Why not a gamut mod?

[mushbert]

So here's the situation. I've been locked on here trying to choose a new monitor, and at the moment it's 60/40 for the HP ZR24W or HP LP2475W. For me, the former's major downfall is the broken scaler, the latter's that it's wide gamut.

My CRT held out until last month, but a few years back I accepted that leaving that wonderful tech behind would be inevitable. So I've been watching long enough to see hundreds of 2nd posts state the following: "By using different backlighting, wide gamut monitors can display more colors - but for most users this is a bad thing since currently little/no consumer content reaches us that way".

The question is this: People replace malfunctioning CCFLs in (mainly laptop) displays all the time. So why can I not find any reference to someone replacing their WCCFL tubes for the sole purpose of changing to an sRGB gamut? I know results would vary, but assuming you pick the right parts and are patient+dextrous, wouldn't this solve the whole issue?

 

[RadXge]

First, physically replacing the backlight looks like a challenge by itself. We are not talking about changing a bulb here...
Secondly, the firmware would also have to be modified in order to handle the new gamut.

BTW, some wide colors gamut monitors emulate the sRGB color space with satisfying results (unless your are a professional that requires near perfect calibration settings).

 

[sethk]

Very, very, very high technical difficulty in achieving this mod and ending up with a monitor that displays regular gamut colors with low delta E color tracking errors.

 

[monitorrat]

So here's the situation. I've been locked on here trying to choose a new monitor, and at the moment it's 60/40 for the HP ZR24W or HP LP2475W. For me, the former's major downfall is the broken scaler, the latter's that it's wide gamut.

My CRT held out until last month, but a few years back I accepted that leaving that wonderful tech behind would be inevitable. So I've been watching long enough to see hundreds of 2nd posts state the following: "By using different backlighting, wide gamut monitors can display more colors - but for most users this is a bad thing since currently little/no consumer content reaches us that way".

The question is this: People replace malfunctioning CCFLs in (mainly laptop) displays all the time. So why can I not find any reference to someone replacing their WCCFL tubes for the sole purpose of changing to an sRGB gamut? I know results would vary, but assuming you pick the right parts and are patient+dextrous, wouldn't this solve the whole issue?

1. Who says it will match the internal tables that turn the HIGHLY non-linear LCD response into something CRT-like once you do that?

2. Expensive to first buy wide vs sRGB and then expensive to buy a new CCFL assembly, which also needs to match the physical set layout

3. Tricky to carry out

4. The internal tables will also expect wide gamut primaries and now you'll have pseudo-sRGB primaries and it will make a real mess there too

So why not just get sRGB to begin with???
I don't get it.
There is a huge chance it will be a mess and it will cost far more than an sRGB that will blow that mess totally out of the water.
It's like buying a Porsche because it's a cool choice but then wanting top gas mileage and a less finicky engine and ripping it out and trying to somehow cram one of those mini cars engines in the back and ending up with a very costly mess that won't perform like Porsche or a mini car and will cost more than either and almost as much as both combined.

 

[daveswantek]

What the Hell!; you might as well make your own display from scratch.

Dave

 

[mushbert]

Ok, so most of your responses have been about the technical aspects. I haven't looked into it deeply, but I don't see that it would be that difficult. You open the monitor up, unclip or unsolder the old tubes, and attach the new ones. Assuming you're using the correct sizes to replace them, there shouldn't be a whole lot of fiddling involved. I would expect the overall cleanliness of this operation to have more of an effect on the final quality: it's difficult enough to get a screen protector onto an MP3 player without trapping dust in between, let alone something with a 24" panel.

As for firmware issues, who can say that there's actually a difference other than the OSD's labels and default profiles? With access to RGB controls and a hardware calibration tool, I don't see why you would not end up with something far closer to the sRGB color space than what you started with. But then for that matter, I don't understand why manufacturers don't just release two versions of their higher-tier products.

For me, the driving issue is that I want a display that falls between "very good" and "excellent" for normal uses. The ZR24W's pricepoint is nice, but I know I'm going to feel bad about paying it for something that has poor support for non-PC sources. This is why I would be willing to shell out the extra cash for the other model, which has myriad other inputs and 1:1 scaling, but why they even implemented these on a device which performs badly with THE standard PC color space is beyond me.

I know this CCFL swap would not be an easy or cheap option, but for people dissatisfied/stuck with a WG display, it doesn't necessarily represent a huge investment. A pack of tubes for a 24" display goes for ~$60 on Ebay at the moment, but like everything else, they can probably be had elsewhere for less if you know where to look. As such it really surprises me that no one seems to have tried this. Thinking about it further, I suppose it's possible that those who were happy with wide gamut were those who read up on it first, and decided it was an acceptable compromise - while those who disliked it enough that they might have done something about it, were less technically enthused people who bought into the hype but didn't really investigate that aspect of the product.

 

[brumwald]

1. Who says it will match the internal tables that turn the HIGHLY non-linear LCD response into something CRT-like once you do that?

The topic is about replacing the backlight. Not the panel.

So why not just get sRGB to begin with???

If there were decent sRGB alternatives to all wide-gamut monitors that would of course be the best solution. That route haven't been available for many years though...

BTW, some wide colors gamut monitors emulate the sRGB color space with satisfying results (unless your are a professional that requires near perfect calibration settings).

All sRGB modes I've tried out are actually worse than just running wide gamut..



Replacing the backlight should not be that hard considering that this is a common warranty-issue. Don't know but finding an sRGB backlight thats a perfect match and that takes the same voltage etc. might be harder (guessing).

I'm intrigued by this mod. Probably won't try it out myself but it would be cool if it could be done. I don't know enough to make a qualified guess but the gamuts are not that far away from each other so I'd say it's plausible.

Originally I thought of a converter-box of some kind (using an fpga or something, haven't looked if thats even reasonable ) that takes an DVI-cable and transforms the gamut and outputs an sRGB DVI stream to the monitor. Though that solution would probably be quite expensive and I haven't seen anything to buy
And you would of course loose quality with it (effectively just throwing out all bits that are outside sRGB and 'expanding' the rest to 3x8 bits).

 

[monitorrat]

1. Who says it will match the internal tables that turn the HIGHLY non-linear LCD response into something CRT-like once you do that?

The topic is about replacing the backlight. Not the panel.

even if it maintained relative linearity it would still be a total mess

So why not just get sRGB to begin with???

If there were decent sRGB alternatives to all wide-gamut monitors that would of course be the best solution. That route haven't been available for many years though...

I don't see that the regular old wide gamut appear to be so special compared to sRGB ones.

BTW, some wide colors gamut monitors emulate the sRGB color space with satisfying results (unless your are a professional that requires near perfect calibration settings).

All sRGB modes I've tried out are actually worse than just running wide gamut..

That is hardly the case with the Dell U2410 or new EIZO, NEC PA or any of the recent releases really.

I'm intrigued by this mod. Probably won't try it out myself but it would be cool if it could be done. I don't know enough to make a qualified guess but the gamuts are not that far away from each other so I'd say it's plausible.

not remotely plausible, the spectrum of light put out it is totally different and they also apply all sorts of special filters adapted to the light source since natively they are not very usable. the internal tables will be a mess. the controls probably won't work well as you'll probably have a totally twisted and warped gamut. I'm not sure that DDC monitors generally let you calibrate at a low enough level to reset everything and give it a chance to work out. Software calibration does nothing for movies and games so what good does it do you plus even for color-managed stuff, even a many entry LUT external calibration wouldn't at likely be able to create a reasonable profile since they expect as least some reasonably response to begin with.

ught of a converter-box of some kind (using an fpga or something, haven't looked if thats even reasonable ) that takes an DVI-cable and transforms the gamut and outputs an sRGB DVI stream to the monitor. Though that solution would probably be quite expensive and I haven't seen anything to buy
And you would of course loose quality with it (effectively just throwing out all bits that are outside sRGB and 'expanding' the rest to 3x8 bits).

as you say, doing that externally you are giving up shades by rescaling certain things lower but you also still need to make 255,255,255 stay as 255,255,255 to show max bright white, etc. I still have a hard time believe that all sRGB monitors are so poor that going through the huge effort to design such a converter and taken along with loss in quality would come out better.

 

[monitorrat]

As for firmware issues, who can say that there's actually a difference other than the OSD's labels and default profiles? With access to RGB controls and a hardware calibration tool, I don't see why you would not end up with something far closer to the sRGB color space than what you started with. But then for that matter, I don't understand why manufacturers don't just release two versions of their higher-tier products.

LCD have very complex, very non-linear low-level calibration tables based precisely upon the monitor response as designed.

I suppose most of the weird stuff is just in the panel response itself, so maybe if it had full RGB controls as well as a full CMS (and many of these monitors do not), it would sort of work. I'd still be quite afraid you might end up with some sort of twisted response, especially since it would have been set to also account for the special filters needed to adjust the wide gamut CCFL. I woudln't risk $500 on it myself, but you can try.

 

[brumwald]

That is hardly the case with the Dell U2410 or new EIZO, NEC PA or any of the recent releases really.

The sRGB mode in Dell U2410 have been quite criticized, haven't looked into that monitor so don't know the specifics. NEC PA is beyond high end and came out like yesterday. Yes Eizo probably have some good stuff but they come at a price that often make NEC look cheap...

The U2410 was hyped to be the first consumer-monitor with decent sRGB mode and it didn't really cut it.

I still have a hard time believe that all sRGB monitors are so poor that going through the huge effort to design such a converter and taken along with loss in quality would come out better.

We have the quite new HP ZR24W, it has some issues but it really is a great find considering the alternatives. And thats the only decent sRGB alterntive in existance if you look beyond apple (most don't want glossy and/or displayport-mini-only and it would be nice with at least some decent ergonomics...).
Unless I've forgotten something thats the only alternatives in the whole 24"-30" range that isn't wide gamut, and it's hard as it is to pick a monitor. To throw out 98.7% of the alternatives doesn't give you a whole lot of options. No monitor is perfect to begin with so you will have to look forward to some serious compromises if you want to go with non-TN and sRGB.

And remember, the ZR24W is quite new - before that we had *nothing*.

 

[mushbert]

On a personal note I bit the bullet and ordered the ZR24W. I'm sure I'll be happy with it for PC use, but I'm not looking forward to the unavoidable scaling on the PS3's picture.

I almost want to get a cheap WG monitor to try this swap, but it probably won't happen. This thread is more about the idea though - and I'm sure there are people out there who have cheap/free sources for hardware (and more disposable income than I do). Until somebody tries it though, we'll never know how well/terribly it would work.

 

[brumwald]

You could look into external scalers, I have no experience with them and they are a bit pricey (cheaper than a WG-monitor though).

 

[mushbert]

I've thought of it. But for roughly $300, you're far worse off if it doesn't work - at least the monitor can be useful.

The cause of the problem with the ZR24W scaler is still unknown. another-user has been helpful enough to do some further PS3 tests for me while I wait for mine to arrive, and he says "[...]Almost everything is listed [in the monitor OSD] as running in H 45hz x V 60hz" (as opposed to showing a recognized preset mode). A hardware scaler might not change this, as it's purpose is resize the input, not mess with the refresh rates. Regardless, he goes on to say that some games do show as using the "1920x1080" preset mode, but the scaler options are still disabled.

So an external scaler working would depend on it providing a signal that the monitor... likes? And that's the best we can figure out so far. *shrug*

 

[Sailor_Moon]

You could look into external scalers, I have no experience with them and they are a bit pricey (cheaper than a WG-monitor though).

Can be solution but feasible devices (DVDO, Lumagen) would cost (partially a lot) more than the screen.

edit: No, my Radiance doesn't support 1200p, but the DVDO machines should support 1920x1200@60Hz (could be a fiddly thing regarding timing).

The U2410 was hyped to be the first consumer-monitor with decent sRGB mode and it didn't really cut it.

sRGB and AdobeRGB mode are quite good, the problems of the U2410 lie elsewhere.

Best regards

Denis

 

[monitorrat]

Can be solution but feasible devices (DVDO, Lumagen) would cost (partially a lot) more than the screen.

edit: No, my Radiance doesn't support 1200p, but the DVDO machines should support 1920x1200@60Hz (could be a fiddly thing regarding timing).


sRGB and AdobeRGB mode are quite good, the problems of the U2410 lie elsewhere.

Best regards

Denis

the NEC PA241W lets you adjust the WB of the sRGB simulation mode now

 

[Sailor_Moon]

the NEC PA241W lets you adjust the WB of the sRGB simulation mode now

Yes that is possible with all screens that offer a flexible color space emulation. But one has to be very careful when choosing the whitepoint. sRGB and AdobeRGB are for example defined relative to D65. If an other whitepoint is desired the primaries have to be adapted chromatically (good: Bradford) which leads to the problem that I don't know any calculator that is completely flexible regarding source and destination white (had to implement our own solution for the colorimetric tests). In our review I have at least offered the correct values for 5800K which is a "usable" whitepoint for many users.

Best regards

Denis

 

[brumwald]

sRGB and AdobeRGB mode are quite good, the problems of the U2410 lie elsewhere.

The reports of banding, grainyness and heavy dithering in the sRGB mode makes it barely usable at all according to many of the reports on the web.

It might be the first consumer product with decent colors when it comes to sRGB mode but we still don't seem to have any consumer product that can do a half-decent job of it. As it looks today buying an consumer wide-gamut monitor to display sRGB content (which 99.999% of all people will do) is vastly inferior to a native sRGB monitor.

 

[monitorrat]

Yes that is possible with all screens that offer a flexible color space emulation. But one has to be very careful when choosing the whitepoint. sRGB and AdobeRGB are for example defined relative to D65. If an other whitepoint is desired the primaries have to be adapted chromatically (good: Bradford) which leads to the problem that I don't know any calculator that is completely flexible regarding source and destination white (had to implement our own solution for the colorimetric tests). In our review I have at least offered the correct values for 5800K which is a "usable" whitepoint for many users.

Best regards

Denis

Well for sRGB mode I think most people would just chose the standard D65.

Anyway, I was more talking about Spectraview II than using multiprofiler custom modes (where if you did set a non-D65 you would have to take that stuff into account).

SVII lets you set the WB to D65 as measured by the calibration probe. (It also lets you chose other white points and I can only assume it would automatically shift the primary locations, not sure.) It doesn't appear to let you use the probe to measure and alter the sRGB locations. As best as I can tell it seems to use locked in factory presets for primary locations (although it does measure them and I assume store those in the ICC profile unless maybe they think the factory values are more solid???).

One amusing thing is that SVII locks sRGB emulation mode into a TRC of power function gamma 2.2 while multiprofiler's factory preset sRGB mode locks in an sRGB TRC. Someone at NEC is having an argument over which makes more sense .

I'm not sure why they don't allow you to chose which TRC you want. Sometimes for movies in a dark room PF gamma 2.3 might be better than 2.2 or sRGB, etc.

(and one messed up thing is that the broadcast video calibration mode of spectraview ii, which does let you freely chose any TRC locks in native primaries rather sRGB primaries which makes it all but useless since pretty much nothing other than Adobe After Effects fully color manages video, i can't figure out why they would not have picked sRGB primaries for broadcast video mode, hopefully they will fix it in the next release).

 

[monitorrat]

The reports of banding, grainyness and heavy dithering in the sRGB mode makes it barely usable at all according to many of the reports on the web.

It might be the first consumer product with decent colors when it comes to sRGB mode but we still don't seem to have any consumer product that can do a half-decent job of it. As it looks today buying an consumer wide-gamut monitor to display sRGB content (which 99.999% of all people will do) is vastly inferior to a native sRGB monitor.

does it still dither with the new firmware?

i think the U2711 one has less banding
(of course for that price you can get a NEC PA with a perfect sRGB mode)

 

[Sailor_Moon]

SVII lets you set the WB to D65 as measured by the calibration probe. (It also lets you chose other white points and I can only assume it would automatically shift the primary locations, not sure.)

I currently have a test sample of the PA271W here and will test what happens - but normally those solutions just invoke fixed transformations and are depeding on the user as "intelligence".

One amusing thing is that SVII locks sRGB emulation mode into a TRC of power function gamma 2.2 while multiprofiler's factory preset sRGB mode locks in an sRGB TRC. Someone at NEC is having an argument over which makes more sense

Curious. Handling of gradation is one of the few little things I could also criticize in Eizo Color Navigator which allows only numerical values and L* out of the software. sRGB gradation is here only possible while using color space emulation with an ICC profile with sRGB TRCs as reference.

Best regards

Denis