October 24, 2014, 09:51:26 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - LetTheRightLensIn

Pages: 1 ... 143 144 [145] 146 147 ... 263
Your comment seems to say the 1DsMkIII had no low ISO noise/banding. As someone considering purchasing one I wanted to confirm that positive endorsement (if that's what it was!)
The 5DMkIII and 6D are a confusing pair of cameras. Like the earlier poster I'd like to mix the features differently, some from each... my 5DMkIII a poor copy, I believe, but wished for more improvement in IQ over the 5DMkII.


Well they all have read noise at low ISO, all a lot more than the exmor stuff, but the 1Ds3 did have among the least low ISO banding that Canon had ever released and some of the better ISO100 read noise for a Canon.

From what I hear and have seen and in some, but not all, cases tested, it seems like for the FF cams from Canon that 1Ds3,1DX,6D all have noticeably less banding than the 5D3/5D2 (5D3 completely cures it in one direction but since it still has it so strongly in the other direction it doesn't really do much anything to help). From what I recall 1Ds3 has no horiz banding and a little bit of vertical and the 5D2 has a lot of both and the 5D3 has no almost vertical but a lot of horiz. Those three (1Ds3/6D/1DX) also all have a bit less read noise than the 5D3/5D2 as well at ISO100 (5D3 is actually the worst of them all for ISO100 read noise, although the degree that it is worse than the 5D2 in that regard is so minor that I don't it is possible to notice real world). Some say the 6D has the least banding of these Canon FF cams, I didn't check that out carefully myself yet.

1DX/6D and even 5D3 (and even 5D2) all have better high ISO DR than the 1Ds3 though (even starting by just ISO800 or even 400 really, not even talking high). And at say ISO3200 the 1Ds3 DR actually does fall quite far behind the 5D3 and bit farther still than the 6D and even a touch yet than the 1DX. Even compared to the 5D2 it doesn't even fair so well at ISO3200.

1DX/6D/5D3 all have better SNR than the 1Ds3 across the range (although SNR is so good that at ISO100 it doesn't matter that much and with the better color filters and less banding and a bit less read noise I'd say 1Ds3 does ISO100 better than the 5D3 even if the SNR is worse outside of the darkest tones).

1Ds3 appears to have the least color-blind color array filter when shooting under outdoor lighting of all those cameras. 6D may be the most color blind. In what ways and what the difference means exactly is very complex and hard to say.

6d, at the same ISO settings, produces far less obviously structured noise lines.  THIS IS A GOOD THING! :)
Canon has actually made some significant improvement in this area over the older cameras!

Actually more like significant improvement in this over over the newer cameras! (since the 1Ds3 and even 40D already did this more than half a decade ago  ;) ). But yeah it is good that they at least have back to their past best banding levels. Certainly a shame that they didn't bother to do that for the 5D3 which was pretty ridiculous.

1. maybe the focusing was a trace off with the 5D3 shot? :)

2. maybe the 5D3 uses the split greens that the 7D did which makes the RAW software not be able to quite extract full detail if it wants to avoid mazing artifacts??? :(  :'(

Lenses / Re: Moving on from my 7D to 5D MK III 24-70mm, 24-105mm or prime
« on: February 06, 2013, 07:58:51 PM »
If budget is not an issue, the new 24-70 II is the way to go. On my copy, Reikan FoCal showed the sharpness at f2.8 is 985. Not many zoom lenses out there have this kind of sharpness at f2.8.

My comment is based on real life shooting.....let me know if you want to see some photos with 5D III.

Many owners of 24-105 claimed their copy is razor sharp, but the Canon MTF chart doesn't indicate that at all. Again...I owned 2 copies of 24-105 in the past and I didn't see that. I ended up shooting alot with 50mm f1.4 @ f1.8 to 2.8.
"Many owners of 24-105 claimed their copy is razor sharp, but the Canon MTF chart doesn't indicate that at all. Again...I owned 2 copies of 24-105 in the past and I didn't see that"  I agree with you on that Dylan!

Yup, that is why most of us were saying to look at the 24-70 II (or 24 2.8 IS, 24 T&S II, zeiss 21mm) and forget about the 24-105L.

(and perhaps tack on a bower 14mm if you want something ultra, crazy wide in addition, sometimes on sale for $299)

Lenses / Re: Moving on from my 7D to 5D MK III 24-70mm, 24-105mm or prime
« on: February 06, 2013, 05:46:19 PM »
The 24-70 2.8 II is the one that will give you edge performance on the wide end on your 5D3 that you were used to getting with your 17-55 on your APS-C.

Some of the wide primes could too if you want to pay less than the zoom (zeiss 21, canon 24/28 2.8 IS, canon 24 1.4 II) but the zoom is a lot more flexible.

Of you could go tamron 24-70 vc or, once price drops a bit, canon 24-70 f/4 IS if you really want IS and will trade a bit of IQ for it.

EOS Bodies - For Video / Re: 6D not usable for shooting video?
« on: February 06, 2013, 02:26:21 PM »
The difference between 5Dm2 (and 7D 60D 600D and 550D - i haven't used a 6D) and 5Dm3 is massive when it comes to moire. Not 100% and 0 but enough to make a huge difference for people who actually use the cameras, rather than those who just bitch about them on the Internet.

I still think that in scenes where Moire isn't an issue, the 5D2 looks better.

I don't know about that, the 5D2 has almost 2 stops worse SNR and once you sharpen the 5D3 files in post....
Aliasing and moire come up pretty easily too.

Third Party Manufacturers / Re: D7000 replacement in April...
« on: February 06, 2013, 02:24:11 PM »
You can ask David Hull, I have lent my raw files to many photographers at Dpreview  (as to the  famous Hans Kruse ), in the end all have agreed with me regarding the benefits of low read out noise.
You can defend Canon in absurdum, there are two different sensors and one is with a modern read out analog to digital chain, and  one from 2004 with 14 times higher read out noise at base iso

To be honest, I'm fine and dandy with you perpetually digging yourself into a deeper and deeper hole...it just proves my case for me. So please, happily keep doing what your doing.  ::)

Yeah sure, you'd come up with some other excuse then that he had manipulated his RAW files or something and how can you trust it unless he flies to the US and retakes the shots in front of you and there are three witnesses to track the file from his camera into your computer. You asked for my detailed DR procedure and went crazy that I was supposedly hiding it and then I quickly revealed it and.... it made no difference, you just made up 30 other excuses, until like two months later, you vaguely admitted it all before going back to more excuses.

Nikon has plenty of problems too but there is no need to try to hide the DR thing, if it never affects you fine, whatever, but every time it comes up do you need to extend every thread to thirty pages just to try to cover it up? You have even admitted yourself, in the past, that there is a difference.

Third Party Manufacturers / Re: D7000 replacement in April...
« on: February 06, 2013, 02:17:17 PM »
As always, if you want to 'prove' your point, it helps to design your 'test' with a bias toward your desired outcome

Fair enough, nobody has ever argued that Nikon doesn't lack some lenses or doesn't have some negatives, many of us has said so ourselves too, repeatedly, the problem is that anytime someone mentions Canon is worse at that something the Canon fanboys go nuts and try to hide it.

Third Party Manufacturers / Re: D7000 replacement in April...
« on: February 06, 2013, 02:15:37 PM »
Oh yes...the obfuscation is so thick you could drown in it.  :P They don't want anyone to know they additionally underexposed the 5D II by another couple of stops to put it in particularly bad light relative to the Exmor cameras. Is that not the fundamental crux of bad science? Hide your data?  :o

There you go again.
Everyone from sensor sites to Fred Miranda, to DxO, to Michael, to myself, to numerous posters on DPR, etc. are all in league in a conspiracy against Canon! Everyone get out your tinfoil hats, quick!

EOS Bodies - For Video / Re: Video 5D III vs. 1DX
« on: February 05, 2013, 10:41:57 PM »
From what i've seen/read, the 1Dx is considered soft and lacks detail like the 5d3, with slightly better low light performance. The s35 crop mode on the 1D-C is much better, and shows what they could have done on the 5D3 and the 1DX.

It's too bad they didn't put an APS-C video mode in the 5D3 because:
1. more reach can help for wildlife
2.a.  it would be closer scale to the natural AA filter and might not need to be quiet as softened up
2.b. perhaps the single digic would even be fast enough to simply reach that whole chunk directly and downscale to provide really high quality
3. they could even perhaps have fit in an even high mag mode where they read out a 2k block directly and process it

Anyway I've heard that the 1DX tends to have a bit more moire in it's good mode, but to offer noticeably better micro-contrast/crisper detail. It will never have ML made for it, at least no by the ML group, and considering all that Canon left out of it that is a shame and makes it less useful.

Lenses / Re: "Native" ISO... is it real and does it make a difference
« on: February 05, 2013, 10:25:57 PM »
For one that guy cares only about the interal 8bit jpg/video engine, important to keep that in mind.

Regardless, a blanket statement to shoot at 1-stop multiples of ISO 160 is not universally applicable advice...
Luckily, the Magic Lantern devs have figured out what iso is "best" - and it's rather surprising and more complicated than one might think because an analog and digital component is involved. Btw, setting the iso via ml gives better iq than via Canon :-)


Are ISO 160 multiples the best to use?

NO. They have harsh highlight rolloff and intentionally clipped details in highlights. I have no idea why.

Are ISO 100 multiples the best to use?

NO. While they do have smooth highlight rolloff, they are digitally pushed by a small amount (the exact value depends on picture style and other settings). What does this mean: a small amount of raw data, which actually has the best SNR possible, is simply thrown away.

Then, what is the best ISO?

To the best of my knowledge, the best ISOs are the ones available in recent Magic Lantern versions (April 2012 or later), obtained from ISO 100 multiples adjusted with a small amount of negative digital gain:

    ISO 85, 175, 350, 700, 1400, 2800 - best for Neutral -4 and other low-contrast styles.
    ISO 80, 160, 320, 640, 1250, 2500 - good for a wide range of situations.
    ISO 70, 140, 280, 560, 1100, 2200 - best for high-contrast styles.

To enable them, set DIGIC ISO gain (in ISO submenu) at -0.2/-0.3/-0.5 EV and dial your ISO from ML menu or shortcut keys.

Third Party Manufacturers / Re: D7000 replacement in April...
« on: February 05, 2013, 10:06:01 PM »
I also encourage anyone who disputes that DXO is truly neutral to download DXO Optics Pro and test it out on Canon raw files. Of all the available CR2 RAW editors, DXO Optics Pro seems to fare the worst, by far...especially in the noise and NR department. (Lightroom performs significantly better with CR2 files and NR, as does DPP, obviously.) The poor performance of DXO Optics Pro with CR2 files makes me wonder whether DXO really knows how to process them correctly, and whether that may somehow factor into the growing rift between anything Nikon and anything Canon in their tests.

What does RAW processed images from DxO have to do with their sensor tests which test the RAW file, untouched, directly?

Third Party Manufacturers / Re: Nikon D4x rumors,
« on: February 04, 2013, 07:43:55 PM »
( I initially posted in the wrong section, sorry for that)

Hey all, I read this on the other rumors site, and it got me intrigued to what canon might have on their test shelves!

The rumors for the d4x are :

    Same body as the Nikon D4
    36 MP sensor without AA filter, designed and made by Nikon, with some new improvements (moire in-camera reduction)
    6 fps (11 fps is the maximum possible fps with the current shutter)
    Few video improvements
    Price should be few hundred dollars above the D4 (currently listed for $5996.95).
    Announcement expected in fall of 2013

(scource: the other rumor site)

Could canon have something similarly planned for the 1dx-s?, or maybe a high mp body based on the 5dIII (Like a 5ds )

6 fps for that many pixels sounds really interesting, if they could include a crop mode ( like the d800) and boost the fps 50% like the d800 with grip in dx mode, it would give you 36mp 6fps and 14mp(1.6 crop) with 8-9 fps...

I hope canon is planning something like that, i would love a '2 in 1' camera with high fps ( and still hoping for better buffer sizes on the cameras, like 60 raw....)

I doubt it. And I'd rather have an AA filter at 36MP FF myself and why would they more action cam be the one without it? Price is way too high. It might let Canon be able to look quickly good again (if they fix up the low ISO DR).

Third Party Manufacturers / Re: Canon may be expensive but...
« on: February 03, 2013, 03:34:43 AM »
@ LetTheRightLensIn - Changing the ISO has no direct effect on the amount of light that hits the sensor.  At ISO 400, enabling HTP does not result in a change in the amount of light hitting the sensor, no difference in the number of photons. Period. What the camera does is apply one stop less gain to the signal generated from those collected photons, then applies a tone curve to the jpg data to boost everything but the highlights back up that one stop.

this may be a mess since i typed this out quickly with no thought but:

Yes hitting the HTP button doesn't change the total light now hitting the sensor, but HTP mode both secretly changes the gain 1 stop less AND the metering normally used for that gain by to meter 1 stop down from normal.

If you want some HTP thing then you have some sort of scene with extra amounts of highlight stuff say a full stop more than typical so you have some gain that manages to let you maintain a high enough shutter speed to stop motion or to handhold. You then go one gain down and that gain is now getting shot letting in 1 stop less light than you'd normally let in for that gain for a scene that had less highlights.

Or you can go tripod and then you do ISO100 and just raise shutter more and more or you add on more and more ND filters and you are saving highlights by exposing the scene to less and less light.

ISO400 HTP EC0 is no longer ISO400 EC0 it is ISO200 EC0 metered to expose (or suggested to be so in M mode meter readings provided) 1 stop less than the camera normally would do at ISO200 EC0, i.e. it is effectively as if you were to shoot ISO200 EC -1.

You are not gaining a stop of highlights at a given gain and keeping the same light coming in, that can't be done of course.

Yeah it is at the face of it just applying 1 stop less gain but normally when you apply 1 stop less gain you'd also let in 1 stop more light too and in this case you are not so you are basically letting in 1 stop less light than you'd normally do for the gain. ISO400 HTP the camera isn't doing ISO400 at all it is doing ISO200 and it decides to do it at EC -1 metering instead of EC 0 metering and because of the latter part you might look at it that is letting in a stop less light. In an M mode scenario where you end up needing in some case to fix both aperture and shutter exactly it might be weird to think of it in terms of letting less light in since in this scenario you always want to let the same light in, but it is still metering in way that is compatible with thinking about it that way. And your scenario below where it adjusts many stops to match your light sounds more like AutoISO button than HTP button to me.

Anyway, you can only save 1 stop more highlights than the prior shot by either now letting 1 stop less light come in at the current gain (set EC -1) or by keeping same light coming in and lowering the gain 1 stop (by either dialing gain down 1 stop and then setting EC -1 or swapping on HTP because that is HTP). HTP lowers the gain one stop BUT to keep the same light coming in it must be set to meter 1 stop less than normal at the new gain that HTP selected under the hood (or simply fail to report the new actual gain being applied). So HTP is metering to let a given gain get 1 stop less light than it would normally be metered for.

So in the sense that it meters 1 stop darker than it normally would for the gain that it is actually using you might say it is letting in one stop less light than normal. If not it would no different than shooting 1 stop lower with normal metering. When you go to replicate it yourself, that is what you'd do set ISO 1 stop lower than what you had it in with HTP and then set EC -1.

If you set your camera to ISO100 EC -1 and then shoot all day in P,Av,Tv you'll get RAWs that can be made to give same results as ones from ISO200 HTP EC0. In M mode you could get files that can be made to deliver the same results using either ISO100 EC -1 as ISO200 HTP EC0 and meterings suggest to you to use would be the same in either case.

It might make more sense to think of ISO200HTP not as any sort of ISO200 at all but as ISO100 and as an ISO100 that gets exposed to 1 stop less light than typical.

You could think of it as exposed to the same light and then has 1 stop less gain applied which it is but since the basis for that decision was based on the meter thinking it had been getting 1 stop more gain....

So you shoot your fountain at f/4.5, 1/15th, ISO200HTP or you shoot it at f/4.5, 1/15th, ISO100 HTP-off and get the same thing, same shutter, same aperture,same SNR,same DR,RAW files are store a touch differently but are basically 1:1.

Suggesting that at ISO 400, HTP reduces the light hitting the sensor is not just misleading, it's plain wrong.

Yes, swapping ISO400 to ISO400HTP doesn't change the light hitting the sensor. Hit the button and the total light hitting the sensor stays the same.

But saving a stop of highlights for a given amount of sensor gain does mean 1 stop less total light hitting the sensor.

Under the hood, HTP it is doing 1 stop less gain and then setting the metering to expose 1 stop less than normal for that gain. If you have an ISO400 HTP shot and then want to do that on a camera without HTP you would set ISO200 and EC -1. If you shot at ISO400 and EC 0 and wanted to save a stop more of highlights while shooting at ISO 400 what would you do but EC -1 and let 1 stop less total light hit the sensor.

Although it is probably simplest to not talk about getting less photons.

What if you wanted motion blur of a fountain, but to preserve the highlights in the scene - would you sacrifice the motion blur you wanted, or stop down and change your DoF or lose sharpness to diffraction?  I would do what the camera does with HTP - underexpose by lowering ISO as many stops as needed, and if that took me to ISO 100, it would be time for an ND filter.

HTP doesn't lower as many stops as needed. AutoISO does.
HTP just shoots one stop lower and then meters that ISO one stop under.

In this particular scenario of M mode, we are not talking about an at any given gain scenario any more. You just set the gain to balance how much shadows detail you want vs highlights saved. Once you get it ballpark you can then +/- 1 it to save more or less highlights. (although you might want finer tuning than whole stops and might want to tweak aperture or shutter 1/3 of a stop unless you MUST have it left as is exactly).

In the ap AND shutter MUST be exactly locked sub-case of M mode scenario deciding to lower gain a stop might be a bit odd to think of in terms of shooting a given ISO with 1 stop less light coming in since you are in a we decided to keep that fixed scenario, granted so yeah a bit weird. It is still true though that if ISO200 HTP is what ended up working for the particular scenario then so would ISO100 instead and if you cared what the meter told you, then setting that EC -1 same as always though.

I disagree with your statement, "...if I want to do HTP myself what do I need to do? To get the exact same result I shoot at my selected ISO, keep my selected aperture and then I raise the shutter speed 1 stop faster."

It is probably best to just say that shooting HTP at a given gain an EC 0 is 1:1 in RAW to shooting without HTP but at 1 stop less gain and metering set to EC -1. That is normally what I say. Of course in the latter case the histogram and image review may be a bit tougher to judge at first.

At high ISO where they disallow HTP is would actually make sense to always have it on since the digital gain just lops off the top stop each ISO you go up.

Third Party Manufacturers / Re: Canon may be expensive but...
« on: February 02, 2013, 10:19:27 PM »
no Im saying by exposure after 400 iso you have create a head room by 2 stops compare to 100iso
what is so difficult to understand Neuro! = exposure after 400 iso = halving the hitting lights twice compare to 100iso

Hypothetical example: I shoot jpg. I am taking a picture of a forest scene. I am in Av mode, and I set f/8 to obtain the desired DoF, and I chose ISO 400 to get a 1/100 s shutter speed at metered exposure to avoid camera shake, because I foolishly left my tripod at home. Please note, I couldn't care less about what exposure settings would be at ISO 100, 50, or 3200, that's tangential and irrelevant - I choose f/8 and ISO 400 for the reasons I stated.  I take a shot, look at the review image, and see blinking highlight alerts where I want detail of the sun-dappled forest floor. I've read that HTP can preserve my highlights.

If that scenario is confusing, I'll summarize - with HTP off, I set the camera in Av mode, f/8, ISO 400, and the metered exposure gave a 1/100 s shutter speed.

Answer these questions about what happens when I set HTP to Enable:

1) Does my selected aperture of f/8 change?
2) Does the camera-selected shutter speed of 1/100 s change?
3) Does the amount of light hitting the sensor change?

Please, no hand-waving, no 'please read my earlier posts', no repeating what you've posted before, no referring to what may happen at some other ISO setting that I didn't select and don't care about - just answer those three, simple questions with a yes or a no.

no, no, no

BUT you forget that when you hit the HTP button you are no longer at ISO400 any more but you are at ISO200 only the camera is still metering at as if it were at ISO400. You are not getting to the point of what HTP really is, it is NOT some magic hardware where a new set of photosensors are enabled to capture extra highlights or something, it isn't a 'real' mode, if you want to replicate the mode yourself you can even with a camera that has no HTP button. If you want HTPISO400 and were shooting scenes at say EC0 then you just use ISO200, lock your aperture and then EC -1 or roll up the shutter speed yourself, if in M mode say, 1 stop.

What the camera is actually really doing is just exposing 1 stop less, collecting less photons for any given auto-metering or placement of suggested metering mark.

I can see where your way of looking at it comes from too, but I think it kinda is a less natural perspective.

It doesn't really matter what the camera labels and calls things but what matters in the end is what the sensor is doing, what the shutter speed is, what you get in the RAW file. I think it makes more sense to look at it from if I want to do HTP myself what do I need to do? To get the exact same result I shoot at my selected ISO, keep my selected aperture and then I raise the shutter speed 1 stop faster.

I have newer said any different, you gays are mixed up with something we call head-room, a 400 iso HTP motive are exposed after 400iso with the less gain. TO CREATE A HEAD ROOM  and  400iso  in its self means 2 stops head room from 100iso = halving the signal twice from base iso.

By you guys are you referring to me too or just the others here???

Pages: 1 ... 143 144 [145] 146 147 ... 263