WEBVTT
00:00:00.080 --> 00:00:05.440
I can make that match perfect for you, but a client may walk in the room and they may see slightly differently.
00:00:05.440 --> 00:00:20.480
They may be an older observer who has more macular degeneration, or they have more yellowing of the lens in their eye, or they may just be genetically different and see slightly different because the distribution of uh rods and cones in their eye might be different than yours.
00:00:20.480 --> 00:00:24.079
So those things make perceptual matching quite difficult.
00:00:24.079 --> 00:00:27.679
Now, what's a lot easier is you have one display in the room.
00:00:27.679 --> 00:00:33.840
If that's all anybody looks at, then typically these discussions about which display is right kind of go away, right?
00:00:33.840 --> 00:00:35.840
Because there's nothing else to reference.
00:00:36.399 --> 00:00:43.039
Welcome to Color and Coffee, a podcast that's focused on the craft of color and the artist behind it.
00:00:43.039 --> 00:00:53.280
I'm your host, Jason Bowdach, and each episode we'll sit down with some of the most talented artists in the industry and have a casual chat from one artist to another.
00:00:53.280 --> 00:00:59.520
We'll share their stories, their insights, their tips, and maybe even a little gear talk.
00:00:59.520 --> 00:01:05.920
Whether you're a seasoned pro or just getting started, join us for some great color discussion.
00:01:05.920 --> 00:01:11.359
Sit back, relax, you're listening to Color and Coffee.
00:01:11.359 --> 00:01:15.840
Hello, and welcome to this episode of Color and Coffee.
00:01:15.840 --> 00:01:21.040
I'm your host, Jason Bowdach, and I can't wait to introduce you to today's guest.
00:01:21.040 --> 00:01:26.319
His name is Bram Desmet, and he is the CEO of Flanders Scientific.
00:01:26.319 --> 00:01:34.799
Now, in this episode, we're going to talk about what goes into creating a reference monitor and where he sees the monitor realm going.
00:01:34.799 --> 00:01:36.480
So, welcome to the show, Brom.
00:01:36.719 --> 00:01:37.840
Thanks so much for having me, Jason.
00:01:37.840 --> 00:01:38.319
I appreciate it.
00:01:38.640 --> 00:01:39.519
Truly my pleasure.
00:01:39.519 --> 00:01:43.120
And first off, I have to thank you so much for being a sponsor of the show.
00:01:43.120 --> 00:01:45.200
We could not be around without your support.
00:01:45.439 --> 00:01:46.319
Happy to do it.
00:01:46.560 --> 00:01:48.079
So let's jump right in.
00:01:48.079 --> 00:01:55.599
For those that are not aware, Flanders Scientifics makes some of the best monitors in the business, whether you're a DIT on set or you're in the color suite.
00:01:55.599 --> 00:01:57.519
That has nothing to do with them sponsoring the episode.
00:01:57.519 --> 00:01:59.599
I just happen to love their monitors.
00:01:59.599 --> 00:02:04.400
So let's jump into what goes into making a reference monitor.
00:02:04.400 --> 00:02:07.280
How do you guys start with doing something like that?
00:02:07.280 --> 00:02:13.840
Because I know a lot of people don't quite understand the complicated process of creating a reference monitor and what makes them so expensive.
00:02:14.080 --> 00:02:21.520
I oftentimes joke that it's it's less about what we put in rather than what we don't put in, right?
00:02:21.520 --> 00:02:28.560
So a lot of the kind of consumer or prosumer type displays on the market, they're all using similar processors.
00:02:28.560 --> 00:02:33.039
Typically, they'll they'll buy a processor from from MediaTech that kind of does everything.
00:02:33.039 --> 00:02:39.680
The problem with those is that their aim is oftentimes a lot different than our aim in the professional sphere, right?
00:02:39.680 --> 00:02:43.759
So we want to show you as much as possible your image for what it really is.
00:02:43.759 --> 00:02:58.400
So we don't do things like noise reduction, we don't do anything like a artificial gamma boost or or saturation boosts or special gamut mapping or anything like this that is meant to uh enhance the picture in quotes there.
00:02:58.400 --> 00:03:11.039
So to do all this without relying on one of these kind of generic chips makes things a lot more expensive and difficult to do because essentially what we're doing is purchasing FPGAs and we are programming everything from the ground up.
00:03:11.039 --> 00:03:18.319
So all the video processing, all the ancillary features, all of that is really code from us from the ground up.
00:03:18.319 --> 00:03:23.919
And that's a lot different than you know, again, using a MediaTek chip and skinning it slightly differently.
00:03:23.919 --> 00:03:33.360
And so you'll see a dozen similar-looking consumer displays that all kind of look and perform the same, and that's not an accident, it's because they all use this kind of underlying similar chipset.
00:03:33.360 --> 00:03:41.120
So that's part of the challenge is doing these things and obviously kind of at a at a custom level, but also obviously in much, much smaller quantities.
00:03:41.120 --> 00:03:53.680
You know, we're not selling millions or hundreds of thousands of displays, even you're typically selling things in the hundreds or thousands in this market because it simply isn't that big, especially when you get to these higher-end HDR grading monitors.
00:03:53.680 --> 00:03:55.919
It's actually a very, very niche space.
00:03:55.919 --> 00:04:03.840
And so to build something for that, it's labor-intensive and requires just a lot of work in terms of both the hardware and the firmware design.
00:04:03.840 --> 00:04:06.159
So, and then our primary aims are always the same thing.
00:04:06.159 --> 00:04:10.080
And when we first build a new series of monitors, what's the first thing that we focus on?
00:04:10.080 --> 00:04:11.520
Is color accuracy.
00:04:11.520 --> 00:04:14.800
So we want to make sure that we are as accurate as possible.
00:04:14.800 --> 00:04:17.519
And then we start typically rolling in ancillary features.
00:04:17.519 --> 00:04:28.879
So if you look at a series like the DM series, which we've had out for 10 plus years, there's tons of additional features in that because we've had time to add those based on user requests or things that we thought might be a good idea.
00:04:28.879 --> 00:04:36.720
So you find a lot of like niche feature sets in these things as well, like a pixel measurement feature, a CIE diagram, all sorts of different scopes.
00:04:36.720 --> 00:04:40.319
So that's kind of how we roll in terms of development.
00:04:40.319 --> 00:04:49.040
Uh, a lot of it is based on, again, user feedback and just, you know, knowing the industry well and building something that really is catered specifically towards this business.
00:04:49.360 --> 00:04:51.120
That is incredibly interesting.
00:04:51.120 --> 00:04:55.199
I was going to be my first question is do you guys make your own chips and are you guys programming them?
00:04:55.199 --> 00:05:04.240
And you essentially answered that is I have right now uh one of the monitors I have in front of me is the ASUS PA32 UCDM, if I uh have my memory correct.
00:05:04.240 --> 00:05:14.800
And a lot of people navigate to this monitor because it's very commonly understood that it has the same panel as a lot of the FSIs, but that's just part of the equation.
00:05:14.800 --> 00:05:20.720
You might have the same panel, but what is the signal going to that panel and how do you prep that signal?
00:05:21.040 --> 00:05:21.360
Exactly.
00:05:21.360 --> 00:05:33.120
And I think the other big thing that goes into this is that we don't think that that's a bad thing, that there are all uh computer displays and or the case of the larger monitors, consumer TVs that use the similar or same panels.
00:05:33.120 --> 00:05:40.480
That ubiquity of the technology is actually a strength, not a weakness, because it means that things translate well to those other applications.
00:05:40.480 --> 00:05:47.519
And also we realize that, you know, not everybody's gonna buy a $12,000 monitor to do every step of the process, right?
00:05:47.519 --> 00:05:55.839
So we love the idea that people can go out and buy these kind of more consumer or prosumer type monitors for VFX, maybe some editorial type applications.
00:05:55.839 --> 00:06:05.120
And then when it comes to really color critical monitoring on set for high-end color grading applications, that's where we think the professional reference monitor really comes in.
00:06:05.120 --> 00:06:06.879
And you do sort of get what you pay for.
00:06:06.879 --> 00:06:08.800
Obviously, these things come in at a premium.
00:06:08.800 --> 00:06:12.800
They also, again, take us a lot of time and effort to do this and to get it right.
00:06:12.800 --> 00:06:21.360
And ultimately, what we always tell people is what we sell, what any other professional reference monitor manufacturer is selling you, is confidence that the image is correct, right?
00:06:21.360 --> 00:06:26.079
So, yes, you can use some of these other solutions, but then it becomes a little bit of a science experiment.
00:06:26.079 --> 00:06:27.600
How comfortable are you calibrating?
00:06:27.600 --> 00:06:28.959
You know, have you used color space?
00:06:28.959 --> 00:06:30.240
Have you usal man?
00:06:30.240 --> 00:06:32.959
Do you have the calibration probes to do this?
00:06:32.959 --> 00:06:34.560
Do you know how to use that software?
00:06:34.560 --> 00:06:40.720
And you can do a good job with those things, but there are just a lot of caveats, a lot of ways where you can go wrong.
00:06:40.720 --> 00:06:42.720
So you just have to be a little bit more careful.
00:06:42.720 --> 00:06:45.360
Where with us, you get something that's correct out of the box.
00:06:45.360 --> 00:06:51.279
We have spent years developing our Gaia Color Auto Cal, which is one of the things that sets us apart more than anything else.
00:06:51.279 --> 00:07:01.279
You plug a probe directly into the monitor, it generates all the test patches, it calibrates for thousands of possible combinations of selections on the monitor, all from this one master calibration.
00:07:01.279 --> 00:07:03.600
You don't have to be a calibration expert to do it.
00:07:03.600 --> 00:07:06.959
You don't have to have standalone software, you don't have to break out a computer.
00:07:06.959 --> 00:07:22.160
And then if you are challenged with operating in a novel space for the first time, like for example, if you've never done an HLG project and someone goes, Hey, by the way, we actually need to finish this for HLG, not PQ or not Rec709 or whatever it is, you're just making a new menu selection, right?
00:07:22.160 --> 00:07:24.560
You're not breaking out a computer and having to recalibrate.
00:07:24.560 --> 00:07:27.759
You just make the new selection and you're instantly there at that correct space.
00:07:27.759 --> 00:07:32.240
So those are kind of some of the things that we think we bring to the table for the post community.
00:07:32.240 --> 00:07:43.120
And then obviously for DITs, we have a lot of niche features like the ability to do four inputs into a monitor, see them all on screen at the same time, have each quadrant have its own unique color space selection.
00:07:43.120 --> 00:07:48.879
Those are really powerful features that you don't find on most consumer, prosumer type displays.
00:07:48.879 --> 00:07:58.319
So again, again, we realize it's not the monitor for everybody, but there's a lot to be said for having confidence in the image and having something that's easy to use, easy to navigate, and that you can trust.
00:07:58.560 --> 00:08:00.639
First off, I've had FSI monitors for a while.
00:08:00.639 --> 00:08:06.720
I don't have one right now, but uh I've had them for a while, and I did not know that you can have different color space settings per quadrant on set.
00:08:06.720 --> 00:08:08.160
That's absolutely incredible.
00:08:08.399 --> 00:08:13.600
Yeah, so that that's new on the XMP series that actually went into public beta uh just a couple of weeks ago.
00:08:13.600 --> 00:08:20.319
So it's available for anybody to download and try, and they'll probably become our official shipping firmware version in a month or so.
00:08:20.319 --> 00:08:25.040
But yeah, it's really cool because you can actually now see HDR and SDR side by side if you want.
00:08:25.040 --> 00:08:27.519
You can see different color gamuts, different EOTS.
00:08:27.519 --> 00:08:29.920
Everything is customizable per quadrant.
00:08:29.920 --> 00:08:40.720
And that's just it's a very powerful feature, especially when you're on set with you know a DP who maybe wants to make sure they're protecting well for HDR and SDR, lighting well for both of those scenarios.
00:08:40.720 --> 00:08:42.879
It just gives you an easy way to visualize this.
00:08:42.879 --> 00:08:45.200
I'm not a fan of doing this in a post environment.
00:08:45.200 --> 00:08:51.840
I think you should look at HDR and then SDR, not side by side and certainly not on the same display because your eyes just don't work that way.
00:08:51.840 --> 00:08:54.240
Adaptation works well to one or the other.
00:08:54.240 --> 00:08:59.600
But for DITs, cost and space onset are really big factors.
00:08:59.600 --> 00:09:01.600
You're not doing final color there anyway.
00:09:01.600 --> 00:09:09.039
It's more about just making sure something hasn't broken in that pipeline using your HDR show LUT, your SDR show LUT, whatever it is that you're doing.
00:09:09.039 --> 00:09:14.799
It's an invaluable tool for these guys to be able to use this in a very compact and affordable way.
00:09:15.039 --> 00:09:20.879
I think you nailed on something that I've personally had trouble with, which is trust.
00:09:20.879 --> 00:09:24.399
And that's something that I have to explain to clients all the time.
00:09:24.399 --> 00:09:32.799
That is when you come into a professional color bay, there is hopefully the colorist has nailed down the monitor and they can trust it.
00:09:32.799 --> 00:09:38.879
As opposed to your MacBook or a random desktop monitor or consumer monitor, and even a high-end consumer monitor.
00:09:38.879 --> 00:09:52.559
There is not the trust with the signal that you can have with a reference monitor where I know that pixel by pixel it's going in there and I am seeing my ones and zeros displayed exactly on screen, as you you mentioned.
00:09:52.559 --> 00:10:04.240
There's no Media Tech chip or any other chip in there doing something behind the scenes to make it look good because everybody thinks that they're doing the best job, but actually they're not doing what I want, which is I just want to see the data.
00:10:04.240 --> 00:10:06.399
I don't want to see what you think I want.
00:10:06.720 --> 00:10:07.120
Exactly.
00:10:07.120 --> 00:10:18.240
And the the issue too with a lot of those is like when you get to the consumer chips, they're what we found in testing some of those is when you select a certain feature to be off, it doesn't actually turn off in a lot of instances.
00:10:18.240 --> 00:10:20.000
It goes into the low mode.
00:10:20.000 --> 00:10:22.000
And so that's always a fun, kind of challenging thing.
00:10:22.000 --> 00:10:27.679
And we've seen this a lot with like doing like uh analysis of noise reduction algorithms that are built into some of these chips.
00:10:27.679 --> 00:10:32.000
You'll go to the off mode and you're like, well, there's still obviously some sort of noise reduction going on here.
00:10:32.000 --> 00:10:33.759
And that's not necessarily what you want, right?
00:10:33.759 --> 00:10:48.000
Because as a professional monitor, if there is noise in the signal, you want to see that because it's your job then to fix it, as opposed to uh having the monitor do it and then having that be done in a variable way on every other display downstream.
00:10:48.000 --> 00:10:54.879
You just want to take care of the signal and make sure the signal is as high quality as possible and the content as high quality as possible when you deliver it.
00:10:54.879 --> 00:11:02.720
So yeah, it's again a niche application, but uh, you know, we've been doing this for 20 plus years now, and there's definitely a market for it.
00:11:02.799 --> 00:11:09.360
So I think not only is there a market for it, I think that there is a lack of understanding of how critical it is.
00:11:09.360 --> 00:11:14.080
So for instance, I went from you guys, I had a DM240, I loved it, to be honest.
00:11:14.080 --> 00:11:19.039
I'm a little sad I got rid of it and didn't put it over to the side because I had that trust.
00:11:19.039 --> 00:11:25.200
And now moving over to consumer monitors, I don't necessarily have the buttons and the menus that I had before.
00:11:25.200 --> 00:11:30.080
I have to fiddle and I have to turn things off, which I'm not really sure are off.
00:11:30.080 --> 00:11:32.720
It says it's off in the menu, but is it really off?
00:11:32.720 --> 00:11:36.080
Because when I'm calibrating, it's not getting quite there.
00:11:36.080 --> 00:11:45.440
And being able to be 100% sure that something is off or that the setting is correct is a critical feature that I can turn to my client and they say, Is this what it is?
00:11:45.440 --> 00:11:48.159
Or is this HDR, is this PQ, or is this HLG?
00:11:48.159 --> 00:11:53.679
And I can say 100% certain, look them straight in the face and say, Yes, this is what we are looking at.
00:11:53.679 --> 00:12:02.799
Because I'm sure you've heard before, people come in from the edit suite looking for a show let or working with log footage, and they go, That's not what my movie looks like.
00:12:02.799 --> 00:12:07.120
And a lot of the times you have to tell them, actually, this is what it looks like.
00:12:07.120 --> 00:12:08.879
But this is not a problem.
00:12:08.879 --> 00:12:09.919
We can go from here.
00:12:09.919 --> 00:12:11.279
This is just a starting point.
00:12:11.279 --> 00:12:18.080
And being able to get them one to trust this golden monitor, and two, to trust you that we're gonna take it to a place that you want.
00:12:18.080 --> 00:12:22.080
And then they leave your suite and go somewhere else and say, Oh, it doesn't look right.
00:12:22.080 --> 00:12:34.639
And you have to basically tell them, well, that's a consumer monitor and it's gonna go out into the world and it's never gonna look quite as good as we saw it in the color suite, is a really important basically trust that you need to establish with your client.
00:12:34.639 --> 00:12:38.000
And getting that with a consumer grade electronic is difficult.
00:12:38.000 --> 00:12:38.960
I have trouble doing that.
00:12:39.200 --> 00:12:44.799
Yeah, and it's not it's not impossible to get yourself in a good ballpark or even to make it spot on in some cases.
00:12:44.799 --> 00:12:51.200
But then the other question is, especially with so many people being kind of freelanced these days, is you know, what's your time worth, right?
00:12:51.200 --> 00:12:53.440
So how confident are you in your abilities?
00:12:53.440 --> 00:12:59.120
Uh and I've met some people who've done fantastic jobs calibrating monitors, but I've also seen people take big missteps.
00:12:59.120 --> 00:13:11.679
And so the question is, yeah, you save a few thousand dollars and then you do a project and you realize you exported a LUT incorrectly or you selected the wrong range when profiling, and now you're doing that whole project over, and what does that cost you?
00:13:11.679 --> 00:13:13.919
So it's again, it's not impossible.
00:13:13.919 --> 00:13:20.240
I don't try to pretend that it's you know impossible to get a consumer TV or consumer monitor to do a decent job.
00:13:20.240 --> 00:13:22.559
It's just again, what's your time worth?
00:13:22.559 --> 00:13:23.919
How confident are you?
00:13:23.919 --> 00:13:26.000
Those are real world considerations as well.
00:13:26.240 --> 00:13:26.720
Absolutely.
00:13:26.720 --> 00:13:32.159
And I think the the more technologically savvy you are, the easier it is to do that.
00:13:32.159 --> 00:13:34.159
But you put the nail on a head right there.
00:13:34.159 --> 00:13:35.519
What would you rather be doing?
00:13:35.519 --> 00:13:41.600
Would you rather be doing another job for a client, or would you rather be second guessing your monitor and having to reconfigure it?
00:13:41.600 --> 00:13:46.080
But there's also another consideration in that there are different monitor types.
00:13:46.080 --> 00:13:52.000
There's LCD, there's OLED, there's QD OLED, and there's innate differences between them.
00:13:52.000 --> 00:13:59.200
And without fully understanding the differences between those monitors, it's also hard to calibrate between those, correct?
00:13:59.360 --> 00:14:00.559
Yeah, it can be difficult.
00:14:00.559 --> 00:14:07.440
And there's a lot to be said for choosing a technology that balances accuracy, but also wow factor, right?
00:14:07.440 --> 00:14:10.080
Especially if you have clients who are ever coming in the room.
00:14:10.080 --> 00:14:15.519
So the nice thing about LCDs, static backlight LCDs, is that they tend to be quite linear.
00:14:15.519 --> 00:14:22.080
They tend to be very easy to calibrate typically, but they usually don't offer the kind of wow factor that clients are looking for.
00:14:22.080 --> 00:14:24.879
OLEDs, on the other hand, have a lot of wow factor.
00:14:24.879 --> 00:14:30.480
Typically, a client will see something on an OLED and they're looking at that contrast and like, wow, my content looks really good.
00:14:30.480 --> 00:14:45.519
The issue with OLEDs are that there are lots of different OLED technologies, and some like W OLED tend to do an okay job of mimicking kind of reference behavior at SDR levels, but for HDR levels, you do have this kind of color volume collapse issue to worry about.
00:14:45.519 --> 00:14:47.600
You don't have that with RGBO LEDs.
00:14:47.600 --> 00:14:53.519
The problem with RGBO LEDs is most of them historically have not gotten bright enough to be used for HDR applications.
00:14:53.519 --> 00:14:54.559
There were some exceptions.
00:14:54.559 --> 00:14:58.720
The X300 from Sony was an exception, but it's also hyper-expensive.
00:14:58.720 --> 00:15:08.960
QDO LED is kind of really interesting because it gives you a lot of the benefit of RGBO LED, the wow factor that people have had from WO LEDs and other OLED technologies, but it is a purely additive system.
00:15:08.960 --> 00:15:12.399
So it's RGB additive for white, you don't have the color volume collapse.
00:15:12.399 --> 00:15:14.720
So it is more of a reference technology.
00:15:14.720 --> 00:15:18.639
And it's an OLED technology that's available in a wide variety of sizes.
00:15:18.639 --> 00:15:35.919
And that's always been kind of a sticking point with a lot of other technologies, whether that's light modulating cell layer, you know, those 31-inch, 1000-knit capable displays that have been used as reference monitors, or if you've had, you know, a variety of different technologies, a lot of times they're stuck with just one or two sizes.
00:15:35.919 --> 00:15:37.919
So QDO LED is really great.
00:15:37.919 --> 00:15:45.759
And one of the reasons we've standardized on it is that we can use it all the way down to a 26.5 inch and all the way up to a 65-inch monitor.
00:15:45.759 --> 00:15:54.960
So that means you can get large monitors for post, you can get smaller monitors for set, you can get kind of in-between sizes for editorial, for VFX, all these different applications.
00:15:54.960 --> 00:16:08.080
And so that I think is kind of new in the industry because typically we've had this issue of pairing a one technology as the consumer or the client viewing monitor, and then another one as the mastering monitor.
00:16:08.080 --> 00:16:10.240
And then you get into the even more challenging thing.
00:16:10.240 --> 00:16:20.639
If you think calibrating one display by yourself is hard, now try doing two technologies or two different technologies in the same room and trying to get them to visually or perceptually match each other.
00:16:20.639 --> 00:16:21.679
It's a whole other thing.
00:16:21.679 --> 00:16:28.799
I mean, they're calibrators who made great money just going around and doing nothing but this, like specializing in perceptually matching two different technologies.
00:16:28.799 --> 00:16:33.600
But an even easier path is to either use two displays of the same technology.
00:16:33.600 --> 00:16:41.440
So if you have a small QDO LED and a large one, or even better, uh, because QDO LED has such great off-access viewing, just use one monitor in the room.
00:16:41.440 --> 00:16:49.360
And that's something we really haven't been able to do previously because there weren't large monitors that were large enough for client viewing that you could trust as your primary reference.
00:16:49.360 --> 00:16:54.240
And now you can because QDO LED offers reference performance in these ginormous sizes.
00:16:54.480 --> 00:17:01.840
Yeah, I noticed recently you guys uh now have up to 65-inch QDO LEDs, which I think is absolutely incredible.
00:17:01.840 --> 00:17:04.799
And all the way down to what's the smallest size you guys have for QDO LED?
00:17:05.359 --> 00:17:06.400
26.5 right now.
00:17:06.400 --> 00:17:10.319
So we have 26.5, a 31.5, a 55, and a 65.
00:17:10.319 --> 00:17:11.759
So four sizes to choose from.
00:17:12.000 --> 00:17:16.880
So the whole range essentially, you don't have to sacrifice on the different display technology.
00:17:16.880 --> 00:17:21.279
Now, uh, can you explain a little bit more about perceptual matching?
00:17:21.279 --> 00:17:23.440
Because I'm not sure everybody is familiar with that.
00:17:23.440 --> 00:17:27.920
Like what is that and how is that different from just plain old calibration?
00:17:28.160 --> 00:17:45.440
Well, so fundamentally, you would think that if you had given two displays, if you calibrate them to the same measured values using even a very high-end probe, like a spectroradiometer or something like this, if you measure the same XY, for example, for white, you would think that they should look the same.
00:17:45.440 --> 00:17:59.440
The problem is that if the spectral power distribution, kind of the underlying fingerprints of the monitor, the light output from the monitor is different, you can get into situations where they might measure the same, but they don't actually look the same.
00:17:59.440 --> 00:18:02.319
And there are a few different things that go into this.
00:18:02.319 --> 00:18:11.519
So one of them is that unfortunately, all of our modern-day video standards are still based around CIE 1931 color matching functions.
00:18:11.519 --> 00:18:15.599
And as the name implies, those are almost 100 years old at this point.
00:18:15.599 --> 00:18:19.279
Uh, and those color matching functions aren't perfectly accurate, right?
00:18:19.279 --> 00:18:32.640
So, what you'll get is that if you use that color science, that underlying color matching function, if you rely on that to get a match between different types of spectral power distributions, different types of display technologies, they won't actually match.
00:18:32.640 --> 00:18:36.079
Now, there are more recent color matching functions.
00:18:36.079 --> 00:18:43.839
Most of the industry has heard about things like the Judd Voss offset and whatnot, which is kind of a halfway implementation of an alternate color matching function.
00:18:43.839 --> 00:18:55.119
But there have been newer color matching functions that color scientists have worked on that gets us better agreement between different types of spectral power distributions, or again, light sources from displays, for example.
00:18:55.119 --> 00:18:56.720
And those have helped.
00:18:56.720 --> 00:19:10.960
But even with the newest ones, you still will get into situations where even the best color matching functions available will still fail to provide a perfect match between different display technologies for all observers.
00:19:10.960 --> 00:19:13.119
And the question is, well, why does that happen?
00:19:13.119 --> 00:19:17.519
And the answer is that there is genuine observer variability, right?
00:19:17.519 --> 00:19:20.319
So you might see color slightly different than I see color.
00:19:20.319 --> 00:19:22.480
So we can do a perfect match for me.
00:19:22.480 --> 00:19:27.440
I could have bronze color matching functions, and those will match any two displays well if we use that.
00:19:27.440 --> 00:19:31.519
But then Jason walks in the room and maybe he's a slightly different type of observer.
00:19:31.519 --> 00:19:38.640
And so these color matching functions are always kind of averages of how people see, but there is genuine variability.
00:19:38.640 --> 00:19:53.759
So perceptual matching is this process of hey, we use the best color science available, we get things as close as possible as a starting point, and then we basically look at some footage, look at some test patches, and we fine-tune one display to be a better perceptual match.
00:19:53.759 --> 00:19:59.359
So they won't measure the same anymore, but they will look the same to that operator.
00:19:59.359 --> 00:20:02.160
And that is actually quite effective and it works quite well.
00:20:02.160 --> 00:20:15.599
The conundrum here, and the reason I prefer to have a single display in the room if I can, as opposed to multiple displays, is that again, I can make that match perfect for you, but a client may walk in the room and they may see slightly differently.
00:20:15.599 --> 00:20:30.559
They may be an older observer who has more macular degeneration, or they have more yellowing of the lens in their eye, or they may just be genetically different and see slightly different because the distribution of uh rods and codes in their eye might be different than yours.
00:20:30.559 --> 00:20:34.160
So those things make perceptual matching quite difficult.
00:20:34.160 --> 00:20:37.839
Now, what's a lot easier is you have one display in the room.
00:20:37.839 --> 00:20:43.920
If that's all anybody looks at, then typically these discussions about which display is right kind of go away, right?
00:20:43.920 --> 00:20:45.519
Because there's nothing else to reference.
00:20:45.519 --> 00:20:51.839
And it doesn't matter so much at that point if you see slightly differently from me, because we're all just looking at this one display.
00:20:51.839 --> 00:20:56.160
And what the human eye is really good at is these comparative judgments, right?
00:20:56.160 --> 00:20:59.920
You put two displays next to each other and even small differences you can see.
00:20:59.920 --> 00:21:05.119
If you're just looking at one thing without anything else to reference, your eye's not a spectro, right?
00:21:05.119 --> 00:21:07.920
It doesn't go, oh, well, that X coordinate's off by blah, blah, blah.
00:21:07.920 --> 00:21:10.400
No, it's just like, yeah, that looks pretty normal to me.
00:21:10.400 --> 00:21:22.880
So unless you're like a severely anomalous viewer, if we're all just looking at one display technology and it's calibrated correctly, these issues of needing to perceptually match or or concerns about actually a display tend to go away.
00:21:23.119 --> 00:21:26.000
That's a really, I mean, I'm glad that we're starting to go away from that.
00:21:26.000 --> 00:21:38.799
I've heard of a lot of people having to match Sonys to Sonys, and even there they're really close, but it's painful because especially when you have a client in the room, especially if the client is female, because females tend to have better color accuracy than males just genetically.
00:21:38.799 --> 00:21:44.319
And it's difficult to overcome that in the sense that what are you supposed to do if they see differently?
00:21:44.319 --> 00:21:50.079
Well, you can make it so that it looks better for them, but I mean, it's you're just arguing against the wall there.
00:21:50.400 --> 00:21:59.359
We've been at this game of having to perceptually match for a long time because there was simply no other avenue because the large displays were always one technology and the smaller displays were another.
00:21:59.359 --> 00:22:02.559
But it And again, with QDO LED, that is no longer the case.
00:22:02.559 --> 00:22:15.440
You can get two different sizes, but even better, just because even if they're the same technology, the other thing we have to be cognizant of is that you may see color differently based on how much of your field of view is occupied by the display, right?
00:22:15.440 --> 00:22:28.640
You could have two entirely similar SPDs, but if you're viewing a 55-inch that's occupying a huge chunk of your field of view versus a 31 that's occupying a much smaller field of view, you can still get perceptual differences there.
00:22:28.640 --> 00:22:31.279
So having one display just makes life a lot easier.
00:22:31.519 --> 00:22:32.000
Absolutely.
00:22:32.000 --> 00:22:35.119
And I mean, that's really nice that we've come such a long way.
00:22:35.119 --> 00:22:41.200
And I mean, obviously, you still have to make differences when you're doing for a theatrical distribution versus uh home.
00:22:41.200 --> 00:22:46.640
But I mean, we now need to have two different monitors in the room now, other than SDR and HDR.
00:22:46.640 --> 00:22:48.559
You can essentially flip the same monitor over.
00:22:48.799 --> 00:22:49.200
Exactly.
00:22:49.440 --> 00:22:53.200
So that's a significant improvement from even, would you say, two years ago?
00:22:53.440 --> 00:22:54.000
Yeah, yeah.
00:22:54.000 --> 00:22:59.440
I mean, we were basically two, two and a half years ago is when we started doing large format QDO LEDs.
00:22:59.440 --> 00:23:00.880
And it's difficult, right?
00:23:00.880 --> 00:23:03.200
Because it changes how you design rooms.
00:23:03.200 --> 00:23:07.200
So it doesn't work in every room and realize a lot of rooms are still going to have to use two displays.
00:23:07.200 --> 00:23:12.880
But for people designing new rooms, there's this at least this opportunity to go to a single monitor room.
00:23:12.880 --> 00:23:27.359
And we're starting to see some adoption of that where it's become a little bit more favorable and it really solves this problem of the colorist having to play color scientist and psychologist to their client and say, don't worry that these don't look the same.
00:23:27.359 --> 00:23:30.160
We're gonna, you know, your content is safe with me.
00:23:30.160 --> 00:23:40.079
If you just look at one display, I mean, I can't tell you how much heard people say that just simplifies their lives because all those discussions go away, and we can really focus on what we care about, which is, hey, what's your creative intent here?
00:23:40.079 --> 00:23:41.119
How do you want this to look?
00:23:41.279 --> 00:23:41.839
That's amazing.
00:23:41.839 --> 00:23:46.880
And I think a lot of people should take that in mind if you're doing a new color suite to try and do a single monitor if possible.
00:23:46.880 --> 00:23:48.160
I think it's a lot easier.
00:23:48.160 --> 00:24:19.440
It's obviously I've worked my entire career with having a close monitor and obviously the client having a further one unless you're doing a projector, but it is something I would love to move away from because it's so much easier to not have to worry about that difference and knowing that the client is looking at a less, almost an inferior image because you are settling down with they're probably looking at a consumer display, or even if it was a reference, it's a reference of a different technology, and you're having to do in your head the difference of slightly more green or it's less contrasty or whatever the difference in the tech is.
00:24:19.440 --> 00:24:20.160
Exactly.
00:24:20.160 --> 00:24:20.559
Yeah.
00:24:20.559 --> 00:24:28.240
So where do you see monitor technology going now that we have sort of settled in on a single technology?
00:24:28.240 --> 00:24:30.720
We've been seeing it get brighter and brighter.
00:24:30.720 --> 00:24:33.119
Where do you see us going now?
00:24:33.119 --> 00:24:38.640
Do you see us continuing to get brighter and brighter up to the 10,000 nits, or do you see us sort of settling down?
00:24:38.640 --> 00:24:40.640
You already have a 4,000 nit monitor.
00:24:40.640 --> 00:24:44.480
Do you think we're gonna sort of settle there, or do you think we're gonna continue to get brighter?
00:24:44.720 --> 00:24:50.720
I think there's gonna be a push for um for brighter displays, especially in the small sizes, right?
00:24:50.720 --> 00:24:56.720
So you're right, we we now have a 55-inch that is capable of 4,000 nits peak uh luminance on that display.
00:24:56.720 --> 00:24:58.799
But on the smaller ones, we're still at 1,000 nits.
00:24:58.799 --> 00:25:02.480
So I think there will be a push to start making those brighter and brighter if possible.
00:25:02.480 --> 00:25:04.880
I do think 4,000 nits is a good benchmark.
00:25:04.880 --> 00:25:13.119
I think there are severe diminishing returns after 4,000 nits, considering the additional cost that goes into trying to get above that.
00:25:13.119 --> 00:25:21.119
And when I say cost, I don't necessarily mean just financially, but also in terms of power consumption, heat generation, these types of things.
00:25:21.119 --> 00:25:25.680
And we see that a lot of clients have actually quite happy at a thousand nits, even, right?
00:25:25.680 --> 00:25:28.000
So 4,000 nits is already a stretch.
00:25:28.000 --> 00:25:31.359
So I think there is some motivation to go brighter and brighter.
00:25:31.359 --> 00:25:34.400
There's more you can do creatively if you're given that opportunity.
00:25:34.400 --> 00:25:42.720
And one thing that I do like about QDOLA technology, which obviously we've anchored ourselves to quite a bit here, is that it has evolved sharply, right?
00:25:42.720 --> 00:25:46.319
So a little over two years ago, we introduced a 2,000 net 55 inch.
00:25:46.319 --> 00:25:48.880
Now we're already at a 4,000 net 55 inch.
00:25:48.880 --> 00:25:51.119
So there's more and more sizes coming.
00:25:51.119 --> 00:25:56.319
We've seen that we started at a 55, then we introduced a 65, then a 31, then a 26.
00:25:56.319 --> 00:26:00.079
So we've gone from one size to four in just a couple of years.
00:26:00.079 --> 00:26:02.480
We've gone to writer, obviously.
00:26:02.480 --> 00:26:06.000
But QDO led it's another important kind of distinction, right?
00:26:06.000 --> 00:26:17.839
So the things that I like about it, obviously, off-access viewing, the fact that it's purely RGB additive for white, the ubiquity of the technology, the fact that it's not just for our industry and that you can get it as consumer devices.
00:26:17.839 --> 00:26:19.839
So there's this huge market for it.
00:26:19.839 --> 00:26:25.119
And then the fact that the technology keeps evolving and evolving at a rather rapid pace.
00:26:25.119 --> 00:26:27.519
And I think those are all healthy signs for the technology.
00:26:27.519 --> 00:26:34.960
Because one thing our industry has been plagued with is that panel manufacturers, semiconductor manufacturers will make something very niche for us.
00:26:34.960 --> 00:26:43.599
They'll make a full array local dimming backlight, or we'll have to make our own full array local dimming backlight, or they'll do a light modulating cell layer panel or an RGB OLED panel.
00:26:43.599 --> 00:26:45.759
And these things are only ever sold in our market.
00:26:45.759 --> 00:26:50.079
And the problem is our market's not big enough to sustain these semiconductor manufacturers.
00:26:50.079 --> 00:26:58.960
So they all give up because they go, look, we can't run a panel production line that is only making hundreds or thousands of displays a year.
00:26:58.960 --> 00:27:02.160
We need to be making hundreds of thousands or millions if we can.
00:27:02.160 --> 00:27:10.720
And with QDOLED, because there are so many applications, you know, medical, GUI monitors, gaming monitors, all these different TVs, outdoors, everything.
00:27:10.720 --> 00:27:13.359
There's all these other applications for it.
00:27:13.359 --> 00:27:19.680
It really means that they are able to do that sort of volume that means the technology is not going anywhere.
00:27:19.680 --> 00:27:25.759
So that should give us a little bit of peace of mind in our industry because we've seen so many technologies come and go.
00:27:25.759 --> 00:27:27.680
This has a lot of staying power.
00:27:27.680 --> 00:27:36.240
And then looking at the long-term horizon, prognosticating a bit here on what is to come, I think the holy grail of display technologies, and this can change.
00:27:36.240 --> 00:27:39.839
These things can evaporate overnight and something else new comes along.
00:27:39.839 --> 00:27:49.039
But ELQD, based on what I've seen, is kind of the most interesting kind of potential future technology, probably not commercially available for at least another five years.
00:27:49.039 --> 00:27:51.519
But that's electroluminescent quantum dot.
00:27:51.519 --> 00:27:56.720
So instead of using an OLED layer to excite the quantum dots, you actually just use electrons directly.
00:27:56.720 --> 00:28:00.319
So you essentially light up the QDs directly.
00:28:00.319 --> 00:28:06.480
And what's cool about that is that it's going to offer a lot of display performance benefits.
00:28:06.480 --> 00:28:18.160
But because it's still ultimately quantum dot-based, it's going to make for a really smooth transition from QDOLA to ELQD if that ever comes, because the SPDs will be relatively similar.
00:28:18.160 --> 00:28:28.480
That's a really kind of neat thing that's happening with the adoption of quantum dot technology, is that you tend to get more similarity as you get an evolution of the technology.
00:28:28.480 --> 00:28:33.839
You don't have this like sharply different SPD where suddenly everything you graded for the last 10 years looks a lot different.
00:28:33.839 --> 00:28:38.559
So I don't know that that's be my guess as to what's coming in the next five plus years or so.
00:28:38.559 --> 00:28:41.440
But in the interim, QDO LED is set to keep improving.
00:28:41.440 --> 00:28:49.599
We're going to see basically every year or two, you see further improvements in terms of peak luminance, size options, color gamut, all these sorts of characteristics.
00:28:49.839 --> 00:28:54.559
That's amazing to hear because I have a little bit of a scar tissue from the J OLED.
00:28:54.559 --> 00:29:00.559
I mean, for those that are not aware, J uh J OLED was a technology that was really perfect for the time.
00:29:00.559 --> 00:29:03.440
The LG EP950 was based off that.
00:29:03.440 --> 00:29:14.000
It only went up to like 700 nits, but they completely went under and basically they were not able to, like Bram said, produce the amount of panels that were needed to stay in business and they just folded.
00:29:14.000 --> 00:29:16.720
And so we lost those monitors entirely.
00:29:17.039 --> 00:29:18.240
Unfortunately, that happens quite a bit.
00:29:18.240 --> 00:29:28.000
I mean, you had the same thing with the Sony RGBO led panels that were used in the E250 and those types of monitors, but also in our own displays, like the DM250, CM250, which were very popular monitors.
00:29:28.000 --> 00:29:39.440
But yeah, the issue again with these semiconductor manufacturers is that they only ever got uptake from our market, maybe a little bit in medical, but there was no big consumer uptake of the technology.
00:29:39.440 --> 00:29:41.680
To give you an idea of the staying power of quantum.
00:29:41.680 --> 00:29:56.000
It's much more than you get than you've had with these other technologies, where maybe a J OLED or Sony dumped maybe $500 million or something like this into a small pilot line to get the technology out into our markets.
00:29:56.000 --> 00:29:58.400
With SDC, the manufacturer of the quantum.
00:29:58.400 --> 00:30:05.519
They've invested a lot.
00:30:05.519 --> 00:30:14.319
And again, the technology continues to evolve, which is a telltale sign that it's not a static technology and that there is a lot of growth for this technology in the future.
00:30:14.559 --> 00:30:21.599
When you say consumer devices, I think their first thought is like, oh, it's the Samsung TV and the consumer TV that we're seeing at Costco.
00:30:21.599 --> 00:30:24.079
Is that what else are you talking about with consumer?
00:30:24.079 --> 00:30:26.160
Because I know, I mean, I'm looking at two Stream Decks.
00:30:26.160 --> 00:30:29.519
If I'm not mistaken, they also use OLED panels for the dots.
00:30:29.519 --> 00:30:30.640
Is it that?
00:30:30.640 --> 00:30:35.839
Is it what else are using these OLED panels, other than obviously the first thing of our TVs?
00:30:36.160 --> 00:30:47.519
Yeah, so TVs are a big one, obviously, like you said, but also just gaming monitors uh are increasingly becoming OLED or QDO LED on the IT side of things with the standard RGBO LEDs so not quantum.
00:30:47.519 --> 00:30:53.359
But RGBO LEDs, which we also use in products like our DM160, for example, which is a 16-inch OLED.
00:30:53.359 --> 00:30:55.599
But there's a huge market for that.
00:30:55.599 --> 00:31:02.240
So SDC, who provides the QDO LED panels for the larger sizes, they're also a manufacturer of RGB OLEDs.
00:31:02.240 --> 00:31:05.200
And one of their biggest markets is actually phones.
00:31:05.200 --> 00:31:10.720
So if you look at you know iPhones, those types of things, they're all using OLED-based displays as well.
00:31:10.720 --> 00:31:16.799
So phones, tablets, regular computer displays, gaming displays, those sort of types of things.
00:31:16.799 --> 00:31:18.960
And then the other big thing is automotive.
00:31:18.960 --> 00:31:24.720
And that's something that it's something most of the semiconductor manufacturers try to cater to, but it's a hard market to break into.
00:31:24.720 --> 00:31:26.559
So this is something that J OLED tried to do.
00:31:26.559 --> 00:31:32.400
J OLED was trying to get uptake into the automotive market, but they just didn't succeed particularly well at that.
00:31:32.400 --> 00:31:40.880
But a very successful semiconductor manufacturer that's making panels for the consumer markets out there, those are the things they're looking at.
00:31:40.880 --> 00:31:45.440
They're looking at automotive, they're looking at computer displays, they're looking at TVs.
00:31:45.440 --> 00:31:46.720
Those are the big markets.
00:32:19.609 --> 00:32:27.689
That makes perfect sense for automotive because you're outside, it needs to be bright, it needs to be viewable from all different accesses, no matter what.
00:32:27.689 --> 00:32:32.649
And I mean, the colors, it would be nice if they're accurate, but it's not the first preference.
00:32:32.649 --> 00:32:35.609
Really, it needs to be bright and it needs to be viewable from off-access.
00:32:35.609 --> 00:32:36.730
That's totally.
00:32:36.730 --> 00:32:37.929
That makes perfect sense.
00:32:37.929 --> 00:32:39.210
Thank you for that clarification.
00:32:39.210 --> 00:32:42.730
Because I think the first thing we think about is oh, it's TVs are selling pretty well.
00:32:42.730 --> 00:32:46.569
I don't know why this is doing it, but I think that's just the first thought about it.
00:32:46.569 --> 00:32:52.649
When in fact, I think phones and automotive and other uses, like you're saying, are actually a bigger seller.
00:32:52.889 --> 00:32:53.289
Absolutely.
00:32:53.289 --> 00:32:54.490
I mean, those are huge markets.
00:32:54.490 --> 00:33:00.329
I mean, the TV business is also not tiny, obviously, but it's a very cutthroat business, very, very thin margins.
00:33:00.329 --> 00:33:16.169
And so what you see is a lot of it actually has been a little good for us in that what's happened is these premium panel suppliers like SDC have started to focus more on these premium technologies and trying not to compete in the commodity space of like really cheap LCDs.
00:33:16.169 --> 00:33:20.089
You know, they're not interested in making panels that are going into $300 TV.
00:33:20.089 --> 00:33:30.569
What they're interested in doing is these higher quality, premium kind of tier products and then eventually working the price down over time so that they can start to sell into some of these more commodity spaces.
00:33:30.569 --> 00:33:35.129
But they always start kind of tip of the spear, you know, what's the kind of highest level of performance we can get?
00:33:35.129 --> 00:33:37.369
And those are the markets that they try to cater to first.
00:33:37.609 --> 00:33:39.609
So let's jump into something a little bit different.
00:33:39.609 --> 00:33:48.329
When you guys are starting to create a new monitor, what is the process that you guys go to to like essentially start doing that?
00:33:48.329 --> 00:33:57.369
Because I've asked you guys for a while, like before you guys came out with the 31-inch, I had a DM240 and I was like, I love this thing, but I need something bigger and in 4K.
00:33:57.369 --> 00:34:00.250
And for years you guys were not able to do that.
00:34:00.250 --> 00:34:01.849
And I get that now.
00:34:01.849 --> 00:34:03.210
It makes complete sense.
00:34:03.210 --> 00:34:05.049
So what are you guys looking for?
00:34:05.049 --> 00:34:10.329
Other obviously you need the panel, but then you start making that custom board, correct?
00:34:10.490 --> 00:34:17.369
Yeah, I mean, really, we can build a monitor or prototype a monitor very quickly once we find the right panel.
00:34:17.369 --> 00:34:24.250
There's really the the biggest thing for us is panel evaluations, looking for new panel technologies, and then trying to get access to it.
00:34:24.250 --> 00:34:36.409
Because that's not a trivial matter for our business as well, because it's very hard to get the attention of somebody selling hundreds of thousands of panels to a TV manufacturer when you go, hey, we'd love to buy 500 or a thousand of these.
00:34:36.409 --> 00:34:38.409
They kind of look at you like, oh, that's cute.
00:34:38.409 --> 00:34:39.450
So it's difficult.
00:34:39.450 --> 00:34:47.769
And so part of the process is one, evaluating things, making sure that they're good enough for our market because we do have probably the toughest clients out there, right?
00:34:47.769 --> 00:34:51.769
So people making color critical decisions that are the most discerning users out there.
00:34:51.769 --> 00:35:02.090
So panel evaluation is the real big one, and then also talking to these semiconductor manufacturers and trying to kind of sell ourselves in terms of, you know, what's the benefit of working for us?
00:35:02.090 --> 00:35:07.930
Well, you get the prestige of working or selling monitors into the really high-end Hollywood space.
00:35:07.930 --> 00:35:10.490
They all kind of want that prestige factor.
00:35:10.490 --> 00:35:13.289
And then the other thing is just the feedback that we offer, right?
00:35:13.289 --> 00:35:22.250
Again, because we have some of the toughest clients and because we are rather tough in our evaluation metrics, we go back to them and sell them all the things that we don't like and that we wish they would improve in the future.
00:35:22.250 --> 00:35:30.170
And so I think we have some key partnerships with that, with a few different semiconductor manufacturers who make the panels, and that's been very beneficial.
00:35:30.170 --> 00:35:31.690
That's usually a long process.
00:35:31.690 --> 00:35:37.210
So it took us 18 months of meetings before we had access to uh quantum.oled panels to give you an idea.
00:35:37.210 --> 00:35:41.610
And in that time, we were, of course, doing evaluations as well and seeing if they met our needs.
00:35:41.610 --> 00:35:46.410
Once we do find a good panel, we try to typically slot it into an existing architecture.
00:35:46.410 --> 00:35:48.410
So the DM series is a great example, right?
00:35:48.410 --> 00:35:51.690
So the DM architectures, it's a very old platform, actually.
00:35:51.690 --> 00:35:53.370
It's been around for quite a while.
00:35:53.370 --> 00:35:59.769
But as we find new panels, those typically slot in pretty easily, and there's usually not too much additional work there.
00:35:59.769 --> 00:36:03.370
We kind of optimize the autocal system for this display technology.
00:36:03.370 --> 00:36:04.650
That takes a little bit of time.
00:36:04.650 --> 00:36:07.450
The XMP series were kind of interesting, that was new for us.
00:36:07.450 --> 00:36:15.930
We really want to develop a new platform to really support what the things that we wanted to do with 4K UHD resolution, HDR displays.
00:36:15.930 --> 00:36:19.450
So that was a good bit of ground up work to redo those.
00:36:19.450 --> 00:36:26.490
And honestly, one of the most time-consuming things on the XMP series was the optimization of the Gaia Color Auto Cal system.
00:36:26.490 --> 00:36:32.250
So that system takes over a thousand readings during the process, and we build out all the targets from there.
00:36:32.250 --> 00:36:37.289
Like I mentioned, this one master calibration essentially that takes you to all positions on the monitor.
00:36:37.289 --> 00:36:51.610
But one of the key things to do, and one of the reasons we're able to make Gaia Color work so efficiently and target these places very accurately, or these targets very accurately, is that we spend a lot of time optimizing the test patch sequence, right?
00:36:51.610 --> 00:36:55.610
So if you look at how AutoCal runs, it's not just a random sequence of events, right?
00:36:55.610 --> 00:36:59.130
We actually start with the monitor kind of at SDR levels.
00:36:59.130 --> 00:37:02.490
We take a certain number of readings, we know what patches we need to read more.
00:37:02.490 --> 00:37:05.050
So we read more low-end patches than high-end patches.
00:37:05.050 --> 00:37:11.610
And then we go through also additional stabilization and warm-up steps that are all automated as part of the process.
00:37:11.610 --> 00:37:29.530
So we stabilize before we start, we warm up before we start for 10 minutes, then the process starts, we generate the SDR test patches, then we bump up a gain level on the monitor and we basically take another series of readings, but we allow for a certain cool down and warm-up period, all things that you have to basically.
00:37:29.530 --> 00:37:36.970
If you are an independent calibrator trying to do the same thing, you could do this, you could learn how to do this, but it would take a lot of trial and error.
00:37:36.970 --> 00:37:44.650
And so that's why I always feel bad for the independent calibrator because you have to be able to calibrate every model of monitor out there on the market.
00:37:44.650 --> 00:37:50.090
Whereas with us, we have kind of the unfair advantage, if you want to think about it, in that we know exactly what it is we're calibrating.
00:37:50.090 --> 00:37:59.370
So we can do a lot of study and research beforehand and know what test patches are worth measuring, you know, what are what's wasted to, you know, test patches, what don't we need to measure?
00:37:59.370 --> 00:38:03.450
And then also we can optimize for being able to use more affordable probes.
00:38:03.450 --> 00:38:05.450
So that's another huge step that we took.
00:38:05.450 --> 00:38:12.090
So when we first released the XMP series, it was only compatible with higher-end probes like the CR100, like Klein K10A.
00:38:12.090 --> 00:38:17.690
But a big goal for us was how can we make like a sub-300 probe work as well?
00:38:17.690 --> 00:38:23.050
And so when you have those sub-300 probes, there are some limitations there, especially on what they can do in the low end.
00:38:23.050 --> 00:38:26.809
So again, we have the unfair advantage that we know what we're measuring.
00:38:26.809 --> 00:38:33.370
So, unlike having to use third-party software, what we can do is we can go, look, we've measured thousands of these at this point, right?
00:38:33.370 --> 00:38:42.890
We know that if they track to this point to the noise floor of that probe, we know that because of how they were tracking above it, it's almost certainly going to track this way below that noise floor.
00:38:42.890 --> 00:38:44.570
So we can supplement that data.
00:38:44.570 --> 00:38:49.370
And it really allows us to get very good results in something that costs almost nothing, right?
00:38:49.370 --> 00:38:53.130
For 280 bucks, no software required, plug it into the monitor, does a trick.
00:38:53.130 --> 00:38:57.530
The only downside being that still at that point is that those probes are slower, right?
00:38:57.530 --> 00:38:59.370
So it's gonna take you longer to calibrate.
00:38:59.370 --> 00:39:05.050
If you own a facility with 100 monitors, you probably do want to invest in the higher end probe and save yourself the time and energy.
00:39:05.050 --> 00:39:11.530
But if you get just one or two displays, then yeah, just run it overnight, who cares, you know, or go out to lunch while you're calibrating and then you're good to go.
00:39:11.530 --> 00:39:14.650
And you've only spent 280 bucks on the probe to do the calibration.
00:39:14.650 --> 00:39:17.050
So those are some of the steps that we've gone through.
00:39:17.050 --> 00:39:22.650
And then once the platform is kind of out there and we have again, calibration is always our number one thing, right?
00:39:22.650 --> 00:39:25.289
Signal support, calibration, that's where we start.
00:39:25.289 --> 00:39:27.530
Then once we have that, we start listening to those feedback.
00:39:27.530 --> 00:39:28.570
What do users want?
00:39:28.570 --> 00:39:32.809
And that's where this, you know, quad view thing came up with different color spaces for DITs.
00:39:32.809 --> 00:39:43.370
DITs kept saying, look, I love this display, but I can't afford, or even if I could afford, I don't want to roll around on a cart with four to eight monitors on it to support the four to eight cameras I had to monitor.
00:39:43.370 --> 00:39:49.289
So like, I would love to have one or two displays where I can monitor either four or eight signals across these couple of displays, for example.
00:39:49.289 --> 00:39:54.090
And so we started listening to those things and what people want and then add features that way over time.
00:39:54.090 --> 00:40:44.950
And again, one of the beautiful things about not being locked into kind of a consumer chipset because everything's built on these FPGs and custom code, is that we can rewrite things, we can add to the code, we can do these new features that might not be possible on other platforms because it is a very, very flexible system for us.
00:40:45.270 --> 00:40:47.750
That's a really, really good description.
00:40:47.750 --> 00:40:58.070
One of the things that I did want to ask you about is something that I think a lot of colorists will be very familiar with if they've ever tried to purchase a consumer-style monitor, and that's what is known as the panel lottery.
00:40:58.070 --> 00:40:58.870
Yeah.
00:40:58.870 --> 00:41:03.030
So how do you guys deal with that internally?
00:41:03.030 --> 00:41:05.830
I'm sure you guys reject a lot of panels.
00:41:05.830 --> 00:41:09.190
Have you ever like measured a percentage of how many panels you turn away?
00:41:09.430 --> 00:41:11.750
So, yeah, it really depends on the technology.
00:41:11.750 --> 00:41:16.070
One of the reasons we were really happy to, because we did some W O LED panels for a while.
00:41:16.070 --> 00:41:20.790
So 55 and 65 inch W O LEDs, which we never really marketed as HDR monitors.
00:41:20.790 --> 00:41:27.030
We said, look, these were SDR reference mastering monitors, and you could use them for like HDR previews, what we'd call it, right?
00:41:27.030 --> 00:41:28.790
But they weren't HDR mastering monitors.
00:41:28.790 --> 00:41:37.830
Even with that caveat, near the end of our run in using these W O LED panels, we were rejecting over 50% of the panels that were being delivered to us.
00:41:37.830 --> 00:41:41.030
So, which is not a figure we like to see because it's costly, right?
00:41:41.030 --> 00:41:50.230
Because we have to go through the process of evaluating all these displays, sending them back to the display manufacturer, arguing with them about things that don't meet our basic requirements.
00:41:50.230 --> 00:41:53.910
We are much luckier when it comes to QDO LEDs.
00:41:53.910 --> 00:41:56.230
The rejection rate on those is much, much lower.
00:41:56.230 --> 00:41:58.070
The quality of those is much better.
00:41:58.070 --> 00:41:59.110
But things happen, right?
00:41:59.110 --> 00:42:03.670
We'll get a panel that just you know doesn't meet the uniformity benchmarks we want or whatever it may be.
00:42:03.670 --> 00:42:07.750
So we do look at, we evaluate all panels before we put them in monitors.
00:42:07.750 --> 00:42:09.030
And it depends a bit.
00:42:09.030 --> 00:42:14.150
You know, some tiers of monitor, we have like really entry-level monitors where our tolerances won't be as tight.
00:42:14.150 --> 00:42:30.710
So if you buy a $2,500 monitor from us, yeah, we're not going to put kind of the same limitations on that that we'll put on a $24,000 monitor, but we do spend an inordinate amount of time uh actually looking at the panels, evaluating them, and not just slapping whatever in and sending it out into the world.
00:42:30.710 --> 00:42:32.710
So yeah, it's a lengthy process.
00:42:32.710 --> 00:42:44.150
The other thing that we do that's kind of related to this, that's different from a lot of consumer brands and not all consumer brands, some consumer brands do some of this, but we do literally calibrate every individual display.
00:42:44.150 --> 00:42:48.310
We're not loading some generic calibration into it and saying good luck.
00:42:48.310 --> 00:42:52.470
We are taking the time to do this auto cal process, run it.
00:42:52.470 --> 00:43:01.270
And then after we're done with that auto cal process, we do the additional evaluations in standalone software with a completely separate probe.
00:43:01.270 --> 00:43:05.590
And then we send that, those monitors to you with those calibration reports, right?
00:43:05.590 --> 00:43:14.390
So we check the HDR performance, we check the SDR performance to make sure that that process went well and that you have a verified calibration report showing that it was done accurately.
00:43:14.390 --> 00:43:23.190
In addition to that, we also spend a lot of time not just with, so we use CalMan to generate those reports internally, but we also do a lot of research using tools like color space.
00:43:23.190 --> 00:43:23.590
Yeah, right.
00:43:23.590 --> 00:43:32.790
So we'll do a lot of display evaluation internally and we'll try to figure out, you know, how is a display behaving in a certain region or operating mode or whatever it may be.
00:43:32.790 --> 00:43:37.030
And then we'll use that to fine-tune things in our own guy of color auto cal process, of course.
00:43:37.030 --> 00:43:39.030
But yeah, a lot of time and effort goes into it.
00:43:39.030 --> 00:43:50.390
Again, part of that QC process is we use one probe to do the actual calibration, then we actually use a completely independent probe to verify because something that we've seen in the past is how do you know your one probe isn't broken?
00:43:50.390 --> 00:43:55.990
So uh so we we cross-check our reference probes, our reference spectrodes against each other weekly.
00:43:55.990 --> 00:44:01.750
Again, we use independent probes for calibration and QC because if something's happened with those probes, you'll see that in that process.
00:44:01.750 --> 00:44:03.750
You go, oh, wait a minute, something's not matching up here.
00:44:03.750 --> 00:44:04.630
And we've had that happen.
00:44:04.630 --> 00:44:12.790
We've had cleaner who was in in the evening accidentally knock over a probe and then you check it in the morning and you're like, oh wait, suddenly this off, the diffraction grading is shifted or whatever.
00:44:12.790 --> 00:44:15.910
So we take a lot of care and effort to make sure that those things are correct.
00:44:15.910 --> 00:44:17.990
And again, it all boils down to the same thing.
00:44:17.990 --> 00:44:19.190
What are we trying to achieve?
00:44:19.190 --> 00:44:24.070
Confidence for ourselves, confidence for our end users that they know that the monitor is correct.
00:44:24.230 --> 00:44:24.710
I love it.
00:44:24.710 --> 00:44:32.310
And honestly, that's what I think it's all about is you guys put so much care and effort into making sure that these monitors are reference grade.
00:44:32.310 --> 00:44:34.950
There's really a lot that goes behind that word reference.
00:44:34.950 --> 00:44:40.710
And if I'm not mistaken, you guys will give a lifetime calibration for your monitors if you send it back.
00:44:40.950 --> 00:44:41.510
Yeah, absolutely.
00:44:41.510 --> 00:44:42.630
So we still offer that.
00:44:42.630 --> 00:44:52.870
We don't have a ton of people who take us up on that for newer monitors because all of our monitors in our current lineup, the DM series and the XMP series, support our Gaia Color Auto Cal.
00:44:52.870 --> 00:45:00.470
And because all of these displays can be calibrated with a $280 probe, it usually costs more to send round trip.
00:45:00.470 --> 00:45:04.550
Certainly, if not that one time, if you do it multiple times over the life of the monitor.
00:45:04.550 --> 00:45:06.150
So almost everybody's gone to just buying.
00:45:06.150 --> 00:45:07.830
Probe and doing it in-house.
00:45:07.830 --> 00:45:17.270
However, if someone really prefers that we do it for them, you can send it to our office in Atlanta, office in Belgium, office in the UK, and we'll do a free calibration for you.
00:45:17.270 --> 00:45:20.950
Mostly, again, monitors are coming in with that are mostly the old monitors.
00:45:20.950 --> 00:45:25.350
And it used to be the fact that there wasn't a good way for us to do this on the old displays.
00:45:25.350 --> 00:45:26.550
They didn't have USB ports.
00:45:26.550 --> 00:45:28.150
You couldn't plug a probe directly in.
00:45:28.150 --> 00:45:29.590
You have to use standalone software.
00:45:29.590 --> 00:45:33.270
And so we understood that that was intimidating and that people preferred to send monitors in.
00:45:33.270 --> 00:45:35.030
And so we offered that for a very long time.
00:45:35.030 --> 00:45:36.550
We still offer it to this day.
00:45:36.550 --> 00:45:40.470
But Gaia Colors really changed the game for people, made things even easier.
00:45:40.470 --> 00:45:43.190
Because now your downtime was also a big consideration.
00:45:43.190 --> 00:45:45.990
Now you don't have to worry about being without a monitor for several days.
00:45:45.990 --> 00:45:48.470
Plug in a probe, an hour later, you're ready to go.
00:45:48.790 --> 00:45:50.070
That's really huge.00:45:50.070 --> 00:45:59.110
I think literally just being able to hook up a probe directly to the monitor without having to have a separate system with calibration software is a major consideration alone.00:45:59.110 --> 00:46:05.270
And the fact that you guys have custom configured it and have set everything up, it's a major feature that a lot of people don't look at.00:46:05.270 --> 00:46:10.230
So really congratulations on spending so much time in engineering to get that so locked in.00:46:10.230 --> 00:46:13.750
I think that is one of the key features that I love about your monitor so much.00:46:13.750 --> 00:46:17.110
And I like that you guys are still standing behind the lifetime calibration.00:46:17.110 --> 00:46:20.150
Sometimes it's just nice to have another pair of eyes on it.00:46:20.390 --> 00:46:20.870
Absolutely.00:46:20.870 --> 00:46:23.670
We're happy to do that for anybody who's ever interested in it.00:46:23.670 --> 00:46:29.590
It's all, again, all about confidence, and we're just here to try to aid our users in having that confidence in what they're looking at.00:46:29.830 --> 00:46:37.030
As we sort of get to the end of our time together, if people wanted to find out more about you and more about Flanders Scientific, where can we find out?00:46:37.350 --> 00:46:39.510
So you can go to Flanderscientific.com.00:46:39.510 --> 00:46:43.270
We have detailed information on all the monitors, all the different features that are there.00:46:43.270 --> 00:46:44.790
We have all the tech specs.00:46:44.790 --> 00:46:49.270
We even have a menu simulator, which is really great if you just want to get familiar with the monitor and how it works.00:46:49.270 --> 00:46:51.910
Uh, if you want to know how to turn something on or off, you can do that.00:46:51.910 --> 00:46:55.510
And then the other thing I would say is, you know, don't be shy about just calling us.00:46:55.510 --> 00:46:58.150
Email us, call us, get us on live chat.00:46:58.150 --> 00:47:02.150
Um, I think we're one of the most accessible companies in our space.00:47:02.150 --> 00:47:08.390
You will call us and you're never more than a few minutes away from talking to somebody who does nothing but displays for a living, right?00:47:08.390 --> 00:47:16.790
So it's not like you're calling a big uh multinational corporation who has a camera department and a router department and a and a display department.00:47:16.790 --> 00:47:18.790
We just live and breathe displays.00:47:18.790 --> 00:47:22.070
And so almost anybody you get on the phone here can answer your questions.00:47:22.070 --> 00:47:27.590
And again, we have a phone tree that is basically one for support, two for sales.00:47:27.590 --> 00:47:31.510
And that's the amount of delay you'll have before you get in touch with somebody.00:47:31.510 --> 00:47:33.510
It's not complicated, it's not a machine.00:47:33.510 --> 00:47:37.990
We're here to help you make the right decision and not to steer you to the most expensive solution, right?00:47:37.990 --> 00:47:48.070
So we are here to find out what solution is going to work for your particular application, and we try to be just really blunt and honest about what we think will meet your needs.00:47:48.310 --> 00:47:50.470
And I have to ask you a personal question.00:47:50.470 --> 00:47:54.870
Out of all of the monitors that you guys sell, what is currently your favorite one?00:47:55.110 --> 00:47:58.870
I think the new 5.51 is definitely my favorite.00:47:58.870 --> 00:48:05.670
I just think there's so much to be said for a display that can be used as your single reference monitor.00:48:05.670 --> 00:48:07.990
So I love the size of that display.00:48:07.990 --> 00:48:09.590
And I still can't get over.00:48:09.590 --> 00:48:13.110
I feel like I'm dreaming every time I look at the off-axis viewing of these things.00:48:13.110 --> 00:48:16.150
You can literally be standing, you know, completely to the side of the display.00:48:16.150 --> 00:48:18.870
There's virtually no color shift, there's no contrast drop.00:48:18.870 --> 00:48:22.870
And that's something that we saw on paper before we even got the panels to evaluate.00:48:22.870 --> 00:48:30.710
And then we started getting these things in, and we're just like, everybody just kept walking around display going, this is the weirdest thing we've ever seen because it just does not move.00:48:30.710 --> 00:48:36.150
And again, I think that's part of why I love it is because it makes room design kind of really interesting these days.00:48:36.150 --> 00:48:39.750
You can really do some creative things because your client doesn't have to sit in line with you.00:48:39.750 --> 00:48:42.150
They don't have to sit in your lap anymore to see the same thing.00:48:42.150 --> 00:48:43.750
They can really be anywhere in the room.00:48:43.750 --> 00:48:45.270
So that one's pretty impressive.00:48:45.270 --> 00:48:48.710
The fact that it gets so bright is also has a lot of wow factor.00:48:48.710 --> 00:48:52.310
But yeah, that's probably my favorite right now would be that display.00:48:52.470 --> 00:48:52.870
I love it.00:48:52.870 --> 00:48:55.830
I'll check in with you in about a year to see what your next one is.00:48:55.830 --> 00:49:00.310
Thank you so much for coming on the show and for sponsoring the podcast.00:49:00.310 --> 00:49:06.550
I really appreciate you and everyone at FSI for being so straightforward and for keeping everybody honest.00:49:06.870 --> 00:49:07.430
I appreciate it.00:49:07.430 --> 00:49:08.150
Thanks, Jason.00:49:08.310 --> 00:49:11.030
And for Color and Coffee, I'm Jason Bowdach.00:49:11.030 --> 00:49:14.870
Until the next episode, happy grading, and we'll see you guys later.00:49:14.870 --> 00:49:20.550
Be sure to follow us on Instagram, YouTube, and your podcast app of choice.00:49:20.550 --> 00:49:26.470
Search for at Color and Coffee or at Color Coffee Podcast and join the conversation.00:49:26.470 --> 00:49:30.630
If you're using Spotify or Apple Podcasts, please leave a review.00:49:30.630 --> 00:49:37.270
Huge thanks to FSI, Demystify Color, and Pixel Tools for sponsoring the show.00:49:37.270 --> 00:49:39.670
Until the next episode.
