RAW

[Alex_B]

@ Alex_B: You are only seeing posterization on that second image because you're looking at the higly compressed 800 pixel wide junky Fotki.com copy and not the 2.5 MB original which looks great. And yes, I did mention Cokin grads if I got more serious about landscape photography. I almost bought a set of those but mostly I do baby/family photography nowadays and can't get out much anymore so I passed.

Ah, ok

what if you set up your light on the baby with a strong gradient? then you could justify the cokin grads because you have to compensate

Anyway, we could agree that RAW is not overrated, but that JPG is often underrated and falsely dismissed as unworthy.

 

[Alex_B]

Im really not understanding why something so simple has to be so philosophical.

Because else we could not discuss and fight :mrgreen:

 

[asfixiate]

Even if I've never even looked at a camera before the only thing I'm saying is Whether you spend 3 hours editing a .jpg or 3 hours editing a RAW file what's the difference.

Your shots are in my opinion very great work but you can't tell me that they aren't touched with software. I dind't read the Rockwell article but I assume he's treating a digital camera as a film camera and he doesn't believe in editing the photos afterwards or something like that. Or at least not editing them as much.

If you twiddle the hell out of your .jpgs I assume that means he's calling you a twiddler?

People here are only stating their preference in what appears to me a non biased way. You, in a biased way, are saying that everyone is wrong for recommending RAW since you can do the same stuff with .jpgs.

You could easily just say you can do anything with a .jpg that you could do in raw and be done but when you call everyone out it starts arguments.

 

[Alex_B]

Even if I've never even looked at a camera before the only thing I'm saying is Whether you spend 3 hours editing a .jpg or 3 hours editing a RAW file what's the difference.

If you edit one image for 3 hours, then either:

a) you have a very slow computer

or

b) you want to print really HUGE, like a whole wall.

or

c) something is wrong with you or the image

Not talking of graphical art here where you create something entirely new (then 3 hrs would not be much) ... talking about photography.

You mention that Rockwell calls people who shoot in .jpg twiddlers. What's the difference in you twiddling your .jpgs? Wouldn't Rockwell frown upon what you do?

What Ken says does not matter too much anyway. He is not the ultimate
and final judge of everything, he is just an individual stating his personal opinion, as we all do.

 

[Helen B]

Why were imaging systems standardized on 8-bits / 256 levels? Because the human visual system is hard pressed to identify even 100 distinct tones.

I think that that is misleading. We may be able to distinguish between only 100 tones at any one instant, but we can distinguish between many more as we scan an image with our eyes.

To test the B&W printer profiles I make, I print a target made up of hundreds of squares and rectangles arranged so that the 256 tone values that an 8-bit printer driver will print can be compared with the numerically adjacent tone as well as the tone 2, 3, 4 and 5 values away - for example I can compare 75 directly with 76, 77, 78, 79 and 80. For prints with a very high density range (over about 2.4) it gets difficult to adjust the profiles so that the one-value steps (eg 75 to 76) are not noticable as a line. The two-values steps (eg 75 to 77) are nearly always noticable as a line. This tells me that if the density range of the print is high enough I can distinguish over 200 shades of grey. That is limited by the print, not by my vision. Anyone can do the same test if they can print to a great enough density range - ie something in excess of a traditional silver gelatin print.

If you are looking at a medium with a density range greater than 2.7 (slide film, for example) you would be able to distinguish even more tones.

You can also calculate the number of discernible tones by dividing the least discernible density difference by the brightness range that humans can handle - it's just theoretical, and the numbers are inaccurate, but the answer will be way over 100.

Best,
Helen

 

[asfixiate]

lol Alex. Sorry I was just exaggerating in efforts to make some kind of point.

I just don't see the difference in supporting someone against "twiddling" if you are doing the same thing jsut with different image format.

Arch
By the way I like the "Sleepy Hollow" look a lot. Cool pic. The original has a good creapy approach to.

Can't the whole raw/jpg thing just be settled scientifically anyways? do what you do to a .jpg and a raw print out in 8x10 take magnifying element and compare or just look at the science behind it?

In my line of work we can't scan certain images in bitonal since the contrast is funky. Instead we use greyscale and it looks 100 times better. Difference between the two is more shades to work with that handle brightness and darkness better.

 

[Mav]

People here are only stating their preference in what appears to me a non biased way. You, in a biased way, are saying that everyone is wrong for recommending RAW since you can do the same stuff with .jpgs.

- A lot of people say you "cannot" adjust WB in JPEG. False
- A lot of peope say you "cannot" adjust exposure in JPEG. False
- A lot of people say you "cannot" recover shadow details in JPEG. False
- A lot of people say you "cannot" re-save a JPEG without losing quality. False
- A lot of people say RAW is automatically better just because it's uncompressed. False
- A lot of people say RAW is automatically better because it's 12/14/16-bits vs 8-bits for JPEG. False
- A lot of people say you "cannot" go back and re-edit old JPEGs with newer/better software. False

I see these reasons being given for shooting RAW all the time and they're just not true. A lot of these myths come from the RAW software that's out there that supports a lot more adjustments with RAW files and very little with JPEG. It doesn't mean that you "cannot" do those things with JPEG. It just means that that specific software cannot do those adjustments specifically with JPEGs. Other software most certainly can. I seriously have had people tell me they shoot RAW because you "cannot" adjust exposure with JPEG. In Capture NX for Nikon exposure is a slider bar adjustment for RAW, but you can only do crude curves to adjust exposure for JPEG thus leading to the "you cannot adjust exposure in JPEG" myth. In DxO exposure adjustment is a slider bar for either RAW or JPEG. Other software is similar.

Like I said above, "a lot" (not all) of the reasons for not shooting JPEG and to shoot RAW instead are outright false at the worst, or exaggerated at best. And if those are your reasons, why are you even bothering with RAW? You don't need to because you can do all of that with JPEG to nearly the same extent you can with RAW. If I'm pointing out that the reasons one gives for shooting RAW don't hold up from a technical standpoint and simply aren't true, then how is that being biased? It's like someone claims 1+1 = 3, and I say no, it's 2, and then they call me "biased" for pointing that out.

You could easily just say you can do anything with a .jpg that you could do in raw and be done but when you call everyone out it starts arguments.

The arguments start when people say things about JPEG or JPEG shooters that simply aren't true AND when a JPEG shooter actually bothers to argue back which does not occur 99.999999% of the time thus leading to a false consensus about JPEG based on bad or highly debatable information. :greenpbl: If people would just stick to the facts then there would be no nonsense to clean up and hence no war.

I shoot RAW when...

- I need maximum leeway to fix any mistakes for critical shots, such as overexposure on a JPEG where you're hosed.
- When I'm rushed for time in a shoot and won't have any time to dial in the camera or lighting and am just going to have to deal with what I get, heck yeah I shoot RAW.

Others shoot RAW because...

- They like to do more than just basic or moderate editing, might do things in multiple steps and save in a lossless higher bit format, and so it makes PERFECT sense to start out in a lossless high bit format in the first place.

- They have no Auto Contrast (Canon) and it's too difficult to adjust it or for your eyes to notice it and your JPEG outputs end up getting blown which you could have easily fixed if you shot in RAW (friend of mine). I'm spoiled by Auto Contrast and other automated JPEG output settings on my Nikon - it's never failed me, and JPEGs come out looking great. Can't say the same for other systems though.


You're correct, I did do editing to the two photos I posted. The originals were actually more on the bland side but here's the before and after.


Original:

After PP in DxO software:

For "Beginners": This is within the leeway of adjustment for getting large quality prints out of JPEG. If you want to tweak things much heavier than this, by all means shoot RAW or use better technique, whichever you prefer. If I wanted to bring the beach up a lot more I really would have needed a 3-stop Grad ND filter rather than the 2-stop that I had. I did all that in just a few minutes. DxO makes it easy, and it doesn't touch your original file so you can go back and re-tweak to your heart's content. In fact I did just that. I haven't printed this one, but I have printed some others that I PP'ed similarly at about 3-feet wide that are waiting to go up in my house and they look spectacular.

changes made:

- leveling and slight crop
- auto sharpening
- applied Velvia film profile (non-linear color adjustments)
- applied mild "Smart Vibrancy" adjustment
- mild lighting correction to bring the beach up just a tad, but not too much
- cooled off the White Balance just a tad
- minor exposure adjustment
- probably some other minor stuff that I forgot

This looks a bit overcooked color wise, but I have it tweaked for the printer that I use that tends to have a bit duller colors. If it's not right, I can go back and re-adjust from the original JPEG again later. No big deal at all, but even on screen I like it. I'm like KR in that I like more wild colors in my landscape shots.


BTW, when I say "without losing quality" in reference to JPEG adjustments, what I'm actually talking about is quality differences that you can see in actual prints viewed at reasonable differences. Some people define a loss of quality as any difference they can make out when viewing something with their nose touching their screen at 100% (nose into a 3 foot wide print) or even 200% (nose into a 6 foot wide print). This is extremely stringent and ridiculously stringent considering most people don't view photos like that. I usually review photos and adjustments at 50% or less. I only view at 100% if I'm checking focus for the most part. The most extreme example I've seen are those that use special photoshop layer/threshold masks to highlight the smallest of differences. These are invisible period, even in prints, and even on the very best monitor viewed at ridiculously high magnification, which is why they have to use photoshop to "see" any difference. I no longer consider that photography but rather computer geekery or analysis paralysis.


@ Arch: Can't see your photo due to firewall issues atm but I'll check it out later.

 

[Mav]

Anyway, we could agree that RAW is not overrated, but that JPG is often underrated and falsely dismissed as unworthy.

Perfectly stated.

What Ken says does not matter too much anyway. He is not the ultimate and final judge of everything, he is just an individual stating his personal opinion, as we all do.

Exactly. He just has 1.5 million unique visitors per month reading his opinions which gives his more sway, which I think upsets a lot of people who don't agree with him. I don't agree with him on everything, and yes I've seen people be misled by him. But more often than not those people are not necessarily doing all of their homework and are taking the advice blindly. Shame on them. From personal experience, Color Mode IIIa and Saturation+ doesn't work out too well for skin tones. :blushing:

 

[Mav]

I think that that is misleading. We may be able to distinguish between only 100 tones at any one instant, but we can distinguish between many more as we scan an image with our eyes.

Yes, that's true, but how do you look at photographs, or your television or computer screen? There are some Apple fans that are upset because the type of monitor used in the 20" iMac is really only a 6-bit display rather than a true 8-bit like the 24" model. To get the extra few bits it dithers and what your eyes see is an average that "looks like" the full 8-bit color range, even though it's apparently really only displaying two 6-bit colors. The 6-bit dithered display has the advantage of being a bit better for motion video or games though since the dithering cuts down on redraw times and fades out the old image quicker. The full 8-bit display goes into the 24" model since it's definitely preferred for photography use over the 6-bit. I don't think you can see the difference, but it still upsets some people anyways on the 20" model. If they had the 8-bit display then they'd complain more about screen lag and slow redraw times and poor video which you can actually see, rather than the difference between dithered 6-bit quasi 8-bit color vs true 8-bit color which you can only really see the difference in spec sheets and not the actual display.

Point taken on greyscale stuff, though. I've looked at 8-bit greyscale maps and I could definitely see more distinct tones on the greyscale map, but definitely fewer on the color maps. In general though, 8-bits is plenty for "most" uses which is why everything was standardized on 8-bits. For the more extreme editing, yes more will probably work better.

 

[Alex_B]

Original:

After PP in DxO software:

hmm, a bit off-topic, but I prefer the original a lot!
(neglecting the tilted horizon)

 

[asfixiate]

Scientifically, mathematically, literally, a RAW image is 65,000 and a .jpg is 8bit....

Scientifically I can drive a ferrari with regular with a bottle of octane boost and still make it to the grocery store but do I want to gamble that my car will handle the gas correctly? Why not just use super?

You can't use the argument "could you even notice the difference" because everyone is basing their views on facts that anyone can go and look up to prove correct. For preferences use what works for you and sure you can do the same but if you compared the 2 ANYONE would see the difference eventually. Starting to print different sizes and eventually you will.

Press Print Screen on your keyboard and paste into Paint. Save that file as .jpg. See what happens?

BMP = non compressed
JPG = compressed

 

[dEARlEADER]

oh hey everyone,

I'm a little late in joining this thread and have never seen anything on this subject before.... and 40 posts would take much too long to read...

so if someone could just tell me what is better RAW or JPEG that would be great...

 

[Mav]

"yes" :lmao:

 

[logan9967]

Entire Post

wow. first reply and probably the easiest to understand and make a decision from

and my 2 cents.

.jpg is perfectly fine for pics, i just found out about raw recently and 95% of my pics are from .jpg and i can still editing a fair bit from them. of course this is nowhere close to what i could get out of the raw files. it's enough. .jpg isn't garbage, but when compared to raw it's basically a honda civic and a bmw.

both will give you what you need.
both have enough capabilities for any normal person.
but one of them has that extra 'umph' the other doesn't.

hope that helps

 

[Jon0807]

oh hey everyone,

I'm a little late in joining this thread and have never seen anything on this subject before.... and 40 posts would take much too long to read...

so if someone could just tell me what is better RAW or JPEG that would be great...

You're gonna start another 40+ posts lol