mosc

Joined on Aug 9, 2012

Comments

Total: 1592, showing: 1 – 20
« First‹ Previous12345Next ›Last »

Man I need to clean my monitor.

Link | Posted on Apr 24, 2017 at 15:05 UTC as 4th comment
On article The Sony a9 is a 24MP sports-shooting powerhouse (1857 comments in total)

Ok, but how does it look with a 400mm f4 lens attached to it? I think that grip is still a little small for that kind of weight don't you?

Link | Posted on Apr 19, 2017 at 20:55 UTC as 174th comment | 10 replies
On article Canon EOS 77D Review (252 comments in total)
In reply to:

MichalEs: DxO sensor measurements for 80D and M5 gave different results for dynamic range or high ISO noise. It' hints at a possibility of Canon using some slighly different sensors for other cameras than the one used in 80D. It would be very interesting to see if DPR dynamic range and ISO-invariance tests could confirm it, but sadly the review doesn't include these tests. Will it be added in the future?

What a world we live in where you can immediately answer that with actual evidence. Kudos Richard and DPR.

Link | Posted on Apr 18, 2017 at 19:26 UTC

To capture the pro market all they need to do is come out with an tweaked A7R2 sensor in an FE mount 1D/D5 type body (or just one that gets there with an optional grip). The A99m2 is close, but it's not e-mount. They also need to reach 15fps and more battery life.

42mp stills with better-than-A7Rm2-AF at 15fps from a big weather sealed body will really impact the sports market.

That, and they need a good answer to the 200-400 f4 with built in 1.4x and a 200mm f1.8 prime or somesuch

Link | Posted on Apr 18, 2017 at 19:22 UTC as 18th comment | 6 replies
On article Canon EOS 77D Review (252 comments in total)
In reply to:

mosc: What's all the naming fuss. It has a pentamirror doesn't it? It's a Rebel T7s. Canon Jerks.

Top plate LCD and a second control dial as a market differentiation makes no sense to me.

I wish DPR would call out pentamirrors as junky viewfinders. They're a pain for landscape photography. EVF's get plenty of hate but at least they're full coverage. I think any sub-100% coverage finder should have viewfinder coverage listed as a negative. There are plenty of sub-$900 cameras that have full coverage viewfinders even if they're EVF (and pentax would remind us plenty that have OVF's too).

Link | Posted on Apr 18, 2017 at 19:08 UTC
On article Canon EOS 77D Review (252 comments in total)
In reply to:

entoman: It's a great shame that the pentamirror viewfinders fitted to this (and most other) "budget" DSLRs are so poor - a bit like peering down a dark tunnel. A proper pentaprism would add to the cost and weight, but surely not by a huge amount? The poor viewfinder experience is quite likely to convince a lot of potential buyers to go mirrorless. Canon, Nikon and Pentax all need to look seriously at ways in which they can improve the viewfinder experience of their budget DSLRs.

When I heard "Canon 77D", I immediately thought T7i with a real pentaprism. The fact that it doesn't have one is what really makes the name bizarre.

Link | Posted on Apr 18, 2017 at 18:49 UTC
On article Canon EOS 77D Review (252 comments in total)

What's all the naming fuss. It has a pentamirror doesn't it? It's a Rebel T7s. Canon Jerks.

Top plate LCD and a second control dial as a market differentiation makes no sense to me.

Link | Posted on Apr 18, 2017 at 18:46 UTC as 42nd comment | 2 replies
On article Panasonic Lumix DC-GH5 Review (1156 comments in total)
In reply to:

mosc: How does it take a 6000x3000 image for it's 6K photo mode (which doesn't make sense anyway because 6K would be 5760x3240) when the sensor is only 5184 pixels wide?

So you say 4992 x 3744 is the resolution in 6K photo mode which is a 4:3 aspect ratio crop of the total sensor of roughly 1.04x (2.08x relative to FF). Total MP is 18.7MP which is I guess why they call it 6K because 16:9 6K is also 18.7MP.

I guess I just answered my own question but perhaps it would be good to indicate that when describing the 6K photo mode?

"6K photo mode is a 4:3 native aspect ratio 1.04x crop of the sensor that gives the same 18.7MP resolution of 6K 16:9. Unlike the 5760x3240 you'd expect as '6K', it's a taller and less wide 4992 x 3744."

That would seem better than 6000x3000... not sure where those numbers came from other than it starts with a 6.

Link | Posted on Apr 11, 2017 at 19:36 UTC
On article Panasonic Lumix DC-GH5 Review (1156 comments in total)

How does it take a 6000x3000 image for it's 6K photo mode (which doesn't make sense anyway because 6K would be 5760x3240) when the sensor is only 5184 pixels wide?

Link | Posted on Apr 11, 2017 at 19:16 UTC as 55th comment | 4 replies
On article At the market: Panasonic GH5 sample video (76 comments in total)
In reply to:

Ebrahim Saadawi: Is it just me, or do these GH cameras produce a Video image, as in broadcast imagery we see on TV. But not "filmic" images, or cinematic, as we see in most movies. I mean the ancient 5D mark II for god's sake made filmic images out of anything!

1- Is it the high resolving power that give detail that I don't percieve as film, but as a new type of hyper-realistic video? (which btw looks amazing). Is this why the lower resolution Alexa is more used and more filmic that the enormous resolution REDs?

2- Is the sensor size being quarter of the 5D sensor giving a deeper DOF than "I" associate with film? (When shot w/o SB)

3- Is it the subtle IBIS warp like effects we see in some movements and robotic rendition that's rare but present bs optical IS?

4- Is it the colour science of Canon and Arri and so on that induce this effect in me. Or simply the picture processing: digital sharpening, halos, contrast, saturation of the highlights, hue, skin, etc?

Do I freakin need a LOW resolution camera (which should be bad) with no digital enhancements (which should be good to enhance) with no IBIS (should be good to stabilize!) with a huge sensor for ability to get ridiculously thin DOF to be satisfied and percieve it as filmic?!!

I think I'll keep shooting the 5D for artsy films and GH4 for impressive videos, and satisfy both clients...

or get a Metabones SB for the GH4 to get a s35/APS-C size DOF, turn to V-LOG and remove all settings, turn off IBIS and add a freaking gaussian blur plug in onto the 4K image and render to 1080p?! Maybe crop to 1:35:1 for a wide screen film look, maybe grade to get lowish contrast like in older movies? Is this too much work to replace just pressing record on a back-up 5D or a Nikon DSLRs or a blackmagic pocket 1080p?

I just CANNOT like the GH5 video images. Trying ti analyze WHY!!!

I feel like a good portion of this is depth of field. I think the Cinematic feel you're talking about is everything in focus. Studio set backgrounds help this (rather than being outside) but a large part is aperture.

I also think higher frame rates can feel less smooth to people.

If you slow down the GH5 to 24fps, stop the lens down to f8, and keep all backrounds within say 30 feet (generally with brighter than daylight lighting from angles that avoid shadows entirely), you'd get the feel you're looking for.

Link | Posted on Apr 7, 2017 at 19:44 UTC
On article At the market: Panasonic GH5 sample video (76 comments in total)

Why do you suppress the video screen size controls on the front page? It's not like we can't find this video on Youtube directly. More annoyingly though, this is intended as a video sample (we're hardly hitting play for a stimulating narrative here) to show 4K video quality... which you're actually implying has some value to look at in a 580x326 window?

EDIT: I see you can go full-screen from the plugin if you're on the actual article page rather than on the front page. I would suggest either preventing videos from playing (and instead just linking to the article page) from the front page or allowing the full screen button everywhere.

Link | Posted on Apr 7, 2017 at 11:39 UTC as 26th comment | 2 replies
In reply to:

citrate: So, people who want to buy a Fuji GFX or Hasselblad X1D should really save their money now and wait the the new sensors?

If you really have interest in such things than the lens line-up is the real expense (along with flashes, triggers, studio backgrounds, tripods, whatever). All those things work fine when you just upgrade the body later on. The current 50mp with middling DR (no better than a D810) is just a placeholder supporting all that gear with a more competitive sensor down the road.

Link | Posted on Apr 3, 2017 at 21:10 UTC
In reply to:

Josh Leavitt: I'm terrified to think about how much the lenses will cost that can sharply resolve 150MP across their full aperture range. Pretty awesome nonetheless, though.

I think the new fuji lenses can already exceed 100mp. They built that camera system more for the 100MP BSI sensor Sony will sell them than the current one IMHO.

Link | Posted on Apr 3, 2017 at 21:07 UTC

Hey Sigma, how about updating your 70-300mm F4-5.6 while you're at it? I wouldn't mind paying $400 for a modern 'C' version of it.

Link | Posted on Mar 31, 2017 at 14:02 UTC as 16th comment
In reply to:

mosc: I understand how the tilt and shift lens needs such wide coverage but if you actually start tilting and shifting with it, wouldn't you then get some unlit parts of the MF sensor? Straight on it would cover but it's no longer a fully functional tilt and shift lens if you're asking it to cover the larger image circle. I guess this will answer a lot of questions about FF lenses though in terms of their coverage.

looks pretty tilted and shifted in the supplied photo.

Link | Posted on Mar 31, 2017 at 11:37 UTC

I understand how the tilt and shift lens needs such wide coverage but if you actually start tilting and shifting with it, wouldn't you then get some unlit parts of the MF sensor? Straight on it would cover but it's no longer a fully functional tilt and shift lens if you're asking it to cover the larger image circle. I guess this will answer a lot of questions about FF lenses though in terms of their coverage.

Link | Posted on Mar 29, 2017 at 17:12 UTC as 2nd comment | 3 replies
In reply to:

iAPX: So sad this confusion between Nyquist frequency, that clearly state the highest resolution that could be obtained, and the ability to have clear representation of lower resolutions.

For example in 1D, having 4000 pixels enables to represent 2000 pairs, but won't be able to correctly represent 1900 paris, in fact quality will be degraded overs 1400 pairs, and will be perfect at exactly 2000 pairs.

Maximum representable resolutions <> All inferior resolutions will be correctly represented!

You guys are talking about two slightly different things.

Link | Posted on Mar 29, 2017 at 16:56 UTC
In reply to:

iAPX: So sad this confusion between Nyquist frequency, that clearly state the highest resolution that could be obtained, and the ability to have clear representation of lower resolutions.

For example in 1D, having 4000 pixels enables to represent 2000 pairs, but won't be able to correctly represent 1900 paris, in fact quality will be degraded overs 1400 pairs, and will be perfect at exactly 2000 pairs.

Maximum representable resolutions <> All inferior resolutions will be correctly represented!

Yes. You have the same issue explaining to people why a 720p signal looks better on a 720p TV than a 1080p TV.

Link | Posted on Mar 29, 2017 at 16:51 UTC
In reply to:

mosc: So here's my idea.

A Brayer pattern except you change the red and blue pixel filters to not-red and not-blue. Essentially you don't filter out green light from either one. You have to calculate the colors through the vicinity and subtraction. Each not-red and not-blue pixel is surrounded by 4 green pixels so you should have a good idea what to subtract out color wise to get red and blue back.

Advantages: Green channel has full spacial resolution. Greyscale resolution would be improved by around 50%. You also have less total light loss as 50% of your pixels are now only filtering 1/3 of the light instead of 2/3. The other 50% are still filtering out 2/3. Still, that's 12.5% more light hitting the photo diodes.

Disadvantages: Processing is going to be worse than x-trans probably in equal proportion more complicated than between brayer and x-trans.

Same issues: The red and blue resolutions are unchanged. Color moire is not any better, but shouldn't be any worse.

I thought of another negative. The full well capacity (major factor in sensitivity) of the not-red and not-blue pixels will limit the image since they're now taking in green light as well. So base-ISO sensitivity would be considerably worse but the spacial resolution is better and the high ISO light gathering is slightly better.

Link | Posted on Mar 29, 2017 at 16:46 UTC

So here's my idea.

A Brayer pattern except you change the red and blue pixel filters to not-red and not-blue. Essentially you don't filter out green light from either one. You have to calculate the colors through the vicinity and subtraction. Each not-red and not-blue pixel is surrounded by 4 green pixels so you should have a good idea what to subtract out color wise to get red and blue back.

Advantages: Green channel has full spacial resolution. Greyscale resolution would be improved by around 50%. You also have less total light loss as 50% of your pixels are now only filtering 1/3 of the light instead of 2/3. The other 50% are still filtering out 2/3. Still, that's 12.5% more light hitting the photo diodes.

Disadvantages: Processing is going to be worse than x-trans probably in equal proportion more complicated than between brayer and x-trans.

Same issues: The red and blue resolutions are unchanged. Color moire is not any better, but shouldn't be any worse.

Link | Posted on Mar 29, 2017 at 16:37 UTC as 46th comment | 4 replies
Total: 1592, showing: 1 – 20
« First‹ Previous12345Next ›Last »