Seeking beta testers for Mac Sync-inator VGA sync converter

[LaPorta]

I’ll clarify, the PaintBoard Li lights up both LEDs as per my photo. That one works with the display as seen as well. The “lines” were due to poor frequency adjustment that I eventually corrected on the LCD itself. I believe the color issue was simply due to my inability to adjust the color via LCD hardware buttons: the color adjustment exists, I believe, in software that isn’t built into 7.1 on my IIfx (the manual says 7.5 or newer is needed to run the monitor).

As for the ColorCard, only one LED was lit; I forget which one now and will have to test again. I could not make my Dell LCD work with it no matter what combination I tried. I tried setting it on 640x480, 67Hz (which is its only output capability) and still nothing. I’ll try more soon.

 

[bigmessowires]

The ColorCard was probably csync-only, I have a ColorCard SE/30 and that's what it does. It should hopefully work in sync mode 8 (csync splitting) and might also work in mode 7 (csync to hsync). Let me know if you see different results between your Dell display and your Apple Studio Display.

 

[halkyardo]

Another tester is seeing some strange horizontal jittering at the top of his screen with a Mac IIsi on an old 15" Dell LCD, when using sync mode 8 (csync splitting). It's like the menu bar is randomly jumping left and right by 1-2 pixels. I never saw a behavior like that in any of my testing, so I'm not sure if it's the Sync-inator or something else. Please let me know if you see anything similar.

I'm not sure how useful this will be since I don't have a Sync-inator, but thought I'd chime in, since I see something similar on the output of my E-Machines Futura SX connected to a modern video scaler with an adapter in separate-sync mode (Mac pins 12, 15 -> VGA pins 14, 13).

I put my oscilloscope on it and from the looks of things, the horizontal sync output is not in lockstep with vertical sync - the hsync pulse 'wanders' back and forth within the vsync pulse. Presumably the PLL in the scaler stabilizes it eventually, but for the first couple of lines I get weird distortion and jittering back and forth.

In my case, composite sync (Mac pin 3) has a stable hsync pulse within vsync, and switching my adapter to composite sync mode (Mac pin 3 -> VGA pin 13) gives a stable picture, so I'm assuming its an issue with the separate-sync outputs on my card. Not directly comparable to your adapter, but the symptoms definitely sound similar.

 

[bigmessowires]

That's probably exactly it. When performing csync splitting, the Sync-inator's hsync output is generated with asynchronous combinatorial logic from the csync signal, so it's very consistent, but the vsync signal is generated in software and there's a small amount of jitter based on the microcontroller clock. But to my knowledge this shouldn't be a problem since vsync isn't used to synchronize horizontal lines, only to signal the start of a new frame. I tested with 11 different monitors and in early testing I did sometimes have vsync problems where the whole picture would bounce up and down by one line, or it would completely fail to sync, but I never saw symptoms like that. I wonder if it's related to particular sync processing chips used in specific monitors?

At any rate, it's good to know that composite sync mode fixed the problem for you, and my tester saw the same result: it worked fine when he put the Sync-inator in mode 7, which routes csync to hsync and adds a new software-generated vsync. So far this seems like a rare problem with a good alternative solution, but if it comes up again then I may need to get my hands on one of the monitors that showcases the issue for more testing.

 

[halkyardo]

if it comes up again then I may need to get my hands on one of the monitors that showcases the issue for more testing.

At least in my case, the 'monitor' that shows it most noticeably is an Extron RGB-DVI 300 scaler - they're pretty cheap, smaller than a monitor, and handy to have around in general, so that might be an option.

On the Extron it shows up as a bit of smearing and/or static horizontal displacement of the first few lines of the menu bar, but I have seen jitter like you describe too. On my other monitors, it's either not noticeable at all, or just manifests as a fuzzy, noisy image that can't be corrected with phase adjustments.

 

[bigmessowires]

I could not make my Dell LCD work with it no matter what combination I tried.

What's the model number of the Dell LCD?

 

[LaPorta]

What's the model number of the Dell LCD?

1907 FPt

 

[bigmessowires]

Have you tried the ColorCard/24 on the Apple Studio Display with any of the sync modes, or with any monitors other than that Dell? It would also be useful to see results from some of your other CRTs and LCDs. Also your IIsi and one of the PowerMacs like the 6300. And your IIgs if you have a 15 kHz monitor. For my own testing I made a big matrix of video sources, monitors, and sync modes, and slowly filled it in. I didn't test every single combo but it helped to make sure I was getting enough coverage of each parameter.

 

[LaPorta]

I have the IIgs: are you looking for the IIgs outputting to a regular monitor, or the IIgs monitor working with another machine?

 

[bigmessowires]

IIgs outputting to a regular monitor. It would need to be a monitor with 15 kHz support though, which aren't very common. Two that I have are a SyncMaster 512N and an Asus VE228. But IIgs testing is secondary in importance to Mac testing - mostly I need as much data as possible about Mac / sync mode / monitor pairings.

I got some more data from another tester today and did some hands-on tests with his equipment. Unfortunately his Dell E173FP LCD shows odd problems with the Sync-inator's actively-generated sync signals that none of my eleven other monitors showed: a sort of jittery flashing near the top of the screen. This happened with a bunch of different video sources when selecting one of the active sync modes, so it's not a problem with one specific source. Auto-adjust or manual fine-tuning can largely eliminate the problem, but not 100 percent. I'm not sure what to do with this one... I could attempt to play with the algorithms to improve the E173FP results, but I'm not sure I'd be successful and any changes I make would risk introducing new problems with other monitors / modes. So I guess it'll depend on whether anybody finds other monitors with a similar problem, and whether this is a rare issue or a common one. I know I can't promise success with every possible combination of computer, sync mode, and monitor, so this might be have to be a "known issue". The E173FP still works great with the Sync-inator in the passive sync modes.

Another puzzling discovery is that when using a Performa 6200 series computer, the Sync-inator's actively-generated sync signals don't seem to work on any monitor. This isn't a huge problem because the 6200 already has separate hsync and vsync outputs, so the Sync-inator passive modes work just fine. But if you stubbornly want to use the active sync modes anyway, it should work, but it doesn't. I haven't yet looked into why.

I must admit to a lack of enthusiasm for diving back into the guts of this again. It's sort of like a house guest who's far overstayed their welcome... I need to get this out the door so I can restore my sanity.

 

[trickydee]

Hi @bigmessowires

I would love to test this, as I have 2 'not so great' VGA solutions for the following machines:

IIGS (Rom 0) - US
IIGS (Rom 1) - UK
Performa 630
Quadra 700

I currently use the IIGS with a home made VGA cable and GBS8800 (to convert the Sync) with a Samsung Syncmaster 940 but the colours are quite washed out.
The Performa & Quadra are currently used with a passive VGA adapter on a Dell WFP-2007 display - however the Q700 doesn't work correctly with the Samsung with the colours being wrong.

I'm UK based, happy to pay postage if needed.

Regards

R

 

[bigmessowires]

I would love to test this, as I have 2 'not so great' VGA solutions for the following machines:

Thanks, I'm going to wait for results from the three beta testers I have now and then see how it looks. You shouldn't expect the Sync-inator to affect washed-out colors, though. It transforms the composite sync, horizontal sync, and/or vertical sync, but doesn't touch the analog RGB signals at all. I would guess that washed-out colors is either a gamma problem, or else a problem where the monitor misinterprets sync-on-green (and possibly also sync-on-red and blue) as being part of the color range.

 

[bigmessowires]

Another puzzling discovery is that when using a Performa 6200 series computer, the Sync-inator's actively-generated sync signals don't seem to work on any monitor. This isn't a huge problem because the 6200 already has separate hsync and vsync outputs, so the Sync-inator passive modes work just fine. But if you stubbornly want to use the active sync modes anyway, it should work, but it doesn't. I haven't yet looked into why.

This is now fixed, but existing beta testers may still run into this problem. It turned out that the 6200 (and probably other PowerMacs) has a composite sync signal structure that's different from any of the other examples I'd looked at previously. It doesn't really matter since you'd normally use passive pass-through sync mode on that computer, but now csync splitting mode will also work if you want it for some reason.

 

[trickydee]

Thanks, I'm going to wait for results from the three beta testers I have now and then see how it looks. You shouldn't expect the Sync-inator to affect washed-out colors, though. It transforms the composite sync, horizontal sync, and/or vertical sync, but doesn't touch the analog RGB signals at all. I would guess that washed-out colors is either a gamma problem, or else a problem where the monitor misinterprets sync-on-green (and possibly also sync-on-red and blue) as being part of the color range.

Agreed, I'm pretty sure the Samsung Monitor is the reason behind the colour issue as the Quadra works fine on other monitors.
I'll be sure to pick up one of your convertors when they are released to replace my current beat up dongle.
R

 

[eharmon]

If you’re running System 7 without Update 3.0 there’s a couple color output bugs with Quadra’s. Might be related. If you don’t have that installed, give it a try.

I just read that changelog yesterday and somehow it’s been relevant twice in two days…

 

[Garrett B]

I finally got some time to do a little testing with my beta adapter. Monitor I am using is a Samsung SyncMaster 171N. I've attached a screenshot of the specification for those interested. The following are tested working well:

IIci - switch settings 1 4 7 8 - comp sync LED only
IIsi - switch settings 1 4 7 8 - comp sync LED only
Radius Pivot IIsi - 1 4 7 8

LCII - switch settings 1 4 7 8 - both LEDs lit (this one should have worked with a regular adapter but did not. No issues with the Sync-inator)

RasterOps 24SX - switch settings 1 4 7 8
RasterOps ColorBoard 364 - switch settings 1 4 7 8
Apple Bungee Video Card - switch settings 1 4 7 8
Apple High Resolution Display Video Card - switch settings 1 4 7 8
RasterOps ColorBoard 108 - settings 1 4 7

Here are two cards I could not get to work (may be the fault of the cards, not the adapter):

Radius Pivot SE/30 Monochrome - not working - both LEDs briefly flash every 3 seconds
Radius Pivot SE/30 Color - not working - both LEDs briefly flash every 3 seconds

 

[bigmessowires]

Awesome data, thank you!

Radius Pivot SE/30 Monochrome - not working - both LEDs briefly flash every 3 seconds
Radius Pivot SE/30 Color - not working - both LEDs briefly flash every 3 seconds

When this happens it means the computer isn't outputting any video data. It's dead, Jim. I've updated the troubleshooting section of the product page to clarify: https://www.bigmessowires.com/mac-sync-inator/

I'm not sure why those Pivot SE/30 cards aren't outputting anything. Do you need to enable them in a control panel first? Maybe the Pivot card requires a different monitor ID than the "14" that you're using for a Mac 640x480 monitor? I've seen some adapters that mention a 24 setting for "15 inch Tilt". Not sure if that means a Pivot monitor, but you could give it a try.

 

[LaPorta]

I got my ColorCard/24 to work with a much older Gateway LCD I have, The 15” FPD1540. It works at native Mac 67Hz 640x480 (set via 14 on the adapter), but with with only 7 selected: the adapter’s “best guess” via 78 produced a garbled, unusable image where the left 40% of the screen didn’t even register anything. Here are the photos of 78 selected, and 7 selected:




I’m glad I got it to work: this means that my other Dell LCD adjust isn’t compatible with 67Hz (the only thing this card outputs), and it is not the adapter.

Nice thing is that’s another monitor we know will work with a classic Mac!

 

[bigmessowires]

Nice! Yes that's normal that some Sync-inator sync modes won't work with some computer-to-monitor pairs. It will depend on the behavior of both the computer (or video card) and the monitor. That's an interesting half-image you got in the first photo, though! Did you try running the monitor's image auto-adjust from its on-screen menus?

Also have you tried connecting an external 5V USB power supply when running with those settings? Most of the time you won't need one, and you should do most of the testing without it, but when you encounter a card-monitor pair that refuses to work then it's definitely worth trying the 5V supply. Some video cards and computers are unable to provide enough self-powered voltage to run the Sync-inator's microcontroller at full speed, so the accuracy of the generated sync signals is reduced. Connecting a 5V supply enables the microcontroller to run at full speed in those cases.

 

[LaPorta]

I did indeed try the auto-adjust (to no avail), but not the 5V. I can give that a go later on to see if that makes any difference. The adapter seems to work very well so far with what ive done. There is one thing I would mention: The board just clears the NuBus slots on my IIfx: I have to finagle it to not have one or two of the under board solder points not scrape the plastic on the case. If you are planning on encasing these, they might not have clearance on such machines as they barely make it as is.