Anyone suspect the Voigtlander 35mm f1.4 MC might be slower than rated?

[ckuang]

Interestingly in Steve Huff's review of the lens, he notes that it also tends to underexpose by a stop on the M9. I'm not sure if he means it literally underexposes or it underexposes relative to another lens of his which is rated at F1.4. Don't get me wrong the Voigtlander is a good lens. Unfortunately for me, on digital and chrome half is stop is a big deal, and given the number of images I need to shoot a week half a stop difference can make hours of difference in post production.

 

[Sparrow]

Interestingly in Steve Huff's review of the lens, he notes that it also tends to underexpose by a stop on the M9. I'm not sure if he means it literally underexposes or it underexposes relative to another lens of his which is rated at F1.4. Don't get me wrong the Voigtlander is a good lens. Unfortunately for me, on digital and chrome half is stop is a big deal, and given the number of images I need to shoot a week half a stop difference can make hours of difference in post production.

As I said, why is it Cosina attracts these “damn with faint praise” type threads?

 

[ampguy]

please clarify

please clarify

I'm not sure I follow, are you saying that other manufacturers are over-rating their f/t stops by 1/3 or 1/2 or more stops? Who? Are my Summiluxes really 1.8 or something?

What about when a lens is compared to another lens and at same f stop, has t stop 1/2 stop lower?

You're following the thread here, I presume??

Do you think it could be that Cosina are simply following the precedent set by the European manufactures over the previous two centuries? It would seem a particularly stupid commercial strategy to not follow Leica's established practice don't you think?

P. S. clearly the MC version would transmit more light than the SC one anyway

 

[Sparrow]

I'm not sure I follow, are you saying that other manufacturers are over-rating their f/t stops by 1/3 or 1/2 or more stops? Who? Are my Summiluxes really 1.8 or something?

What about when a lens is compared to another lens and at same f stop, has t stop 1/2 stop lower?

You're following the thread here, I presume??

Well, let me explain more clearly, for hundreds of years the lens speed has been expressed as the ratio of focal-length divided by the diameter of entrance pupil, Zeiss, Leica and Voigtländer traditionally used that ratio to express lens speed.

Clearly that is not an expression of light transmission, if your lenses are physically f1:1.4 then they cannot be t1:1.4 as some light is, no must be, absorbed and some reflected by the glass and it's coatings. Are you understanding it so far?

My point is not that, but, why is Cosina being required to reach a standard higher than the traditional manufactures? and why do we get these threads which while purport to be one thing but are in fact simply inviting criticism of the only company who is providing us with entry level RF equipment?

 

[palec]

Well, let me explain more clearly, for hundreds of years the lens speed has been expressed as the ratio of focal-length divided by the diameter of entrance pupil, Zeiss, Leica and Voigtländer traditionally used that ratio to express lens speed.

Clearly that is not an expression of light transmission, if your lenses are physically f1:1.4 then they cannot be t1:1.4 as some light is, no must be, absorbed and some reflected by the glass and it's coatings.

I'm off topic, but is this still valid in this time? I've read several times (also in outdated Ansel Adams's Camera) that because of optical advances in design and coating the light transmission losses are very low and thus T-stops are not used to describe lenses as often as they used to be. What would be the purpose of specialised external light meters (spot and incident) if every lens had different tolerances for actual light transmission?

 

[ampguy]

yes, I am understanding so far...

yes, I am understanding so far...

For the Leitz Summarit, we have this data:

[The values for Xenon/Summarit and Sonnar
are T=1.89 and T=1.72 at 4.5% loss per surface for
uncoated surfaces, and T=1.66 and T=1.59 at 2.0%
loss per surface for (not multi-) coated surfaces.2]

What is the data for the CV 35/1.4, or the overall t-stop? That is all I want to know.

When I use my VC II or Sekonic meter with 100 ASA film, and it says for my given shutter speed to use F2, where do I adjust the CV 35/1.4? What if the meters say 1.4??

Well, let me explain more clearly, for hundreds of years the lens speed has been expressed as the ratio of focal-length divided by the diameter of entrance pupil, Zeiss, Leica and Voigtländer traditionally used that ratio to express lens speed.

Clearly that is not an expression of light transmission, if your lenses are physically f1:1.4 then they cannot be t1:1.4 as some light is, no must be, absorbed and some reflected by the glass and it's coatings. Are you understanding it so far?

My point is not that, but, why is Cosina being required to reach a standard higher than the traditional manufactures? and why do we get these threads which while purport to be one thing but are in fact simply inviting criticism of the only company who is providing us with entry level RF equipment?

 

[smillie]

are you saying that other manufacturers are over-rating their f/t stops by 1/3 or 1/2 or more stops? Who? Are my Summiluxes really 1.8 or something?

I'm pretty sure he's saying that all manufacturers mark f-stops, and don't mark t-stops, regardless of what (if any) differences exist between those two measures.

What about when a lens is compared to another lens and at same f stop, has t stop 1/2 stop lower?

Is this a real concern, or a purely theoretical "what if?" Honestly I'd be surprised if any modern prime had a t-stop that varied from the f-stop by as much as 1/3.

I mean, let's get into what that would imply in concrete terms: for a difference of 1/3 stop between f and t measures, the lens would have transmission of around 80%, and around 70% for a 1/2-stop difference.

This isn't plausible to me, and I'll explain why: uncoated glass reflects ~4% per air/glass interface. Modern coated glass is typically 1% or lower. The Nokton has 6 groups, so seven air/glass interfaces, and we can work out plausible transmission values as follows:

uncoated: 75%
coated: 93%

So while a completely uncoated lens would give a t-value of around 1/3-1/2 stop difference from the f-value, a coated lens would come in closer to 0.1 of a stop, the difference between f/1.4 and f/1.46.

IF these lenses have a dramatic difference between marked and actual f/stops - and I don't honestly think they do - then t-stops and the coating are complete red herrings in tracking it down.

 

[Juan Valdenebro]

If a minimal difference exists, and if someone shooting slide film uses the lens and cares for that small difference, that person must have seen it on the slides from the first time the lens was used, and as everything in photography and other member said, it really doesn't matter because you know your equipment, just like when a shutter speed is a bit different in the high speeds... Anyway if there's difference, for sure it's very small...

Cheers,

Juan

 

[ckuang]

Well, let me explain more clearly, for hundreds of years the lens speed has been expressed as the ratio of focal-length divided by the diameter of entrance pupil, Zeiss, Leica and Voigtländer traditionally used that ratio to express lens speed.

Clearly that is not an expression of light transmission, if your lenses are physically f1:1.4 then they cannot be t1:1.4 as some light is, no must be, absorbed and some reflected by the glass and it's coatings. Are you understanding it so far?

My point is not that, but, why is Cosina being required to reach a standard higher than the traditional manufactures? and why do we get these threads which while purport to be one thing but are in fact simply inviting criticism of the only company who is providing us with entry level RF equipment?

I don't really see why you think it's a criticism of Voigtlander just because i'm wondering if it's slower than it's rated speed and seeing if anyone else has the same experience. It's just something that is good to know so that every time i throw on this lens at this aperture, i remember to overexpose the reading on my light meter. Lots of camera equipment has little quirks that is good to know to get the best out of it rather than just shooting blindly. Isn't this the beauty of the internet and sharing of information?

 

[antiquark]

If you have a digital camera, it should be easy to compare the T-stops. Just take different lenses, set them to the same F-stop and shutter speed, photograph a grey card, and see how similar the histograms look.

If the grey card is evenly illuminated, the histogram should look like a sharp peak in the center.

 

[ampguy]

bingo

bingo

Better yet, post the images. I can understand why people don't want to share their distorted images, or their images from f2 to f8 with focus shift, but jeez, just a couple of snaps to see how the light behaves relative to any other 35 of a grey card, with cam on tripod at same aperture/shutter speed should quell the whole issue, maybe we can't easily notice it, maybe summiluxes like Sparrow is implying behaves the same (but I'm really beginning to doubt it ...)

If you have a digital camera, it should be easy to compare the T-stops. Just take different lenses, set them to the same F-stop and shutter speed, photograph a grey card, and see how similar the histograms look.

If the grey card is evenly illuminated, the histogram should look like a sharp peak in the center.

 

[smillie]

Also, why is Sparrow saying MC lens will trasmit less than SC?

Above I read it saying the opposite, that MC transmits more light than SC, which is correct. Assuming you're still curious: The major loss of transmission in lenses is due to reflection, SC reduces reflection around one particular wavelength, MC around several, hence MC transmits more light. Wikipedia has a decent article, though I prefer this one:
http://rick_oleson.tripod.com/index-166.html

Still curious why the OP is seeing the 1/3 to 1/2 stop differences with his CV 35/1.4 lens (and others have too) compared with his Leica lens.

Don't know. I lack a Leica lens to compare, but a quick sanity check (on a µ4/3) against a Nikon 50/1.4 and a white wall shows my 35/1.4 SC to give an identical reading (within 1/3 stop). Stopping down the 35 also behaves sanely, with no unexpected changes between f/2 and f/2.8 as the original post would imply (in that it can't be correct at f/2.8 and slow at f/2 without some change between the two conditions).

 

[ampguy]

Thanks

Thanks

I read and understand the info in your link.

Can I ask which lens, the CV, or the Nikon transmitted more light at 1.4 within that 1/3rd stop?

I only have 2 35 lenses presently, a pre-asph lux 35, and LTM J12 35/2.8, but if anyone is interested in me testing these out, or with the 75/1.4 lux on a white wall, I'd be pleased to, just ask, also specify what meter(s) you'd like me to test with.

Above I read it saying the opposite, that MC transmits more light than SC, which is correct. Assuming you're still curious: The major loss of transmission in lenses is due to reflection, SC reduces reflection around one particular wavelength, MC around several, hence MC transmits more light. Wikipedia has a decent article, though I prefer this one:
http://rick_oleson.tripod.com/index-166.html



Don't know. I lack a Leica lens to compare, but a quick sanity check (on a µ4/3) against a Nikon 50/1.4 and a white wall shows my 35/1.4 SC to give an identical reading (within 1/3 stop). Stopping down the 35 also behaves sanely, with no unexpected changes between f/2 and f/2.8 as the original post would imply (in that it can't be correct at f/2.8 and slow at f/2 without some change between the two conditions).

 

[ckuang]

i'll try and shoot some comparison images tomorrow

 

[Sparrow]

I don't really see why you think it's a criticism of Voigtlander just because i'm wondering if it's slower than it's rated speed and seeing if anyone else has the same experience. It's just something that is good to know so that every time i throw on this lens at this aperture, i remember to overexpose the reading on my light meter. Lots of camera equipment has little quirks that is good to know to get the best out of it rather than just shooting blindly. Isn't this the beauty of the internet and sharing of information?

Yes, I know you made no criticism, I chose my words very carefully, I said “damn with faint praise” it was left for others to do the criticism, I think they talked it down to at least 1.8 in the end.

Yes the beauty of the internet is the ease with which we share information, and no doubt some of it is even correct.

PS smillie: I said “clearly the MC version would transmit more light than the SC one anyway” previously

 

[Andy Kibber]

If you find it a half stop slower pump up the ISO or shutter speed? What's the fuss?

Maybe I'm missing something here, but most M-mount cameras use through-the-lens metering. Why would you have to make changes to ISO or adjust the shutter speed if the lens is "slower than rated"?

 

[Dave Wilkinson]

Yes the beauty of the internet is the ease with which we share information, and no doubt some of it is even correct.

LOL! prize for quotation of the week, Stewart!😀.....now it's time to head for the airport! - back in a fortnight!....Yippee! 😛
Dave.

 

[pvdhaar]

There's probably a logical explanation for all this that doesn't require any formulas..

On a digital camera (the OP mentions using the M8/M9) the light fall-off towards the corners has a tremendous effect on measured exposure. That's without even considering digital processing to compensate for light fall-off.

If the camera is going to compensate, it can't just crank up the brightness away from center as that would result in excessive noise. Instead, with a bit more liberal exposure, the corners' brightness doesn't have to be amplified as much, while the center will only need toning down a bit. This balancing act will make for a better overall result.

So, while the 35/1.4 may have a Tstop of 1.4 on film, it may well drop to 1.8 effectively because of the above..

 

[ampguy]

Hi Andy

Hi Andy

While the later M's do have meters, and the M7,8, and 9 have AE, many of us don't use that mode, but set our cameras manually.

Especially for backlit and dark scenes where I am intentionally over and under exposing, and sometimes I will use an external meter which simply provides an EV or shutter/f-stop combo, based on a given ISO set.

So when a given lens lets in more or less light at a marked f-stop than my other lenses which are very consistent and provide similar histograms wide open and at stopped down apertures, it is helpful to know if a lens is deviating, so that special consideration during manual exposure settings can be used.

Maybe I'm missing something here, but most M-mount cameras use through-the-lens metering. Why would you have to make changes to ISO or adjust the shutter speed if the lens is "slower than rated"?

 

[umcelinho]

I cant afford a 35mm summicro, even less a summilux. If the 35/1.4 nokton is after all a 35/1.8 or a 35/2 I don't really care. it is still half the price of a summicron. is it worth it? i don't know, never really used one. their ergonomy is better, they don't need an extra hood but, still, at least double the price.

it gives me pleasant results and that is what matters to me. I use cameras with TTL metering, so the lightmeter will give me reading regardless of what is marked on the lens. it reads light, not numbers. I've never experienced that underexposing issue when changing apertures whilst maintaining shutter speed.

for those with meterless cameras, the reported could be annoying though.