mgrum

Lives in United Kingdom United Kingdom
Joined on Jan 14, 2009

Comments

Total: 364, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On article Sony FE 100mm F2.8 STF bokeh demystified (298 comments in total)
In reply to:

mgrum: I find it very interesting from a technical point of view and it's good for the company to have more unique (well almost, Fuji have their APD lens) products to attract people to the brand...

but I personally strongly dislike the bokeh it produces, especially the way highlights and point lightsources are rendered, in the examples above I much prefer the image on the right.

Oh well it saves me some money I guess!

@Marko2

No this lens does not have the same DOF as a regular f/2.8 lens - aperture diameter is not the only factor here, the way that the aperture darkens toward the edges also affects the DOF.

Imagine you had an f/2.8 lens with a filter next to the aperture that had a clear circle in the centre but outside of that cut out 99.99% of the light - it would clearly not give the same DOF as an f/2.8 lens...

Link | Posted on Feb 19, 2017 at 15:53 UTC
On article Sony FE 100mm F2.8 STF bokeh demystified (298 comments in total)
In reply to:

jonny1976: i agree is smooth but terribly fake..really. it's interesting. i was interested in buying the laowa for app effect...how does it compares to sony?

The Laowa lens has a fairly weak APD element compared to the Sony (it's f/2.0 and T/3.2, whereas the Sony is f/2.8 and T/5.6) this means the Laowa only has slightly smoother bokeh than a regular lens whereas the Sony is much much smoother.

Link | Posted on Feb 17, 2017 at 19:17 UTC
On article Sony FE 100mm F2.8 STF bokeh demystified (298 comments in total)
In reply to:

mgrum: I find it very interesting from a technical point of view and it's good for the company to have more unique (well almost, Fuji have their APD lens) products to attract people to the brand...

but I personally strongly dislike the bokeh it produces, especially the way highlights and point lightsources are rendered, in the examples above I much prefer the image on the right.

Oh well it saves me some money I guess!

@ tkbslc

That's ok! It's also why it's good to have choice. I can think of maybe a few cases where I'd want this effect, but most of the time I prefer there to be some texture to the bokeh.

Link | Posted on Feb 17, 2017 at 19:10 UTC
On article Sony FE 100mm F2.8 STF bokeh demystified (298 comments in total)
In reply to:

zeratulmrye: After seeing a lot of samples from this lens I have the feeling that the bokeh is so perfect that it starts to look fake (or a result of photoshop)......

"Perfect" is not the right word, I would have said "smooth" instead, but I agree - you're spending a lot of money to make your images look like you cut out the subject and applied a Gaussian blur to the background in Photoshop!

Link | Posted on Feb 17, 2017 at 16:11 UTC
On article Sony FE 100mm F2.8 STF bokeh demystified (298 comments in total)

I find it very interesting from a technical point of view and it's good for the company to have more unique (well almost, Fuji have their APD lens) products to attract people to the brand...

but I personally strongly dislike the bokeh it produces, especially the way highlights and point lightsources are rendered, in the examples above I much prefer the image on the right.

Oh well it saves me some money I guess!

Link | Posted on Feb 17, 2017 at 16:09 UTC as 80th comment | 16 replies
On article Canon debuts EOS M6 mirrorless with optional EVF (610 comments in total)
In reply to:

Donnie G: Mirrorless fanboys constantly bash Canon for not abandoning its long-established best-selling lineup of DSLRs in order to produce nothing but shiny new unestablished mirrorless replacements. So what does Canon do? They build both types of ILCs and sell them side by side. Now those same mirrorless fanboys are huffing and puffing and throwing hissy fits because fully half of all ILCs (DSLR and mirrorless) sold in 2016 were Canon.

But Canon isn't worthy, they say. Their specs are not as numerous and glittering as those of their competitors, they say. They don't innovate, they say. Yet Canon ILC sales and profits say that soccer moms love Canon's DSLR and mirrorless camera controls, Dual Pixel AF, and superb touchscreen tech. Who do you think Canon is listening to? 😎

Straw men constantly bash Canon for not abandoning its long-established best-selling lineup of DSLRs in order to produce nothing but shiny new unestablished mirrorless replacements.

So what does Canon do? They completely half-a** their mirrorless line (pretty much everyone else has 4K video and more than 7 lenses) and carry on churning out the same DSLRs with minimal innovation. Now the same non-existant straw men are huffing and puffing and throwing hissy fits because fully half of all ILCs (DSLR and mirrorless) sold in 2016 were Canon.

But Canon isn't worthy, they say. Their specs are not as numerous and glittering as those of their competitors, they say. They don't innovate, they say. Yet Canon DSLRs sales (which are good right now) are shrinking faster than mirrorless sales overall.

Who is Canon listening to? People who expect the world to stay exactly the same has it has been for the last 15 years.

Link | Posted on Feb 15, 2017 at 14:39 UTC
In reply to:

Favorable Exponynt: Development costs were sunk already. They just admitted they cannot compete with Sony and Panasonic in this segment.

That depends on whether they consider tooling as part of the "development costs", if so that's not a sunk cost and could be what makes the line unprofitable.

Link | Posted on Feb 14, 2017 at 10:06 UTC
On article Google AI adds detail to low-resolution images (145 comments in total)

It clearly works very well with faces, which are highly constrained in the way they appear in images, but what if you photo contains an 8x8 region which roughly matches the brightness pattern of a face and the algorithm decides to insert a high resolution face in the middle of a patch of dirt...

I've read a lot of papers that present very promising results on their own datasets but in the general case fall down, this seems like another example. Creating more data from less data relies heavily on assumptions, which if they are wrong will destroy the results.

Link | Posted on Feb 8, 2017 at 20:37 UTC as 67th comment | 1 reply
In reply to:

nerd2: "The optical design is said to minimize vignetting"

"bokeh produced by the FE 100mm F2.8 STF OSS GM, which it says is improved by the use of an apodization (APD) element"

"An apodisation element is simply a radial neutral density filter (that gets darker towards the edges) "

????

@Rambazamba

It's not a typo. There are two different features - vignetting is reduced by using a larger front element to reduce optical vignetting and cat eye bokeh.

The apodisation element is completely unrelated, it sits next to the diaphragm and turns the bokeh discs from flat circles to circles that get darker toward the edges, so the background blur is closer to gaussian blur (i.e. it is smoother).

Link | Posted on Feb 7, 2017 at 17:34 UTC
In reply to:

Mike99999: 100mm T/5.6 GM at $1,500 and 700 grams is outrageous. The 90/2.8 Macro is $900 and 600 grams and an optical marvel. I'm not sure why this lens was a priority.

The FE 85/1.8, on the other hand, comes at a good price, a good weight (370 grams), and supposedly an excellent dual-motor focusing system. I'm looking forward to see how the IQ compares to the Batis.

What is outrageous? The fact that a 100/2.8 is slightly heavier than a 90/2.8? Or the price? Do lens prices have to be determined solely by maximum aperture or is it possibly that the image quality justifies the price tag?

Link | Posted on Feb 7, 2017 at 17:20 UTC
On article Sony SLT a99 II Review (1444 comments in total)
In reply to:

Hellraiser: tip to Sony - make a proper DSLR with A mount and stop beating this dead horse (SLT)

@zeratulmrye

The Canon 1DX can move its mirror at 14fps...

Link | Posted on Jan 31, 2017 at 15:57 UTC
On article Sony SLT a99 II Review (1444 comments in total)
In reply to:

HFLM: "Here, we found the a99 II to be about 1/3 EV behind the a7R II, closer in line with the expected 1/2 EV noise cost expected and reported by DXO. All in all, the mirror's impact on image quality is remarkably small." Since the mirror is removable, the effect of the mirror could be measured and determined more accurately. The smaller deviation compared to the DXO value could be due to better or adjusted NR algorithms or a different sensor "tuning". Looking at Bill Claffs DXO derived data, it looks like the high ISO part was improved at the expense of the base ISO performance.

@cresulka

I'd guess they had to make a tradeoff in the ADCs to support 12fps readout from the sensor and that results in increased read noise and hence lower DR at base ISO.

The DR of the A7RII drops when shooting at 5fps for this reason.

Link | Posted on Jan 31, 2017 at 13:05 UTC
On article Sony SLT a99 II Review (1444 comments in total)
In reply to:

Eloise: Now I may be missing the point ... but doesn't this camera offer the worst of all worlds. It doesn't have a optical view finder, yet has the translucent mirror in the way and taking up room - or is the fact that they still use the mirror for the focusing sensor an admission that mirrorless focusing still isn't as good?

Is this camera just for existing A-mount users? Wouldn't anyone buying into a new system do better with Sony FE-mount?

@mosc

You don't get 12fps shooting or a big battery with the A7RII

Link | Posted on Jan 31, 2017 at 13:01 UTC
On article Sony SLT a99 II Review (1444 comments in total)
In reply to:

Hellraiser: tip to Sony - make a proper DSLR with A mount and stop beating this dead horse (SLT)

They'll never compete with Canon and Nikon by offering the same products, they have to offer people something different to encourage them to switch (EVF, IBIS, high framerates).

Link | Posted on Jan 31, 2017 at 12:57 UTC
In reply to:

Prairie Pal: Those aren't girders laying on the floor. What we're looking at are storage racking and black storage boxes that fell from the walls. The exaggerated depressing lighting in the photo gives the impression of catastrophe. Tragic loss of lives aside, there really wasn't as much to overcome as Sony needs us to believe.

Erm you don't think there could be damage to other parts of the factory that haven't released images of, perhaps?

Link | Posted on Jan 30, 2017 at 13:15 UTC
In reply to:

eyeport: The biggest disappointment:::: Flash Sync 1/125 or slower! WTH??!

@StephanBG

Global shutters that don't come at the price of sensitivity and dynamic range are still a way off, we haven't seen any APS-C sensors of this type, so it will be a while before we see a medium format (CMOS) global shutter sensor.

1/125s is not unusable I think they are just aiming more at landscapers than professional studio fashion photographers who are likely to chose Hasselblad or Phase gear with leaf shutters.

Link | Posted on Jan 21, 2017 at 20:18 UTC
In reply to:

nikoreste: Is the sensor normal come or X trans?

Normal

Link | Posted on Jan 20, 2017 at 13:08 UTC
In reply to:

eyeport: The biggest disappointment:::: Flash Sync 1/125 or slower! WTH??!

Big slow focal plane shutter, what did you expect?

It's the reason a lot of MF cameras (like the new Hasselblad mirrorless) have leaf shutters.

Link | Posted on Jan 20, 2017 at 13:07 UTC
In reply to:

lacikuss: Regardless of IQ 3 fps is a serious limitation for any photographer.

"Regardless of IQ 3 fps is a serious limitation for any photographer."

Any photographer? Like those landscape photographers trying to nail a shot of that mountain as it flies past the camera?

Link | Posted on Jan 20, 2017 at 13:05 UTC
In reply to:

nerd2: So its sensor is only 1.2~1.4X linear size of now-cheap FF cameras, and the fastest lens it has is only f2.8.

Isn't it just better to use 40-50MP FF sensor (i.e. sony's excellent 42MP BSI sensor) and sharp 1.4 lenses we already have instead? That setup has more light gathering than GFX50s+2.8 (or even 2.0) setup, and we have tons of legacy lenses, as well as other goodies (IBIS and faster sync than pitiful 1/125!)

I remember that's the VERY logic fuji used to argue that "Our system with good and fast lenses is just as good as FF". Isn't the logic working the other way around this time?

@Gray Harman

"68.5% larger is meaningless if you're talking about a cell phone sensor."

Actually it's the opposite, relative differences are what matter in photography in almost every case (I can't think of single counter example right now, absolute resolving power, absolute diffraction disk, all meaningless). 100% more area in a cell phone sensor gives a one-stop advantage in light gathering, same with any other sensor.

"And it's patently absurd to think that taking a FF sensor, and adding the additional light-gathering area of another APS-H sensor would only make a marginal difference in image quality"

8x10 film is rarely cut exactly to size. If it's cut just 2mm larger, then you're adding the additional area of another FF sensor. It's patently absurd to think that you could tell the difference between 8x10 film cut exactly and 8x10 film cut 2mm larger or smaller in terms of image quality, despite there being a whole full frame sensor's worth of difference.

Link | Posted on Jan 20, 2017 at 10:46 UTC
Total: 364, showing: 1 – 20
« First‹ Previous12345Next ›Last »