[whojammyflip]

I've just picked up a RMC Tokina Doubler for significantly less than the Pentax doubler T6-2x I've been seeking and I've been looking for reviews to justify the purchase

Something I saw on photo.net was the comment that enlargement reduces resolution by the factor of enlargement (obviously) whereas a teleconverter might only lose 20% of the resolution. I thought about this statement and realised that, indeed, as the doubler is exposed to the resolution coming out of the back of the lens, not looking at the image on the film/sensor, there are different physical effects in play. If the aerial resolution of the lens is miles better than the sensor (as is normally the case with a decent prime), then the degradation from the doubler may well be less than that due to cropping after the image has been captured.

If the doubler has the same resolution as the prime in use, then the resolution of the combination will be 1/sqrt(2) of the prime alone. If the prime lens' aerial resolution, for example, is 200lp/mm, then the combination will still deliver ~140lp/mm, way beyond that which the sensor or film can discern.

Only once the resolution of the doubler itself starts approaching the resolution of the sensor will there be any real concerns. So I thought I'd post something to the forum along this line, as I don't think I've seen anything like this written down here, just statements along the lines of TC's degrade the image just as much as cropping, so you might as well crop. Either way, something immediately obvious is that distant focussing is far more accurate with the Doubler. As I shoot film, I'd rather just use the Doubler, than have to move the enlarger head and crop at the time I come to print, so its a bit of a moot point for me, I just thought I'd share a thought.

---

Last edited by whojammyflip; 09-27-2010 at 03:04 AM.

[Digitalis]

using teleconverters brings in a whole new set of variables to lens performance. In my experience the canon EF 200mm f/1.8L USM with a 2X teleconverter does cause some loss of contrast in comparison to using the EF 400mm f/2.8L IS USM. However, the resolution is basically identical. and as an experiment I stacked two 2X teleconverters on the 200mm f/1.8 and one 2X on the 400mm f/2.8 and still, there was not as much difference between the two (though the 200mm f/1.8 lost even more contrast, but the resolution was very close.). These two lenses are pretty much as good as it gets in terms of optics. your mileage may vary, not to mention what you consider to be an acceptable loss of image quality would be considered serious degradation of image quality to others.

There are many makers of teleconverters and generally the apochromatic ones are in all probability going to be the ones that provide the best results. I don't know much about pentax teleconverters, but I do recall that they make ones that are lens specific: One for lenses under 300mm but no shorter than 100mm and one for lenses over 300mm.This implies that Pentax is aware of the degradation that can occur and are taking specific measures to combat it. As opposed to releasing a "general" teleconverter that delivers great performance with one lens but absolutely trashes the performance of another.

[bdery]

You are mixing up concepts here. A lens does not have a "resolution", per say. It has a resolving power, a modulation transfer function, it has aberrations, distorsion, etc. Resolution is a property of the sensor.

You can simplify by saying that a TC is a magnifier, enlarging the center of the image created by the lens. If you assume a perfect TC (more on that later), then what you will get is an enlargment of the center of the image created by the lens, accompanied by the enlargment of all the aberrations and flaws of the lens, and you loose a part of the light passing through your lens (you "cropped" it out).

Now, you have to consider that your TC has its own flaws and aberrations, its own resolving power, etc. And its starting material is not the object you are photographing, it's the image created by the first lens. So the "object" the TC is seeing is a flawed image created by your lens. Its OWN image is then seen by the sensor, which has a resolution limit to capture all the details.

What does it mean? It means the output will depend on the quality of your lens and TC. I've played with two TCs some time ago, and decided they were not worth it for me. In part because I did not like how they enlarged image flaws (including, for the record, motion blur). But YMMV.

[whojammyflip]

The key point remains that the combination of a TC and lens is not the same as cropping the centre of the image recorded on the sensor. Looking on Ebay, it seems there is a lot of dros available in terms of TC's, so there will be a wide variance of reported performance of TC's according to their individual sample quality.

I agree I simplify "resolution" for the lens; its indeed a decreasing function of contrast for increasing frequency of line pairs. Often, this is summarised by the objective description of the maximum spatial frequency with satisfactory contrast/modulation. Significantly, considering MTF charts, the best bit of a lens' MTF is typically in the centre, so focussing the TC on the centre will take the best the lens can offer.

It will be interesting to do a test print from a cropped 200mm tele shot vs a test print of a TC+200mm.

[reeftool]

I have an older Vivitar 2x converter (m42) and use it from time to time with a 200/4 Tak and have had mixed results. A lot of it is probably me as the Tak is a pretty hefty lens and adding the converter doesn't help. I've never done a real test using a tripod and careful focusing to really determine whether or not the converter or crops are better. For my real world shots, which are hand held or with a monopod, I think the crop is better. But back in the day, with my Spotmatic and film, the 2x converter was all I had at one point and I was thrilled with the shots I got with the converter on my 55/2 Tak. Back then, we weren't routinely enlarging 100% crops of every photo either. A converter is another lens and it's quite possible for it to degrade an image some. It's very likely that what happens is that the converter will work better with some lenses than with others. Canon sells some converters that are specificly designed to work with certain lenses as does Sigma. When people use them on other lenses and get poor results, it would be wrong to make a blanket statement that converters always degrade the image. Another issue is that often people buy third party inexpensive converters and expect great results which may or may not happen. It would be dumb of me to say "teleconverters are junk" based on the shots from my $10 Vivitar. I have had some nice ones from time to time so I'm not throwing it away.

[Lowell Goudge]

I have a whole collection of TCs.

- Soligor 2x M42
- Tamron Multi Coated @x M42
- Vivitar 4 element 2x22
- Vivirat 4 element 3x22
- Takumar A 4 element 2x TC
- sigma APO 1.4x EX
- sigma APO 2x EX DG teleconverters.
- 2 element optically corrected nikon to K mount and Canon to K mount adaptors (1.25 and 1.4x TCs effectively, and
- SMC-F 1.7x AF TC

So you would think with all these, I could give a definitive answer.

Well, yes, sort of.

The Sigma TCs are excellent when used on my APO-70-200F2.8 EX (non macro non DG) and if there is image degradation, I can't see it. BUT these were designed exclusively for sigmas long telephotos and zooms.

The SMC-F 1.7x AF TC is an interesting beast and has produced great results when used on my SMC-300F4. Yes there is CA but really only what was present in the lens itself, the sharpness is there. It works similarly well with my Vivitar 85mmF1.4 to give me a 145mmF2.5 lens. No real issues either wide open or stopped down.

I have not ever been satisfied with the vivitar TCs but perhaps that is unfair, as I tried them mostly with a 400F5.6 and should really try them better. A similar comment goes to the takumar A 2x. I have not really tested it properly,

The M42 TCs I have only played a little with, but the tamron seems pretty good, at least on my 200F3.5.

The bottom line is that the sigmas are good, but they only fit on relitively few sigma or pentax lenses that can accept a protruding front element on the TC, the SMC-F 1.7x AF TC is a good all round performer with the advantage of getting AF, but will amplify any lens defect, so really it is only worth while on a good lens to start.

Most 4 element TCs are a real gamble, and best used only in a pinch.

The optically corrected nikon and canon lens adaptors suffer badly from color blooming or bleading in high contrast shots when the lenses are wide open but are OK when stopped down. they are probably only good for emergency use or on telephotos.

[whojammyflip]

Yeah, I guess there is as much variation in TC performance as there is in lenses. I'm in no doubt that the dedicated T6-2x works better than the Tokina RMC, but the Tokina was 20% of the cost. I think its a 7 element, and multicoated, but I bought it more as an experiment to test out the idea that digital cropping vs using a 2x converter is superior. Will have to wait for better weather as its grim here in the UK today, and I cannot see clearly through the viewfinder at f8

[dadipentak]

I regularly use the Pentax AF 1.7x, A 2x-S and A 2x-L with my DA*300mm A*400mm and SMC 1000mm lenses. The strategy is to fill as much of the frame with the subject as possible (although I rarely stack.)

These are high quality and well-matched optics and I'm quite pleased with the results. I don't have a super discerning eye and there may be some degradation but I suspect this is more than offset by an enhanced ability to nail the focus.

[Clinton]

One mor thing of note is that when used with a macro lens, the tc can increase the magnification ratio. My sigma 180mm macro becomes a 360mm 2:1 macro with the sigma 2x tc attached.

[whojammyflip]

Well, I've recently picked up a M 100mm and can now compare the M 100 on the Tokina Doubler vs the plain vanilla at M 200mm. To my eye, the images are very similar, but this study is a quick and dirty one. The limitations to this comparison are: auto flash, ISO 800, not tripod mounted (hence change in size). The following images are 1000 pixel wide crops out of RAW files which were processed identically. Just looking at the grain in the wood at the lower left of the watch, and the text in the central button on the face of the watch, the M200 has more contrast, but otherwise the IQ is quite similar.

M100 with Doubler effectively at f8



M200 at f8



Will have to do some real testing at f11 with the M200.

[dadipentak]

That looks pretty good but you really need to use a tripod for credible testing.

[whojammyflip]

yeah, was quick and dirty, as i was thinking i had not tested it at all yet. will eventually get round to comparing a crop vs tc shot using the m200mm

[whojammyflip]

Quick tripod mounted test. Both shots taken at 6m from target, so that it filled the frame for the one with the doubler. I used a split screen too. The doubler most definitely increases the resolution at the expense of a lot of ugly chromatic aberation, looking at the vertical lines at least. I suppose for such a trivial amount of money, this is a useful increase in resolution, and with some colour filters on the front, its possible to use this for BW film, but the contrast looks pretty poor without a filter, thanks to the CA.

without doubler, 500 pixels wide, M200mm at f11


without doubler, resized to 1000 pixels wide, M200mm at f11


with doubler, 1000 pixels wide, M200mm at effective f11 (f5.6+doubler)

[dadipentak]

The doubler most definitely increases the resolution

I wouldn't say that: the cropped w/o doubler sample looks significantly sharper than the one with.

[whojammyflip]

The contrast of the crop is without doubt better, but check the resolution limits on the vertical converging lines...the doubler is still going at 800, whereas the crop is a blur by around 650. This could potentially be thanks to the sensor reaching its limits (about 60lp/mm), as I think the 800 mark represents 30lp/mm for the doubler, but 60lp/mm for the crop, and nothing to do with the performance of the lens. On a better sensor, it might be possible to get more detail out of a crop.