Peter Cockerell

Joined on Mar 7, 2012

Comments

Total: 33, showing: 1 – 20
« First‹ Previous12Next ›Last »
In reply to:

Suntan: After a couple of years of people rioting in the streets calling for dead cops, as well as numerous occurrences of cops being gunned down at random, out of no-where, across the country; now cops are quite jumpy.

Imagine that.

There's a difference between being "quite jumpy" and opening fire with zero warning (and presumably zero confirmation that there actually was a threat, seeing as there wasn't). Do these cops receive any screening/training at all? And, BTW, the number of cops killed by firearms in 2016 was 64. 64 too many, perhaps, but hardly an epidemic.

Link | Posted on Sep 5, 2017 at 19:00 UTC
In reply to:

Bambi24: "police are trigger happy"

You'd be too if every lunatic walked around with a gun, able to take your life in a split second.

Don't point cameras at cops in the middle of the night -> you won't get shot.

Why is there no "Unlike" button here?

Link | Posted on Sep 5, 2017 at 18:55 UTC
In reply to:

Okapi001: It seems the Deputy is in fact a Wesen. That would explain everything;-)

If only he'd woged before he opened fire, the photographer would have stood a chance.

Link | Posted on Sep 5, 2017 at 18:53 UTC
In reply to:

Ivan Paganacci: What are the odds that the sun and the moon appear the same size in the sky? That's what I wonder whenever there's an eclipse. Maybe I should just take a picture instead.

It's just the time we happen to occupy in history. Previously the moon appeared larger than the sun, and before long (in cosmic timescales) it will be too far away to fully eclipse the sun. Just another reason to be thankful to be alive at this point in the life of universe. Beats heat death any day.

Link | Posted on Jul 8, 2017 at 01:51 UTC
In reply to:

Cary Knoop: From the article: "4:2:2 10-bit codec is 64x as much data as before"

No it isn't!

10 bit is 4 times as much data as 8 bit and 4:2:2 is 1 1/3 times as much data as 4:2:0, so it is 5 1/3 times as much not 64 times as much.

J A C S you are definitely confusing the phrase "amount of data" with "number of representable values". A ten-bit value is able to represent four times the number of values of an eight-bit one, but no-one (other than you) would claim it amounts to four times as much "data". It takes 25% more storage, requires 25% more time to transfer (serially, all things being equal) etc. Maybe the problem is more one of language than math, but you're definitely using very non-standard terminology.

Link | Posted on May 9, 2017 at 06:49 UTC
In reply to:

Zdman: You forgot the main advantage which is lens design. To achieve the same depth of field and light gathering you can design a lens one stop higher than you would for full frame. So instead of making design trade offs with a F2 design you can go to F2.8 with less spherical aberration and chromatic abberations etc. Go to large format and you can design crazy good F5.6 lenses and still get the depth of field of a full frame at f1.4. Why do you think large format hasn't got lenses much faster than F5.6? Because they don't need it (the DOF would be too small) and the results at F5,6 are spectacular. Its not just about the silicon its the lenses. Extrapolate the MF F2.8 (which would probably only need 6 elements) and light gathering to M43 and you'd need an F1 lens which is just not feasible to design.

" Go to large format and you can design crazy good F5.6 lenses and still get the depth of field of a full frame at f1.4."

Only if the "large" format camera had a crop factor of 0.25x and a sensor area 16 times that of a full-time camera. Which model is that, and how much does it cost?

Link | Posted on Mar 23, 2017 at 17:02 UTC
In reply to:

Greg Gebhardt: Kinda shows how China is getting way ahead of us. All the wonders in the world are no longer here.

Only the most parochial-minded of Westerners would ever have believed all the wonders of the world were "here".

Link | Posted on Feb 25, 2017 at 17:34 UTC
On article Sigma Announces 100-400mm F5-6.3 DG OS HSM lens (154 comments in total)
In reply to:

janist74: Sigma is very clever.

They address the "size disadvantage" of FF/APS-C cameras (instead of Canon/Nikon). This lens is barely bigger than the Pana 100-400 MFT lens ( a tad slower and heavier though). If the IQ is good enough for an APS-C 24mP camera, then it will sell very good.

ps.: my only reservation: if the Pana 100-400mm has quite strong vignetting with a filter thread of 72mm then what this lens can do with 67mm...

Yep! ! was using 1.5-1.6 as the APS-C:MFT factor instead of the correct (2 over that number), i.e. 1.25-1.33, so my MPixel counts for APS-C should have been 20-21.3M pixel. So as you say, current APS-C sensors do have a slight pixel density advantage over older MFT ones, but not, I think, over a 20.3M pixel GH-5

Link | Posted on Feb 22, 2017 at 02:26 UTC
On article Sigma Announces 100-400mm F5-6.3 DG OS HSM lens (154 comments in total)
In reply to:

janist74: Sigma is very clever.

They address the "size disadvantage" of FF/APS-C cameras (instead of Canon/Nikon). This lens is barely bigger than the Pana 100-400 MFT lens ( a tad slower and heavier though). If the IQ is good enough for an APS-C 24mP camera, then it will sell very good.

ps.: my only reservation: if the Pana 100-400mm has quite strong vignetting with a filter thread of 72mm then what this lens can do with 67mm...

+NetMage No, because to confer the kind of digital zoom advantage that you imply, the larger sensor camera would need a greater pixel density, not just greater pixel count. Even for a trailing-edge 16M pixel MFT sensor, an APS-C sensor would need to exceed 36-41M pixels and a full-frame would need to exceed 64M pixels. I don't know of any available cameras that meet those requirements (and of course the numbers are even greater for a newer 20M pixel MFT sensor).

Link | Posted on Feb 21, 2017 at 15:47 UTC
In reply to:

Jefftan: These adapter won't work with EF-S lens as far as I know
They are for full frame EF lens

any EF-S to E adapter for Sony APS-C camera

Thanks

Though of course you wouldn't want to use a speed booster adapter with an ef-s lens on an aps-c body, or you'll get severe vignetting.

Link | Posted on Jan 25, 2017 at 18:36 UTC
In reply to:

Paul_B Midlands UK: Clearly Nikons in the light not sound business. This video is a lot more enjoyable with sound on mute. Some great and poerrful imagery, Nikon is a global brand everyone knows the name. I've never owned one yet.

Absolutely. The voice-over, music and lyrics were all execrable. OK visuals, though.

Link | Posted on Jan 11, 2017 at 09:59 UTC
In reply to:

Peter Cockerell: What assurance is there that these Chinese-manufactured phones aren't sending your Google contacts and captured passwords to a site in China?

I was thinking of this story specifically, for those who haven't seen it:

http://mobile.nytimes.com/2016/11/16/us/politics/china-phones-software-security.html

Plus Huawei (manufacturer if the Nexus 6P) routers, of course:

http://www.techeye.net/business/huawei-products-do-have-backdoors

Personally I'd rather take my chances with the NSA than the Chinese equivalent under Xi Jinping.

Link | Posted on Dec 2, 2016 at 03:39 UTC

What assurance is there that these Chinese-manufactured phones aren't sending your Google contacts and captured passwords to a site in China?

Link | Posted on Dec 1, 2016 at 00:38 UTC as 9th comment | 6 replies
In reply to:

Peter Cockerell: For the 65mm "macro", where are they measuring the minimum focus distance from? If it's the focal plane, as is usual, then the magnification will be 0.43 (1:2.3), far short of a true 1:1 macro. That's actually the best case. If they're measuring it from, say an effective lens plane about halfway through the lens, the magnification will be around close to 0.31 (1:3.2). To be a true macro lens, the minimum focal distance would need to be 0.26m or less (4x the focal length). I always thought that if a lens mentioned "macro" in its name, then it's a true macro lens.

+Henning W, thanks for the clarification. The only macro lens I've owned is a Tamron which is specifically 1:1.

Link | Posted on Oct 10, 2016 at 16:14 UTC

For the 65mm "macro", where are they measuring the minimum focus distance from? If it's the focal plane, as is usual, then the magnification will be 0.43 (1:2.3), far short of a true 1:1 macro. That's actually the best case. If they're measuring it from, say an effective lens plane about halfway through the lens, the magnification will be around close to 0.31 (1:3.2). To be a true macro lens, the minimum focal distance would need to be 0.26m or less (4x the focal length). I always thought that if a lens mentioned "macro" in its name, then it's a true macro lens.

Link | Posted on Oct 6, 2016 at 18:36 UTC as 5th comment | 3 replies
On article Throwback Thursday: Canon PowerShot Pro70 (111 comments in total)
In reply to:

trenzterra: As a Singaporean, I found the photos pretty familiar, then I realised: cool! The sample photos were taken in Singapore.

Was Phil based in Singapore back then?

Interesting, I guess just because a camera produces subjectively appealing shots, that doesn't mean the colors are necessarily accurate!

Link | Posted on Sep 17, 2016 at 02:02 UTC
On article Throwback Thursday: Canon PowerShot Pro70 (111 comments in total)
In reply to:

trenzterra: As a Singaporean, I found the photos pretty familiar, then I realised: cool! The sample photos were taken in Singapore.

Was Phil based in Singapore back then?

Yes, Phil lived in Singapore for several years at the start of dpreview. The sample photos often featured his lovely Singaporean wife, but Phil was never shown. I think I only ever saw one photo of him, where his face was obscured by the camera he was holding!

I still remember his review of the Pro70, which I definitely coveted at the time. Even then, though, the resolution was lower than the cheaper competition (I ended up getting a Casio!) But there's no denying how nice the colors are, and I often wonder why Canon didn't persevere with that alternative to the Bayer mask.

Link | Posted on Sep 16, 2016 at 03:34 UTC
In reply to:

piratejabez: The full PDF has some great examples: http://arxiv.org/pdf/1608.05148v1.pdf
I'm going to extract the images from it for better comparison :)

I wonder what compression was used for the images in the PDF. Lossless, hopefully...

Link | Posted on Aug 25, 2016 at 01:28 UTC
On article Bentley creates a 53 billion pixel car commercial (189 comments in total)

My god, what an utter load of billocks. How does dpreview fall for this kind of crap?

Link | Posted on Jul 21, 2016 at 22:31 UTC as 27th comment
Total: 33, showing: 1 – 20
« First‹ Previous12Next ›Last »