zkz5

zkz5

Joined on Dec 1, 2011

Comments

Total: 230, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On article iOS 11 will cut photo, video size in half (81 comments in total)
In reply to:

bartjeej: Would HEIF also allow even lower compression (or atleast, less lossy compression) than current low-compressed jpegs, or more than 8 bits, at sensible file sizes?

"Willl the day of mass-market lossless images be a thing?"

It has been a thing for around 20 years now thanks to PNG.

Link | Posted on Jun 12, 2017 at 20:34 UTC
In reply to:

virtualreality: Creativity is no longer required for making interesting photographs. The camera will be smarter than you are.

For most people, creativity never had anything to do with it in the first place. They're trying to preserve a memory or communicate, not make a work of art. This is just a tool to get a job done.

Link | Posted on May 20, 2017 at 00:13 UTC
In reply to:

webber15: Who gives a ****!!

Around 166 million people, according to their IPO documents.

Link | Posted on May 17, 2017 at 10:06 UTC
In reply to:

piratejabez: Very cool solution—I like it!

"Sure I could take a screenshot, or piece the images together in PS"

The later could be done programmatically, and I wouldn't really say that taking a screenshot was all that much work in the first place.

Link | Posted on May 3, 2017 at 01:06 UTC
In reply to:

piratejabez: Very cool solution—I like it!

Why? It doesn't even remotely work. The comments here are filled with people successfully copying the sample image using a variety of methods.

Link | Posted on Apr 30, 2017 at 00:24 UTC
In reply to:

derfotograf: There is a free solution for the stealing problem: don't be lazy and modify your .htaccess file on your server.

Just do a google search for 'htaccess prevent file download'...

Nope, that doesn't work either. It is based on the Referer request header, which is trivial to spoof. Several methods for copying an image send a legitimate Referer header anyway. A total waste of time like every other technique used to prevent copying images off of the web.

Link | Posted on Apr 30, 2017 at 00:22 UTC
In reply to:

HowaboutRAW: Ah yes, "disorderly conduct" redefined to mean "not doing whatever a cop says", glad that garbage criminal prosecution was dropped.

Now the civil fine will just anger authoritarian cops/types even more.

"Try expressing an opinion at a university the "mob" doesn't agree with. Good luck..."

Milo yiannopoulos had been doing that frequently. I don't remember any instances of him being held in a choke hold, thrown to the ground and then arrested for "disorderly conduct". Do you?

Link | Posted on Mar 10, 2017 at 03:35 UTC
In reply to:

unbelievable: This review shows exactly why - for me - the MFT system makes the most sense as a mirrorless camera system. The penalty of (slightly) reduced image quality is compensated big time by a tremendous increase in portability over APS-C and FF.
(I fully understand that other antropomorphic entities have a different set of priorities)

Your post is incomplete. Please, continue your comparison for the following lenses:

Pany 35-100 f4/5.6 (50mm, 135g)
Oly 14-42 EZ (23mm, 90g)
Pany 20/1.7 (25mm, 100g)
Pany 15/1.7 (36mm, 115g)
Oly 45/1.8 (46mm, 115g)

Link | Posted on Feb 8, 2017 at 22:08 UTC
In reply to:

Romasito: I've been subscribing to Adobe's LR+PS plan since Nov 2013 paying €12.09/month.
Looks like so far I've already paid 39 * 12.09 = €471.51. It is a lot of money until you realise that maybe it is not thaat much for 3+ years of use.

I don't really use Photoshop albeit it is included. If Adobe could create a cheaper monthly subscription plan just for LR - I would happily take it.

If it increases additional £10/month - I will probably cancel the subscription, as I don't use the offering to the full extent (i.e. I don't need Photoshop) and will look for a better alternative (I hear Affinity Photo is a good one)

"€471.51. It is a lot of money until you realise that maybe it is not thaat much for 3+ years of use."

As opposed to CS6's infinite years of use for about the same price?

Link | Posted on Feb 3, 2017 at 10:42 UTC
On article Fujifilm X100F steps up to 24.3MP, adds AF joystick (34 comments in total)
In reply to:

Triplet Perar: This is what more honest companies deliver in the first version of the camera.

If the D300 was "bang, perfect", then why did Nikon issue 11 firmware updates for it? And if the 5D is "bang, perfect", why does it have as many hardware revisions as the X100?

Link | Posted on Jan 20, 2017 at 06:38 UTC
On article Fujifilm X100F steps up to 24.3MP, adds AF joystick (34 comments in total)
In reply to:

Triplet Perar: This is what more honest companies deliver in the first version of the camera.

What other company delivers a camera like this on *any* version?

Link | Posted on Jan 19, 2017 at 22:07 UTC
In reply to:

Suntan: Good in theory, horrible in use. Besides the overhead of encryption and its negative impact on burst modes, do you want to have to input a pin number to "unlock" your camera everytime you want to take a picture? Or would you be fine with only having to enter a pin when you want to review a taken picture?

Not to mention the negative impact on workflow. Who wants to run all their pictures through a weakly supported program, made by their camera manufacturer, to "decrypt" their images before opening them in pshop?

What does a 5 digit code have to do with anything? As several others pointed out, the right way to implement this is with public key cryptography, where they encryption key and decryption key are stored separately.

Nevertheless, you don't need a touch screen to enter 5 digits into a camera.

Link | Posted on Dec 16, 2016 at 01:31 UTC
In reply to:

VENTURE-STAR: What a load of idiotic crap. Clearly those writing on here have never been in a situation where someone extremely ugly, menacing and heavily armed orders you to hand over your equipment. Do you really want to accept the consequences of protecting a few images on a memory card? Speaking from personal experience, I promise you there are situations where you do exactly as you are told and you will give up whatever you are asked for.

"you will give up whatever you are asked for"

That's only if you have it, and the adversary knows that you have it.

You don't necessarily need to have the *decryption* key with you, only the encryption key. And if this is a standardized and widely implemented optional feature, the adversary might actually believe that you don't have the decryption key as well.

Link | Posted on Dec 15, 2016 at 23:26 UTC
In reply to:

Suntan: Good in theory, horrible in use. Besides the overhead of encryption and its negative impact on burst modes, do you want to have to input a pin number to "unlock" your camera everytime you want to take a picture? Or would you be fine with only having to enter a pin when you want to review a taken picture?

Not to mention the negative impact on workflow. Who wants to run all their pictures through a weakly supported program, made by their camera manufacturer, to "decrypt" their images before opening them in pshop?

"While they are adding all the options, they should probably add an AM/FM radio receiver."

You can implement an AM/FM radio receiver entirely in software? Please do share the code...

Link | Posted on Dec 15, 2016 at 23:20 UTC
In reply to:

EskeRahn: The only safe practise would be to upload the pictures through an encrypted link to somewhere physically out of reach, and to remove it from the recording device.

OR some slightly less safe variant that needs less bandwidth: An advanced version where the files are encrypted with a system generated key (UNKNOWN to the photographer), that are sent out of reach, so the files can not be unlocked before the photographer gets back home.

As several points out, a key known to the photographer could well lead to threats/torture to reveal the password, thus just escalating things...

ADD: A little more on the second idea. Imaging some small server-box with a GPS that receives the keys through the internet, but ONLY unlocks the keys when the box and the camera are physically close - say simple NFC. and only unlock at the position it got the keys. This way anyone really wanting the pictures would have to know that place and to go there with the camera and break in.

Way too complicated. The solution already exists: public key cryptography. See this post:

https://www.dpreview.com/news/4106017651/photojournalists-and-documentary-filmmakers-ask-camera-manufacturers-for-encryption?comment=5347519752

Link | Posted on Dec 15, 2016 at 23:16 UTC
In reply to:

ProfHankD: I think the big problem with this is that good encryption isn't computationally cheap enough for the relatively slow ARM32 processors used in most cameras. Using CHDK, ML, or OpenMemories, one could probably put code into a camera to do this, but it would almost certainly be as a slow in-camera postprocessing step. It would be pretty hard to convince companies to add encryption hardware....

Actually, a lot of Flash memory cards also have ARM32 cores that could be programmed to do it... but then the camera probably wouldn't be able to review the files once written.

"What you'd really want is a public key system where the camera could encrypt images so that ONCE ENCODED they could ONLY BE DECODED BY A PARTICULAR SEPARATE COMPUTER, and NOT by the camera that made them"

Best idea I've read here so far. For any of this to work it has to be well known that an adversary cannot coerce the key out of the photographer.

Link | Posted on Dec 15, 2016 at 07:16 UTC
In reply to:

whyamihere: I'm surprised camera data encryption isn't already a thing. We're talking a few KB of data added to each image & minimal processing power on the behalf of the camera to ensure that information remains secure and/or confidential. The media most cameras use - Secure Digital - supports encryption; a SD card's ROM could probably be modified to secure erase after too many password attempts.

With that said, encryption is only as good as its design and implementation. Companies get hacked all the time, and their cipher data could be stolen and distributed. Backdoors are made because some companies willingly hand the keys over to any government who politely asks. Heck, even the few bits of information passed back & forth between camera & the card could leak enough data for a crack. Nothing is foolproof, and encryption might just wind up being a mild inconvenience instead of a huge detriment.

Before anyone cracks wise: This is part of my job. I think about security in practical terms a lot.

"Secure Digital - supports encryption"

SD cards have a locking feature whereby the card will reject all read/write attempts until a password is given. That might be what you were thinking of. The data in that case is not *encrypted*, however - the controller in the card is simply rejecting requests to read it.

Extracting data from card locked in such a manner is probably very difficult but not beyond the ability of a government.

Link | Posted on Dec 15, 2016 at 07:08 UTC
In reply to:

scottcraig: Undoubtably there are some genuine concerns for the safety of persons such as photojournalists who need to protect sensitive images. However one should consider the negative impact it can have which can include a significant price increase in camera models for average photographer or professional who has no need to encrypt their images via camera. Instead I would urge such persons/journalists to take the necessary measures to download and encrypt their sensitive data using the tools already available. To make such a request from camera manufacturers is in my opinion reflects an unwillingness to take full responsibility for securing their own data. What we need to understand here is any act to protect our privacy or that of the persons we work with cannot be dependent on any one piece of hardware or software but rather by engaging in sound practices ourselves and simple common sense.

Just saying.

Why would this mean a significant price increase in the camera? The feature can be implemented entirely in firmware running on the existing camera hardware. The poster right after you pointed out that it has been implemented after the fact in the Samsung NX300 & 500.

Link | Posted on Dec 15, 2016 at 06:55 UTC
On photo Towering in the Your City - Dark and Gloomy challenge (2 comments in total)
In reply to:

thejohnnerparty: Very nicely done. It's got that industrial kind of look.

Thanks!

Link | Posted on Nov 23, 2016 at 03:43 UTC
In reply to:

maxnimo: I doubt that this lens can outsharp a Fujinon prime at F2.

"Well by trying most of them..."

Usually that isn't an objective test. It is good to hear an opinion based on actual *use*, at least.

That IR summary of the Lens Rentals test confirms this Leica is indeed outsanding: "the super-expensive Leica APO-Summicron bested every other lens in this test". I couldn't find anything like this - thanks.

Link | Posted on Nov 18, 2016 at 05:29 UTC
Total: 230, showing: 1 – 20
« First‹ Previous12345Next ›Last »