[Openicc] [argyllcms] Re: Helping with colord

Mon Mar 7 17:49:55 PST 2011

There is a funny part and a sad part to the EDID story.

The funny part is that standalone panels which implement EDID really
really well and trustably are usually capable of simulating sRGB or
another standard space in hardware, and so can be manually
"calibrated" by the user just telling the OS which space it is.

The sad part is that a standard got written which allows decent
automatic calibration for zero money, and now nobody apart from Apple
seems to be implementing it.

What

On Tue, Mar 8, 2011 at 2:21 AM, Chris Murphy <lists at colorremedies.com> wrote:
>
> On Mar 7, 2011, at 5:45 PM, edmund ronald wrote:
>
>> I agree that incorrect EDID is a fraud. But then as many graphics
>> cards drivers under Windows cannot read EDID, the fraud is often
>> undetectable.
>
> In which case a bogus profile cannot be made, and at least you aren't worse off than without one.
>
>> Apple EDID is correct, and actually used by software.
>> Nicomsoft makes software that reads EDID under Windows if it can be read.
>>
>> Edmund
>>
>>
>> On Tue, Mar 8, 2011 at 1:32 AM, Chris Murphy <lists at colorremedies.com> wrote:
>>> Yes. Not such a small problem. But then, incorrect EDID information is a kind of fraud, isn't it?
>>>
>>> Chris Murphy
>>>
>>>
>>>
>>> On Mar 7, 2011, at 3:39 PM, edmund ronald wrote:
>>>
>>>> compelled - hah! :)
>>>> and what instrument do you expect your average Taiwan corp will use to
>>>> measure the primaries ?
>>>> not only that, but my experiments a year ago indicated that many
>>>> laptops cannot get the EDID data from an attached monitor under
>>>> Windows, because the driver has not implemented this data channel. I
>>>> don't know how it is with Linux.
>>>>
>>>> Edmund
>>>>
>>>> On Mon, Mar 7, 2011 at 10:56 PM, Chris Murphy <lists at colorremedies.com> wrote:
>>>>> Historically that has been expensive. At least if the filters are both accurate and consistent unit to unit. My preference is that manufacturers be compelled, one way or another, to insert the correct information in EDID. And then users don't have to deal with it at all.
>>>>>
>>>>> We do have the small problem of backlight decay over time.
>>>>>
>>>>> Chris
>>>>>
>>>>> On Mar 7, 2011, at 2:33 PM, edmund ronald wrote:
>>>>>
>>>>>> Chris,
>>>>>>
>>>>>> As usual, you have hit the nail on the head, colorimeters are not
>>>>>> going to be much fun when the primaries are spiky and unknown.
>>>>>> I knew this when I did my colorimeter design.
>>>>>> But I think a low-cost spectro will not be very happy with spiky LEDs either.
>>>>>> I think the only really reliable way to measure a color LED backlight
>>>>>> panel is probably a really good high-end filter instrument.. There is
>>>>>> an ex-east german company which I believe has an accessible
>>>>>> colorimeter with "real cone" RGB filters.
>>>>>>
>>>>>>
>>>>>> Edmund
>>>>>>
>>>>>>
>>>>>> On Mon, Mar 7, 2011 at 9:51 PM, Chris Murphy <lists at colorremedies.com> wrote:
>>>>>>> I wish a grant were possible to determine the accuracy of EDID primary information. With display technologies all over the map, a colorimeter is ill equipped to do a good job all of the time due to mismatching primaries to the color matrix in the colorimeter (or its software). I am presently unconvinced that most people need a colorimeter in this category. It really takes a spectro to be sure. I have seen colorimeters, including the Huey, do a worse job than default EDID based profiles, with common displays.
>>>>>>>
>>>>>>> Chris Murphy
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Mar 7, 2011, at 2:24 AM, Richard Hughes wrote:
>>>>>>>
>>>>>>>> On 7 March 2011 09:14, edmund ronald <edmundronald at gmail.com> wrote:
>>>>>>>>> Of course I didn't :)
>>>>>>>>
>>>>>>>> Heh :)
>>>>>>>>
>>>>>>>>> However the technology and calibration issues are the same as with any
>>>>>>>>> colorimeter; I wouldn't expect any surprises, other than the perennial
>>>>>>>>> matrix issues. In fact, I would expect my design to perform better
>>>>>>>>> than Huey, which at the time I did this was the low-cost instrument to beat.
>>>>>>>>
>>>>>>>> Right, and I guess the Huey is still the low cost instrument to beat.
>>>>>>>> It's certainly what I advise people who ask to buy.
>>>>>>>>
>>>>>>>>> My guess is that the technology I used was intended for, and will be
>>>>>>>>> or is found in some self-calibrating screens, where of course the
>>>>>>>>> calibration issues can be solved at manufacture time, and later by
>>>>>>>>> remeasuring with a spectro.
>>>>>>>>
>>>>>>>> I guess if someone was to manufacture the devices in small quantity
>>>>>>>> they should just calibrate them at manufacture time with something
>>>>>>>> like a ColorMunki. I assumed you wanted to post the PCB and BOM online
>>>>>>>> and let people make their own colorimeter, open source style.
>>>>>>>>
>>>>>>>>> I should stress that I did the hardware design myself, and had some
>>>>>>>>> protos fabricated, and wrote the Mac instrument drivers, I did not
>>>>>>>>> write full calibration software.
>>>>>>>>
>>>>>>>> Right. It sounds like you could commercialize this and make a bit of
>>>>>>>> pocket money, on the assumption you can undercut the huey by a
>>>>>>>> significant enough margin for production volumes < 1000 or so. That's
>>>>>>>> the tricky bit.
>>>>>>>>
>>>>>>>> Richard.
>>>>>>>> _______________________________________________
>>>>>>>> openicc mailing list
>>>>>>>> openicc at lists.freedesktop.org
>>>>>>>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> openicc mailing list
>>>>>>> openicc at lists.freedesktop.org
>>>>>>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> openicc mailing list
>>>>> openicc at lists.freedesktop.org
>>>>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>>>>
>>>> _______________________________________________
>>>> openicc mailing list
>>>> openicc at lists.freedesktop.org
>>>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>>
>>> _______________________________________________
>>> openicc mailing list
>>> openicc at lists.freedesktop.org
>>> http://lists.freedesktop.org/mailman/listinfo/openicc
>>>
>> _______________________________________________
>> openicc mailing list
>> openicc at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/openicc
>
> _______________________________________________
> openicc mailing list
> openicc at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/openicc
>