1" chip vs 3 x 1/3" chips at DVinfo.net

Charlie Steiner · June 20th, 2018, 12:34 PM

Has anyone been able to compare the recent Sony's with 1" sensor to older or current models with three 1/3" chips? I'm wondering in particular about low light. I have a set of NX5U's (1/3" chip) and thinking to upgrade to something newer with the 1".

Mark Watson · June 20th, 2018, 10:47 PM

I compared the FDR-AX100 to lots of cameras, including the Canon XF305 and XL-H1A. If the iris was set to same f-stop, same frame rate and shutter speed, the AX100 did better than both those 1/3-inch cameras. I am sure it is not only due to the physics of the sensors, but also newer chips and image processors are less noisy, IMO. My PXW-Z90 seems to have the same sensitivity as the AX100. Have not published a video on it, but I did a low-light test between the two.

Mark

Ron Evans · June 21st, 2018, 10:18 AM

Quote:

Originally Posted by Charlie Steiner

Has anyone been able to compare the recent Sony's with 1" sensor to older or current models with three 1/3" chips? I'm wondering in particular about low light. I have a set of NX5U's (1/3" chip) and thinking to upgrade to something newer with the 1".

I have a NX5U and an AX100. They are about the same light performance with maybe the NX5U better in really low light but a LOT more noisy. Mainly because the lens can open up to F1.6 rather than f2.8 of the AX100 but only of course at wide angle as both camera lens ramp and close down. The AX100 is a far better camera, sharper and a lot cleaner image. No comparison in picture quality. I also have GH5 and GH5S and these then are better than the AX100 !!! If you need XLR etc like the NX5U the new Sony NX80 would be a far superior camera to the NX5U.

Paul Anderegg · June 21st, 2018, 05:50 PM

You cannot compare same aperture on different sensor sizes, they are not the same. A 2/3" B4 lens for instance, will have an f1.9, but the exact same piece of glass with 1/3" mount, will accomplish an f1.4. The smaller the sensor, the larger the "apparent" aperture becomes. The 1" sensor cameras may be clean for any given gain value, but they perform very poorly in low light, especially with f4.0 at full telephoto.

Paul

Ron Evans · June 21st, 2018, 06:11 PM

I am comparing cameras I have and can say that the AX100 is better than the NX5U and the GH5's are better than both. That is at wide angle and full zoom on the lenses that are fixed on the camcorders and with 12-60 or 14-140 on the GH5's. Accept the technical aspect just saying my view from what I have.

Mark Watson · June 22nd, 2018, 01:03 AM

Charlie,

With a 4K/UHD-capable 1-inch sensor, my testing has shown that the noise/grain seen in the 4K video sample is finer than the grain of my HD resolution video. When I apply Neat Video to the clips, the "finer" grain footage cleans up better than the HD's more coarse grain. Bottom line, there's an advantage to shooting low light with a 4K sensor over an HD sensor. Sensitivity-wise, when I set the AX100 to HD mode and shot a scene, then re-shot that same scene in UHD mode, there was no difference.

A test is available here:

If you have a camcorder that ramps up the aperture when zooming in, such as the AX100 does, there is another low-light advantage that can be gained by shooting in 4K, as long as your project resolution is less than 4K, i.e. HD or SD. You can zoom out from the framing you need in order to achieve a more open aperture, then crop back in during editing to get the framing you need. This gives you maybe a 1/3 of a stop advantage. This is one of those things that works in theory and in tests, but which I would rarely bother to actually worry about otherwise. But it's nice to know about. You can see the "ever so slight" low light advantage in the video below. The right side of the screen is a little brighter.

Mark

Noa Put · June 22nd, 2018, 03:13 AM

I use the ax100 for wedding ceremonies as a safety wide set in the back, often in darker rooms, usually I have to zoom in to get the best frame meaning the iris will close down a bit. Considering the camera is allready 4 years old it still is holding up very well among the latest generation of camera's with some great looking and detailed footage in good light conditions, eventhough it's a decent performer in lower light it is the only camera I have now that needs neatvideo treatment now and then because it is so noisy at higher iso's.

Cliff Totten · June 22nd, 2018, 01:37 PM

Y'know, this is a damn good thread question! I have been wondering this exact thing now since NAB.

I have to guess that Sony is using the very latest generation Exmor RS stacked sensor with gapless design. In the past, it seemed that XDCAM would often use older Sony Semiconductor sensors while Alpha always had the newest models.

These 1/3 chips are to going to need all the technolgy help they can get to perform well. Anybody guess what the cell size will be on these guys? 1.2 microns...1.5 microns?

Just that info alone will give you an idea of the sensor's raw performance. I suspect this camera will rely HEAVILY on a very powerful noise reduction algorithm.

I just hope that its 0db wont need to have high amounts of noise reduction and that it competes with 0db on Sony 1inch-type cameras. (I expect high noise reduction ar higher gains)

Waiting for Doug to get one and tell us whats up with it.

CT

Ron Evans · June 22nd, 2018, 04:37 PM

I think the OP was wanting the difference between his NX5U and new 1" sensors. We now have newer 1" sensors than in the AX100 and new 1/3" sensors in the new PXW Z190. Certainly the AX100 is better than the NX5U in practice because it has a sharper far less noisy image. Be interesting to see how the NX80 compares to the Z190. No price yet on the Z190 but I expect it will be twice the NX80.

David Barnett · June 26th, 2018, 04:25 PM

Charlie I went from a Sony Z5U to a Sony X70. It's definitely an upgrade as far as low light performance is concerned. Much less grain comparitively as well. I can bump it up to maybe 18-21 whereas on the Z5 I would try to stay around 9, otherwise I would need to break out Neat Video or something, which was a pain.

I mostly shoot weddings, and it was a noticeable difference, albeit not a be all end all. Sortof a sidegrade with a step up, but its no a7s low light beast. Still, its a decent little pocket rocket jawn.

One drawback however, as to its littleness, is the zoom & focus sortof share a ring, with a level to switch between the two (I like to zoom in, focus up quick, and zoom out at times, was a habit of mine). I've since broken out of that habit a bit, or use the handle zoom to do so, plus relying a bit on AF when I can.

If you're NX5 is becoming dated, or you've just gotta move on from it, I'd recommend the little ones. However, I wouldn't suggest it if you're looking for a marked improvement into low light ability, its better, and better in handling gain, but I don't think it's mindblowing fwiw.

June 20th, 2018, 12:34 PM	#1
Charlie Steiner Regular Crew Join Date: Jul 2008 Location: New Orleans, LA Posts: 66	1" chip vs 3 x 1/3" chips Has anyone been able to compare the recent Sony's with 1" sensor to older or current models with three 1/3" chips? I'm wondering in particular about low light. I have a set of NX5U's (1/3" chip) and thinking to upgrade to something newer with the 1".

June 20th, 2018, 10:47 PM	#2
Mark Watson Major Player Join Date: Oct 2008 Location: Whidbey Island Posts: 873	Re: 1" chip vs 3 x 1/3" chips I compared the FDR-AX100 to lots of cameras, including the Canon XF305 and XL-H1A. If the iris was set to same f-stop, same frame rate and shutter speed, the AX100 did better than both those 1/3-inch cameras. I am sure it is not only due to the physics of the sensors, but also newer chips and image processors are less noisy, IMO. My PXW-Z90 seems to have the same sensitivity as the AX100. Have not published a video on it, but I did a low-light test between the two. Mark

June 21st, 2018, 05:50 PM	#4
Paul Anderegg Trustee Join Date: Jan 2013 Location: San Diego, Califonia Posts: 1,559	Re: 1" chip vs 3 x 1/3" chips You cannot compare same aperture on different sensor sizes, they are not the same. A 2/3" B4 lens for instance, will have an f1.9, but the exact same piece of glass with 1/3" mount, will accomplish an f1.4. The smaller the sensor, the larger the "apparent" aperture becomes. The 1" sensor cameras may be clean for any given gain value, but they perform very poorly in low light, especially with f4.0 at full telephoto. Paul

June 21st, 2018, 06:11 PM	#5
Ron Evans Inner Circle Join Date: Mar 2003 Location: Ottawa, Ontario, Canada Posts: 4,220	Re: 1" chip vs 3 x 1/3" chips I am comparing cameras I have and can say that the AX100 is better than the NX5U and the GH5's are better than both. That is at wide angle and full zoom on the lenses that are fixed on the camcorders and with 12-60 or 14-140 on the GH5's. Accept the technical aspect just saying my view from what I have.

June 22nd, 2018, 01:03 AM	#6
Mark Watson Major Player Join Date: Oct 2008 Location: Whidbey Island Posts: 873	Re: 1" chip vs 3 x 1/3" chips Charlie, With a 4K/UHD-capable 1-inch sensor, my testing has shown that the noise/grain seen in the 4K video sample is finer than the grain of my HD resolution video. When I apply Neat Video to the clips, the "finer" grain footage cleans up better than the HD's more coarse grain. Bottom line, there's an advantage to shooting low light with a 4K sensor over an HD sensor. Sensitivity-wise, when I set the AX100 to HD mode and shot a scene, then re-shot that same scene in UHD mode, there was no difference. A test is available here: FDR-AX100 Low-Light Test HD vs 4K on Vimeo If you have a camcorder that ramps up the aperture when zooming in, such as the AX100 does, there is another low-light advantage that can be gained by shooting in 4K, as long as your project resolution is less than 4K, i.e. HD or SD. You can zoom out from the framing you need in order to achieve a more open aperture, then crop back in during editing to get the framing you need. This gives you maybe a 1/3 of a stop advantage. This is one of those things that works in theory and in tests, but which I would rarely bother to actually worry about otherwise. But it's nice to know about. You can see the "ever so slight" low light advantage in the video below. The right side of the screen is a little brighter. FDR-AX100 Low-Light Test - an advantage of cropped 4K on Vimeo Mark

June 22nd, 2018, 03:13 AM	#7
Noa Put Inner Circle Join Date: Jun 2007 Location: Belgium Posts: 9,510	Re: 1" chip vs 3 x 1/3" chips I use the ax100 for wedding ceremonies as a safety wide set in the back, often in darker rooms, usually I have to zoom in to get the best frame meaning the iris will close down a bit. Considering the camera is allready 4 years old it still is holding up very well among the latest generation of camera's with some great looking and detailed footage in good light conditions, eventhough it's a decent performer in lower light it is the only camera I have now that needs neatvideo treatment now and then because it is so noisy at higher iso's.

June 22nd, 2018, 01:37 PM	#8
Cliff Totten Trustee Join Date: Aug 2010 Location: Florida Posts: 1,197	Re: 1" chip vs 3 x 1/3" chips Y'know, this is a damn good thread question! I have been wondering this exact thing now since NAB. I have to guess that Sony is using the very latest generation Exmor RS stacked sensor with gapless design. In the past, it seemed that XDCAM would often use older Sony Semiconductor sensors while Alpha always had the newest models. These 1/3 chips are to going to need all the technolgy help they can get to perform well. Anybody guess what the cell size will be on these guys? 1.2 microns...1.5 microns? Just that info alone will give you an idea of the sensor's raw performance. I suspect this camera will rely HEAVILY on a very powerful noise reduction algorithm. I just hope that its 0db wont need to have high amounts of noise reduction and that it competes with 0db on Sony 1inch-type cameras. (I expect high noise reduction ar higher gains) Waiting for Doug to get one and tell us whats up with it. CT

June 22nd, 2018, 04:37 PM	#9
Ron Evans Inner Circle Join Date: Mar 2003 Location: Ottawa, Ontario, Canada Posts: 4,220	Re: 1" chip vs 3 x 1/3" chips I think the OP was wanting the difference between his NX5U and new 1" sensors. We now have newer 1" sensors than in the AX100 and new 1/3" sensors in the new PXW Z190. Certainly the AX100 is better than the NX5U in practice because it has a sharper far less noisy image. Be interesting to see how the NX80 compares to the Z190. No price yet on the Z190 but I expect it will be twice the NX80.

June 26th, 2018, 04:25 PM	#10
David Barnett Major Player Join Date: Apr 2009 Location: Philly, PA Posts: 951	Re: 1" chip vs 3 x 1/3" chips Charlie I went from a Sony Z5U to a Sony X70. It's definitely an upgrade as far as low light performance is concerned. Much less grain comparitively as well. I can bump it up to maybe 18-21 whereas on the Z5 I would try to stay around 9, otherwise I would need to break out Neat Video or something, which was a pain. I mostly shoot weddings, and it was a noticeable difference, albeit not a be all end all. Sortof a sidegrade with a step up, but its no a7s low light beast. Still, its a decent little pocket rocket jawn. One drawback however, as to its littleness, is the zoom & focus sortof share a ring, with a level to switch between the two (I like to zoom in, focus up quick, and zoom out at times, was a habit of mine). I've since broken out of that habit a bit, or use the handle zoom to do so, plus relying a bit on AF when I can. If you're NX5 is becoming dated, or you've just gotta move on from it, I'd recommend the little ones. However, I wouldn't suggest it if you're looking for a marked improvement into low light ability, its better, and better in handling gain, but I don't think it's mindblowing fwiw.