Rifle Scopes Why isn't glass objectively measured?

[koshkin]

Re: Why isn't glass objectively measured?

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: dbooksta</div><div class="ubbcode-body">Thank you Ilya. I think you've given good reasons why manufacturers haven't released technical data.

Note, however, that the data aren't "proprietary:" Anyone can take a sample and run it through all these tests. I like valise's mention of dpreview.com, which has done a great deal of this for camera glass.
</div></div>
The data is most certainly proprietary. If you take the same scope and run it on a different piece of equipment, that is an entirely different data set that may or may not agree with what the manufacturer has measured. To acquire that data (and make sense of it) you need to know what you are doing.

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: dbooksta</div><div class="ubbcode-body">
And I like the way at least some camera lens makers have done a lot more in terms of publicizing this information. Why does one lens cost $1500 when another with identical speed and range costs $500? Here are the MTFs. Heck, they'll not only tell you how many lenses are in the system but even the type and function of each. Even if you don't understand these data at least you have something concrete backing up the experienced users offering their subjective, "that $1500 glass is way sharper at all apertures."

Granted, I imagine there are a lot more professional photographers than professional shooters, so maybe your point is that you need an expert user base of a certain size to make these data worth releasing. Heck, the camera user base is evidently large enough to support significant <span style="font-style: italic">third-party</span> objective testing of lenses. Are rifle scopes that fringe of a market?
</div></div>
Photography people, on average, have no idea how to read an MTF curve either. However, the big difference is that there you can measure system MTF, which includes both the lens and the image acquisition device (the imaging sensor). That results in data you can compare between different products. With riflescopes, since the image acquisition device is your eye, you can not easily generate system MTF. Even with all that in mind, photography people make all sorts of assinine decisions about their gear simply because they do not understand published data (I worked in commercial imaging world for about five years and it is a major pain in the ass).
Most importantly, as Lowlight pointed out, getting appropriate measurements on the camera is fairly cheap. Setting up a proper metrology lab for characterizing riflescopes would cost me upwards of $50k. How do you propose I finance it? More specifically, if someone offered that service, how much would you be willing to pay for that?

On an off topic, now that I am curious, what is your engineering degree in?


<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: dbooksta</div><div class="ubbcode-body">
If the reason manufacturers don't provide better optical data on 4-figure scopes is that shooters are, on the whole, too stupid or immature to handle it, then that's a damned shame. If the reason we don't have third-party analysis is that the market is too small, then shoot.

Perhaps we could provide a "safe-harbor" warehouse for this information online: Manufacturers, third-party testers, and anyone: submit your optical data to be served up by the scope warehouse, and then disclaim any responsibility for "supporting" those data. I.e., if somebody calls to discuss your MTF curves, do what you do to me when I call asking for even basic figures like resolving power: Say, "I don't know if we have those numbers; I'll try to see if somebody in engineering does and never get back to you."

Like you said, a lot of buyers never look for the data, and those who do may not know what to do with it. But there's a lot of us with engineering degrees and know-how who think long and hard before dropping 4-figures on a scope. And we don't all have the luxury of lining up $20k worth of candidates in circumstances where we can subjectively evaluate their relative optical merits. And we shouldn't have to given that objective measures can tell us what we're looking for. </div></div>

 

[Grump]

Re: Why isn't glass objectively measured?

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: ILya</div><div class="ubbcode-body">Interesting thread.

As a matter of background: I work for a company that makes optical and electro-optical test sets for all sorts of weapon sights, thermal cameras, etc.

[snip]
- Resolving power: how do you know what the limiting factor of the system is? is it the riflescope? or is it your eye? how are you going to account for the variations in resolving power of your eye with respect to eye pupil dilation, hydration on that day, etc?
ILya </div></div>
That's the biggest red herring in this whole thread. Differing eyeballs/retinas/brain housing units to process the image is a given, and it will cancel out. Hotshot eyeballs will see better with EVERYTHING. So, THAT'S why we want an objective instrument to measure resolution.

My left eye resolves noticeably sharper than my right eye. But I demand so much from my eyes, it drove my ophthomalogist crazy almost. I detected and objected to image "smearing" from astigmatism which most of his patients don't care about or don't even notice.

If Ransom can make a machine rest adaptable to virtually every handgun made, the interface between a rifle scope and test equipment is just a matter of fixtures and 16th-century mechanical engineering.

And yes, this will test ONLY the optical qualities, NOT tracking or accuracy of reticle hashmarks or shock resistance in three separate planes or many of the other factors critical to good total performance in a rifle scope.

But to avoid the "Ford Taurus Factor" (crappy car but they sold so many there is a HUGE herd of satisfied idiots who also were either lucky enough to not get a specimen defect, OR conveniently forget how many times the piece has been in the shop...all = "most popular car in America!), I vote for objective measurements. Then I can decide for myself whether I care in my own uses about the difference between 92% and 96% "light transmission".

 

[Grump]

Re: Why isn't glass objectively measured?

BTW, I forgot to profusely thank ILya for bringing so much informed and useful information to this thread.

And others. It's like the good old days before my time when many NRA members were also machinists and contributed fascinating articles about projects (gun mods or reloading tools or whatever) they had done themselves because they not only understood the scientific process but also had the technical know-how to pull it off. My Dad's old stacks of The American Rifleman really opened up the world for me.

That said, the other red herring here is comparing a digital camera image SYSTEM evaluation with a rifle scope evaluation. Here, we are talking about the equivalent of just measuring the camera's LENS.

I for one understand that, just like audio speakers, there can and are measurable performance differences between devices which a particular individual might not be able to perceive.

But since I know that I still have near-perfect hearing up to about 12 Khz, don't insult me by calling all music listeners too stupid or immature to use objective performance data. That data is often available. Optics evaluations are more involved, but I hope that the analogy rings true.

If there were a will, there would be a way to bring this data to the market. Challenges and difficulties with the *current way* of doing things do not equal impossibility or inevitable un-affordability. Just look at semiauto rifle accuracy--what used to cost $10-12,000 in 1985 dollars in the performance level of the H&K uber-sniper G3 variant is now available for $3,500-$5,000.

So, ILya, it might be impractical right now based on current conditions, but I'm sure a less-comprehensive data set could be offered now, and something to satisfy yours and Lowlight's demands could roll around within our lifetimes.

 

[koshkin]

Re: Why isn't glass objectively measured?

Red herring? Your Ransom rest comparison is kinda like comparing apples to pine cones. Unfortunately, most mechanical analogies are simply not applicable to optics or to the science of perception.

Grump, how much does your visual acuity vary day to day? how about during the course of the day?

How do you plan to cancel that out.

As far as testing photography lenses goes, how do you interpret the results of those test without knowing which imaging sensor was used to test them?

ILya

 

[dbooksta]

Re: Why isn't glass objectively measured?

You don't try to cancel out user visual acuity and variations; like Grump said that's something for an individual to take up on his own or with an ophthalmologist -- if at all. We'll have made a lot of progress if we can <span style="font-style: italic">objectively</span> say a S&B is optically better than cheaper scopes, and how.

As for bench testing: Isn't it trivial to adjust for anomalies of a test sensor? Presumably you're going to put subject scopes in between a CCD or CMOS sensor and test targets. Before you start taking readings you calibrate the sensor using a high-grade reference lens: you note sensor anomalies and subtract them out of test images or else get a better sensor if they're too extreme to compensate for.

In any case, given the demands regular photography and high-grade lenses put on sensor quality I doubt riflescopes are going to push the envelopes of even prosumer image sensors.

I won't challenge your assertion that it would cost ~$50k to setup a good testbed for general riflescope evaluation. My initial query was why higher-end manufacturers who presumably have these data don't release them. It would be nice if somebody thought it was worth the investment to fund third-party tests, but until then I thought demanding a little more of the industry would benefit us.

(To answer your other question, Ilya: I have degrees in Computer Engeineering and Math.)

 

[dbooksta]

Re: Why isn't glass objectively measured?

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: dbooksta</div><div class="ubbcode-body">I won't challenge your assertion that it would cost ~$50k to setup a good testbed for general riflescope evaluation. ... It would be nice if somebody thought it was worth the investment to fund third-party tests, but until then I thought demanding a little more of the industry would benefit us.
</div></div>

Note that Bryan Litz seems to have carved out a profitable niche for himself doing independent external ballistics evaluations that typically exceed those offered by manufacturers.

 

[koshkin]

Re: Why isn't glass objectively measured?

What Bryan Litz does is a lot less equipment intensive and is very easy to confirm empirically without any background ballistics. Besides, if memory serves me right Bryan now works for Berger bullets.

Optics does not work this way, since direct confirmation is difficult.

The reason I asked for what your education is was to see if you have any background in optics or mechanics. Since you do not, I can kinda see why you are struggling with understanding the problem at hand.

It is very common for people to excel in a particular discipline and the go ahead and assume that the same tools they used will help them tackle problems in completely different fields. That is often not the case. In my line of work, one of the biggest problems I have to face is when excellent mechanical engineers decide that they can apply the same deterministic principles they hold so dear to optoelectronics. That usually results in a number of very costly problems.

Either way, you did not answer my question: if someone were to offer this service, how much would you be willing to pay for that?

How many people do you think would be willing to pay for this service?

On image sensors: adjusting for differences between imaging sensors is not trivial at all. Accounting for "anomalies", as you put it, is a very non-linear problem.

Variations in visual acuity happen with everyone. You see differently in the morning and in the evening. You perceive differently in different lighting conditions. You perceive differently with one eye depending on what your other eye is doing. You perceive resolution differently with different targets even if they ostensibly show the same resolution and contrast, but utilize different patterns. You resolve differently depending on colors presented and that perception also changes through the day.

Everything on a scope can be tested. The problem is translating those test results into something actionable for the user and that will still require someone to translate the results into actionable information on a case-by-case basis.

Hence, the question remains: how much would you be willing to pay for that? Unless there is a market for it, there is no compelling reason for anyone to do it.

Here is another kink to ass to this argument: once we start talking about actionable data, a skilled observer can arrive to effectively the same conclusions after spending some time with the scope and without resorting to instrumental testing.

ILya

 

[Grump]

Re: Why isn't glass objectively measured?

I think I read last week that a system has been developed in Germany to objectively evaluate the density or some characteristic of beer suds.

Granted it's a single-purpose application, but it's optical and has already proven more repeatable and precise than eyeballing the "lacing" of foam on the inside surface of a glass.

Again only a single-element/phenomenon measurement, but I know that one speaker system has a high-end response that clips down a significant number of dB above 14,000 Hz, *I* won't be able to tell it from one with a frequency response all the way up to 20KHz. Those response curves are objective whether or not I can perceive the differences they describe.

 

[koshkin]

Re: Why isn't glass objectively measured?

I do not think I can explain it any better that I have, so I will accept failure and give up, I think.

I will re-iterate one thing though: I can objectively (well, with error bars, of course) measure any and every optical parameter of a riflescope with the equipment my company makes. I make my living on selling similar test sets from off-the-shelf to full custom jobs. What I can not do is make the test results from one of those test sets into understandable and actionable information for people who do not do this for a living (which just about everyone) without having a human there who can both understand technical data AND translate it into normal english.

This is not something you can easily boil down to couple to a couple of numbers.

ILya

 

[noname]

Re: Why isn't glass objectively measured?

grump- not sure what your are inferring, human auditory signals generated by the ear's mechanical system are linear until they get to the brain- and deterministic, as are any light measurement before they reach the eye, but as Ilya pointed out after that they nonlinear and non deterministic after that. Both systems show hysteresis in their operating range, with the auditory adding system adding a significant amount of deterministic chaos (dithering)in characteristic frequency ranges.
But I have the answer, My colleagues and I have developed a single event test that will solve all of the above problems. We have been able to get time on on one of the largest MRI scanners known to man. Select the scopes you wanted tested and join us for a weekend. During the test using a S$B as a standard we will run brain scans on any one noting which areas of the brain are using the most oxygen and any other statistical similarities the customer wishes.
Please arrive early (about 1/2 a day) to fill out the legal forms. Currently we are trying to get the cost down as each test runs about 4K. Also we are unable to bill your health insurance as the test has not been assigned a medical code as of yet.

 

[dbooksta]

Re: Why isn't glass objectively measured?

Jacob Gottfredson covers this subject in the June issue of Guns Magazine (pp 28-9) -- MTF, aperture, exit pupil, color correction, et. al. His intro:

<div class="ubbcode-block"><div class="ubbcode-header">Quote:</div><div class="ubbcode-body">One source of irritation for this author is the lack of technical data published for consumers about binoculars, riflescopes, and spotting scopes.</div></div>

 

[dwtaylorpdx]

Re: Why isn't glass objectively measured?

My wife used to work at Leupold assembling rifle scopes, I asked a engineer once why they could not give a chart that showed the quality of the scopes.

First thing is that with all the regulations in the world glass is now only made (Optical quality) in a couple of factories oin the whole world. They make production runs which are cast into ingots then sold at auction to the highest bidder, companies like Zeiss and Nikon buy most of the top shelf up for the Medical and scientific industries. Basically its like buying diamonds. So the relative quality between scope of different product runs can be significant if the manufacturer cant win auctions. Then the ingots are re-melted into the rough lens blank material and sawed into slices that are ground into the final lens. Lots of opportunity to mess up even a little bit along the way. Then they go to get coated, yet another opportunity...

Second he told me that to quickly compare two scopes grab a piece of white paper and look backward through both at the paper under good light. The scope with better glass will transmit the color as "whiter" than the other. Since hearing this I've tried it quite a bit with friends scopes and it seems to line up, not real scientific, but if its what your eyes see and it sorts the final choice then cool beans.. He said it wont work as well looking the right direction due to the way the coatings and lenses pipe light through the tube...

Dave