I've spent a few days running tests and profiling. Wasn't sure whether to add to the previous post but will add more details here. The N170G was set far too bright previously, leading to a much elevated black level.

I have two i1D3 probes, so just putting them side by side I can see different shadow responses - the rest seems to match pretty closely. I am trying to work out whether this is a probe issue or a monitor issue or whether one is better than the other. I've enclosed a PDF report with them side by side.

My question is whether there is anything I can do to get these probes better or determine is one is better than the other, if this level of difference is normal and whether this is just the nature of these? In the PDF I am very simply profiling the native 709 mode of the monitor through Resolve > SDI and that set up is all correct with the two probes.

I have read the guide about Probe inaccuracy. I have them both on Intelligent Integration, so they take longer on darker patches. I am wondering if there is a clever way I can test for accuracy in the sense that an inaccurate probe would compound the issues when profiling. So profile - create LUT - profile again - create LUT - profile again - sure the inaccuracy would come up as a constantly shifting error? Maybe?

There's no correlation file for a Postium N170G - I could grab something similar but I don't know if that's just going to make things worse. I assume the backlight is W-LED.

Any pointers. And yes the obvious one would be test the probes with a spectro and maybe I can find somewhere to hire me something but aside from that how can I make the best with what I have?

Thanks
Paul

Attached file: N170G Calibration Native709M.pdf

A man who wears two watches never knows what time it truly is.
You can't determine relative accuracy without a reference. Without one, you're down to whichever meter you like the results from best - personal preference. And THAT does not guarantee accuracy, for many reasons.
A better use for said reference would be profiling either meter to the display. You would not have to worry then about relative accuracy, because the profiled meter would be the more accurate for that display.

Clearly the better option is to use a spectro to check against and it's looking more and more likely that this really is the only way. I've looked in the UK for renting but I can only find FSI who will rent to their customers. So will keep checking for rentals.

There is the more cost effective i1Pro3 though, and I've read that I could use that to check the i1D3 - I don't see much mention of the i1Pro3 on here, so perhaps it really isn't a good choice?

I am currently going through correlation files as well, been reading up on those as something closer to the display seems like it's better than nothing.

Also in the profiling section where you can set an Active LUT - what's the difference between No LUT and Unity - are they not going to be the same thing?

thanks
Paul

As the bigger variation is in the near blacks renting a spectro only will likely not help as they do not read well near black.
So you really need a good tristimulus and a good spectro to match the tristimulus to before then doing the comparison tests...

And yes, No LUT and Unity are the same within Active LUT.
The idea of Active LUT is to test a real LUT before you upload into a display, etc.
(All is explained in the User Guides.)

Steve

Steve

Yes, I can see that. I've got an iPro2 coming in as a poor mans way to keep the i1D3 in check on each screen. But as you say I know they don't do darks very well but it should help me gain confidence with the rest of the profiling.

What would be useful (and you're probably going to say there's a page for that already) is a place to collect users samples of their monitors to give a comparison. I am see what I think is poor tracking in the blacks but then I think all LCDs suffer to a greater of lesser degree here, but I can't quantify whether mine is normal or sub par. Is the example sources just what I can find in the forums here?

Also, have you considered renting spectos?

Kindest
Paul

No, there is no collection of different monitor profiles, mainly as that would take a lot of memory.
But it is well known that all LCDs have a colour tint in the blacks that tends towards blue.
(You can see that in many of the graphs within the website.)
Your plots are very much as expected in the blacks.

And no, we do not rent any hardware as the insurance costs are prohibitive.

Steve