What's new
By:
Filters

Arcade Monitor calibration guides

Since I run linked cabs side by side I really try and dial the colors in on them to match as best I can. No special tools, just some eyeballs, but I do a pretty good job. Hard to get it exact, no two tubes really match in my experience, and it's especially harder when one is an MS8 and the other an MS9.

One thing I'd correct in this thread since I've seen it mentioned a few times—unless someone can point me to how I've been missing it in which case I'd be grateful—is the MS8 chassis doesn't offer a contrast control, only the MS9.
 
Nebula said:
Pinging @ShootTheCore about Contrast and Brightness adjustments.

Seeing MS9 manual, I noticed that there is some section where it explains how to calibrate Brightness:

brighness_adj.png



It also involves tunning Sub-Contrast, ABL (Automatic Bright Limiter), Screen, Brightness and Contracts pots to adjust general brightness, but in a different manner.

As I don't really have many knowledge in luminance and technical bright concepts, I don't know if this would be a proper way to adjust it.
Do you think this method could result in a similar final adjustment as yours?

It talks about 0.8 ft-L in the first step, while setting up brightnes pot to the maximum. That doesn’t really make much sense for me, seeing the magnitude values for luminance in your method, or even in the second and third steps in picture above...
Click to expand...

First off, generally speaking, there's certainly more than one procedure for adjusting a monitor. No matter how you do it, the idea is that you have a target goal in mind, and you adjust each component (RGB drive, RGB cutoff, brightness, contrast) repeatedly until you've homed in on your target. Adjusting one component shifts another, so each component is usually adjusted more than once. You use a pattern generator of some kind to display a steady pattern on the screen, and you use a measurement device to determine both what component to adjust next and whether the target has been reached.

For most reasonable folks, the Target, Signal Generator, and Measurement Device are:

Target: "Looks pretty good to me"
Signal Generator: Test patterns from a favorite game PCB
Measurement Device: eyeballs

For OCD weirdos like me, the Target, Signal Generator and Measurement Device are:

Target: neutral white balance at 6500 kelvin color temperature, and a maximum brightness between 30 and 40 ftL (30 for dark room use, 40 for bright room use)
Signal Generator: 240p Test Suite from a high-color console or PGenerator running on a Raspberry Pi
Measurement Device: X-Rite I1 Display Pro or Datacolor Spyder 4 colorimeter

With that in mind, let's look over the procedure in the MS9 Service Manual. There seems to be multiple manuals out there - the one I found seems to have slightly different instructions than the screenshot you posted. If you have a link to yours please share it. Here's a link to mine:
https://wiki.arcadeotaku.com/images/3/34/MS-2931_Service-manual_full.pdf

First off, Page 47-49 of the manual lists the tools they want you to use - a Sega Model 3 arcade board set to 31hz resolution for the test pattern generator and a brightness meter for the measurement device. A modern colorimeter acts as a brightness meter in addition to reading the specific R, G, and B values. No target is specified in the Service Manual from what I can see, but from context it looks like the target is a neutral white balance.

Given that a brightness meter doesn't measure individual R, G, and B values and that the Drive and Cutoff adjustments are made to match only white balance targets, following the procedure in the service manual will give you a good white balance result (the darkest darks and lightest lights will consistently match across monitors) but an inconsistent color temperature (color tones will vary across monitors).

That doesn't surprise me given that the tools for professional grade monitor calibration were very expensive back in the 90s when the Nanao monitors were being manufactured, and arcade operators didn't care about color accuracy as much as photographers and video producers (Sony PVM & BVM customers) did back then either.

TL;DR - my procedure will give your a more accurate and consistent outcome then following the procedure in the service manual.
 
Last edited:
Aurich said:
Since I run linked cabs side by side I really try and dial the colors in on them to match as best I can. No special tools, just some eyeballs, but I do a pretty good job. Hard to get it exact, no two tubes really match in my experience, and it's especially harder when one is an MS8 and the other an MS9.

One thing I'd correct in this thread since I've seen it mentioned a few times—unless someone can point me to how I've been missing it in which case I'd be grateful—is the MS8 chassis doesn't offer a contrast control, only the MS9.
Click to expand...

Hmm ... well, I'm much more familiar with the MS9 and Wells-Gardner chassis than the MS8, but it looks like some MS8 chassis have Contrast and some don't.

The MS8 user's manual here lists both Contrast and Brightness dials:
https://wiki.arcadeotaku.com/images/e/ee/Nanao_MS8-25F,_MS8-29F.pdf

I'll update my colorimeter calibration procedure - if you don't have a Contrast dial, using the Screen dial on the flyback to set the initial high target and then adjusting the Brightness, Drive and Cutoffs correctly will derive the correct Contrast.

EDIT: don't use the Screen dial on the flyback to accomodate the lack of a Contrast dial - it's a coarse Brightness adjustment, not a Contrast adjustment. Instead, adjust the Red, Green and Blue Drive pots evenly - they are essentially individual Contrast adjustments for each color.
 
Last edited:
ShootTheCore said:
if you don't have a Contrast dial, using the Screen dial on the flyback to set the initial high target and then adjusting the Brightness, Drive and Cutoffs correctly will derive the correct Contrast.
Click to expand...
Isn’t the contrast adjustment pot just like turning all RGB drive pots at once? and the brightness dial is like all cutoff pots turned in unison? My monitor chassis doesn’t have a discrete adjustment pot for contrast but I can get same results by boosting all RGB drive pots together.
 
kikaso said:
Isn’t the contrast adjustment pot just like turning all RGB drive pots at once? and the brightness dial is like all cutoff pots turned in unison? My monitor chassis doesn’t have a discrete adjustment pot for contrast but I can get same results by boosting all RGB drive pots together.
Click to expand...

Yes, you're correct. The breakdown is as follows:

Brightness is the control for adjusting black level. Setting it too low causes dark details to be lost (black crush). Setting it too high causes black to appear gray. Brightness affects all colors at the same time.

Contrast is used to adjust white level (peak light output). Setting it too low gives you a dim picture. Setting it too high causes lighter details to bleed into white (blooming / smearing). Contrast affects all colors at the same time.

RGB Cutoff is essentially an individual brightness control for each color.

RGB Drive is essentially an individual contrast control for each color.

Once calibration is set, you don't want to adjust the Contrast pot for any reason. You can adjust the Brightness pot to accommodate for dark or light game boards.

Now that I rethink it, the Screen dial on the flyback is essentially a coarse Brightness dial so my earlier response to @Aurich about using it to accommodate a missing Contrast dial is incorrect. But you should still be able to reach the target via RGB Drive adjustments since they are Contrast adjustments.

I'll do some further research and will then report back. I honestly haven't yet calibrated a monitor without a Contrast dial.
 
Last edited:
ShootTheCore said:
First off, generally speaking, there's certainly more than one procedure for adjusting a monitor. No matter how you do it, the idea is that you have a target goal in mind, and you adjust each component (RGB drive, RGB cutoff, brightness, contrast) repeatedly until you've homed in on your target. Adjusting one component shifts another, so each component is usually adjusted more than once. You use a pattern generator of some kind to display a steady pattern on the screen, and you use a measurement device to determine both what component to adjust next and whether the target has been reached.

For most reasonable folks, the Target, Signal Generator, and Measurement Device are:

Target: "Looks pretty good to me"
Signal Generator: Test patterns from a favorite game PCB
Measurement Device: eyeballs

For OCD weirdos like me, the Target, Signal Generator and Measurement Device are:

Target: neutral white balance at 6500 kelvin color temperature, and a maximum brightness between 30 and 40 ftL (30 for dark room use, 40 for bright room use)
Signal Generator: 240p Test Suite from a high-color console or PGenerator running on a Raspberry Pi
Measurement Device: X-Rite I1 Display Pro or Datacolor Spyder 4 colorimeter

With that in mind, let's look over the procedure in the MS9 Service Manual. There seems to be multiple manuals out there - the one I found seems to have slightly different instructions than the screenshot you posted. If you have a link to yours please share it. Here's a link to mine:
https://wiki.arcadeotaku.com/images/3/34/MS-2931_Service-manual_full.pdf

First off, Page 47-49 of the manual lists the tools they want you to use - a Sega Model 3 arcade board set to 31hz resolution for the test pattern generator and a brightness meter for the measurement device. A modern colorimeter acts as a brightness meter in addition to reading the specific R, G, and B values. No target is specified in the Service Manual from what I can see, but from context it looks like the target is a neutral white balance.

Given that a brightness meter doesn't measure individual R, G, and B values and that the Drive and Cutoff adjustments are made to match only white balance targets, following the procedure in the service manual will give you a good white balance result (the darkest darks and lightest lights will consistently match across monitors) but an inconsistent color temperature (color tones will vary across monitors).

That doesn't surprise me given that the tools for professional grade monitor calibration were very expensive back in the 90s when the Nanao monitors were being manufactured, and arcade operators didn't care about color accuracy as much as photographers and video producers (Sony PVM & BVM customers) did back then either.

TL;DR - my procedure will give your a more accurate and consistent outcome then following the procedure in the service manual.
Click to expand...

The one I was taking a look is the MS9-29SU section from Sega Touring Car Championship manual. In particular, page 154: https://www.gamesdatabase.org/Media...ga_Touring_Car_Championship_-_1996_-_Sega.pdf

I was researching about ABL (Automatic Beam Limit or Automatic Bright Limiter), because it seems it could be related to that blooming effect that uses to appear when sudden bright flashes occur during gameplay, producing some kind of zoom-in and zoom-out effect. And that's why I ended finding that method in the manual, where talks about using ABL pot in some of the Brightness tunning steps (but I cannot figure out nothing about it yet...)


I'm agree with you about the method of the manuals where they only talk about adjusting white levels, but no RGB levels independently in order to make pure white consistency. But it surprised me a bit regarding the ft-L values mentioned in those steps and what are the actions to perform. That's why I want to ask to someone with more knowledge in this matter ;)
 
Last edited:
Nebula said:
The one I was taking a look is the MS9-29SU section from Sega Touring Car Championship manual. In particular, page 154: https://www.gamesdatabase.org/Media...ga_Touring_Car_Championship_-_1996_-_Sega.pdf

I was researching about ABL (Automatic Beam Limit or Automatic Bright Limiter), because it seems it could be related to that blooming effect that uses to appear when sudden bright flashes occur during gameplay, producing some kind of zoom-in and zoom-out effect. And that's why I ended finding that method in the manual, where talks about using ABL pot in some of the Brightness tunning steps (but I cannot figure out nothing about it yet...)


I'm agree with you about the method of the manuals where they only talk about adjusting white levels, but no RGB levels independently in order to make pure white consistency. But it surprised me a bit regarding the ft-L values mentioned in those steps and what are the actions to perform. That's why I want to ask to someone with more knowledge in this matter ;)
Click to expand...

Thanks for the link to the Sega Touring Car manual. It looks like the calibration procedure in that manual is less detailed than the procedure in the manual I linked earlier. Odd.

I haven't touched the ABL adjustment on my chassis yet but since its function is to set a threshold on peak brightness, my thinking is that you'd do the following:

1) Set ABL pot to maximum.
2) Calibrate the display.
3) Pull up the IRE 100 pattern (so that you're displaying 100% peak brightness for the calibrated display target).
4) Slowly adjust ABL pot down until the the brightness starts to decrease, then back it off slightly.

That should stop the display from going past the calibrated peak brightness with screen flashes and what not.

That said, it's normal with CRTs for the geometry to fluxuate a bit with screen flashes - the chassis is handling the rapid voltage shifts from zero voltage (black) to peak voltage (white) as best it can. But if the geometry distortion is still very bad on screen flashes, here's a few other things you can do:

  • Check that the monitor chassis is well-grounded.
  • Check the B+ voltage for the flyback - it should be around 74 volts on an MS9.
  • Recap the chassis if it hasn't been done already - fresh caps handle the voltage fluctuations more effectively.
  • Decreasing Brightness via the Screen dial on the flyback or Brightness pot will reduce the amount of geometry distortion on flashing since the voltage spikes won't be as extreme.
 
Last edited:
If you guys want a good core workout, try adjusting the geometry with the chassis still in the cab from the main door... :) "It's close enough" pffft. I wish there was a remote board for all the geometry pots! EDIT, think I just talked myself into soldering some long leads w/ pots to the chassis....hmmm.
 
Last edited:
digitron said:
If you guys want a good core workout, try adjusting the geometry with the chassis still in the cab from the main door... :) "It's close enough" pffft. I wish there was a remote board for all the geometry pots! EDIT, think I just talked myself into soldering some long leads w/ pots to the chassis....hmmm.
Click to expand...
CPS3 really makes me wish the pincushion adjustment on the ms8 was on the remote board.


Also, @ShootTheCore, I read through your guide, but how do you recommend setting the Screen pot? I noticed on my MS8 if what's on screen isn't bright/busy enough, like simple text on an otherwise black screen, the raster lines come back, so I've got it too high. Obviously the answer is to just turn it down a little but just wondering if you had any suggestions to integrate the screen setting into the overall procedure.
 
I added an adjustment procedure for the Screen pot on the Flyback to my guide. Here it is:

1) On the cab adjustments, turn Brightnesss and SubBrightness all the way down. Set Contrast and SubContrast to center. Set all three Drive adjustments to center. Turn all three Cutoff adjustments all the way down.

- Some chassis don't have Contrast, SubContrast and/or SubBrightness adjustments.

2) In the 240P Test Suite, back out and go to "Test Patterns" ->"White Screen". Tap R until the screen is all-black.

3) Turn the Screen dial on the flyback down (counter-clockwise) until the display is completely dark and pitch black.

4) Slowly nudge the Screen dial on the flyback up (clockwise) until the display becomes slightly visible from complete black.
 
Last edited:
ShootTheCore said:
I added an adjustment procedure for the Screen pot on the Flyback to my guide. Here it is:

1) On the cab adjustments, turn Brightnesss and SubBrightness all the way down. Set Contrast and SubContrast to center. Set all three Drive adjustments to center. Turn all three Cutoff adjustments all the way down.

- Some chassis don't have Contrast, SubContrast and/or SubBrightness adjustments.

2) In the 240P Test Suite, back out and go to "Test Patterns" ->"White Screen". Tap R until the screen is all-black.

3) Turn the Screen dial on the flyback down (counter-clockwise) until the display is completely dark and pitch black.

4) Slowly nudge the Screen dial on the flyback up (clockwise) until the display becomes slightly visible from complete black. Then nudge the Screen dial back down (counter-clockwise) to where the display barely goes completely dark again.
Click to expand...
Thanks for this bud =)
 
@ShootTheCore, which software would you recommend to use with a X-rite i1 pro Spectrophotometer and the i1 Display Pro Colorimeter? (Windows 10 user).

Thanks!
 
You're getting into fancier equipment then what I have experience with. I've been meaning to dive deeper into calibrating modern displays but have not yet done so - you'll want to look for a better mentor. :D

FWIW, from what I have researched, ColorHCFR is as good as it gets for free software - I'd start with that and only look into paying for other software if you need it. On the paid-software side, assuming that you're calibrating a modern flat-panel display, Calman and ChromaPure Professional are both supposed to be excellent software solutions.

There's a pretty good tutorial on calibrating modern displays with spectrometer here:
http://www.curtpalme.com/forum/viewtopic.php?t=35322

Also, reach out to Keith Raney (@khmr33) on Twitter - he's a very smart dude on all things display calibration, and it one of the contributors to the 240p Test Suite.
 
ShootTheCore said:
You're getting into fancier equipment then what I have experience with. I've been meaning to dive deeper into calibrating modern displays but have not yet done so - you'll want to look for a better mentor. :D

FWIW, from what I have researched, ColorHCFR is as good as it gets for free software - I'd start with that and only look into paying for other software if you need it. On the paid-software side, assuming that you're calibrating a modern flat-panel display, Calman and ChromaPure Professional are both supposed to be excellent software solutions.

There's a pretty good tutorial on calibrating modern displays with spectrometer here:
http://www.curtpalme.com/forum/viewtopic.php?t=35322

Also, reach out to Keith Raney (@khmr33) on Twitter - he's a very smart dude on all things display calibration, and it one of the contributors to the 240p Test Suite.
Click to expand...
Thanks for the info! I'd use these stuff to calibrate my CRTs, specially arcade tubes.
 
I was calibrating my MS9 with the SNES 240p test suite on a Mister, double checked with the Mister DDP core, and then my real Tekken Tag board.

PXL_20221014_015034913.jpg

I followed Mike's instructions and it looks good in game but in test patterns my 4 highest color values are crushed. The monitor is in the cabinet so I have to run back and forth to mess with the chassis pots (hence the post whining instead of starting over). Which adjustment or step in the procedure could I have done wrong to produce the crushed highs? I also noticed two middle values are crushed/the same on the Mister but not on Tekken so I wonder if that's a Mister issue.

I think I might just get a colorimeter.

Edit: I decided to stop being a bitch and try again and now the highs look good but the lows might be a bit dark. I bought a colorimeter on ebay.
 
Last edited:
@ShootTheCore thanks for the great guide, i used it this past weekend to calibrate my MS-2931. Do you know if the 240p Test Suite for the NeoGeo Mister core has the same issues as the SNES core? Reason i ask is my gamma did come out a bit wonky.

Mind you i'm extremely happy with my first attempt at this. Was using the Calibrite Display SL which is detected by HCFR as a Colormunki
 

Attachments

  • APC_0250.JPG
    APC_0250.JPG
    102.7 KB · Views: 23
  • APC_0269-r.jpg
    APC_0269-r.jpg
    406.9 KB · Views: 25
  • Screenshot 2024-03-20 062341.png
    Screenshot 2024-03-20 062341.png
    58.9 KB · Views: 25
  • upload_2024-3-17_18-49-16.png
    upload_2024-3-17_18-49-16.png
    347.2 KB · Views: 22
You must log in or register to reply here.
Back
Top