AI-S won't work in "S" or "P" modes

[theodoros_fotometria]

<p>" I just want to be clear (or corrected!) that the mechanism of metering and aperture control is identical for AI and AF lenses, and that it's just the AI/AI-S distinction of whether the camera knows the lever position for the selected aperture that's distinct - hence my claim that AI-S lenses should "just work" if the camera knew it had one attached."</p>

<p>Actually Andrew, the camera "knows" when it has an AI/AI-s lens attached by detecting the movement of the ring around the aperture mount which also provides the aperture setting as info to the metering... Notice that when an AI or AI-s is attached, the ring is "grabbed" from the lens and is moved, the same doesn't happen with a G lens. What opposes the automatic aperture lever on the back of the lens to set the aperture automatically, is that the lever <em>cannot be </em><em>controlled by the coupling ring around the mount, it must be controlled by the camera dials because the dials are connected with the metering, </em>Thus, if the system (the way it is designed) would give the ability to the ring around the mount to be an extra control to the metering, both the other controls on the camera (the aperture and shutter dials) would have to be disengaged, then the camera would be able to do S & P modes, ...but not M&A (!!!). Obviously they could design a more complex system and have everything working, but this would cost much more, because other that it is more complex, they would need to have a different (much cheaper) system to control the cheap cameras that have no AI coupling... while now, they have one (cheap) system for all cameras and only add the coupling ring to the more expensive models.<br /><br />"To clarify, there's no "disengaging": all the cameras that can do this at all have purely electronic controls. The aperture following ring around the lens mount, which engages with the aperture ring, just feeds a number into the camera meter electronics to tell it what aperture you selected. An equivalent number may come from the system of spinning a dial (and the number is just stored in software in that case - which is why there's no hard stop on the control dials when you use them for aperture, and why you can change the stop size change for each click in a menu) or from the camera's selection in auto-aperture mode. All this tells the camera what settings to use for the exposure and whether the exposure is "correct". I don't actually know whether the camera will try to move the aperture lever to where it thinks it should go even if the aperture ring is being used to set the aperture, or whether in this case the camera will fully release the aperture lever and let the aperture ring catch it - does anyone know? Either way, the meter is unaffected - the camera thinks the aperture will be set to the value it expected when the shutter was pressed. Low end DSLRs with no aperture ring following tab (though they do have a switch to detect whether minimum aperture has been selected, I believe) still control the aperture lever in the same way."</p>

<p>Exactly! Only that the AI coupling ring is not engaged as a dial (it only provides info to the metering)... if it would (be engaged) the camera dials would have to be disengaged (which would lead in S&P working and A&M not working) ...look above.<br /><br /><br />"I'm not sure whether we were actually disagreeing about how this worked, but if we were, I hope that's clarified things!"</p>

<p>We fully agree on how it works, my comment is only on <em>why there is no full compatibility on all modes... </em>Just another point if I may... The metering is in full communication with both the camera dials, if the coupling ring was to act as a dial, the <em>matrix metering mode would also work (as it does with the FA), </em>the reason that matrix now doesn't work with AI/AI-s lenses, is that the coupling ring only gives aperture information, it cannot act as a dial!</p>

<p>I have to say that I love the system the way it is! I never use S or P modes and have plenty of AI-s lenses in my possession. In another conversation I'll present a "trick" on how I have convert (harmless and easily reversible) my 14mm f2.8 to an AI-s lens. I would do so with the 17-35 too, but AF zooms don't have a "real" DOF scale on them, pity! I may still do it though, I love using the aperture ring manually the traditional way.</p>

[theodoros_fotometria]

<p>Aahh... Andrew, another proof that the camera knows that there is an AI/AI-s lens attached, is that it automatically turns non-CPU lens data to ON! Only that the system is a bit "stupid" ...and the aperture on the top screen is the one detected from the coupling as if the lens was the last AI-s used... if the lens is different you have to "tell" it! ...unless of course if max aperture is the same... (In this later case all the exif data will only show wrong focal length). LOL...</p>

[andylynn]

<p>That one, at least, I can clear up. The camera isn't detecting the presence of an AI lens, it's detecting the lack of an AF lens. If you have the non-cpu lens data on a D800 set to 50/1.4, remove the lens, and turn it on, it gets no electronic signal, assumes the 50/1.4 lens is there with the aperture ring wide open (because the AI feeler is all the way to the right) and displays f/1.4. If you move the AI feeler with your finger the displayed aperture changes.</p>

<p>The AI feeler is all the way to the right no matter what lens is mounted and 1 stop corresponds to the same amount of rotation on all lenses. All it's doing is, it has the wide open aperture because you input it, it knows how many stops the lens is stopped down, and it does that math. Since the understands how much light will enter the stopped down lens as a function of the amount coming in wide open and the number of stops the lens is stopped down, it doesn't actually care what you've entered in the AI lens data in the menus, it will get the exposure right anyway. The information in the menu is only there so the aperture readout and exif data will give the correct information.</p>

[jay_drew]

<p> </p>

<p > </p>

<p >Wow! 3 pages of answers. I don't believe I've ever asked a question that generated so much activity!</p>

<p >Thanks to all, I really appreciate all the answers.</p>

<p >Actually the answer to my question is that Nikon's definition (& many others as a result of trickle down) of Ai-S incorrect. At my age, I'm not as sharp as I once was. I was afraid that I was making some error when trying shutter-priority. I've mes-exposed quite a few shots due to this misunderstanding. I mean the reality was butting up against Nikons word on this matter. The answer also is that I should get all auto focus lenses, if I want to work in shutter priority.</p>

<p >I can always shoot manually, (I have incident, reflected, & spot meters & know how to use them) or aperture-preferred. But I prefer to let the camera set the exposure in situations where I trust it to be accurate for expediencies sake.</p>

<p >The same way w/ flash, I hate to use guide #s, much too slow (I know, event & press photographers get so that they can manually set flash exposure instantaneously, but I can't).</p>

<p >I prefer shutter-priority so the shutter speed doesn't creep too slow for sharp photos, I can control shutter speed w/ the f-stop, but it's not my preference.</p>

<p >I'm really glad I sat down w/ most of my cameras & lenses & made a matrix of which lens will do what w/ which camera & after I knew for sure what was what, that I asked this question. I should have done this a long time ago. What I did was to take each lens in increasing focal length, attach it to each camera in a brightly lit area to see if the aperture blades ever left the smallest setting (largest #) & I would also set it to almost the max f-stop to see if it would move from there. I did this for 3 cameras; 6 lenses; & 4 modes, all the while continuing to rewind the old 'play' roll of film I had & marked the results on a matrix I had drawn up.</p>

<p >This was to a limited extent the question I had asked some time back about what modes work w/ what equipment, but I expanded it to include more cameras & Nikon Flashes & cords. This (the question I just asked) narrowed it down to chewable bite. Rodeo Joe said, back then, that it looks like I have a bit of research ahead of me, & he was right. The 1st part is to figure out what question(s) to ask.</p>

<p >Thanks to all, again, Jay Drew</p>

<p > </p>

[Uhooru]

<p>I think, based on how they have functioned on my Nikon/s Bela describes the way the AI/S lenses work correctly. Aperture Priority and manual only. Lens is wide open till shutter pulled and it stops the lens down then. If you want to preview the DOF you need to use the preview lever depending on which body you are using. Jay, in aperture priority mode, the camera does set the exposure, by adjusting the shutter speed to adjust to the meter reading through the lens. In cameras from FA and FE forward (I think). Its just that on AI/S lenses the camera can not set the aperture. The index mechanism only closes the aperture to its pre set (set by user via the aperture ring). Depending on the camera, the exposure mode doesn't have anything to do with program mode. In other words, on my, for instance D200 or D700, I can use an AIS in matrix meter, spot meter or center weighted in camera after setting parameters in the "non-cpu lens" setting. But I can't use any P mode where the camera chooses aperture and shutter. Only manual or Aperture priority. Clear as mud? As I recall, some Nikon models required pre-AI lenses to be shot stopped down.</p>

[theodoros_fotometria]

<p>Mind you that P will also work correctly.... but in reality, it's exactly the same as using A mode because aperture can't be adjusted from the camera... Still, exposure will be correct.</p>

[richard_driscoll]

<p>Gosh that was all very interesting.<br>

In fact the ring on the AI/AIS lenses doesn't feed the aperture to the camera; it feeds in the amount below maximum aperture the lens is (or rather will be) stopped down. If you put almost any lens on the camera at full aperture the teller on the camera will assume the same position confirming this fact. I said almost since the very fast lenses (f/1.4 and faster) don't seem to quite follow this rule - I read some guesses some while ago about why this was so but don't remember the details. For centre weighted and spot metering the camera doesn't need to know the <em>actual</em> aperture since it is a TTL meter. For matrix metering the camera does need to know since it needs to know the <em>absolute</em> scene brightness and the lens maximum aperture is fed to the camera mechanically on AI/AIS lenses and via the contacts on AF lenses. I believe that AFD lenses do this mechanically too so presumably they'll matrix meter on a Nikon FA.<br>

My understanding is that an AI/AIS lens on a modern autofocus body is stopped down by the stop down lever moving all the way like it always did; in fact I recall reading that setting a body to work this way with an AFD type lens if a good idea since the accuracy is better than the normal 'dead reckoning' method.<br>

Also, I can't see why S and P mode couldn't be offered with AIS lenses on a modern DSLR if the body had the mechanism that reads the cutout on the lens to tell an AIS lens from an AI one or it might try S or P mode on an AI lens. That wouldn't work since the camera body can't use the slow 'light feedback' closed loop stop-down method.<br>

Jay, to answer your last question. My guess is that the F4 and F5 are probably not capable of telling an AIS lens from an AI one and since they can't stop the lens down using the slow closed loop method of stop down (used by the FA and FG for example) they don't offer the option of P and S mode because they might be used on an AI lens whch would in turn result in incorrect exposure.</p>

 

[theodoros_fotometria]

<p>"In fact the ring on the AI/AIS lenses doesn't feed the aperture to the camera; it feeds in the amount below maximum aperture the lens is (or rather will be) stopped down."<br>

This is the "perfect" way of explaining it... but I don't know if it will be understandable by all... In fact, the coupling ring of AI/AI-s lenses (which is still present on the models that offer Non-CPU lens coupling), does nothing more than replacing the old prong coupling of the 60's cameras to the build in metering and it does so mechanically. Nikon, is the only imaging company at present, which when the body sets shutter speed automatically this is done electronically, but when aperture is set automatically, only the order is electronic, the final action is mechanical and is performed by the lever behind the lens which is there originally to provide the DOF preview!<br>

Richard is also right when he says that "it can be offered" ("I can't see why S and P mode couldn't be offered.."), IMO it is purely the additional cost and complexity that is involved to make the coupling (mount) ring communicate as an extra command <em>with the camera's control system,</em> (not <em>only with the metering as it is today) </em>and the fact that a second (cheaper) control system would be necessary for the low cost cameras that not offer AI coupling. Also, (I like this) the fact that the company thinks of P & S modes as non important for photographers.... at least for the ones that would use an AI/AI'd/AI-s lens. After all, FA was never able to compete with FE-2 sales, despite its "fancy" metering, the modes and the "stupid" F3-like viewfinder info... didn't FE/FE-2/FM-3 offer the <em>best viewfinder info of all times? ...and all the features that a photographer will ever need?</em><br>

</p>

[Uhooru]

<blockquote>

<p>Mind you that P will also work correctly.... but in reality, it's exactly the same as using A mode because aperture can't be adjusted from the camera... Still, exposure will be correct.</p>

</blockquote>

<p>Thanks Theodoros, I wasn't aware of that. So in P, it will effectively act as in A....makes sense.</p>

[paul_wheatland]

<p>I've experienced that a Nikkor manual focus AI/ AIS lens will only work and meter properly on multi mode DSLR or 35mm SLR bodies in A or M modes only! I do have a 45mm f2.8P that works in all modes with exception of auto focus on all Nikons!</p>

[Michael R Freeman]

<blockquote>

<p>"I do have a 45mm f2.8P that works in all modes..."</p>

</blockquote>

<p>And therein lies the solution. Add a CPU chip (such as the "<a href="/nikon-camera-forum/00ZatY">Dandelion</a>") to your manual focus Nikkors if you really want to use them in P or S modes on bodies other than the FA, FG, etc. :-)</p>

<p>Technically I can think of no reason why Nikon could not have allowed for P and S exposure mode compatibility with <strong>AiS</strong> lenses on <strong>all</strong> current and past Nikon DSLR bodies <strong>without</strong> the need for <strong>any</strong> additional mechanical hardware. All that should be required is the necessary programming to allow entry of non-CPU lens data (focal length and maximum aperture) and the supporting exposure control firmware. Once the camera has that info it is in essence working with an Ai-P CPU lens like the 45mm f/2.8P. Lower level bodies without an Ai tab could control aperture via the command dial in all exposure modes, just as they do for AF-D lenses. Higher level bodies with an Ai tab could control the aperture via command dial in S and P modes, and by aperture ring or command dial in A and M mode, also as they can do with a AF-D lens.</p>

<p>Nikon has chosen to completely/partially cripple DSLR compatibility with non-CPU manual focus lenses by not including the necessary firmware for P and S modes. If one believes Nikon is a benevolent dictator with their customers best interests at heart, then we might guess that perhaps this option was chosen for maximum backward compatibility with pre-AiS lenses, and to prevent problems that would occur if the earlier <strong>Ai</strong> or non-Ai lenses (with their non-linear aperture motion and resulting exposure errors) were inadvertently used with body controlled aperture stopdown. After all, the vast majority of Nikon users are completely clueless as to the difference between Ai and AiS lenses. Heck, I've come across a few who assumed that you could autofocus a manual focus lens by mounting it on an autofocus body.</p>

<p>On the other hand if one is cynical, then we might guess that perhaps this decision was made to increase the probability of customers upgrading to new autofocus CPU lenses if P and S mode compatibility was important. After all, Nikon is primarily in the business of selling lenses, not designing legacy support for older dead product lines. >:) ;-)</p>

<p>You decide. :-)</p>

<p>Of course the real kick in the teeth with respect to the F4 was that it <strong>does</strong> have <strong>all</strong> the necessary mechanical devices for P and S mode compatibility with AiS lenses, just like the FA. It has an AiS detection pin on the mount to activate PHi mode, and a lens speed post lever inside the mirror box (4 to 6 o'clock position) to mechanically communicate the actual maximum lens aperture to the camera (hence the reason why the F4 is the only autofocus 35mm camera other than the F6 that can matrix meter with Ai or AiS lenses). Nikon for whatever reason simply chose not to include the necessary programming to make full use of those two mechanical parts.</p>

<p>Perhaps there was no room left on the ROM chips for closed loop metering and all that other good stuff? ;-) :-)</p><div></div>

[bebu_lamar]

<p>But if the F4 works in S and P mode with AI-S lens it can not display the aperture in viewfinder. The FA can't either but may be that's unacceptable for a professional camera.</p>

[richard_driscoll]

<blockquote>

<p>But if the F4 works in S and P mode with AI-S lens it can not display the aperture in viewfinder.</p>

</blockquote>

<p>Yes it could. As Michael points out above, the F4 can read the lens speed post on an AI-S lens and since it will then know the lens' maximum aperture and also how far it will be driving the stop down lever it could display the final aperture on the LCD. The lens aperture ring will be set to minimum aperture and so the optical display will display the minimum aperture but that is no different from the display with an autofocus lens.</p>

<p> </p>

[richard_driscoll]

<p>Michael,<br>

You wrote that the F4 has an AiS detection pin on the mount to activate PHi mode. When I wrote earlier why I thought the F4 was unable to do S and P modes with an AI-S lens I assumed that it lacked this pin but after reading the instruction book I saw that it had. Please can you explain why/how it makes use of this pin to enable/disable high speed program - I just don't get it.</p>

<p>While the backwards compatability with the new DSLRs is a little lacking I'm still pretty impressed. Having recently bought a D7000 at last I have a camera that is reasonably portable and has the required fittings (aperture feeler, ability to set lens info etc.). I was then able to pop on my 1980's 50 mm f/1.8 AI-S and low and behold it works <em>better</em> than it did on my FE2 and F90x since I have both aperture clearly visible in the viewfinder and an excellent electronic rangefinder which with my eyesight I find better than the old split image or microprism!</p>

 

[richard_driscoll]

<p><strong>AI-S zoom lenses with non-constant aperture</strong><br>

I haven't ever looked closely at one of these but I bet the lens speed post is fixed. If so it means that the camera will not be told the true maximum aperture of the lens at all zoom settings; it probably indicates the max. aperture when zoomed out. If so then if P and S modes were offered the result would be underexposure on the DSLRs and most of the film cameras too. Is that the reason why it isn't available on the F4 I wonder? There probably aren't more than about 5 Nikon lens designs like that.</p>

<p> </p>

[Andrew Garrard]

<p>That'll teach me to have an internet-free weekend...<br />

<br />

Fortunately Michael did a sterling job for me.<br />

<br />

Richard: Bear in mind that variable-aperture zooms that aren't G already have a problem with their aperture rings - the ring doesn't move, and there are just two marks on the lens indicating the aperture at the extremes of the zoom range. I've no experience with these. However, the aperture lever does just say how far to stop down from the current "wide open" position - if you hold down the DoF preview on a D700 (say) and zoom a variable aperture lens (the D700 can't move the aperture lever without cycling it back to wide open, unlike the D3 or D800), the brightness changes visibly. Release and re-press the DoF preview button and it moves to a different place; I'm sure the varying aperture is communicated only electronically, though how much difference the "wrong" aperture would really make to an exposure is debatable, given that it's relative to a metered wide-open value: I'm not sure how valuable it is to report the exact aperture to the camera. AI (and maybe even AF non-D?) lenses have no way to tell the camera that the aperture is varying with focal distance either, yet the wonders of TTL metering mean that everything just works.<br />

<br />

I now want an F4 to play with. Although I did anyway...</p>

[jeff_livacich]

<p>Theodoros is incorrect in saying that only Nikon sets the aperture mechanically. Pentax does also. The lever has always been there in both makes primarily to close the auto-diaphragm, with DoF preview secondary.</p>

[nikos_k]

<p>Dear All,<br>

Sorry for digging out this old thread. I have an F4 and have trouble operating it in P and S modes. For some reason, the camera seems no matter what to always stop down the lens to the minimum aperture possible, say if I read f/2 or f/5.4 in the viewfinder the lens gets stopped down to f/22 and I get a blinking red locking the camera.<br>

This happens with the AF D lenses, and in the P/S modes only. Also, when I press the DOF/mirror lock up button I see the lens stops down to f/22 (min aperture value locked). Am I doing something wrong, or it is a fault of the camera?</p>

<p>Thank you</p>