pretty intent on showing off Pentax k-mount gear but he's also stated he doesn't actually shoot with it day to day and mostly just does it because it's a niche that he felt he could fill.

[cport]

You really need to read the lens rentals.com page on sample variation. If you replace Ricoh's QC with standard camera industry QC then you're on to something. You can buy a $15,000 lens from canon, Nikon or Sony and have QC issues.

So, that begs the question, if people don't understand industry wide QC and base a whole report one copy of a lens, is it not incumbent upon them to pay at least lip service to sample variation. After all, many have sent a lens in to a repair depot and have it sent back as "up to spec." when it still performed poorly. Lensrentals.com points out in their article most facilities do not have the equi[ent to test for decentering or mis-aligned lens elements. Such equipment would be extremely expensive. The guy at lens rentals makes adjustments with shims using pretty much informed trial and error. He knows of no facility to send his lenses out where repairs can be done to his satisfaction

While it would be great if they only shipped lenses from the top half of the QC curve, that would double the cost of less as we'd still have to pay for the bottom half. It's sad.

On the positive note, my first DA 55-300 was broken in half my second one is much better right out of the chute. I paid for two lenses, but the one I have now, I'll never part with. I also suspect my 18-135 is from the high end of the curve, but I got lucky.

The sad thing is when people take your advice, look at your pictures, buy the lens and then don't get the results you did. It really is a crap shoot.

Looking at my own expereince, most of my lenses performance is acceptable. My first DA 55-300 was on the low end of the scale (but I was still happy using it, my new one is just better) and is the only one of my 10 or so Pentax lenses where a noticeable performance was likely to improve replacing the lens. My conclusion is industry QC keeps lenses affordable. Adding a whole new level of QC to make sure every lens is above the current median for sample variation would be very expensive and not necessarily all that worthwhile. Most image don't depend on good sharp lenses for their IQ. That being said, if you get an excellent lens at any focal length don't sell it. The thought that you can buy another that's just as good may not in fact be the case without two or three tries.

What shall i read about QC?


Looking at my own experience, I have never had problems with quality of my lenses, so I am quite confident about Pentax lenses.


But it woud be good to know where the problem in the test was. It is possible that the reviewer did a mistake (very unlikely - it is his business), or that he had one bad copy of 70-210/4 lens. And I can hear it quite often that even very expensive Pentax lenses had to be exchanged because of this. Of course, I have no statistics about this issue.

Btw. 11-18/2,8 is on the way to my home right now.

[normhead]

I take the review statements from the Cameraville guy with large grains of salt. I do like that he shows images from his tests but sometimes they don't support his statements at all.

Even with scientific papers, a very high number of studies have titles suggesting theories that are not supported by their data. It's really hard to stop the researchers opinions from creeping in.

---------- Post added 03-25-20 at 09:47 AM ----------

What shall i read about QC?

Lens Rentals | Blog

[pres589]

Even with scientific papers, a very high number of studies have titles suggesting theories that are not supported by their data. It's really hard to stop the researchers opinions from creeping in.

---------- Post added 03-25-20 at 09:47 AM ----------



Lens Rentals | Blog

I seem to remember him being pretty negative in comments about the Pentax 150-450 but his tests showed it to be very good indeed except that he thought it should exhibit less focus breathing. In every other way it bested the lens it was compared to, I think a 100-400mm Sony that had flare and fringing issues well in excess of what the Pentax produced. It was almost night and day, but you know, that breathing...

[cport]

Even with scientific papers, a very high number of studies have titles suggesting theories that are not supported by their data. It's really hard to stop the researchers opinions from creeping in.

---------- Post added 03-25-20 at 09:47 AM ----------



Lens Rentals | Blog

Yes, this is quite expected - a single lens is not a representative for all lenses of the same kind. As well as it is expected that high-end lenses will keep higher standards.

[normhead]

Yes, this is quite expected - a single lens is not a representative for all lenses of the same kind. As well as it is expected that high-end lenses will keep higher standards.

I'd also point out that higher test results do not correlate exactly to higher customer preference for the lens. Pentax's old design philosophy, which was actually a bit mis stated took that into account. The idea was the test chart lenses are not necessarily preferred by people when looking at images, something they discovered with their own research. The simple fact is, high test cart performance does not guarantee more customer satisfaction. And low test chart scores don't mean there won't be instances where you don't prefer the lens.

Some of my "I only bought it because it was cheap" lenses have produced some great images, maybe you can't pixel peep them or blow them up to 48x32 inches, but if you don't do that their rendering is clearly superior to some that will absolutely blow them away on the test charts.

[cport]

I agree.


Ephotozine recently tested some very old and cheap Pentax lenses and they were great.


Btw. it seems that in the video above he really did not have a good piece of this lens. Here you have another (today published) review on Ephotozine

Pentax HD Pentax-D FA 70-210mm f/4 ED SDM WR Review | ePHOTOzine

claiming that its optical performance is really close to D FA 70-200/2,8.


So let's wait for other tests ...

[blackcloudbrew]

Ok, I have not watched the video -yet. I also own a copy of the new Pentax 70-210 and like bdery, so far the only negative I've observed is the bokeh, otherwise it's produced great results so far. However, whatever conclusions one might gain from the comparison above, what's the point? We don't have an option to use the Tamron on our K-mount, and whether or not there may have been QC problems with the test Pentax lens aside. I'm just not sure I need to be concerned about the performance of a lens I can't use.

Ok, ok, I can see this from a curiosity standpoint, not much else.