vincent__l

Lives in United States Los Angeles, CA, United States
Has a website at http://500px.com/uglycowboy
Joined on Apr 10, 2010

Comments

Total: 52, showing: 1 – 20
« First‹ Previous123Next ›Last »
On article Medium-format mirrorless: Hasselblad unveils X1D (1190 comments in total)

Margin still seems high considering the BOM. It's basically a sensor in a box. No mirror, prism, dedicated PDAF, or shutter. Very minimal hardware buttons and dials. Nobody will use this for JPG so they could even have removed the image processor. A $0.50 ARM processor with Linux for the UI and CDAF is all they need. Obviously the biggest cost is the low volume, lower yield, sensor.

Link | Posted on Jun 22, 2016 at 17:56 UTC as 123rd comment
On article Sony Alpha 7R II Review (2161 comments in total)

"When setting Auto ISO, users can pick a minimum and maximum ISO for the camera to work with .."

If this is correct how do you set a max ISO in auto ISO mode ?

Link | Posted on Sep 24, 2015 at 21:31 UTC as 407th comment | 2 replies
On article Sony Alpha 7R II Review (2161 comments in total)

The pages summarizing the customization choices were very helpful. These should have been included in the Sony user guide.

There are some erratas you may want to look at:

Introduction, (Lossy) Compressed Raw section: "Although, at lauch, the a7R II only offered ..." Typo in "launch"

Bodies & Design: "Users can also set the function of the 'AF/MF' and 'AEL' button as well as all four direction keys on the click wheel ..." The up direction (DISP) cannot be customized

Controls & Handling: "The RX100 IV offers much-improved Auto ISO control." I assume you meant to say A7RII

Link | Posted on Sep 24, 2015 at 17:45 UTC as 420th comment
On article What difference does it make? Sony uncompressed Raw (621 comments in total)
In reply to:

vincent__l: Richard,

Thanks for the early heads up on the uncompressed ARW progress. Do you know whether or not the new uncompressed ARW format will require RAW converter updates (Lightroom, ACR, C1, etc ?)

Richard,

Posterization, in smooth sky gradients, was the other issue with 11+7. Fine tonal variations require as many bits as you can get especially if you're going to push the files. If you're open to suggestions for samples that might be a good thing to test.

Link | Posted on Sep 23, 2015 at 22:53 UTC
On article What difference does it make? Sony uncompressed Raw (621 comments in total)

Richard,

Thanks for the early heads up on the uncompressed ARW progress. Do you know whether or not the new uncompressed ARW format will require RAW converter updates (Lightroom, ACR, C1, etc ?)

Link | Posted on Sep 23, 2015 at 19:38 UTC as 69th comment | 10 replies
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)
In reply to:

marc petzold: A7S Mark II comes with uncompressed RAW Option, A7R II via Firmware Update.

Proof: by the Phoblographer (A7S Mk. II Preview)

http://www.thephoblographer.com/2015/09/15/first-impressions-sony-a7s-mk-ii/

Thanks for sharing. Given that the A7RII will likely have the same implementation as the A7SII we can expect that uncompressed literally means uncompressed (not losslessly compressed). No need to waste any more time speculating.

"What we were also told is that the color depth and dynamic range also won’t be affected–and instead the difference in image quality will really just been see in the artifacts that happens in high contrast areas."

I guess this attitude is why Sony has been so reluctant to give us 14-bit RAW. Corner case artifiacts with 11+7 are indeed rare but that's not the main reason for getting away from compressed RAW. 14-bit RAW, with about 4x more latitude, is extremely useful when editing. After manipulating the file many people may export to an 8-bit JPG file and it looks just fine for viewing. However, anyone who has tried to manipulate an 8-bit JPG realizes the detriment of having less latitude in the source file.

Link | Posted on Sep 16, 2015 at 21:22 UTC
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)
In reply to:

vincent__l: Performing a lossless compression on an image file is something that has to be done at the end of the imaging pipeline as the image processing ASIC won't run on a compressed file. Losselessly compressing a file at the end of the chain should be dead simple to implement. It's no different than people zipping a text file before they email it. Just a simple post processing step. The A7 series is already stuck using UHS-I with one of the slowest bus speeds among its peers (write speed around 40MB/s). Unless Sony firmware engineers are completely incompetent I can't see any reason why they would dramatically slow down the write times with a larger file for no reason. Based on my faith that Sony engineers have at least some level of competence I will have to assume that uncompressed = lossless compression. Sony marketing probably uses the term "uncompressed" to mean that the processing doesn't discard data at the beginning of the pipeline. We'll see.

Everlast66, you bring up valid points about the additional decompression that would be required if the file is losslessly compressed. Perhaps leaving the file uncompressed is one way to facilitate the man hours required to update the code.

Link | Posted on Sep 16, 2015 at 02:08 UTC
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)
In reply to:

vincent__l: Performing a lossless compression on an image file is something that has to be done at the end of the imaging pipeline as the image processing ASIC won't run on a compressed file. Losselessly compressing a file at the end of the chain should be dead simple to implement. It's no different than people zipping a text file before they email it. Just a simple post processing step. The A7 series is already stuck using UHS-I with one of the slowest bus speeds among its peers (write speed around 40MB/s). Unless Sony firmware engineers are completely incompetent I can't see any reason why they would dramatically slow down the write times with a larger file for no reason. Based on my faith that Sony engineers have at least some level of competence I will have to assume that uncompressed = lossless compression. Sony marketing probably uses the term "uncompressed" to mean that the processing doesn't discard data at the beginning of the pipeline. We'll see.

chlamchowder, those are valid points. I've considered that a lack of hardware compression in the SoC might be one reason for the delay of 14 bit RAW. If the "BIONZ" SoC lacks hardware compression and optimized instruction set for lossless compression there would be a large penalty in both power and image pipeline latency. Given that 14 bit RAW is a late stage software hack it's likely the hardware isn't properly spec'ed. If the compression is done at the last stage the buffer will likely be inadequate. We'll have wait until Sony publishes the cons to their implementation but I expect there will be a large penalty. The good news is that Sony will likely implement this feature properly on the series 3 and the hardware pipeline will be designed for it.

Link | Posted on Sep 16, 2015 at 02:00 UTC
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)
In reply to:

vincent__l: Performing a lossless compression on an image file is something that has to be done at the end of the imaging pipeline as the image processing ASIC won't run on a compressed file. Losselessly compressing a file at the end of the chain should be dead simple to implement. It's no different than people zipping a text file before they email it. Just a simple post processing step. The A7 series is already stuck using UHS-I with one of the slowest bus speeds among its peers (write speed around 40MB/s). Unless Sony firmware engineers are completely incompetent I can't see any reason why they would dramatically slow down the write times with a larger file for no reason. Based on my faith that Sony engineers have at least some level of competence I will have to assume that uncompressed = lossless compression. Sony marketing probably uses the term "uncompressed" to mean that the processing doesn't discard data at the beginning of the pipeline. We'll see.

Marketing is responsible for creating a high level PRD but implementation details are up to engineering. A lossless compression post processing step seems like a no brainer. However, my inner cynic allows the probability of engineering actually not compressing the data at all just to give the middle finger to all the people harassing them for a crippled RAW file.

Link | Posted on Sep 16, 2015 at 00:13 UTC
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)

Performing a lossless compression on an image file is something that has to be done at the end of the imaging pipeline as the image processing ASIC won't run on a compressed file. Losselessly compressing a file at the end of the chain should be dead simple to implement. It's no different than people zipping a text file before they email it. Just a simple post processing step. The A7 series is already stuck using UHS-I with one of the slowest bus speeds among its peers (write speed around 40MB/s). Unless Sony firmware engineers are completely incompetent I can't see any reason why they would dramatically slow down the write times with a larger file for no reason. Based on my faith that Sony engineers have at least some level of competence I will have to assume that uncompressed = lossless compression. Sony marketing probably uses the term "uncompressed" to mean that the processing doesn't discard data at the beginning of the pipeline. We'll see.

Link | Posted on Sep 15, 2015 at 23:55 UTC as 56th comment | 7 replies
On article Sony brings uncompressed Raw to a7S II, a7R II and... (562 comments in total)

If they willing to concede speed / power for image quality I hope the firmware will also allow 14-bit in AF-C, bulb, bracketing, and silent shutter mode for the A7RII.

Link | Posted on Sep 15, 2015 at 16:29 UTC as 136th comment
On article Analysis: Sony a7R II and RX100 IV autofocus systems (752 comments in total)
In reply to:

Mister Joseph: I am so impressed by those videos, being an owner of a Nikon FF that can only AF on the center of the frame!

"Sure AF-fine tuning is required and takes time. But after it the lenses work great."

How's that working out for you on zoom lenses ?

Link | Posted on Jul 2, 2015 at 21:47 UTC
In reply to:

BobYIL: "Mirrorless" has been a great convenience for lens design. Most of the mirrorless lenses are still retrofocus design to some degree, however the lens designers do not need to take into account the mirror box anymore. So, the register distances are minimal, like rangefinder designs, more compact sizes, allowing also the use of a wide range of adapters for the majority of the brands on the market, new or old.. Better lenses for the mirrorless bodies will be realized, especially those shorter than 60mm, than those specifically designed for the DSLRs. (The 25/2 Batis, for instance, seems to be better than any available 24 or 25mm lens..)

a lens designer will optimize the features for the market that a lens is designed for so making a lens as small as possible is not always the top priority. That's why the FE 35mm f/1.4 is so large while the 35mm f/2.8 is much smaller.

I never said 40+mm is limited. I said that increased flange distances increases your limitation (ie less flexible). That's big difference in meaning. so really... try to read for comprehension.

Link | Posted on Jun 21, 2015 at 22:06 UTC
In reply to:

BobYIL: "Mirrorless" has been a great convenience for lens design. Most of the mirrorless lenses are still retrofocus design to some degree, however the lens designers do not need to take into account the mirror box anymore. So, the register distances are minimal, like rangefinder designs, more compact sizes, allowing also the use of a wide range of adapters for the majority of the brands on the market, new or old.. Better lenses for the mirrorless bodies will be realized, especially those shorter than 60mm, than those specifically designed for the DSLRs. (The 25/2 Batis, for instance, seems to be better than any available 24 or 25mm lens..)

A short flange distance (ie 18mm) increases lens design flexibility. It does not create a limitation. Lenses can be designed from 18mm up to infinity. To facilitate mental comprehension, designing a lens with a longer distance is no different than having a built in adapter. Lens designers are now able to leverage the shorter flange distance for focal length where it is an advantage. With a flange distance of 40+ mm you limit your flexibility. This isn't something to brag about as a selling point.

Link | Posted on Jun 21, 2015 at 20:48 UTC

Just noticed that the AF sensitivity (on paper) goes down to -2 EV. A7II is rated to -1 EV. Seems the increased sensitivity of the BSI sensor is helping the AF.

Link | Posted on Jun 10, 2015 at 19:14 UTC as 254th comment

Interesting engineering approach from Sony. Full frame sensors don't gain as much, for low light performance, from BSI compared to smaller sensors. However, the BSI sensor likely allowed Sony to use much wider (thus faster) metal routing for the pixel interconnect. This greatly improves the readout speed which is important for both video and liveview performance. The faster speed also allowed them to have silent shooting with such a large and high resolution sensor. I'm curious to see the impact on color accuracy due to the BSI approach as there may be some degradation. There may also be an impact to battery life and noise from the higher readout speed. I'm also curious as to what caused the additional weight vs the A7II.

Link | Posted on Jun 10, 2015 at 18:08 UTC as 303rd comment | 4 replies
On article Sony a5100 First Impressions Review (589 comments in total)

The A5100 has three significant improvements vs the A6000 - full sensor readout in video, XAVC codec, touch screen for focus point selection. The full sensor readout, in this 24MP APS-C sensor, is a pretty significant feature and seems to be ignored. The RX100 mk3 and RX10 have full pixel readout but they are small 1" sensors. The A7s can do it with a FF sensor but the resolution is only 12MP. Also, video in FF results in 33ms of rolling shutter The A5100 seems to be the first ILC that can do full pixel readout using a relatively large and high resolution sensor. I'm interested to see the rolling shutter effect in the A5100.

Link | Posted on Aug 18, 2014 at 07:29 UTC as 156th comment | 2 replies
On article Drone lighting could be coming soon to your studio (127 comments in total)

Let me know when it's possible to mount a strobe, battery pack, and large soft box on the drone. That would take "voice activated light stand" to a new level.

Link | Posted on Jul 17, 2014 at 08:02 UTC as 49th comment | 1 reply
On article Benchmark Performance: Nikon D810 review (1984 comments in total)

Does the D810 fix the interpolated crappy live view, when zoomed in, from the D800 ? This makes manual focusing with live view very difficult with the D800.

Link | Posted on Jun 26, 2014 at 04:57 UTC as 503rd comment | 1 reply

It's about time Nikon. The DPR forums are constantly saturated with people screaming for an f1/4 standard lens for close to $2k and you finally delivered. Good call on ignoring the very small minority of whiners who think an updated 24-70 f/2.8 is necessary.

Link | Posted on Oct 17, 2013 at 06:45 UTC as 45th comment | 4 replies
Total: 52, showing: 1 – 20
« First‹ Previous123Next ›Last »