cainn24

Lives in Australia Australia
Joined on Dec 9, 2012

Comments

Total: 241, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On article 2016 Roundup: Compact Enthusiast Zoom Cameras (209 comments in total)
In reply to:

photominion: Does the equivalency-comparison account for the fact that the LX100 uses a crop of the 4/3s sensor and not the entire surface at any given time?

That illustration doesn't show anything that relates to how much bigger the lens would need to be if it covered the entire sensor. What is shows is the following, from left to right:

1) the LX7 lens elements
2) the relative size of the LX100 lens elements
3) how big the LX100 lens elements would be if they had simply scaled up the LX7 lens design instead of utilizing new design principles and fabrication techniques

(note: the LX7 lens/sensor combination is a multi-aspect affair as well, which means that its image circle doesn't fully cover its smaller 1/1.7"-type sensor either)

Link | Posted on Apr 30, 2016 at 10:41 UTC
On article Power Zoom: Panasonic Lumix DMC-ZS100/TZ100 Review (280 comments in total)
In reply to:

Mike FL: It seems SONY is very happy with RX100s' sales, and has no plan to make:

- 1" longer zoom/travel zoom like Panasonic TZ100, and
- 1" UWA like NIKON DL18-50

We will see.

Canon is on the game of catching up while Olympus and FUJI is kind of gunshy for making 1" P&S.

It seemed to me that Sony were pretty happy with the RX10's focal length range. So much so they released an RX10 II with the same lens.

But then, right out of left field, the RX10 III.

So I'd say that Sony could do just about anything at any minute.

I don't own any Sony cameras (although I might one day soon) but I do love the way they shake things up.

Link | Posted on Apr 11, 2016 at 06:34 UTC
On article First sample images from 20MP Sony Cyber-shot RX10 III (197 comments in total)
In reply to:

utomo99: Sony need to improve the jpeg engine.
If needed hire best programmer to improve it for all the Sony products. It is cheap investment

What's genuinely "lame" is the inability of some people to understand that their own subjective perceptions can not be legitimately elevated to the status of objective fact. This is, in fact, an absurdity from which countless ridiculous arguments emerge on these forums just about every day.

Link | Posted on Apr 5, 2016 at 12:38 UTC
On article First sample images from 20MP Sony Cyber-shot RX10 III (197 comments in total)
In reply to:

utomo99: Sony need to improve the jpeg engine.
If needed hire best programmer to improve it for all the Sony products. It is cheap investment

The JPEG engine settings that make things look the least horrible at the pixel level, where pixel peepers congregate, are rarely the same settings that make things look the most pleasing at more typical print/display sizes. So whichever crowd you cater to there will be criticism coming in from all over. And this is true without even addressing the innumerable number of other variables that can also affect one's perception of IQ.

In any case you still have two options here:

1) change the default JPEG engine settings. They aren't set in stone

2) shoot RAW and bake your JPEGs just the way _you_ like them. Seriously. I realize there are a lot of people out there who scoff at RAW, but it's a bit silly to do that if you are simultaneously so _refined_ in your tastes that you find yourself easily dissatisfied with a modern JPEG engine, and feel that the scope for adjustment might be insufficient to bring things in line with your expectations

Link | Posted on Apr 2, 2016 at 03:15 UTC
On article First sample images from 20MP Sony Cyber-shot RX10 III (197 comments in total)
In reply to:

Orcio14: Really good pictures. I've a question: are they directly from camera, or after any PP, rather?

Do you really have to wait for a pre-release version of ACR before making RAW files available? I realize you probably want to put up some actual renders at the same time, but there's plenty of other ways for people to usefully play with RAW files in the meantime ;)

Link | Posted on Mar 30, 2016 at 23:23 UTC
On article Hands-on with the Sony RX10 III (302 comments in total)
In reply to:

maxnoy: How do you get 2.35mp from 1024x768?

The missing part of that explanation (I realize now) is an RGB backlight. That's how each "dot" can be more than one colour.

Link | Posted on Mar 30, 2016 at 07:40 UTC
On article Hands-on with the Sony RX10 III (302 comments in total)
In reply to:

orion1983: If there was only a stills-only version for 500 bucks less... But one cannot have all of his wishes fulfilled. Looks like THE travel and nature cam for a looooong time...til the next update in 2017 ;) Time to save money and/or soll Nikon DSLR gear...

It wouldn't be cheaper without the video features because wider appeal gives you economies of scale (lower per unit production cost, which can more than offset any additional components and development, particularly when video-centric features already exist in sensors and DSPs anyway).

Link | Posted on Mar 30, 2016 at 07:35 UTC
On article Hands-on with the Sony RX10 III (302 comments in total)
In reply to:

Donald B: Are you guys for real "compact" it weighs nearly as much as a 1.25 litre bottle of coke. could you imagine carrying that around your neck all day.

And the 600mm equivalent lens you're using on your EPL5 has no wide end.

Link | Posted on Mar 30, 2016 at 06:01 UTC
On article Hands-on with the Sony RX10 III (302 comments in total)
In reply to:

maxnoy: How do you get 2.35mp from 1024x768?

It's standard practice for all field-sequential EVFs, and pretty much all EVFs are field-sequential EVFs. And it's not quite marketing BS either. I used to think so myself until I read more about it. Here's how I understand it now, in a nutshell:

With a traditional design a single pixel consists of RGB subpixels, but with a field-sequential design full colour information can be displayed by "dots" that are the same size as those RBG subpixels. This is achieved by showing red, green and blue in sequence, and if you do it fast enough the human visual system perceives full colour information as if it was being presented all at once (so it's like having 3 dots for every one dot). One of the side-effects however is a rainbow/tearing effect when panning, but that can be minimized and possibly even avoided with a sufficiently high refresh rate.

Does this really provide _3 times_ the resolution? No, it seems that they probably don't quite get there. But they do in fact provide more.

Link | Posted on Mar 30, 2016 at 05:56 UTC
In reply to:

falconeyes: I must say I am disappointed by the optical performance at the long end.
I am not sure that cropped images (aka digital zoom) from e.g., LX100 or RX100 (or DL or GX) would be visibly worse (at same FoV).

I know that I can heavily crop my RX100m3 images at 70mm and still obtain very usable images. Somebody should make the comparison, that would be a useful task when comparing all those recent 1" cameras ...

And processed in DxO this time:

https://www.dropbox.com/s/jhiv7obrsu43mzx/comparo_sharpened.png?dl=0

RX100 III left, ZS100 right

DxO Optics Pro seems to do some sort of deconvolution sharpening (the Lens Softness module), and it's sometimes a very effective operation. The Sony still has a clear edge, but it appears to have diminished slightly.

Interesting.

Link | Posted on Mar 9, 2016 at 19:06 UTC
In reply to:

falconeyes: I must say I am disappointed by the optical performance at the long end.
I am not sure that cropped images (aka digital zoom) from e.g., LX100 or RX100 (or DL or GX) would be visibly worse (at same FoV).

I know that I can heavily crop my RX100m3 images at 70mm and still obtain very usable images. Somebody should make the comparison, that would be a useful task when comparing all those recent 1" cameras ...

Here, I did the work for you. These are center crops from IR test charts all shot at full tele for all cameras. The best performing fnumber for each camera was chosen. The RAW files were rendered with RawDigger which means no sharpening and no noise reduction. So we have an even playing field on which to try to judge optical performance:

https://www.dropbox.com/s/pcym24e08e3h120/comparo.png?dl=0

RX100 III left, ZS100 middle, LX100 right (upsampled to match)

The RX100 III crop is clearly sharper, as expected. And the ZS100 shot is lacking a little contrast, which also isn't exactly shocking. But the differences here aren't remotely significant enough to lend any credence to the idea that a 70mm RX100 III shot could be cropped to a 250mm EFL and remain in the game.

1.6MP just isn't enough. The remaining sensor area just isn't enough. And the remaining image circle area just isn't enough.

But again, that's not to say that there isn't _something_ wrong with DPRs ZS100 shots.

Link | Posted on Mar 9, 2016 at 18:41 UTC
In reply to:

falconeyes: I must say I am disappointed by the optical performance at the long end.
I am not sure that cropped images (aka digital zoom) from e.g., LX100 or RX100 (or DL or GX) would be visibly worse (at same FoV).

I know that I can heavily crop my RX100m3 images at 70mm and still obtain very usable images. Somebody should make the comparison, that would be a useful task when comparing all those recent 1" cameras ...

I've done exactly what I suggested that you do, and a whole lot of other very similar things. I like to try to be diplomatic but you're so far off the mark here it really is a bit silly.

Go ahead and try it yourself. And go check the Imaging Resource resolution chart samples as well. Or don't, and stick with your RX100 III, and crop away. If you're happy, you're happy.

Link | Posted on Mar 9, 2016 at 18:05 UTC
In reply to:

falconeyes: I must say I am disappointed by the optical performance at the long end.
I am not sure that cropped images (aka digital zoom) from e.g., LX100 or RX100 (or DL or GX) would be visibly worse (at same FoV).

I know that I can heavily crop my RX100m3 images at 70mm and still obtain very usable images. Somebody should make the comparison, that would be a useful task when comparing all those recent 1" cameras ...

Go get the sharpest 70mm RX100 III shot you can possibly find. Crop out a 1.6MP portion of it (1532x1021, which corresponds to an area of the sensor that is smaller in size than the smallest smartphone sensor). Resample the image back to 20MP. Now look at it. After you've done that see if you can come back here and maintain your position that cropping the cameras output to match a 250mm EFL will get you even remotely close to matching the IQ of the ZS100 at the same EFL, with a straight face.

The LX100 wont do any better because it's not quite as sharp as the Sony. In fact if you head over to Imaging Resource and look at some resolution charts you will find that the ZS100 matches the LX100 for peak detail resolution when both are shot at full tele anyway so this is over before it begins, and definitively so.

You're gonna have to blame whatever softness you see on something other than optical performance. Camera shake, flare, AF inaccuracy, whatever, and who knows...

Link | Posted on Mar 9, 2016 at 15:20 UTC
In reply to:

snapa: As long as people do not print larger than 4x6 or view images on anything other than cell phones, these types of pictures should be OK. My old Sony Cyber-shot DSC-H5 (2006) with 7MP really put his camera to shame with much better IQ.

"So you think pixels that are 2.5 times smaller on the same size sensor can take better pictures, lower noise, has better IQ? I think not."

Put a modern 18MP 1/2.3"-type sensor in that Sony of yours, set it to output 7MP JPEGs (or just downsample in post), and you'll have a camera that produces more detailed images with less noise, particularly at higher ISOs. This is because sensor technology has advanced quite a lot in 10 years.

The 30x TZ-series cameras are at a bit of an optical disadvantage in some respects but the sensor resolution makes up for that (that doesn't work forever but it's certainly a factor in this case).

If you don't understand this I think it would be wise to take a step back and rework your perspective on the IQ equation based on the wealth of performance data that is available for perusal.

"Are they something YOU would be proud of?"

These look respectable: http://www.dpreview.com/forums/post/57390016

About the same resolution as your Sony as well.

Link | Posted on Mar 5, 2016 at 18:04 UTC
In reply to:

snapa: As long as people do not print larger than 4x6 or view images on anything other than cell phones, these types of pictures should be OK. My old Sony Cyber-shot DSC-H5 (2006) with 7MP really put his camera to shame with much better IQ.

"My old Sony Cyber-shot DSC-H5 (2006) with 7MP really put his camera to shame with much better IQ."

Nope. An all too common classic mistake on your part. 18MP TZ-series output downsampled to 7MP will be more crisp and more detailed than what your Sony produces because the higher pixel count simply wins out even when up against the additional diffraction introduced by the slower lens.

Mistake? Yes. The only way you can reach the conclusion you reached is by doing a comparison at 100% which is just plain silly. If a 7MP image is a point of comparison, then it's a point of comparison.

You can fall back to complaints about colour/contrast/WB differences if you like but the ZS60 is the first TZ to offer properly configurable JPEG engine. And if you shoot RAW it simply doesn't matter.

My ZS40 even outresolves my Coolpix 8800, which has an 8MP 2/3"-type sensor and is of roughly the same "golden era" vintage as your Sony.

Nostalgia quickly evaporates when the comparison is fair.

Link | Posted on Mar 5, 2016 at 17:01 UTC
In reply to:

Dragonrider: For what it's worth, DXO Optics does a much quieter conversion than ACR on a ZS40, so when DXO gets the ZS60 profiled, expect to see considerable improvement there as well. The prime NR process is too slow for general use, but if you have a keeper that you want to look good, it really cleans up the noise without much loss of detail.

Panasonic made a travel zoom camera that shoots RAW. Three in fact (5 if you count the LF1 and the ZS100). Lightroom supports them all. DxO Optics Pro already has support for some, and will very likely support more.

Such cameras are being made. People are buying them. And clearly some of those people are using them to output RAW files that they then process in programs like Lightroom and DxO Optics Pro. I know of many such people personally. I am one of them. We may be a small portion of the market, but not only are we out there, we are _in here_ reading these reviews and posting in the comments section. As such Dragonrider's comment is right at home.

Link | Posted on Mar 5, 2016 at 15:47 UTC
In reply to:

Ridethelight: Why does anyone need such a massive zoom in this type of camera ?
When I take photos, I like to be in the near vicinity of the subject, I'm seriously intrigued to what these super zooms are supposed to be for .
Tiny sensor and a slow zoom in 2016 is actually depressing .

A casual birding option that fits in a pocket. Really. These cameras can produce decent grabs in the right hands, just like any other camera.

It's that simple really.

Link | Posted on Mar 5, 2016 at 15:27 UTC
In reply to:

pacnwhobbyist: It seems like a step back in IQ over its predecessor.

Anyone who wants 12MP JPEGs can just set the camera to output them. You'll get a tiny bit more detail than if the camera had a 12MP sensor instead (as evidenced by ZS40 vs ZS50 comparisons) and there will only be a significant noise difference at the highest ISOs (and nowhere near as significant as people seem to believe).

This is really a non-issue I think.

Link | Posted on Mar 5, 2016 at 15:18 UTC
In reply to:

piratejabez: Decent detail, acceptable noise for such a small sensor, but RAW doesn't appear to be able to add any useful benefit apart from exposure (and ACR doesn't know what to do with the colors). I think it's a great option for casual or traveling JPEG shooters.

@ breivogel

Noise is the achilles heel of a small format camera like this. Optical performance is a close second partly due to miniaturization and partly due to diffraction. That's why RAW is so profoundly useful. You need to sharpen a lot which accentuates the already prominent noise which then necessitates the use of sophisticated noise reduction tools if you want reasonably clean and detailed output. Same is true when pulling shadows where small format output is even more in need of such treatment.

The bottom line is that I am a whole lot more likely to be satisfied with a SOOC JPEG produced by a larger format camera than one produced by a small format camera. And I suspect that most people would feel the same. Hence RAW is in fact more critical for small formats for the _discerning_ user.

Unfortunately not many people have seen what tools like DxO Optics Pro can do for small format output (most camera review sites included) so the benefits continue to be understated.

Link | Posted on Mar 5, 2016 at 15:10 UTC
In reply to:

Ross the Fidller: @Rishi
Why the put down of a smaller system when reviewing a lovely lens.

"Naturally, the small size and weight do come at a cost: with an equivalent maximum aperture of F8 and F12.6 on the 200mm and 800mm equivalent ends, respectively, subject isolation and noise performance on a MFT camera won't be the same as what you might get with a full-frame 800mm F5.6 lens."

You sound like a 'FF fanboy' with that sort of "equivalence" talk & it is unwarranted. Sure larger sensors have less noise because they typically use larger pixel cells in their array. Whoop-de-doo! This lens is a 100-400mm lens that has to be specified with an equivalent 35mm format focal length for reference so FF fanboys might understand the reach or angle of view it has on the MFT format system. For exposure purposes it is an F4-F6.3 lens that equates to F8-F12.6 for DoF of that for 35mm system. Please leave the rest of the unnecessary BS out of it because it only shows your bias.

The Stylus 1 is a great little camera.

Link | Posted on Jan 11, 2016 at 08:04 UTC
Total: 241, showing: 1 – 20
« First‹ Previous12345Next ›Last »