GF 100-200/5.6 with 1.4x TC on GFX 100s

[yaqbpl]

then the lens seems to be able to provide details for resolution over 100mp.

 

[Crispy_Bee]

Thanks for sharing!

Considering the price of both the 100-200 and especially the TC, that image quality at f8.0 is a real let down. I know the 100-200 isn't that strong of a lens but if the TC is compatible, it should at least deliver OK corners wide open.

:-|

I think I've seen a better performance on a 70-200 Canon L lens with the 1.4x TC II back in the day than this. Even the centre looks pretty weak and shows spherical aberrations. It even looks like there's some "swirl" going on in the bokeh...

 

[JimKasson]

the way teleconverters work they require lens with unresolved sharpness. meaning lens that is able to produce finer details then sensor resolution is able to read. otherwise there is no gain and it could be done just with digital zoom as well. so in general teleconverters has more chance to have gain with 50mp bodies. for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body. anyone tested it on 50mp? do you know where to find lens tests with information about sharpness in mp? dxomark was giving that, but i think they didnt test fujinon gf lenses.

I don't think MP is a good metric for sharpness.

i wasnt very precise with words, i was trying to simplify. maybe i should say resolution

So you want to see MTF curves for both situations?

 

[yaqbpl]

the way teleconverters work they require lens with unresolved sharpness. meaning lens that is able to produce finer details then sensor resolution is able to read. otherwise there is no gain and it could be done just with digital zoom as well. so in general teleconverters has more chance to have gain with 50mp bodies. for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body. anyone tested it on 50mp? do you know where to find lens tests with information about sharpness in mp? dxomark was giving that, but i think they didnt test fujinon gf lenses.

I don't think MP is a good metric for sharpness.

i wasnt very precise with words, i was trying to simplify. maybe i should say resolution

So you want to see MTF curves for both situations?

 

[JimKasson]

the way teleconverters work they require lens with unresolved sharpness. meaning lens that is able to produce finer details then sensor resolution is able to read. otherwise there is no gain and it could be done just with digital zoom as well. so in general teleconverters has more chance to have gain with 50mp bodies. for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body. anyone tested it on 50mp? do you know where to find lens tests with information about sharpness in mp? dxomark was giving that, but i think they didnt test fujinon gf lenses.

I don't think MP is a good metric for sharpness.

i wasnt very precise with words, i was trying to simplify. maybe i should say resolution

So you want to see MTF curves for both situations?

i wanted to know how large resolution it could fill with details. but probably MTF curves could be used to judge if it is "sharp" enough.

You can certainly use MTF curves to see if there's much energy above Nyquist.

 

[yaqbpl]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality. on 100mp body it would have to be 196mp. on 50mp it needs 96mp.

that would mean if 100-200mm is sharp enough for 100mp sensor then it is also sharp enough for 50mp with 1.4x tc. if that is the case, i dont know.

 

[JimKasson]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

The dimensions of resolution - the last word

Back when I was in high school physics class, the teacher taught us to analyze the dimensions involved in any problem we were trying to solve before plugging in the numbers and grinding out the calculations. If the dimensions weren't right, you'd set up the problem wrong, and no amount of fancy...

blog.kasson.com

on 100mp body it would have to be 196mp. on 50mp it needs 96mp.

100MP allows 1.4x the resolution of 50MP, limited by aliasing.

that would mean if 100-200mm is sharp enough for 100mp sensor then it is also sharp enough for 50mp with 1.4x tc. if that is the case, i dont know.

 

[yaqbpl]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

let me calculate to see

8256×6192=51121152 (51 mp)

looks aerial, but im not the expert. any way it seem we agree to the numbers just calling it differently, 50*1.4^2 is around 100

 

[JimKasson]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

let me calculate to see

8256×6192=51121152 (51 mp)

looks aerial, but im not the expert. any way it seem we agree to the numbers just calling it differently, 50*1.4^2 is around 100

Did you read the linked post above?

 

[yaqbpl]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

let me calculate to see

8256×6192=51121152 (51 mp)

looks aerial, but im not the expert. any way it seem we agree to the numbers just calling it differently, 50*1.4^2 is around 100

Did you read the linked post above?

 

[JimKasson]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

let me calculate to see

8256×6192=51121152 (51 mp)

looks aerial, but im not the expert. any way it seem we agree to the numbers just calling it differently, 50*1.4^2 is around 100

Did you read the linked post above?

half of it. anyway, my math was wrong when i said 70mp is needed. not 70mp (50*1.4) but 98mp (50*1.4^2). am i right with that?

Pretty much, if you want to find the total pixel count of the sensor. If you use sqrt(2) instead of 1.4 the numbers will come out closer. But pixel count is not resolution. If you have twice the pixel count, you don't have twice the resolution, although the product managers would like it if you believed that you did.

 

[yaqbpl]

for 100mp sensor and 1.4x tc lens would have to be able to produce details on level of 140mp. 100-200mm maight not have it. but i guess it has at least 70mp and it is fine for 1.4 tc on 50mp body.

well, my math was wrong.

1.4x tc enlarage both dimensions, so it requires 1.4^2=1,96 more resolution from lens to not loose quality.

Resolution is not an aerial metric. So all you need is 1.4x the resolution.

let me calculate to see

8256×6192=51121152 (51 mp)

looks aerial, but im not the expert. any way it seem we agree to the numbers just calling it differently, 50*1.4^2 is around 100

Did you read the linked post above?

half of it. anyway, my math was wrong when i said 70mp is needed. not 70mp (50*1.4) but 98mp (50*1.4^2). am i right with that?

Pretty much, if you want to find the total pixel count of the sensor. If you use sqrt(2) instead of 1.4 the numbers will come out closer. But pixel count is not resolution. If you have twice the pixel count, you don't have twice the resolution, although the product managers would like it if you believed that you did.

 

[yaqbpl]

i have bought this lens. wanted to use that 500e discount + there was additional discount from trade show. without teleconverter, it is not available. once it will be i will try to test. did first tests - looking bare eye i dont see much difference in sharpness between focal lenghts and apertures between 5.6-8. if so, slightly sharper at 200mm. but that is on 50mp.

compared it to 110mm at 5.6-8 - also not much difference. that is on same close distance about 2 meters. cropping to same size of test chart object.