Sony RX100 IV comparison on DPR

[Mt Spokane Photography]

http://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_dscrx100m4&attr13_1=sony_dscrx100m3&attr13_2=panasonic_dmclx100&attr13_3=canon_g7x&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=800&attr16_1=800&attr16_2=800&attr16_3=800&normalization=full&widget=1&x=0&y=0

When looking at raw images, at higher ISO's, the Lumix LX100 is a walk away winner. Noise is virtually identical to the RX100 III, so the stacked sensor is not reducing noise. It must be a improvement for video, or just a excuse to raise the price to $1000.

The Lumix results were pretty spectacular, but then, its a 4/3 sensor. When you add in the G1X MK II, it has very close noise to the Lumix. However the Lumix is on the large side as is the G1X. The G7X definitely has more noise at higher ISO's, and the images are not as sharp. However, it has a much wider zoom range, so that compromises a lens design.

 

[rs]

All this hype surrounding the recent Sony announcements has been quite overwhelming. A BSI full frame sensor and stacked 1" sensors. Quite staggering, with worship levels of expectations. And then these results come along. The mk IV is if anything worse than the mk III at high ISO. I'm curious to see the A7R mk II now.

 

[expatinasia]

Interesting, but hasn't Sony openly admitted a problem with their RAW files which they say can - and will - be fixed by a firmware update.

If you play around with the test settings and switch to JPEG and move the image to the girl's face on the right the Panasonic looks terrible and the RX100 Mark IV looks much better.

The mysteries of camera technology.

 

[psolberg]

http://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_dscrx100m4&attr13_1=sony_dscrx100m3&attr13_2=panasonic_dmclx100&attr13_3=canon_g7x&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=800&attr16_1=800&attr16_2=800&attr16_3=800&normalization=full&widget=1&x=0&y=0

When looking at raw images, at higher ISO's, the Lumix LX100 is a walk away winner. Noise is virtually identical to the RX100 III, so the stacked sensor is not reducing noise. It must be a improvement for video, or just a excuse to raise the price to $1000.

The Lumix results were pretty spectacular, but then, its a 4/3 sensor. When you add in the G1X MK II, it has very close noise to the Lumix. However the Lumix is on the large side as is the G1X. The G7X definitely has more noise at higher ISO's, and the images are not as sharp. However, it has a much wider zoom range, so that compromises a lens design.

unless the stacked sensor was 4/3 I'm not sure how you can even suggest it had or didn't have an effect. The effective sensor area on the LX100 is really 1.5X larger than 1". The area difference is not trivial at all not to mention, did you bother to calculate the pixel density of 20MP at 1"? You're talking not just apples and oranges. You're comparing a freaking watermelon to a grape.

To be honest, there is some marketing ocus pocus. Everybody does that. But just as BSI is standard, stacked will become. Sony is just the first to get there. I'm sure we will see the benefits as the technology is perfected. Who knows, maybe the chip design is not as good but it would have been worse if it had just been BSI.

To really determine if BSI is worse than stacked, you'd have to have the same chip made both ways. Given you will never get this, I have to think sony is simply following the natural evolution:
-front wiring, BSI, stacked.

 

[3kramd5]

The area difference is not trivial at all not to mention, did you bother to calculate the pixel density of 20MP at 1"? You're talking not just apples and oranges. You're comparing a freaking watermelon to a grape.

It's probably more like a blueberry to a grape

 

[Mt Spokane Photography]

Interesting, but hasn't Sony openly admitted a problem with their RAW files which they say can - and will - be fixed by a firmware update.

If you play around with the test settings and switch to JPEG and move the image to the girl's face on the right the Panasonic looks terrible and the RX100 Mark IV looks much better.

The mysteries of camera technology.

Sony has been using lossy compression in their raw files for some time. They seem to feel that no one would notice. It reduces resolution and detail a tiny amount, but using lossy compression does not increase noise, but might hide it by a very tiny amount.

 

[neuroanatomist]

Interesting, but hasn't Sony openly admitted a problem with their RAW files which they say can - and will - be fixed by a firmware update.

Not quite. They stated they are aware that customers are asking for uncompressed RAW files, which could be implemented with a firmware update...but they haven't committed to doing so.

Also worth noting that Sony stated just that – uncompressed RAW. Not sure if that's what people are actually asking for, or if Sony is misunderstanding the message. The problem isn't compression, it's lossy compression. If Sony implements what they said people are asking for, those people can look forward to shrunken file buffers and slower frame rates.

 

[JoeDavid]

The Lumix results were pretty spectacular, but then, its a 4/3 sensor. When you add in the G1X MK II, it has very close noise to the Lumix. However the Lumix is on the large side as is the G1X. The G7X definitely has more noise at higher ISO's, and the images are not as sharp. However, it has a much wider zoom range, so that compromises a lens design.

Another Panasonic comparison to do is with the FZ1000. It has a 20MP 1" sensor and has been out for about a year. Above ISO 800 the older Panasonic sensor (assuming Pansonic made that sensor) appears to have a very slight noise advantage over the new Sony sensor.