Can someone recommend a meter? - Page 3

Someone needs to get a tripod mount made asap.

Guys, I am having a hard time understanding this with my thick head. Why on Earth would you profile ANY tristim against an i1 Pro when you can have the tristim profiled against a 5 nm reference probe instead? Assuming for the moment that it is done *correctly* (I don't want to enter the debate as to whether SpectraCal is doing it right or not), and you have offsets generated for most all display types, why do you insist on profiling against a low end spectro like the i1 Pro instead???? And are you really afraid of the tristim drifting off next week? You don't think a quality Tristim like the Chroma 5 will be good for one full year? And do you really think that your i1 will be good for more than one year?

I really don't get it...

one argument could be that some (many?) people like to own both meters so they can do it themselves, as opposed to having to send theirs in for "enhancement". Many people own for example i1Pro + i1 DT/LT (using the quicker tristim for actual work and the Pro as the profiling reference and for final touch-ups). For a dedicated "enthusiast" or even a professional calibrating consumer displays, this is a powerful and flexible combo...


I don't think that's quite fair -- from what I understand the i1 Pro is a professional grade meter and nearly as accurate as many $$$ meters.


you seem to be sticking with this assumption that the spectro's will "drift" as much as filter-based tristims. That totally flies in the face of everything I've ever heard on the subject; it is a basic tenet of any "which type is better?" comparison that this IS in fact a major, inherent advantage to the spectro (fundamental to the difference in technology).

BTW - I'm not arguing against your thesis necessarily, I am not nearly qualified to do so... just speculating on some counter arguments.

Because the spectral response of every TV is different. The profile for a tristim is most accurate on the display it was profiled against. The greater the difference in spectral response the less useful the profile is.

So the refrence profile could be for a display that is subtly (or substantially) different from the one you are working on. By profiling it against an i1 pro you've now created a profile for the exact display you are working on.

You would use these profiles only on the display you're working on, and probably would be served best by creating a fresh one everytime went to calibrate the display.


The reason you'd use a tristim at all if you have an i1 pro is due to the low end repsonse and speed.

You mean every single display device has a different spectral response than another? So if I have a profile for my JVC RS-35 front projector, it will be worthless when I try to calibrate my JVC RS-20? Or do you mean that each type of display (like front projector versus direct view LCD) has a different spectral response?

I am certainly no expert either, but I have been told over and over again that my i1 Pro should be recertified at least once a year...that anything less would be just fooling myself. I have also read about many of the cheaper tristims (like the Spyders) either being very inaccurate from the start and/or drifting very quickly, but what anyone means when they say "very quickly" has never been defined. Is it a week, a month, a year...how long before it drifts significantly? And how much information is available about a better tristim like the Chroma 5? Heck, I don't even know how long the C5 has been out and if there is any *reliable* information as to how quickly and severely it begins to drift.

What I would really like is to hear from people who actually know and work with these things day in and day out for some unbiased factual information...you know, someone like Greg Rogers who has no financial or otherwise invested interest in ANY meter. I have heard so many conflicting things from so many people that I am totally confused at this point. And from my personal experience, so far the Chroma 5 Pro is proving to be a significantly better and more useful meter than my i1 Pro, but then again, my Chroma 5 Pro is only about a month old.

Yes your RS35 is different than another RS35 with a different bulb with different amount of hours on it.

How much different?
Are the bulbs from the same source?

Worthless and not as accurate are two different things.

Since I purchased a C5 from SpectraCal about a year ago (although it was updated with LED tables at SpectraCal in Oct 2009) and I have an i1Pro I purchased from SpectraCal, I thought I'd check them against each other.

I've attached a couple of PDFs comparing the two on a Pioneer Pro-111FD. I've attached notes at the bottoms of the PDF reports to give you the particulars about how I did the calibration runs. (The C5 NoProfile and wProfile actually contains a mistake in the top portion. It says "Meters: i1Pro (left), C5 with Profile (right)". It should read Meters: C5 No Profile (left), C5 with Profile (right). Sorry)

Let me know if something is unclear. Have a happy 4th!


Best regards,
Greg

All High Def LLC

 

C5_NoProfile_vs_C5_wProfile(i1Pro)_PioneerPro111FD_AllHighDefLLC_070310.pdf 373.0927734375k . file

 

i1Pro_vs_C5_wProfile_PioneerPro111FD_AllHighDefLLC_070310.pdf 361.8046875k . file

Thanks for the response, sotti!

But my whole question comes down to "What is your reference source?" As far as I am concrened profiling ANY meter to an i1 Pro is almost worthless, as the i1 is NOT a ***REFERENCE*** meter...a PR 655 5 nm unit IS a reference meter, so when you profile a tristim to a non reference meter like the i1 Pro, then you can only obtain results as accurate as the i1 non-reference device is capable of providing, and IMHO that MAY or MAY NOT be more accurate than a C5 is (according to Tom Huffman who owns a true reference meter the i1 and C5 factory units measure very closely), so why would you profile a C5 against a meter that may or may not be more accurate than the one you are already using?

I only have 2 questions...please address them separately, if you would be kind enough:

1. If the Chroma 5 factory unit is no better than the lowest price tristim out there and needs to be profiled against a low end spectro to have any usefulness at all, then why does SpectraCal sell it? I mean, you guys should be doing your clientele a favor and sell ONLY the lowest price tristims out there since they will perform almost EXACTLY like a C5 once profiled. Or better yet, ONLY sell the lowest priced tristim in a package with a spectro, as a tool used to gain speed and convenience.

2. Furthermore, why sell an "enhanced" version of the Chroma 5 where you profile it against a true reference meter, because according to your own post above, every display is different than every other display, thus making the "enhanced" feature useless and irrelevant? In fact that would pretty much make **EVERY** "enhanced" meter pointless, so why is this done at all?

Greg, thank you! You have very effectively demonstrated that the profiling feature in CalMAN works exactly as it should... Once a C5 is profiled against an i1 Pro, then the C5 measures similarly to an i1 Pro, just as it is supposed to work.

What is does not prove, however, is which meter is more accurate (without any profiling), as you would need to compare each meter's results to the results of a 5 nm reference meter such a PR 655 individually to see which meter measures closest to the PR 655. Without using a reference meter you have no idea which is the more accurate meter "out of the box" so to speak.

I hope that no one is getting upset with my posts, as I am just trying to stimulate good healthy debate, as well as to provoke people to think for themselves rather than simply "go with the herd" because you have heard the same information over and over again. Sometimes that information is correct, but sometimes it is not, and with a little luck maybe we can all learn a little something in the process.

Not getting upset, but IMO you are trying to get a simple answer to a very complex question. I also think you could do a little research on your own on the differences between spectrophotometers and tristim colorimeters. This subject has been addressed many times by people more knowledgeable than us. A start could be to read this paper from Konica Minolta http://www.konicaminolta.com/instrum...munication.pdf
They have been in this business for many years and make both types of instruments and certainly know what they're talking about, and might help you understand some of the answers you've been given already.

The difference between meters has also been discussed many times on AVSFORUM. On an open forum like this, however, you have to know the difference between those who are truly knowledgeable and those who think they are. I have my opinion about who to believe and you have yours, and I'll bet the farm our opinions don't agree.

Agreed... Though I bet we have at least 2 people in common that we both trust...Greg Rogers and Bill Blackwell.

I think you are right and it is best to do my own research, so I will withdraw from this thread. Thanks for putting up with me.

Edit: for clarification.

And also answering a number of Bob Sorel's posts....The above post is pure FUD.

Spectracal has a Minolta CS-2000 and could generate new tables for a Klein or any other meter in literally seconds if that were the issue.

The fact that you think you have zero issues is really comical in itself - and everyone should bookmark this thread for the future.

Calibrators tell us that we cannot take settings from one calibrated display and move them to another set of the same kind and have it calibrated correctly.

Yet here is a calibrator, taking a set of reading from one meter and transfering them to another meter and he expects them to work on a totally different set.

This is equal to one taking the settings from a Pansonic Plasma and expecting them to work in a Samsung LCD.

Bob, you are correct. There are different Spectral Responses for different LEDs, CCFL, Plasma, etc etc etc. Profiling a meter will ONLY basically work correctly (as correct as sharing one setting works for set to set) on the same model number set you are using.

You can't have it both ways.

You're welcome, Bob. Healthy debate is good, so don't apologize.

As far as determining which meter is more accurate (if either one really is), I think you're right. Unless you use a reference class instrument, you just don't know for sure. Then again, even the reference class instruments have limitations that we need to be aware of to use them correctly.

Of course, I'd LOVE to have a K-10 AND a PR-655 (or PR-670!) to use to profile it! I don't doubt for a minute that D-Nice gets terrific results using his K-10 profiled against his PR-655 by generating his own tables through Klein's software. That's a big investment, though.

Just for your info, I tried profiling my K-10 against my i1Pro (both with CalMAN and Klein's software) with unusable results, at least in my hands.

The only point of my test on the Pioneer was to see how my year-old C5 (which, BTW, I haven't used at all in about 7 months) performed since the question was raised. I don't use the C5 for my professional calibrations. (I've been using a K-10 until recently and I'm going to switch to a CS-200 soon - for the reasons cited above.)

I think that ANY tristim colorimeter needs to be profiled against a true reference, for each display, in order to be considered 'accurate'. Unfortunately, 'accurate' seems to be a somewhat nebulous term. I used to think that 'accurate' was some hard and fast number, but even the two major calibrator training organizations and calibration meter vendors don't agree on how accurate 'accurate' is! Certainly not as clear cut as I'd like. It seems to depend on who's talking, why they're talking, when they're talking, and what they're trying to sell or keep someone else from selling. It's very unfortunate, really, and renders decision making extremely difficult since there doesn't seem to be an unbiased source of information available.

So, in the end, deciding on equipment is about looking at data for yourself and making a decision that works for you (and your clients if you're a professional calibrator). BIG sigh...

Love to see what data you get if you accumulate some! Keep us 'posted'.

Best,
Greg

An i1 pro is more accurate (especially across display types) than a C5. But a c5 is faster, more accurate at lower light, has more dynamic range for readings (reads darker and brighter) and doesn't require dark offsets.

The reason you calibrate is to fill in the features that each probe lacks.

Tristims are not homogenous, there are features and accuracy within the class. The c5 is a much better meter than an i1D2 or a spyder3. It is more accurate than those meters.


Because "more" accurate counts for something.
This is where a strong background in the math helps you understand what's going on. In spectrum land a meter gets the whole spectrum and then translates that to XYZ using the standard observer function. A tristim shortcuts that process by using filters (tuned to the standard observer function) and sensors that produce XYZ data directly. Being that creating perfect filters is impossible you end up with a device that is more or less sensitive to certain parts of the spectrum that it should be.

So while it's best if you could have 1nm resolution of the LED based LCD you're standing infront of, knowing that in general most LED's spectral response has this sort of shape means you can compensate a bit for what the meter can't see. This process is exactly the same process XRite uses to put in the CRT, LCD and UHP Lamp tables that ship with the meter. Everything I said about LEDS applies to those technologies as well.

There aren't very many questions in this feild that have yes or no answers. Most of the time the answer is, it depends and getting a good feel for the technological reasons behind that will help you from blindly leading any company.

JustVisitingHere,

I'm loading tables into my K-10 the same manner Klein Industries (the manufacturer of the K-10) does with a reference meter (I use a 5nm PR-655 and they use a PR-650) and their software. You would know this if you actually had a K-10 and read the maunal. SpectraCAL does the same thing except they are using their own software instead of Klein's software (which may be the problem).


In the future, I suggest you research information before you attempt to make such a foolish post as you did below.

This Thread reminds me of a past Thread (debate?), though low to mid range meters here vs high end meters in the past Thread

Some info relates to a few of the meters being discussed here




Of course, this is what the two Companies are doing, spend more time on the tables (and/or adding new ones) for these Colorimeters.





other:
















For Enthusiasts, I'll still recommend the EyeOne Pro in the under $2000 range. Since the ColorMunki's release, my experience with it and reports from others who have used it, it's the next I recommend.

Though SpectraCal has support for the ColorMunki now, I can assure you that others want to support it as well with their software based on it's performance and the fact that it's available Retail at a good price point (was even lower a few months back ).


.

As always the more money you spend the greater precision you get with engineering tolerances.

If you buy an inexpensive sensor you may as well assume that its accurate within its range of tolerance.

If a given range of tolerance gives you concern then you have to decide on balance whether you are prepared to spend more money to mitigate that concern. Otherwise someone could pick up a K10 or hubble and offer a profiling service that would magically give you the same accuracy in a 100 dollar puck.

Consider that less than 10 years ago if you wanted to calibrate something you were talking thousands of dollars to even get you on the starting grid when it came to metering solutions. And even then few of those solutions were much better than the plethora of inexpensive pucks that you can pick up today.

I do get slightly amused watching someone chase a target down to the nth degree using meters that are less accurate by an order of magnitude. Especially when they find it never reads the same twice down there.

You do get some weird ones. I have a spyder 2 (second hand) that must be at least 7 years old now . At the time it measured very favorably against both a philips color analyser and a minolta CS200 referencing a calibrated ( by myself) film workstation.

I dusted it off the other week and it still seems to perform better than either an LT (2 years old and to be honest it was always a bit sucky) or a spyder3 (brand new out of the box bit better than the LT). In fact the spyder 2 seems to give results that were at least visually on par with a brand new colormunki photo, and this was for 3dlut profiling also correcting for gamut.

Swings and roundabouts you pay your money you takes your choice.
If you have a $10000 home theater I do kinda question the need to buy something like a K10 unless you plan on doing pro calibration on the side. To each their own though.

I have to agree with your here, Mr. D!

I also think that people have to keep in mind that the meter vendors, like SpectraCal, would be remiss if they catered only to professional calibrators since there are MANY enthusiasts who want to have the pleasure of calibrating their own equipment. (I use SpectraCal as an example b/c I have had nothing short of a fantastic experience with them! They are a dedicated group of people who care deeply about customer satisfaction.)

So, vendors like SpectraCal are filling a need and providing a service to EVERYONE who wants to try their hand at calibration - according to their needs and what they can afford. As you say, if they didn't do that, DIY calibration would be financially out of reach for many more home theater enthusiasts.

I don't think anyone can realistically expect a $500 instrument to perform like a $20,000 instrument. I really have to applaud the efforts of these vendors to make the fun of calibrating affordable for most people!

Best,
Greg

+1 for both Mr. D and Greg.

It seems there is a market opportunity for something in the $1500 suggested retail price that may not deliver at the K-10 level but is better than eye one pro on down.

One with Chroma5 like speed and low light level capability, the long term stability of the a Spectro and works well with a wide range of displays including new technology.

How much can the X-rite Hubble be bought for now?

On Curt Palme they do not have the price but have it bundled with Chromapure and there is a price with the DUO.

I looked online but can't find a price for it on its own.

Athanasios

I have found this discussion very illuminating. Bob has raised some valid questions. It has almost got me thinking that it is better to go in with about twenty people and buy a high end meter like the PR.

This sounds like a great idea to me. I think it could work as long as I could have some degree of comfort with the other people time sharing the meter.

Bob

That is two in Atlanta. It probably wouldn't be to hard to round up another twenty people in the Atlanta area.

After doing a lot of research on my own, I have come to the conclusion that any meter less than a PR 670 is a compromise (and even it is a compromise), but how much of a compromise seems to be the crux of the debate. Maybe other 5 nm spectros would be acceptable, but certainly ANY of the meters we have discussed in this thread will just provide "ballpark" calibrations that may or may not be close to what a 5 nm spectro measures. There are even disputes as to which 5 nm meter should be considered the "reference standard" (I just gave it to the PR 670 because it was the most expensive meter I ran across).

I have also been playing with my i1 Pro and Chroma 5 quite a bit and my personal conclusion is that the C5 is the better meter, regardless of the differences in how spectros and tristims measure light and what other people think. Whether I use CalMAN or ChromaPure, the readings of the i1 Pro are simply not consistent enough to be of much value, especially in low light, making calibration much more of a chore than with the more consistent, faster, easier to use C5. Maybe it is just my i1 Pro and not all of them...I really don't know, or to be honest with you...care.

What I would really like to see is a properly conducted shootout between a bunch of the popular low, mid, and high end meters to see how much of a difference there really is between properly working meters. I would like to see not only the difference in measurements, but the difference in real world viewing of material. It is one thing to measure static pattern differences and quite a different matter to see the differences when video is flying by. I mean, the JVC RS-1 was one of the most ill calibrated machines that I know of, and the differences between it and a properly calibrated machine were quite dramatic, but I have a feeling (just a feeling, as I haven't gotten to see it yet) that the differences between an RS-35 calibrated by a $20k 5 nm spectro and the same RS-35 calibrated by a $400 tristim would be extremely subtle at best and totally impossible to see at worst, especially if you are able to bypass the bias and "placebo" factors and conduct a true double blind test.

So there are a few things that I now consider:

1. Not even a $20k meter is truly a "reference standard".
2. How much of a difference perceptibly is there between a calibration from a $20k meter and a calibration from a properly working $400 Chroma 5? Where is the "sweet spot" between accuracy and price in meters?
3. ANY meter can be made somewhat more accurate by profiling, whether it be to a more accurate meter on the spot or with offset tables from a 5 nm "reference" meter.
4. And here's the big one - The calibration that you make today will start drifting the next time you use the display, or at least this is true with lamp based front projectors, and that is all that I really care about right now. A calibration on such a display is only at its most accurate at the completion of the calibration....from that point on it begins to deteriorate. Different displays with different lamps and lamp ages will drift differing amounts, but the one thing you can count on is that the spectral balance will change. Why spend thousands of dollars, as an enthusiast, on equipment to get the nth degree of accuracy when you could probably maintain a more accurate display over time by simply calibrating more often?

True

Depends. In your situation with your FP, the differences would be on the low end of the grayscale.


I doubt you will get any objective answers to this question.



What you have now is the best choice for your situation

Perhaps not for your situation, however you would be amazed at how little difference there is between an i1pro and a PR-655 or 670 when the measurements are within the light capabilities of the i1pro.

Actually I should have asked "Where is the sweet spot between accuracy, convenience, ease of use, speed, and price in meters?"
Time is one of my most valuable commodities. I can dial in my RS-35 so that the dE's are all around 1 (or under) with both grayscale and gamut, with a highly dialed in gamma curve, using a Chroma 5 in about 2-3 hours from a stock RS-35 color mode. I am certainly not the most efficient calibrator, as I have to make MANY runs with MANY small changes in order to get that degree of accuracy. I tried doing the same thing with my i1 Pro and gave up after 8 hours...

When I double checked my work using the i1 Pro after calibration with the Chroma 5, I found that the dE's were still within acceptable limits (like in the 3 to 4 range, when the i1 Pro was able to measure at all), so even if the i1 Pro is more accurate (and I am convinced that my particular unit is not), the difference in speed, ease of use, and convenience make it a better choice for my particular set of priorities. Maybe I will change my mind in a year, especially if a new spectro is released that is fast and handles low light a lot better, and is sold for a reasonable price (under $1k).

This is the only thing I would disagree with bob as well. Every other point in his list seems spot on to me.

It could be that he's down at the lowend of the i1 pro's range. But with enough light I find the i1 to have superior accuracy and repeatability than a c5.

But for lowlight, that's why we recomend profiling the c5 to the i1 and going from there (Being that the profile is created at 100% WRGB).