[UncleVanya]

I've used a APS-C camera for years, and have dealt with all of the shortcomings of using a cropped sensor. I want better features, better bokeh, better low light capabilities, better noise control, faster AF, and Faster FPS.

I was following your logic until you said faster FPS. Generally the apsc cameras have that not the ff. But perhaps that’s the point? Using both?

I understood what UncleVanya meant. I just don't feel it is worth wasting a extra grand on a 2MP difference when I already have a APS-C that I can use for more reach, if need be.

Actually that’s my confusion. I mistakenly thought you had said the A7RIV was your target body. I agree that if it isn’t you shouldn’t buy it just for that. Having an existing apsc body is a reasonable approach.

[Prince Harbinger]

That's why I recommend TCs be used only with ƒ2.8 glass, or possibly ƒ4.



Sounds like you've used some pretty crappy APS-c cameras.

You need to attach some specs to the features you want improved. Better, suggests we know what you find inadequate, and we don't. The fastest I've used my K-1 with my DA*55 1.4 in the field is ƒ2. I can match that on APS-c by shooting at ƒ1.4. Sometimes these numbers don't mean what you think they will.

Sony doesn't make a 600mm that has a aperture of f/2.8 They make a 600mm F/4 prime that is 13,000.00 While I could save a little and buy that lens. I'm not interested in spending that much money on a lens. I've been using the A6300. I'm done with APS-C cameras. I'm ready to move on to a better more supported format. I'd go over what I'm tired of with APS-C cameras. However I feel it would be like debating which religion is right. You're content with your format and I'm not content with it anymore. There's no convincing me to stick with a format that I've lost interest in. I don't like it for landscape and portraiture. Let alone for tight event spaces. I don't like not being able to see my vintage glass the way it was intended to be seen. You could say than use a focal reducer but that degrades the photo quality. It also changes the character of the bokeh. Giving it a unnatural swirl that I don't want to see on a lens that isn't suppose to have swirl to begin with.

[normhead]

Sony doesn't make a 600mm that has a aperture of f/2.8 They make a 600mm F/4 prime that is 13,000.00 While I could save a little and buy that lens. I'm not interested in spending that much money on a lens. I've been using the A6300. I'm done with APS-C cameras. I'm ready to move on to a better more supported format. I'd go over what I'm tired of with APS-C cameras. However I feel it would be like debating which religion is right. You're content with your format and I'm not content with it anymore. There's no convincing me to stick with a format that I've lost interest in. I don't like it for landscape and portraiture. Let alone for tight event spaces. I don't like not being able to see my vintage glass the way it was intended to be seen. You could say than use a focal reducer but that degrades the photo quality. It also changes the character of the bokeh. Giving it a unnatural swirl that I don't want to see on a lens that isn't suppose to have swirl to begin with.

My solution is FF (K-1) for landscape and larger wildlife, APS-c for birding and smaller wildlife. (K-3)

I'm content with my formats, I've figured out what they re good for for my own use, and my suspicion is you're not going to get what you think you're getting. But the frame rate with Sony is such that you may be happy with a one format solution. My main objection is the weight of FF long glass and the expense. You don't want to spend the money on a 600 ƒ4 for FF, but I have a 300 2.8 that with a 1.4 TC is 420 mm APS_c or 630 FF mm at ƒ4. So, I'm assuming you have no experience on using FF glass, or understanding why my FF body gets left home when I'm birding. You have to carry those beasts to understand. If you never plan to use long glass, FF is a viable one format solution. But Aps-c is more useful in many situations. That's why I shoot both.

But long story short, if you aren't buying $7k-$20k glass for your FF, you won't be getting what I get from my $2k APS-c glass.

Sometimes my camera bag has both my K-1 and K-3 in it with appropriate lenses attached. The DFA 28-105 on The K-1 for landscape. The DA 55-300 PLM on the K-3 for wildlife and birding.(8 FPS isn't too shabby.). And the tiny Sigma 24mm FF lens in the bag , just in case. It doesn't even take a large heavy bag to carry them, and the three lenses can all be used on both formats so 3 lenses but 6 distinct fields of view.

So seriously, I can't entertain your anti-APS-c opinions. Every camera I own (of the 6, with 4 different formats) has a use for which it is the best option. These are tools, popularity contests are futile. They each do what they do.

[Prince Harbinger]

Nothing wrong with a heavy tripod, really! As long as it's what you're comfortable carrying that's what matters. It's all about compromise.

As for your comments on the 60MP sensor etc. Bear in mind this - you're using a 24MP APS-C with a pixel pitch of 3.93µm. A 42MP A7Riii has a pixel pitch of 4.51µm (bigger pixels, more light per pixel, theoretically lower noise, lower pixel density). Only if you're finding your current APS-C out-resolves your vintage lenses, in the portion covered by the APS-C sensor, does stipulating that a 60MP FF sensor would do the same (A7Riv for comparison 3.73µm). Ergo the crop mode equalling 26MP in the latter.
Hope I'm not offending if you already considered that, but it wasn't evident from your response above.

Back to your T/C argument - if you can boost the range at no detriment versus an APS-C camera, whilst not having to carry said APS-C camera, then I'd say it's beneficial. As you rightly point out, whether it's '$1,000' beneficial (what-versus-what what exactly?) then that's a financial choice (but would seemingly fit into your $7k brief).

It's all fairly speculative anyway, given the A7iv has not been released yet, so whether that will actually fit the bill for you remains to be seen. Difficult for us to give valuable recommendations in the meantime when we don't know a definite price for the camera you want.

"bigger pixels, more light per pixel, theoretically lower noise, lower pixel" If that's the case than why does the A7III outperform any A7R, A9, A6xxx camera in the noise department? No I don't see 24MP outresolving old glass and I believe it would be fine on a 32-36MP A7IV. Also the A7IV information was leaked by a trusted source that was spot on with his A1 information. The A7IV is expected to be around $2,500.00 Likely will be a little less with my student discount access.

---------- Post added 02-11-21 at 09:27 AM ----------

I was following your logic until you said faster FPS. Generally the apsc cameras have that not the ff. But perhaps that’s the point? Using both?



Actually that’s my confusion. I mistakenly thought you had said the A7RIV was your target body. I agree that if it isn’t you shouldn’t buy it just for that. Having an existing apsc body is a reasonable approach.

"Generally apsc cameras have that not the ff" That's not true at all.
I'll give you the latest example. The A1 is capable of shooting at*30 fps*with no blackout.

---

Last edited by Prince Harbinger; 02-11-2021 at 09:27 AM.

[Benz3ne]

"bigger pixels, more light per pixel, theoretically lower noise, lower pixel density" If that's the case than why does the A7III outperform any A7R, A9, A6xxx camera in the noise department? (1) No I don't see 24MP outresolving old glass (2) and I believe it would be fine on a 32-36MP A7IV. Also the A7IV information was leaked by a trusted source that was spot on with his A1 information. The A7IV is expected to be around $2,500.00 Likely will be a little less with my student discount access.

---------- Post added 02-11-21 at 09:27 AM ----------



"Generally apsc cameras have that not the ff" That's not true at all (3).
I'll give you the latest example. The A1 is capable of shooting at*30 fps*with no blackout.

Remember BIGGER pixels is not the same as MORE pixels.
A7R - smaller pixels than A7iii.
A6xxx - much smaller pixels than A7iii.
A9 - same sized pixels as A7iii, so processing. Marginal difference at best.

If it is a 24mp camera with FF then the pixels are much bigger than a 24mp APS-C (which is comparable to a 60mp FF, to expand on your point (2)).
You have used an example that (almost) exactly makes my point.
IMO 24mp FF is a nice sweet-spot (I'm not alone in this). 36MP as on the K-1 is also great (as I've used previously).
Point (3) - Generally. I.e. until quite recently. I'd imagine the next line of Sony APS-C Mirrorless will use better processors so can have faster electronic shutters than previously (versus mechanical shutters where generally APS-C has been quicker).

$2,500 sounds like a reasonable estimate. It'll be up to you at the end of the day but it sounds like you've made your mind up.

[normhead]

No I don't see 24MP outresolving old glass and I believe it would be fine on a 32-36MP A7IV. Also the A7IV information was leaked by a trusted source that was spot on with his A1 information. The A7IV is expected to be around $2,500.00 Likely will be a little less with my student discount access.

The only lens I have that actually is out performed by any sensor is a 40 yer old Vivitar 135 2.8. Every other lens I own gets better with a denser sensor. My suspicion is that the whole "you need better lenses for modern sensors" is vastly over-stated by marketers. They do make better lenses now than they used to (for a lot more money). But they are overall just better lenses. It's not about the older lenses being out resolved by the sensor. Pentax has battled the noise / high ISO thing in APS_c with the accelerator chip. (one stop noise reduction in the K-P, two stops of noise reduction in the K-3iii )The new one coming out in the K-3iiI will push that camera well ahead of many FFs for high ISO and noise reduction images. There is a lot more than sensor size and pixel size to noise and high ISO performance. Nothing but the image comparisons matters. Thinking you can make an assessment based strictly on format is a mistake. Internal processing can make a huge difference.

That's where sites like Imaging Resources and DxO come in.

[Prince Harbinger]

My solution is FF (K-1) for landscape and larger wildlife, APS-c for birding and smaller wildlife. (K-3)

I'm content with my formats, I've figured out what they re good for for my own use, and my suspicion is you're not going to get what you think you're getting. But the frame rate with Sony is such that you may be happy with a one format solution. My main objection is the weight of FF long glass and the expense. You don't want to spend the money on a 600 ƒ4 for FF, but I have a 300 2.8 that with a 1.4 TC is 420 mm APS_c or 630 FF mm at ƒ4.5. So, I'm assuming you have no experience on using FF glass, or understanding why my FF body gets left home when I'm birding. You have to carry those beasts to understand. If you never plan to use long glass, FF is a viable one format solution. But Aps-c is more useful in many situations. That's why I shoot both.

But long story short, if you aren't buying $7k-$20k glass for your FF, you won't be getting what I get from my $2k APS-c glass.

Sometimes my camera bag has both my K-1 and K-3 in it with appropriate lenses attached. The DFA 28-105 on The K-1 for landscape. The DA 55-300 PLM on the K-3 for wildlife and birding.(8 FPS isn't too shabby.). And the tiny Sigma 24mm FF lens in the bag , just in case. It doesn't even take a large heavy bag to carry them, and the three lenses can all be used on both formats so 3 lenses but 6 distinct fields of view.

So seriously, I can't entertain your anti-APS-c opinions. Every camera I own (of the 6, with 4 different formats) has a use for which it is the best option. These are tools, popularity contests are futile. They each do what they do.

I'm not sure where you got the impression that I wasn't intending to use my APS-C camera for wildlife. When I stated that I'm done using it for landscape, portraiture, and events. Sony does have a 70-200mm F/2.8 GM that would be 105-300mm on a APS-C body. With a 2X TC it would become a F/5.6 in aperture. Than again with a higher aperture the photos should be more sharper. The noise control on the A7IV should be more superior than the A7R series of cameras. So I likely don't need to revert from mirrorless to DSLR. Also if I need more reach I can always use a feature called clear image zoom that doesn't affect the aperture at all.

[biz-engineer]

"bigger pixels, more light per pixel, theoretically lower noise, lower pixel" If that's the case than why does the A7III outperform any A7R, A9, A6xxx camera in the noise department?

Doesn't outperform anything. The total noise in the image is the same because the sensor area is the same, assuming shutter speed, lens aperture and ISO are the same. Some people believe that larger pixels produce less noise, which is true at pixel level but untrue for the whole image captured. Most noise in a pixel is related to photon noise, that is why the difference between sensors of the same size is negligible, and the difference of 1 stop between apsc and full frame only happens if exposure value is the same (ISO, shutter speed and aperture must be identical).

[Prince Harbinger]

Remember BIGGER pixels is not the same as MORE pixels.
A7R - smaller pixels than A7iii.
A6xxx - much smaller pixels than A7iii.
A9 - same sized pixels as A7iii, so processing. Marginal difference at best.

If it is a 24mp camera with FF then the pixels are much bigger than a 24mp APS-C (which is comparable to a 60mp FF, to expand on your point (2)).
You have used an example that (almost) exactly makes my point.
IMO 24mp FF is a nice sweet-spot (I'm not alone in this). 36MP as on the K-1 is also great (as I've used previously).
Point (3) - Generally. I.e. until quite recently. I'd imagine the next line of Sony APS-C Mirrorless will use better processors so can have faster electronic shutters than previously (versus mechanical shutters where generally APS-C has been quicker).

$2,500 sounds like a reasonable estimate. It'll be up to you at the end of the day but it sounds like you've made your mind up.

Thanks for the refresher. I'll likely upgrade the A6300 as a secondary camera when I feel a worthy successor is available. My cup is half full for Sonys faith in APS-C cameras. They really need to improve upon the dynamic rage, noise control and color science.

---------- Post added 02-11-21 at 10:29 AM ----------

The only lens I have that actually is out performed by any sensor is a 40 yer old Vivitar 135 2.8. Every other lens I own gets better with a denser sensor. My suspicion is that the whole "you need better lenses for modern sensors" is vastly over-stated by marketers. They do make better lenses now than they used to (for a lot more money). But they are overall just better lenses. It's not about the older lenses being out resolved by the sensor. Pentax has battled the noise / high ISO thing in APS_c with the accelerator chip. (one stop noise reduction in the K-P, two stops of noise reduction in the K-3iii )The new one coming out in the K-3iiI will push that camera well ahead of many FFs for high ISO and noise reduction images. There is a lot more than sensor size and pixel size to noise and high ISO performance. Nothing but the image comparisons matters. Thinking you can make an assessment based strictly on format is a mistake. Internal processing can make a huge difference.

That's where sites like Imaging Resources and DxO come in.

Sounds interesting. I know the benefit of mirror sensors is lower noise. digital mirrorless sensors tend to add noise due to the sensors needing to always be on to function. As well as a few other things. I'm not going to mention as I'm sure you already know this. I wouldn't rely on DxO bench marks. Lens rentals is far more accurate imo.

[Benz3ne]

Doesn't outperform anything. The total noise in the image is the same because the sensor area is the same, assuming shutter speed, lens aperture and ISO are the same. Some people believe that larger pixels produce less noise, which is true at pixel level but untrue for the whole image captured. Most noise in a pixel is related to photon noise, that is why the difference between sensors of the same size is negligible, and the difference of 1 stop between apsc and full frame only happens if exposure value is the same (ISO, shutter speed and aperture must be identical).

This reads as conflicting statements to me. May have misread but;
Larger pixels = less noise at pixel level.
Larger pixels = fewer pixels across the sensor size.
Yet same total noise across the sensor?

[normhead]

Thanks for the refresher. I'll likely upgrade the A6300 as a secondary camera when I feel a worthy successor is available. My cup is half full for Sonys faith in APS-C cameras. They really need to improve upon the dynamic rage, noise control and color science.

---------- Post added 02-11-21 at 10:29 AM ----------



Sounds interesting. I know the benefit of mirror sensors is lower noise. digital mirrorless sensors tend to add noise due to the sensors needing to always be on to function. As well as a few other things. I'm not going to mention as I'm sure you already know this. I wouldn't rely on DxO bench marks. Lens rentals is far more accurate imo.

What I like about DxO is not their sharpness calculations, which are completely misleading in that there is no known correlation between lw/ph and image appreciation, is compared to DxO , which tell you what the largest size you can produce a completely noise free image is. With their software. With conventional lw/ph measurements, you have no idea what size you can print. Especially since som many companies seem to o for test chart resolution without regard to what noise reduction has to be done to produce clean images.

After all, I don't want the original resolution, I want the resolution after the noise has been cleaned up.

It's only the finished product that counts. Not the resolution of the original image before its cleaned up.

When DxO tells me I can get a clean 8 MP image out of a 24 MP K-3 file at 2400 ISO, that's useful info.

[biz-engineer]

Yet same total noise across the sensor?

Yes. It's because if you read 4 small pixels (each having random noise of top of signal) and average out the 4 values, you get about the same S/N result as the S/N of a single pixel as large as the 4 pixels combined. On the larger pixel, the light is being averaged before A/D conversion, whereas with the 4 smaller pixels the values (that represent collected light amount) are being averaged after A/D conversion. The S/N can be slightly better with the 4 smaller pixels (as opposed to 1 larger pixel), because A/D quantization noise is averaged as well as photon noise, but that difference in S/N is negligible because the quantization noise over 14bits is very small compared to photon noise. And there is also that the readout of smaller pixels come with an additional sample/hold error due to the small hold capacitance of the smaller pixels. So all in all, it's the same.

[repaap]

Since Sony has super good AF already in A7III I'd get that instead and get as good lenses I could instead in that budget. Then after you'v made back some of that investmet, upgrade body to A7IV in the future and you have nice backup body to go.

just my 2 cents.

[biz-engineer]

Since Sony has super good AF already in A7III I'd get that instead and get as good lenses I could instead in that budget. Then after you'v made back some of that investmet, upgrade body to A7IV in the future and you have nice backup body to go.just my 2 cents.

That's a very good approach. Sony cameras depreciate quickly given how frequently they introduce model upgrades. Good glass on the other hand can serve you a long time, hold their value, and sensor only reflect in image file what the lens give it in the first place... it's a myth that you get much more out of the same glass when moving from 42Mp to 61Mp, at 61Mp you just get a bigger picture of lens imperfections at the edges.

[Benz3ne]

Yes. It's because if you read 4 small pixels (each having random noise of top of signal) and average out the 4 values, you get about the same S/N result as the S/N of a single pixel as large as the 4 pixels combined. On the larger pixel, the light is being averaged before A/D conversion, whereas with the 4 smaller pixels the values (that represent collected light amount) are being averaged after A/D conversion. The S/N can be slightly better with the 4 smaller pixels (as opposed to 1 larger pixel), because A/D quantization noise is averaged as well as photon noise, but that difference in S/N is negligible because the quantization noise over 14bits is very small compared to photon noise. And there is also that the readout of smaller pixels come with an additional sample/hold error due to the small hold capacitance of the smaller pixels. So all in all, it's the same.

So it’s less noise per pixel for smaller pixels?
It’s all semantics though as you rightly point out. Real-world use shows that sensors of this era are typically very good performers with respect to noise anyway.
An interesting clarification though, thank you. I’ll have to spend more time looking at it when I’m less tired.