deep7

Lives in New Zealand (Aotearoa) New Zealand (Aotearoa)
Works as a writer/photographer/ecologist
Has a website at deeppics.com
Joined on May 10, 2008
About me:

God makes it, I see it and photograph it. Sometimes that works well!

Comments

Total: 1270, showing: 61 – 80
« First‹ Previous23456Next ›Last »
On article Sony Cyber-shot RX10 IV review (522 comments in total)
In reply to:

MrBrightSide: Question from we animal photographers: Does the eye AF work on animals very well? Or only humans?

I have read of it working on dogs.

Link | Posted on Nov 15, 2017 at 05:14 UTC
On article Sony Cyber-shot RX10 IV review (522 comments in total)
In reply to:

camcom12: "there are times this reviewer craved something a tad wider than 24mm."

"And some landscape photographers will find 24mm not wide enough for their needs."

It is surprising there are relatively few <24mm equiv. fixed-lens zoom models available these days. I can think of one tiny-sensor FZ model from Panasonic. It is a feature some would like. For example, the DL18-50 might have filled that niche. But it's probably no easy task to make an affordable zoom that starts at ~20mm and ends above 200mm, for any sensor 1" or larger.

I think it's amazing they (three manufacturers) can go this wide on a largish sensor super-zoom and still get pretty decent quality. I realise a fair few people want wider but I'd hardly call that a negative with such a huge range!

Personally, I find nearly all my better landscapes (the ones people go "wow" over) were taken with a telephoto lens. "Wide" doesn't equate with "landscape" most of the time. Cityscape maybe...

Link | Posted on Nov 15, 2017 at 05:13 UTC
On article Sony Cyber-shot RX10 IV review (522 comments in total)
In reply to:

proxy: Not taking anything away from Sony RX10IV, which is still a fine camera... this review shows how great of a deal is FZ1000. Almost negligible degradation is sharpness vs. a saving of about $1000. Sharpness can be improved in post if one chooses so. FZ2500 seems to have weakest optics of the bunch. I like RX10IV but did not see a compelling argument to upgrade from FZ1000 which still has a killer 4K and 1080p + v. good photo, especially for the price. Does RX10IV have so much more to cost about 2.5x the FZ1000 (today)? I don't think so. Phase detection sensor, slightly better algorithms and a notch better optics do not amount to this extreme price difference, especially when we talk about a upgraded existing camera, not a new product.

Proxy makes a fair point. I wanted an RX10/3 but bought an FZ1000. $900 as opposed to $2,500 at the time. The FZ1000 is a bulbous, plasticky lump next to an RX10 but, it turns out, it's a super convenient camera which can produce the goods - sometimes even very well. For sure I'd like more reach and better quality at the long end but it seems like a heck of an indulgence to pay that much to get it! IF it was my only camera, that would be different...

Link | Posted on Nov 15, 2017 at 04:41 UTC
In reply to:

davidodd: Confused. If this mode moves the sensor 1 pixel then how do we get an increase in resolution above that from the actual photo-sites? If we are; then how is this different from enlarging a JPEG file (for which we have RGB at a pixel level)?

An "emoticon" would be wasted on me. Aside from that, I wonder if you are looking at this back to front. Think of it this way - a traditional Bayer doesn't always achieve full spatial resolution. Foveon and pixel shift go a long way to fixing that. They don't add more resolution than is already there (except for the Olympus and Panasonic systems which, effectively, overlay a second, slightly offset, image) but simply produce what is there more accurately, unlike a simpler Bayer sensor/grid. So no, not more photosites with the Sony or Pentax system but the photosites are more accurate. That's all.

Link | Posted on Nov 14, 2017 at 03:02 UTC
In reply to:

davidodd: Confused. If this mode moves the sensor 1 pixel then how do we get an increase in resolution above that from the actual photo-sites? If we are; then how is this different from enlarging a JPEG file (for which we have RGB at a pixel level)?

The "magical sensor" you refer to already exists. Called the Foveon and used by Sigma.

As for the issue you are struggling to understand, you have to realise that the difference comes about because there is up to four times the information at each pixel. There is no more resolution in the image per se (same number of pixels) but each pixel produces more information and contributes more precisely to detail in the image. The "guessing" produces things like colour pattern moire and, sometimes, imprecise edges, which near-enough vanish with pixel-shift technology.

Link | Posted on Nov 13, 2017 at 17:36 UTC
In reply to:

Androole: I don't quite understand why people are pixel-peeping an 80MP JPEG and expecting it to compete at a 100% crop with 36-45MP images from FF cameras processed from RAW.

At the very least, we should wait and see what the RAW results look like. But even more realistically, we should look at a downsampled version of the image. There simply isn't 80MP worth of data with a half-pixel overlap. If I'm looking at this correctly, it should be more like 1.5^2*20MP, so 45MP. This is why Olympus gives you a 50MP JPEG only, and why Panasonic has a 40MP option with the G9.

That is the image that you should be looking at if you want to compare it with a magnifying glass to an A7r III or D850.

Download one of these images, resize to 40Mp and, bang, all the little aberrations vanish. You still give away a little detail but the quality is really good.

Link | Posted on Nov 13, 2017 at 17:27 UTC
In reply to:

DavidBleeker: Very nice resolution gain but movement still poses a problem. See person in one of the windows.
Also in the studio shot the paint tray in the right bottom corner shows a lot more colour fringing!

You answered your own question! No need to go through the upsizing part of the process with the Sony. I thought that was obvious, whoops. Sorry if I confused you.

Link | Posted on Nov 12, 2017 at 23:58 UTC
In reply to:

obsolescence: Since the tripod & head are essential equipment for this demonstration, I think it would be appropriate for the author to say what brand(s) and model(s) were used.

Unusual for me to defend Rishi but why snap at him? He answered your question accurately. The brand/model is not that relevant. What is relevant is that the tripod/head combination was steady enough and there are a vast number of tripods that will do that, if used carefully.

Link | Posted on Nov 12, 2017 at 23:44 UTC
In reply to:

eno2: Please Dpreview team, do a proper test like this also on the Panasonic G9 pixel shift mode. What you did in that previous article (with a lot of people moving) was very disturbing and non representative for what this function can achieve.
Also, please test the G9 pixel shift option with a very sharp lens at it's sharpest aperture (probably around f/4) to get the maximum out of this feature. 80 MP is very demanding on the lens side of things and using a mediocre lens at a diffraction ready aperture (over f/4) won't show what this function is really capable of.
Thanks!

Do you not expect them to do just that? The article you referred to was to demonstrate another possible use for the technology - and a very interesting article at that. The dpreview testers may not test everything and some of what they test may be contentious but they do tend to investigate technology like this pretty thoroughly. Patience, Grasshopper.

Link | Posted on Nov 12, 2017 at 23:34 UTC
In reply to:

DavidBleeker: Very nice resolution gain but movement still poses a problem. See person in one of the windows.
Also in the studio shot the paint tray in the right bottom corner shows a lot more colour fringing!

As I mentioned above, you can fix these issues by blending with a "normal" exposure. That's harder to do with the Olympus/Panasonic system but straightforward with the Sony.

Link | Posted on Nov 12, 2017 at 20:03 UTC
In reply to:

Mescalamba: Thats, impressive..

..4th lowest tier window on right side shows some artifacts (maybe someone went in during taking pics and went out).

Also A7RIII can make feathers in test scene magically appear, thats something..

Plenty of artefacts in those windows (computer screens, people moving). However, in a situation like that, you have the option of taking a "normal" image and blending in software later. Sounds like a pain but you'd only use a system like this for super-fussy work which would have you in an image manipulation programme anyway. Unless you were some sort of nut case, of course!

Link | Posted on Nov 12, 2017 at 20:01 UTC
In reply to:

davidodd: Confused. If this mode moves the sensor 1 pixel then how do we get an increase in resolution above that from the actual photo-sites? If we are; then how is this different from enlarging a JPEG file (for which we have RGB at a pixel level)?

Because, normally, only a part of the information from each pixel is recorded, the rest being derived from neighbouring pixels. Take the guesswork out and replace it with actual information from that pixel site and resolution appears better. Still the same number of pixels, just more accurate pixels.

Link | Posted on Nov 12, 2017 at 19:53 UTC
In reply to:

User9362470513: I tested the Hi Res mode on the Olympus em1 mk2 with the 12-40 Pro lens against a Sigma Dp3Merrill.

The Sigma DESTROYED the Olympus.

I took a photo of a bookshelf in gloomy light from about 15 feet. The Olympus was good, but the Dp3Merrill picked up individual threads and dust on the glass front of the bookshelf that were just not visible with the Olympus. The Sigma had incredible fine detail of the texture on the book spines, which were just blur with the Olympus.

If you want to take high resolution shots with a small mirrorless camera and you are going to use a tripod, save time and money and try a Sigma Foveon camera.

Interesting. However, that does go against a lot of evidence to the contrary. Was the focus point identical in each photo? It sounds a bit like a plane of focus miss.

Shame your Olympus got destroyed. Expensive to replace...

Link | Posted on Nov 12, 2017 at 00:41 UTC
In reply to:

deep7: I just clicked on the market pic. Three observations:

1) It's been getting trendy to get some of the subject sharp and blur out everything else. This is, refreshingly, the opposite. Will it therefore become trendy too? Of course not but the thought made me smile.
2) The predictable whiners in the comments below had me expecting a real mess. Actually, the in-camera processing isn't perfect but it's a long way from a disaster.
3) The promise of improved dynamic range isn't seen in that photo. Very blown highlights. I'd love to see how much can get recovered when raw files are supported (also how that affects point 2)!).

It might just be the default camera settings though. I've been amazed, over the last four or five years, how a few of my cameras throw away the highlights like they don't matter, yet the information is there if you process the raw files.

Of course it could be because one channel will clip early and, using all colours at all pixel locations, that clipping gets universally applied. I don't know, just speculating fairly wildly!

Link | Posted on Nov 12, 2017 at 00:33 UTC

I just clicked on the market pic. Three observations:

1) It's been getting trendy to get some of the subject sharp and blur out everything else. This is, refreshingly, the opposite. Will it therefore become trendy too? Of course not but the thought made me smile.
2) The predictable whiners in the comments below had me expecting a real mess. Actually, the in-camera processing isn't perfect but it's a long way from a disaster.
3) The promise of improved dynamic range isn't seen in that photo. Very blown highlights. I'd love to see how much can get recovered when raw files are supported (also how that affects point 2)!).

Link | Posted on Nov 11, 2017 at 19:49 UTC as 57th comment | 3 replies

Great choice!

Link | Posted on Nov 10, 2017 at 20:05 UTC as 103rd comment
On article Panasonic Lumix DC-G9 first impressions (402 comments in total)
In reply to:

(unknown member): Well, WHY? Like recent Fuji and Sony E offerings, it is very expensive relative to new full-frame competition (not just APS-C) because its lenses are extremely expensive relative to sensor size, and I frankly see not enough of an improvement over my fantastic FZ1000 to bother!! The 1600iso close photo of the two dogs shows the limit of the M4/3rds systems. This is 2017 state of the art imaging, and it is worse still than APS-C was a long time ago. Lens cost for a poor result like this is prohibitive. NO pros use m4/3 any more: its why Olympus killed off their original 4/3 pro system. They couldn't fool all of the people ALL of the time!!

I have an EM1 and an FZ1000. The FZ1000 has great image quality but the EM1 is very noticeably better.

Link | Posted on Nov 9, 2017 at 19:04 UTC
On article Sigma's new 16mm F1.4 will cost $450, ships this month (359 comments in total)
In reply to:

WillWeaverRVA: That's a steal. Should be a no-brainer for Sony E-mount users looking for a good wide angle prime. There's probably better options for M43 since you lose the wide angle advantage here but this is still not bad at all.

There are a handful of m4/3 lenses in the 14,15 and 17mm range, so this sits right in there. I don't think any are this heavy though!

Link | Posted on Nov 9, 2017 at 18:46 UTC
On article Panasonic Lumix G9: What you need to know (245 comments in total)
In reply to:

justmeMN: I didn't know that Panasonic made DSLRs. :-)

I wouldn't know. I only use it at 100. With modern raw converters, it's actually quite special when you do that.

Link | Posted on Nov 9, 2017 at 09:27 UTC
On article Panasonic Lumix G9: What you need to know (245 comments in total)
In reply to:

DingieM: Can they please remove the LCD plate???

One of the best features?? Seriously, why remove such a useful item?

Link | Posted on Nov 9, 2017 at 04:56 UTC
Total: 1270, showing: 61 – 80
« First‹ Previous23456Next ›Last »