Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Just to be clear his job is to professionally calibrate displays for money. I’m not saying he’s wrong just an important consideration.

My first thought as well was "professional TV calibrator says this device isn't as good as professional TV calibrator" um ok. But if you want to judge the effectiveness of TV calibration tools who else are you going to ask? Either way I wasn't planning to use the feature anyway, my televisions are already properly set up.
 
Knew it. A gimmicky, flawed feature.
Does Apple claim to make the picture "more accurate"? Or to make it "better"?
What *most* people (as opposed to video snobs) want from a "better" picture is more pop, a wider color range, and more brightness (NOT at the expense of contrast).
If Apple can deliver that, and do so without the user having to spend an hour fiddling buttons, mission accomplished.

Yes, it's not for the sort of person who would pay for video calibration. But it is a nice improvement for the average user with an average TV.
 
Had high hopes, bummer. Maybe the software will improve over time, but this a letdown.

I get my main displays ISF calibrated anyways, so not a huge deal. However, this would have been a great feature to have when setting of friends and family TVs.
I used it on two TVs - very happy with the results. Of course I can always turn it off on the AppleTV, so why is the biggy? It seems people here are thinking that it actually resets the TV or something, NO, it modified the signal going to the TV. It might be that the impact of the "mis-calibration" are the current TV settings not being in some neutral territory.

But in any event, we can all stop whining now, and just turn it off, if you are not happy with it.
 
I find it an odd choice to calibrate an LG OLED TV seeing as how it is Dolby Video-compliant. I have an LG OLED TV and went to calibrate it and it was grayed out because I already use Dolby Vision. Who buys a $1500 TV, $200 ATV, and doesn't use Dolby Vision?
 
Does Apple claim to make the picture "more accurate"? Or to make it "better"?
What *most* people (as opposed to video snobs) want from a "better" picture is more pop, a wider color range, and more brightness (NOT at the expense of contrast).
If Apple can deliver that, and do so without the user having to spend an hour fiddling buttons, mission accomplished.

Yes, it's not for the sort of person who would pay for video calibration. But it is a nice improvement for the average user with an average TV.

You're going to have detail loss then, since you would have to clip parts of the range to achieve what you describe there. Most of the buttons shouldn't exist anyway. They're a carry over from ancient crt televisions, which could be meaningfully be tuned. The biggest impact with modern ones is just buying a better television and replacing when it gets too far out of spec.

I hate the environmental aspect of that, but I don't bother owning a television anymore.
 
  • Like
Reactions: peanuts_of_pathos
Reminds me of the Sonos "Trueplay" tuning gimmick for their speakers where you wave your phone around the room while the speaker(s) play some white/pink noise. I never had it do anything but make the sound worse.
 
I have a stupid question. I've been reading this thread and everybody keeps saying to start with the "most accurate" preset. How do you know which preset is the most accurate? They all look different in different ways. Some have humans who look like humans but everything around them looks like it's underwater, some have lobster humans, some are so dark you can't see anything to know if color is accurate or not... Is there a standardized naming scheme? Since they're all ugly do I just assume my TV sucks, pick a random one, and hope for the best?

... or I guess try the calibration on each preset and see which ends up looking best. Still, I'd love to know if there's an obvious way to identify the "most accurate" preset. Subjective ratings are hard when you have the decisiveness of a puppy.

For that matter, what's actually different about the presets that would result in one calibrating differently from another? Shouldn't it be able to adjust any "wrong" colors until it gets to a standardized result? My assumption was that each preset just had a different range of wrong colors, but that doesn't seem to be the case if one would be more difficult to get up to snuff than another.

I'm starting to think I need a 'for dummies' book...
 
The best-calibrated TV is always the best-looking for everybody. People are used to uncalibrated colors so they will automatically believe that a calibrated TV is crap because they aren’t used to it. Allow them to adapt and it changes their views.
No they are of course not "best-looking for everybody", they are calibrated to adhere to standards.
 
  • Like
Reactions: jhwalker
Would it?
i’m not so sure.
lets start with construction. The majority of TVs that people have in their homes are made of plastic. With Apple, that wouldn’t be the case. It’s Apple, they would want to make it out of fully recyclable aluminum and glass, and be as thin as possible. so add $500 for that.
next, there’s the panel. Now the majority of regular consumers would be fine with just a simple 1080P or 4K panel.
Not Apple.
they would most certainly want to include all of the technology from there XDR Display. Now the XDR display is 32 inches and 6K. I’ll just take a rough estimate, and say that Apple‘s television set, if it were to exist, would be about 64 inches and 8K or more. So that’s a $5,000 panel, plus the Apple tax, so let’s just bump it up to $6,000.
so with the panel, and the fully recyclable aluminum and glass construction, we’re already up to $6,500.
but we’re not done yet.
of course you would need a cool way to mount it. Apples not just gonna have you hang it on your wall like a regular TV, they want it to have a cool floating design. So… $1,999 mount.
We’re up to $8,499, rounded up to the nearest hundred, because of course you would know they would do that.
Now, we would need the processors to be able to both power this insane display, and power tvOS.
now Apple wouldn’t cheap out and include the A12, it probably wouldn’t even be able to power this entire thing.
We’re going with the M2, whatever it may be. Whatever the most powerful processor apple makes at the time that’s not in a Mac Pro is.
Let’s say, $1,000.
$200 for all The included cables and Power supply, $100 for tvOS, $100 for the special remote, rounded up to the nearest thousand…
oh and did I mention, of course this thing needs to have the best speakers. $1,000 for the speaker set up.
and I’m sure there will be other things that would cost a little extra money, so we’ll just bump it up to be safe.
we’re looking at a television set that costs upwards of $12,999.
now, do you think customers would be willing to spend that much money on a television? A television that most wouldn’t be able to fully appreciate, that costs way more than the majority of customers would be willing to spend on a television?
They weren’t willing to pay $350 for a home speaker, so I’m gonna go ahead and say no.
on top of all this, the majority of consumers only upgrade their TVs every… what... 8-10 years? Some even longer than that.
not saying that an Apple television wouldn’t be beautiful, because of course it would be. But would it sell a lot? I don’t think so.
If anything, Apple should be working on a cheaper version of the box. It’s already hard enough trying to sell a $200 box from Apple, when the competition is basically giving their streaming sticks away
I mean true, but if we used that thinking for all the other things Apple makes, they’d never make anything.
 
Exactly. He's an expert in the field, who is in the perfect position to confirm what users of the feature have already complained about. I calibrate displays as part of my job and, while I don't have an Apple TV to try it, assumed it wouldn't work well. Glad to see others with the proper equipment and experience are able to back that up.
He's also in the perfect position to claim that what they are offering for free could put him out of business. Street goes both ways.
 
  • Like
Reactions: Unregistered 4U
I used it on two TVs and both of them have a better picture. If the picture didn't look better to me, I would have told the AppleTV not to change the settings. It's a pretty impressive feature and was very easy to use. I'm glad it worked for me.
 
FTA: "When the tests were re-run with the TV's default color presets, the Apple TV's calibration feature improved color accuracy on all three models."

Wellll....the on-screen directions accompanying the feature advise that you should measure your TV in its base, default configuration. So, if one follows that simple yet important direction, there's a good chance that the color accuracy of your display will be improved.

Following directions. Go figure.
That parsed statement really doesn't say what you're implying. It doesn't imply the color accuracy of the display will be improved. It only implies the color accuracy will be improved over the base config; not improved overall. They are not the same thing.

Essentially you're saying to make the Color Balance Feature (CBF) one would need to ignore the more accurate presets that are available and set it to the default. Wouldn't it make more sense to just use one of the more accurate presets?

Moreover, you only presented a partial quote to support your narrative. What that quote actually says is:
"When the tests were re-run with the TV's default color presets, the ‌Apple TV‌'s calibration feature improved color accuracy on all three models. -you left this out→ Even so, Teoh found that when color accuracy was improved, image quality was degraded in other ways. Since the ‌Apple TV‌ had to shift the color output so much in these instances, the resulting image suffered from unwanted color banding lines, known as posterization."

The basic takeaway should be use the color balance feature if you want, but understand you could make your picture worse... even if you follow directions. Go figure.;)



 
  • Like
Reactions: neander
Exactly - brightness / contrast / gamma / sharpness, etc., all impact texture / detail and are just as important as color.

You should never touch settings marked gamma, contrast, or sharpening. Those should pretty much be set by the factory and never touched again.
 
  • Haha
Reactions: jhwalker
I have a stupid question. I've been reading this thread and everybody keeps saying to start with the "most accurate" preset. How do you know which preset is the most accurate? They all look different in different ways. Some have humans who look like humans but everything around them looks like it's underwater, some have lobster humans, some are so dark you can't see anything to know if color is accurate or not... Is there a standardized naming scheme? Since they're all ugly do I just assume my TV sucks, pick a random one, and hope for the best?

... or I guess try the calibration on each preset and see which ends up looking best. Still, I'd love to know if there's an obvious way to identify the "most accurate" preset. Subjective ratings are hard when you have the decisiveness of a puppy.

For that matter, what's actually different about the presets that would result in one calibrating differently from another? Shouldn't it be able to adjust any "wrong" colors until it gets to a standardized result? My assumption was that each preset just had a different range of wrong colors, but that doesn't seem to be the case if one would be more difficult to get up to snuff than another.

I'm starting to think I need a 'for dummies' book...

Typically, TV makers use terminology like 'THX (something)' or 'Cinema' or 'Accurate' or something like that for the presets purportedly closest to the standards.

My TV (one of the last great Panasonic plasmas) has a preset called "THX Cinema" that has been measured as the closest to the standard. It gets pretty darned close. I wound up tweaking that a bit; i.e., changed the tint just a hair, fine-tuned the brightness / contrast and even further tuned the gamma curve to get almost 100% accurate.
 
  • Like
Reactions: neander
You should never touch settings marked gamma, contrast, or sharpening. Those should pretty much be set by the factory and never touched again.
What makes you think that? Even among the various presets, the settings are different "from the factory". Of COURSE you can change them.

I've only been doing this for about 20 years :rolleyes: and I've never ONCE heard ANYONE say you shouldn't change these; in fact, the FIRST step is to set brightness, then contrast, THEN you can start messing with the colors. Gamma can also be changed and / or fine-tuned to match the standard(s).
 
I'm not an expert in image but isn't this a feature that will get people 80% of the way there?
for the average layperson who buys a big grand TV and has no idea about setup, this fits the bill because god help they try to adjust it themselves...we've all had those relatives with nicer TV than your car with a colour balance that looks awful. If the app get's people into a ballpark then there will always be room for improvement but at least you're in there.
 
  • Like
Reactions: jhwalker
I'm not an expert in image but isn't this a feature that will get people 80% of the way there?
for the average layperson who buys a big grand TV and has no idea about setup, this fits the bill because god help they try to adjust it themselves...we've all had those relatives with nicer TV than your car with a colour balance that looks awful. If the app get's people into a ballpark then there will always be room for improvement but at least you're in there.
These days you will almost inevitably do worse than the factory.

What makes you think that? Even among the various presets, the settings are different "from the factory". Of COURSE you can change them.

Your perception is not that uncommon. I blame the ****wad manufacturers for this.

The manufacturers add interfaces that should not exist at times. Since those interfaces are exposed and user adjustable, people mess with them.

Settings such as gamma and unsharp masking have unambiguously correct values that generate reasonable output responses. If you mess with them, typically it just maps the range to fewer output values. There are various tests to see how well this is working, but beyond brightness levels, you really shouldn't be manually adjusting things on most televisions. If you buy it and it looks wrong, you should return it. You will be much happier in the long run that way.

edit: Keep in mind I wouldn't be telling you this on all parameters if they had meaningful controls. For example, if the hardware came with a way to adjust the color distribution of the backlight to emit a warmer or cooler tone, I wouldn't be against displaying color temperature as an option. The way it is now, these things can't do a good job of emulating such settings. Things like unsharp masking are different and should never be touched under any circumstance. There are rudimentary browser tests that ballpark how well your screen performs on some of these things though.
 
Last edited:
Not surprised that a first iteration of a new feature has room for improvement. There's nothing magical about the Apple process for doing this. Colorimeters have been used for monitor adjustment for years by photographers. I use one. The phone in the case of the Apple TV is just running a simple routine to try to improve color balance. The feature will improve over time. I love this notion that any and all new features must be released with absolute perfection otherwise they will be declared an utter failure. And, honestly, a guy that makes his living overcharging for monitor calibration is never, ever going to say anything positive about a feature like this no matter who it is from.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.