martian1

Joined on Jul 11, 2007

Comments

Total: 52, showing: 1 – 20
« First‹ Previous123Next ›Last »
In reply to:

Bananas in Paris: I am a longtime LR user and am using the last copy of the Standalone Edition Version 6 without any monthly fees. Eventually I will have to give in and pay monthly for LR software that I already bought..... unless:

But, there seem to be Open Source solutions out there, which are mentioned in this article.

Having many photos with their editing history available anytime in LR, I am hesitant to ditch LR just because I may want to do some more adjustments based on this history in the future.

Question: Is there any LR alternative out there which will not charge a monthly fee, and will be able to migrate my entire LR catalogue with editing history included?

Basically looking for the smoothest transition possible from the LR pay by subscription model to either free or just one time fee model..

Currently all of my LR edits have been made to Canon, Olympus and Fuji files.
I would like to have the edit history and full migration of the LR catalogue into the new Raw Editor.

Thanks
Andrzej

@Ranger 9
Please not that 'Bananas in Paris' clearly mentions that he still uses LR6, which is many more than 2 years old.

Consequently, your 'nice' suggestion simply does not work for him.

And it also does not work for any others who do not 'version upgrade' every 2 years.

Link | Posted on Oct 24, 2022 at 22:32 UTC
In reply to:

Dan Monroe: is there an optical lens that captures more light than what's coming into the front of the lens?

@Karroly
I understand the comment of Cax: 'Tele lenses throw away most of the light that hits them' in the way that a tele lens blocks any incoming light that hits the lens outside of the lens' angle of view, which is most of the light for a tele lens.

Therefore, Cax' statement is actually correct.

Taking into account that the blocked light is outside the angle of view and therefore not useful for the lens' design intent, your explanation why a tele lens needs a bigger aperture to compensate its magnification is also correct.

Link | Posted on Sep 21, 2022 at 22:02 UTC

Yasuo Baba/Sony states: 'appropriate adaptations for each industry need to be made', while at the time of Sony's announcement there is simply no such adaptation available or announced.

Additionally, at least for now this seems unusable as a customer does not (yet or ever) have any 'customer’s own certificate server during examination.'
- Is a user required to run and maintain his own certificate server?

I acknowledge that Sony makes a start in this field, however I fear they might be left largely alone, and after a while the system could die before getting alive.

Link | Posted on Aug 8, 2022 at 23:22 UTC as 27th comment | 1 reply
In reply to:

lnsmr: Would be interesting to read more about why NASA designed it with 68GB storage. At $10Bn with all the top scientists, hardware costs was pretty much a non-issue.

@JeffinVancouver

Of course, I do not know for certain that there is not a backup SSD, or two or three.
However, DPR's article 'only' mentions presence of a single SSD of 68GB size, no mentioning or hint at any backup SSD at all.

This simply leaves the question open, whether NASA has actually installed any backup or not. I do not see this as a shortcoming of NASA when assuming they actually installed a backup. But if there is no backup SSD installed, then in my view this would be an actual shortcoming.

Do you know whether there is a backup SSD, or two or three?

Link | Posted on Jul 19, 2022 at 22:58 UTC
In reply to:

lnsmr: Would be interesting to read more about why NASA designed it with 68GB storage. At $10Bn with all the top scientists, hardware costs was pretty much a non-issue.

A 2nd SSD of the same type and size is not adding any development cost, also no additional development time.

Therefore it would have been 'cheap' to add a backup copy of the SSD.

To me it sounds strange to omit a backup of a mission-critical component, which does not add significant weight/size, especially taking into account that the only installed component is expected to substantially deteriorate during the planned mission time.

Link | Posted on Jul 18, 2022 at 22:30 UTC
In reply to:

papajohnny: "The mobile imaging sensor features the industry's smallest pixels, at 0.56-micrometer (μm), reducing the overall image sensor and camera module size."

That doesn't sound like an achievement. 5600A is in the visible range, the photons will be spread over multiple pixels (uneducated people need not to reply), resolution is limited by diffraction, not by the number of pixels. The noise is determined by the sensor size, which they reduced.

I fully agree to your statement about the significant diffraction limitation of such a small pixel pitch and corresponding reduction in resolution.

Samsung's claim about better cross-type phase detection autofocus can be real and needs to be demonstrated in real-world application.

The claims about extended dynamic range appear exaggerated as essentially no low light capability is available at more than the standard 12.5 MP (combining 4x4 neighbouring pixels to yield a pixel pitch of 2.24μm), or would require substantially longer exposure times which introduce blur by motion of the subject and/or camera.

Additionally, note that combining 4x4 neighbouring pixels for a pitch of 2.24μm ('tetra' mode) provides less pixel area and therefore light gathering capability than 1 pixel of 2.24μm, as there is a non-zero gap between the pixels.

Link | Posted on Jun 24, 2022 at 22:06 UTC
In reply to:

TonyPM: Now let's see.

This sensor is bigger than the one in the Nikon P1000 and similar cameras.
This sensor has better af capabilities, possibly better low light performance, and more mpx to boot.
It will even allow a high-res mode.

Question is, will it make a better super zoom?
Why aren't they making it yet?

Actually, the Sony RX10 IV has an optical 24-600mm zoom (35mm equivalent) - while this might sound crazy, I own this camera and can confirm it provides quite good resolution across the entire zoom range and edge to edge (left-right and top-bottom, not quite corner to corner) at 20 MP using a 1" stacked CMOS sensor.

Link | Posted on Jun 24, 2022 at 21:16 UTC

Generally interesting data and thanks for publishing it!
Just a question on the text for SSD drives in 2021:
'Backblaze uses only 20 Crucial drives'
Backblaze's table actually shows a drive count of 80 Crucial SSDs (all the same model), not 20.
Which number is correct?

Link | Posted on Mar 4, 2022 at 21:55 UTC as 31st comment
In reply to:

Analog4Ever: Canon allows profiles of 4 lenses to be stored in camera for jpegs.

I use DPP4 to process RAW files. You can apply a recipe and Batch process, save to TIFF with the lens correction baked in. I usually like include an Auto Gamma along with that.

If you're just doing a lens review displaying an uncorrected version is fine as long as longs as it can be compared to a corrected version.

According to the EOS Utility 3.13 manual, Canon R5, R6, R, RP, 1D X III, 1D X II, 5D IV, 90D, Rebel T8i, SL3, 850D, 250D, 200D II, M6 II, M200, M50 II, M50 have correction data and data for the Digital Lens Optimizer (DLO) of the complete lens lineup pre-stored at time of the camera's release.
Other cameras compatible with EOS Utility may have only some lens data pre-stored, however normally can store much more than data of 4 lenses.

Link | Posted on Jan 4, 2022 at 21:27 UTC
In reply to:

martian1: The announcement includes a new firmware version 1.5.0 for the EOS R5, likely also in late Dec. 2021 or early 2022.

I wonder what other improvements Canon will add in this firmware - for me a high resolution mode using sensor shift and multiple exposures in rapid succession would be useful, e.g.:

A) 2 exposures, the 2nd exposure with the sensor shifted by 1/2 pixel in x and 1/2 pixel in y direction (for a 4x larger image with interpolation for missing pixel positions)

B) 4 exposures with the sensor shifted by 1/2 pixel in x, then 1/2 pixel in y, then 1/2 pixel in x and 1/2 pixel in y direction (for a 4x larger image, with sampling of each pixel).

In an easy implementation this could just rely on high speed capture (20 fps with electronic shutter / 12 fps with mechanical shutter) to minimise motion artifacts, or in a more sophisticated version apply detection of movement between frames and corresponding removal of artifacts, e.g. by local interpolation using unaffected frames.

Dear kb2zuz,

Thanks for your comments to which I generally agree.

However, a 1/2 pixel shift in x and y direction (above Case A) can substantially increase image resolution as the Bayer pattern which includes 1/2 of the resolution-important green pixels, 1/4 red and 1/4 blue pixels essentially provides a doubled spatial (apparent) resolution.

Waiting for 4 images instead of 2 images doubles the time and consequently doubles motion artifacts, which can occur even using a tripod, e.g. due to movement of leaves or grass in a mostly stationary landscape scene, and for any even slowly moving subject like people or an animal.

Some 3-CCD / 3-CMOS video cameras (e.g. JVC GZ-HD3) use a 1/2 pixel shift in x and y direction of the green channel thereby substantially increasing resolution as primarily green defines spatial resolution. Of course, colour accuracy of such a system producing 3 separate full images in green, red and blue is much better than a single sensor with a Bayer filter.

Link | Posted on Oct 13, 2021 at 22:02 UTC

The announcement includes a new firmware version 1.5.0 for the EOS R5, likely also in late Dec. 2021 or early 2022.

I wonder what other improvements Canon will add in this firmware - for me a high resolution mode using sensor shift and multiple exposures in rapid succession would be useful, e.g.:

A) 2 exposures, the 2nd exposure with the sensor shifted by 1/2 pixel in x and 1/2 pixel in y direction (for a 4x larger image with interpolation for missing pixel positions)

B) 4 exposures with the sensor shifted by 1/2 pixel in x, then 1/2 pixel in y, then 1/2 pixel in x and 1/2 pixel in y direction (for a 4x larger image, with sampling of each pixel).

In an easy implementation this could just rely on high speed capture (20 fps with electronic shutter / 12 fps with mechanical shutter) to minimise motion artifacts, or in a more sophisticated version apply detection of movement between frames and corresponding removal of artifacts, e.g. by local interpolation using unaffected frames.

Link | Posted on Oct 6, 2021 at 23:08 UTC as 32nd comment | 3 replies
In reply to:

whitelens: Like me, they got sick and tired of waiting for another Canon R3 release tease.

@MikeRan
Just an excerpt from the A9 II camera review on this site (1 Introdution):

'One of the most significant of updates is a new mechanical shutter mechanism that allows for 10 fps bursts: this is significant because if you find yourself in a situation where you can't use the electronic shutter for risk of banding or other artifacts, the older model could only muster 5 fps which is a bit uninspiring on a camera meant to specialize in sports and action.'

Actually, also the A9 II (and original A9) can show significant banding or other artifacts with electronic shutter - the banding may be less likely for a faster readout speed, but remains a significant risk.

As stated in the A1 camera review on this site (2 What's new ...) even its extremely fast readout speed of 3.84ms does not always ensure absence of banding, although it becomes so unlikely that it is fine for almost any situation.

May be you should also do some more research?

Link | Posted on Aug 26, 2021 at 07:19 UTC
In reply to:

whitelens: Like me, they got sick and tired of waiting for another Canon R3 release tease.

@MikeRan
Thanks for your facts on readout speeds in electronic shutter, which substantiate that the A1 at 3.84ms surpasses the A9 II at 6.25ms (almost x2).

Please note that this proves your claim:

'The A9II is in a different class compared to any other full frame camera on the market with respect to stills performance in electronic shutter.'

... false.

It appears odd that you suggest that I need to become better informed, when your above statement is false ?!

I do not want to enter into a ping-pong game with you arguing that only you or only me is right.

I agree that my statement on video read-out rates is not directly transferable to stills, while the R5 read-out of 15.4ms in 8K/30p (35.4MP) scaled to 24MP gives 10.4ms which is in the same order as 6.25ms of A9 II in stills (similarly 11.8-16.7ms 45MP stills scale to 8.0-11.3ms).

For mostly static interviews/conferences when the silent electronic shutter is preferred both, R5 and A9 II, provide sufficiently low readout speeds.

Link | Posted on Aug 25, 2021 at 22:59 UTC
In reply to:

whitelens: Like me, they got sick and tired of waiting for another Canon R3 release tease.

@MikeRan
I have substantiated my statements about R5 vs. A9 II speed with facts as published by Sony and Canon (both achieve the same 20 fps in electronic = silent shutter mode) and actual measurements about sensor read-out speeds (as published on this site).

In contrast, you simply state a claim without any substantiation:

'The A9II is in a different class compared to any other full frame camera on the market with respect to stills performance in electronic shutter.'

Actually, Sony's A1 is on the market for several months with 30 fps electronic shutter and clearly surpasses the A9 II with 20 fps while it additionally provides much higher resolution at 50 MP (similar to the R5's 45 MP), which can well be used for cropping.

Apparently your claim is simply false.

Coming back to the R5, with mechanical shutter the R5 achieves 12 fps, while the A9 II achieves 10 fps, so actually less when artificial lighting may cause banding or other artifacts with the electronic shutter.

Link | Posted on Aug 24, 2021 at 23:08 UTC
In reply to:

whitelens: Like me, they got sick and tired of waiting for another Canon R3 release tease.

Just a note that Canon's R5, which is on sale for almost a year now, offers completely silent shutter (electronic shutter) with 20fps, i.e. the same as Sony's A9 Mark II.
Also note that the sensor's read-out speeds are 2x faster for Canon's R5 than Sony's A9 II at 4K and similar at 1080 (important for rolling shutter), see below.

Canon's R5 read-out rates (45 MPixels):

9.7ms for 4K/30p, 4K/60p, 4K/120p
8.7ms for 1080/60p
15.4ms for 8K/30p, oversampled 4K (from full sensor)

Sony's A9 II read-out rates (24 MPixels):

23.2ms for 4K/24p
17.9ms for 4K/30p (1.2x crop)
6.3ms for 1080/60p

This shows that Canon does not need the new R3 to achieve (fps) and clearly surpass (read-out) the speed of Sony's A9 II, and this even with 45 MP vs. 24 MP.

Link | Posted on Aug 23, 2021 at 21:36 UTC
In reply to:

Jeff Greenberg: -----
> " 'three-way gravitational tug-of-war between interacting galaxies"

there's a 3 way tug of war between Sony Canon Nikon & as
with galaxies its easy to know which is irrefutably swallowing the others

As Weia pointed out above:

'In fact they mostly pass through each other, with some gravitational effects but hardly any star to star collision. So empty a galaxy is.'

However, in this case to me there are significant gravitational effects visible in the image captured by Hubble, e.g. several of the galaxy arms appear to extend and likely move farther out of their parent galaxies compared to single galaxies. Possibly the outmost portions of such arms will leave their parent galaxies in the future.

Link | Posted on Aug 5, 2021 at 20:21 UTC
In reply to:

MinAZ: Hmm, no update for the R5?

@Stig Nygaard:
Actually on the R5 the user can already select dual-recording (at least with firmware 1.3.1/1.30), i.e. RAW to Card 1 (CFexpress) and MP4 to Card 2 (SDXC).

Link | Posted on Jul 9, 2021 at 23:30 UTC
On article Adobe releases Photoshop for Windows ARM64 devices (142 comments in total)
In reply to:

Lessiter: Quite strange that it's such an effort porting windows software to different architectures. Basically all Linux applications can be crosscompiled and are made available for lots of cpu architectures.

I assume Photoshop uses assembler language (which is close to machine code) to speed up specific, time-consuming functions, such code cannot be ported via cross-compiling and instead needs to be re-programmed in ARM code.

Link | Posted on May 18, 2021 at 16:18 UTC
On article Hubble captures star ‘on the edge of destruction’ (155 comments in total)

'According to NASA the STAR was created from giant eruptions that occurred 10,000 years ago, and which blew out the star’s outer layers.'
- I assume this should actually state:
'According to NASA the CLOUD/NEBULA was created from giant eruptions that occurred 10,000 years ago, and which blew out the star’s outer layers.'

Link | Posted on Apr 29, 2021 at 23:59 UTC as 15th comment
In reply to:

Tazz93: Why leave out the only other sensor that was created with in the last 3 years?

Why not compare to Canon's R5 ?
- similar high resolution (50 vs. 45 MPixel is 5.4% increase in linear resolution)
- 2nd fast at 20fps electronic shutter and faster at 12fps mechanical shutter
- similar dynamic ranges (actually also two ranges similar to Sony's A1)

Link | Posted on Mar 13, 2021 at 10:51 UTC
Total: 52, showing: 1 – 20
« First‹ Previous123Next ›Last »