I've been publishing my results in the Nikon SLR forum at DPReview for a while now, but I hadn't registered as a Nikonian until today. However, I've been wanting to share with you my findings.
After noticing "underexposure" in my photographs with the D100, I decided to do several tests. I wanted to find out if the camera was indeed underexposing. So I tested my D100's meter against my faithful Nikon FA's meter which has always given me near-perfect exposures. I found that both camera's metered the same. Then I photographed evenly lit walls with the default tone settings on my D100 and found that the histograms were centered at a point midway between the lower quarter and the midpoint of the histogram range. After taking 30 photos with the same results, I concluded the tone curve was off.
I then experimented and created 3 tone curves with resulting histograms that matched almost exactly the histograms created when setting +0.3, +0.5, and +0.7 EV corrections with the normal tone curve setting. The tone curves curves were further tested thoroughly to ensure they did not introduce any color changes or blow out highlights any more than the default tone setting.
You can download my custom tone curves here (recommend using +0.5 ev equivalent curve for general use): http://fotogenetic.fade.to
Try them out if you are unsatisfied with the default tone settings. They've worked well for me.
In the D100, there are 4 possible tone settings you can choose from at any one time: Normal, Less, High, and Custom. When there is no custom tone curve loaded to your camera, when you set the camera to the custom tone setting, it is actually the normal default setting that is used, so returning your camera to the default setting it came with is a waste in the first place. The default tone settings are always on the camera even if you upload a custom tone setting.
Also, you can only store one custom tone setting in the camera.
I also saw these custom D100 curves on the DP Review site but they were being advertised by a gentlemen calling himself Raul Apolinario. In fact, many people have downloaded the tone curves and have referred to them as the Apolinario curve. My questions to you are:
1. Are these your curves or Apolinarios'? Apolinario's thread for anyone who wants to follow where I am going here is listed below. Or are you and Raul the same person ?
2. Did you take these curves from Apolinario, or did Apolinario take them from you? I guess I'm trying to find out where the true origin is of these curves.
I just want to clarify this scenario because of the confusion. And if there was fraud, I'd like to narrow it down to who perpetrated it. I also downloaded the curves - but they were originally from the Apolinario site. And of course, if Apolinario took them from you, what are you doing to correct that - or get the word out on the DP Review site?
I believe Apolinario and Photogenetic is the same person. If you check out his gallery from the link he has provided, you will see the same images that he used to show you the differences in his custom curves. Also if you click on the link he has provided to download his curves, it takes you to the same download site as the one he provided on DP Review.
I may be wrong, but I think this is the same person with a different name. Why two different names? Only he knows that.
BTW, I also downloaded his curves the other night, but I haven't had a chance to try them out.
Over a week I had responded to one of "Apolinario's" threads because I assumed that was his real name and he answered right back. On his original custom tone download site he had posted a picture of himself and his girlfriend. He appeared to be Filipino - I know, I've been there many times when I was in the Marines - and, when I commented on him and his uniform, he responded back quite promptly. His first name is Raul - or so it appeared.
Problem is that FotoGenetic's given Profile name is Atma bin Imam. Of course, this could be a joke in reference to a modified "Osama Bin Laden." If Atma is his real name then "Apolinario" either submitted a fake photograph to his original download site, orrrrr... a Filipino is using an Arab name.
Ya gotta love this. So now I'm curious if he'll fess up and let us know who he really is.
When you download this zip file, it contains three custom tone curves: the +0.3, the +0.5, and the +0.7 curve. The author - whoever that is at the moment - had recommended that overall picture taking would be best with the 0.5 curve.
So I uploaded the +0.5 curve and tried it out. It definitely does a good job of bringing up the midtones without blowing out highlights. It made a noticeable difference in many of the photos I've taken so far.
I don't know of the specific curve you are talking about so I can't make comparisons.
I've just now downloaded this JLMcurve that you referred to. Before I upload it into the D100, what has been your experience with this - that is, surmising you've uploaded this curve into your D100? If so, has it really helped with the underexposure scenarios that many of us have experienced with the D100.
it is like Night and Day the difference i saw, i did it about 1 week ago i even sent some samples to a good friend of mine who has been shooting lot longer than i have he compared them and even took the new one into photoshop to see if it did anything in the auto mode he informed me it did very little to nothing , color on the pic's is 100 times better, i have been very pleased with it plus i felt confy after readying a little background on the curve was developed, i personally find i can shoot and print now with no need for touch ups, just a little crop here or there
give it a shot only takes a sec to upload it into the camera and try it out , and then only another sec to switch back if u so desire
let me know how u compare the 2 curves as i have not downloaded the one form Mr. Laden
Do us a favor and download the zip file as listed on this posting. After unzipping, choose the +0.5 custom curve and upload it into your D100. Take some photos and if you can, report back later and let me know how it compares with the JLMCurve. I'll do the same on this end. I've only downloaded the Apolinario - whoops - the fotogenetic - whooops..."Da curve," so I haven't compared it to anything else yet.
I found this amusing so far. I use the name Apolinario at DPReview. You can call me Raul if you like even if it may not be my real name!
My curves are not the same as the JLMcurve or the e20 curve someone else had posted. I tried out several other curves and most of them did increase low-midtone brightness, however, they did so by sacrificing contrast. What I tried to do is create a curve that created histograms that exactly match the histograms created when you add +0.3 and +0.5 EV corrections in the camera while using the normal default tone setting. With the curves on my website, they do exactly that without any noticeable color shifts. My +0.7 curve, however, does have an increased red and yellow saturation which looks good for some photos(some landscapes), but not others (portraits).
I was working on more custom tone curves for people who wanted less contrast than the normal setting, basically a tone curve that mimicked adding +0.3 and +0.5 EV with the less contrast tone default tone setting. Unfortunately, my USB cable broke! I'm waiting for Christmas to get a new one. Warning: the USB cable does not stand up well against someone taking thousands of test photos while connecting and disconnecting the USB cable in between photos! ;-(
Thanks for clarifying the issue. I became a fan of your work when you listed on the DP Review website. And of course, I assumed that the photo listed on the download page was of you and a babe. Of course, what do they say about the word, "assume?" When you spell "assume," it breaks down to, "it makes an Ass of U and Me. Ya gotta love it.
Naturally when I noticed your tone curves listed on Nikonians, and then looked up the profile of fotogenetic and saw this variation of Osama Bin Laden...well, ya gotta allow me some leeway here.
Speaking of the JLMcurve, have you tried that one yet, and how does your +0.5 curve compare with that curve if you have in fact tried it. Let us know.
Of course, if that mideastern sounding name is really yours, then I'm sitting here ready to do a hundred pushups sir!!!
I kid you not - as a retired Marine, I still pound out ten pushups when needed and still bench press 375! Keeps me young, now that my sex-for-hire quotient has taken a dramatic downturn - read "age" here. So pushups work!!!
Here's a sample someone else posted in the JLMcurve thread of the mid-tone lack of contrast which I also experienced while photographing geese. After looking at the points on the curve, I realized it is attributed to the downwards concavity of that curve at the midpoint.
By the way, to the JLMcurve's credit, it does do a good job in photos such as silhouettes where there are not as many midtones.
I tried out this curve originally and found that it indeed boosts low tones, however, the midtones from some of my shots (specifically, the brown spots on geese)lacked contrast. You really have to understand what the different concavities along a tone curve correllate to when you create custom tone curves. I've attached a comparison of the different curves.
The black dots indicate the waypoints on my curve while the red dots indicate the waypoints on the JLMcurve. One thing I've noticed about the JLMcurve that I do like and want to try and change on my own curve is how the midtones are darker. This has the effect of making skies look deeper. For now, I use a circular polarizer to create this effect. I haven't been able to find a way to create the same effect while maintaining contrast and brightening low-mid tones. http://genji.image.pbase.com/u17/atmabini/upload/7888628.D100TONES.JPG
In summary, whenever a portion of the tone curve is concaved downwards, there will be a loss of contrast because tones will be clumped around that portion, whereas, wherever there is a straight line or concavity upwards, there will be an increase in contrast. One other thing to note with Nikon's implementation of custom tone curves is that you cannot alter separate RGB channels, thus, when you bend the curve, you risk altering the curve for some colors more than others. What I've noticed, subjectively, is that red seems to react to the bending less than the greens or blues. In order to avoid these color shifts, as well as maintain contrast in the low-mid tones, I kept that portion of the curve straight, but increased its slope. In order to avoid blown highlights, however, I started to introduce a concavity downwards "tail" at the end of the curve. This has the effect of making highlights softer and retaining slightly more information in the highlights. In any case, try out the curve yourself and see if you are satisfied.
Again, you never fail to impress. However, as much as I've been into photography, I'm by no means the high end technoweenie that you are. Your reponse, explanations, and layouts are fantastic. I appreciate those comments and in appreciation for your time will now begin my pushups right now...one thousand one, one thousand two. I'll get back with you when I'm done. I should be recording these for RickD.
You wrote, " . . . I then experimented and created 3 tone curves with resulting histograms that matched almost exactly the histograms created when setting +0.3, +0.5, and +0.7 EV corrections with the normal tone curve setting."
Is there a difference between using the +0.5 ev custom curve (without using any exposure compensation), and using the default curve with a +0.5 exposure compensation in the camera? I can't tell from your description, but it sounds like you end up in the same place If so, why wouldn't one simply just set the camera's exposure compensation function?
The difference is that I get the same highlight details as I would without setting any EV correction and I also get a smoother transition to blowouts. So, basically, my custom tone curve has the same brightness as setting an EV correction in camera, but with the highlight detail of photos without the EV correction--the best of both worlds.
In that test photo of the fish jumping out of the water, I noticed that the photo with the JLMcurve was brighter. If you like that brightness, use my +0.7 curve. It has mid-tone contrast, however, like the JLMcurve, it has an increased red and yellow saturation that won't work for portraits, but for landscapes works great. Adds pop to photos.
On another thread I recommended a link to a master tone curve page where enterprising folks like yourself could post the tone curves, along with explanations of each.
As I mentioned on the DPReview forum, you are providing a tremendous service. Not to give you a big head, once you get that USB cable for Christmas - or whatever, could you keep your links posted on this and perhaps the DPReview site so that those of us who've downloaded your curves can keep in touch with newer ones that you've modified?
I'm hoping, however that with enough response threads to my:
Well it's obvious that when Fotogenetic enters the forum arena, there is much to talk about, dispute, exchange, and what have you.
I do know that he mentioned that his USB cable is "kaput," and that he'll wait till Christmas to get another one.
Hmmmmm, that's a shame and quite unusual. He's obviously an intelligent and ingenious person. Waddaya say we find out where he resides and we send him a Nikon USB cable or two for X-Mas so that he can continue making curves for us to review and try!!!
Our holiday present to an enterprising individual.
Thanks but it's really not a big deal. I'm just trying to find the curve that I like for my own personal use and make it available for everyone else to use if they want to. I actually enjoy hearing that it has worked out for at least some people more than anything else. This is definitely not the fix-all.
I'm still a little unsatisfied with the curve, which is why I'm trying to tweak it.
By the ways, Happy Thanksgiving!
Don't forget to watch the Army/Navy game on December 7th. Help Navy break a 10 year losing streak! That's right, 10 years!