Any other probes you have found good times for?

Any other time-settings (i.e. per slide) suggestions?

All probe settings are variable based on the display being profiled, and the probe in use.
There are no standard settings that are guaranteed to be correct.
It is a question of trying a few combinations to get the best response (stable measurements) with the correct tables selected (LCD/Oled/CCFLetc).

'Time Per Frame' is only used for DIP mode, and is also a variable that has no guaranteed values based on the probe abd display being used.

See: http://www.lightillusion.com/profiling_manual.html#select_time_per_frame

Close Loop has no such setting

Steve

I see, so Time is user-fixed for DIP mode (playing back an edit-list of slides on a hand-synched external system, so the length must be matched to what is needed for the probe and what was thus set in the EDL) but automatically varied as needed when driving slides directly via Lightspace?

If we are using the Resolve plugin, is that considered a DIT mode or closed-loop by the software?

Could I make an ease-of-learning suggestion that value be greyed out when in Direct Loop Mode?

DIP duration is set by finding the longest time taken for any given patch - usually black or very dark blue.
This is then the duration for DIP mode.

With Closed Loop the duration for each patch is variable depending on the time the probe takes.

Any 'plug-in' is obviously Closed Loop by default, as the patch generation is being controlled directly via LightSpace.

Steve

OK! Thanks!

I was 99% sure that's how it worked, but not 100% as I did not know what level of inter-action the slide playback had with the Lightspace host computer at all times. I've seen such plugins in other software where the only communication between computers (other than probe readings) was a trigger to start synch playback of slides at a predetermined speed. This sounds like it is much more effective.

Still wondering about integration time. How is it different from "per slide" time? In my understanding this is an internal probe thing, and it's difficult to know which time is "right" for any given probe. What observable behavior would allow to know we used to short or two long an integration time (if calibration is slightly off because of too short a time, there would be no way of knowing that for sure, so there must be some other way)? How is integration different from "per slide"? What is 'integration' exactly? I have read the docs, but they seem to assume the user already knows what integration is and its purpose, and that the user would know when the time is set too low. Does calibration time occur just once at the beginning of the cube 'slideshow'?

The Hubble when used with Calman asks for a 15 minute calibration time with the Hubble lens cap on. Where is this process within Lightspace (maybe it is not needed because Lightspace uses meters differently?)

Oops I meant "Does integration time occur just once at the beginning of the cube 'slideshow'."

Integration time is a probe function, not a LightSpace one - LightSpace just provides a control for it.
Different probes use integration differently, with some probes not using it at all.
You need to talk to the probe manufactures to gain understanding how they use it specifically.
Integration occurs for the same duration for every patch reading.
It is the 'integration' of multiple reading over the given time to provide a better, more accurate result.
If set too low the results will be 'unstable'.
If set too high the probe can over saturate, and provide garbage readings.

If Calman asks for 15 minutes for dark calibration (lens cap on) there is a major problem with Calman.
LightSpace does the same function when the probe is used for the first time - but it takes a few seconds.

Steve

That's only one problem with Calman - the list is long

Thank you for that great explanation - now I know.

Are there any probes you happen to know proper integration time for? Do you know what it is for the Hubble?

Contacting X-Rite is not frequently fruitful.

Please see previous replies - integration time depends on the display as well as the probe.

Steve

Sorry I got confused - I thought that was just for Slide Time.

So I guess I just keep raising it until readings in "Live" mode become garbage?

Or rather should I keep lowering it until things look jumpy/unstable? What is the most smart way of finding the right time per probe/display combination?

Is there often a wide range of "Ok" values?

Should I use dark red or dark blue to find this value as well?

Patch time has nothing at all to do with Integration time.

Taken from the User Manual:

Integration Time: sets the amount of time the probe will integrate light reception over, if applicable. Some probes will ignore these setting, other will use them as time settings, others will select from probe pre-sets. In general, the longer the Integration Time the more accurate the reading, but the slower the profiling. But, setting too long a time can cause the probe to over saturate... You can think of integration time as being like exposure time in a camera.

With plasma profiling there is a variation with respect to integration time. Due to plasmas PWM (Pulse Width Modulation) the optimum integration time is calculated as an integer multiple of the PWM frequency of the corresponding display.

So, as an example, in PWM-based Display that has a frequency of around 180Hz, the integration time should be an integer multiple, i.e., n*(1/180) so if n were 500 (and n can be any large integer value) then we obtain the following:

Integration time 500*1/180 = 2.78 seconds.

The more accurate the integration time, the better the results.

Ok thanks for taking the time to answer. I did understand that patch time is completely different, but I still don't quite know the best way to set the ideal time other than integers for plasmas. Are they are methods/tricks for finding the sweet spot (thats' why I brought about dark blue patches - I was thinking maybe they could be used to see if the measurement has time to stop "dancing")? How does one determine the true refresh rate of a given plasma? There must be some good procedure for determining the right Integration Time number, some step by step iterative process.

I assume that total time per patch = Patch Time + Integration Time?

Integration time has nothing to do with patch time, so no, "total time per patch = Patch Time + Integration Time" is incorrect.

I do not know what you mean by 'true refresh rate for plasmas' - the only thing mentioned for plasmas is PWM frequency.

As has been said, there is no single answer for integration time - it depends on the display/probe combination.
Same as there is no single setting for other probe variables - some Plasmas work better with LCD settings, white some LCDs work better with Burst mode selected (i1D3 probe).

It all depends on the attributes of the given display.

Steve

Ok Steve, I do understand. Sorry about the bad terminology with "refresh rate" I meant PWM frequency (tired). How could a user find out what that is for a given plasma they are trying to calibrate in order to be able to use the math you suggest?

I'm not looking for a single answer (i.e. in the form of a particular number) , but for some methodology for figuring out what the Integration Time good number might be for a particular probe/screen combination. What would I look for to tell me "now I got it right".

For example, to set brightness within the display's controls, I know to turn the value down until the last darkest square in the brightness test pattern is "just visible". That is something I can fine tune based on observation of a particular phenomenon, that I can see on the display. But to set integration time, I;m really not sure what I could look at/measure (why I was guessing it was maybe when the interface cross fully stopped wiggling at the end of every measure) to know when I got it right, and I have not found this answer anywhere on the net or in the documentation. How do you efficiently determine how much time to set for this parameter? What can a user look for? What is a smart trick for fiddling this? I know it says "sufficient time to stabilize" but what does that really look like vs insufficient time when we are adjusting the parameter? What are the tell-tale signs of insufficient time? (Besides slightly inaccurate profiling - as we'd never really know unless there was another accurate screen in the same room)

Since Integration Time, in the manual, is stated as adding length to the time for creating a profile (when set to something longer), and so is Patch Time, one can only assume they add up in some way to a total time consumption figure, as both, when longer, slow down the calibration. Its' hard to understand how it doesn't work that way. I;m not arguing in any way, just failing to understand. Maybe it's not simple addition though.

PWM for any given plasmas can only be found from the display specification from the manufacturer...

There is no easy way to 'set' integration time - you just have to use the basic rules outlined above.
And, if you were to try different settings, you would find it's impossible to set patch time to be lower than integration time!



Steve