Testing Retron 5 lag vs SCART to HDMI

[thetallguy24]

I was wanting to test the Retron 5 lag on an ASUS VE247h monitor vs SNES/GENESIS SCART To HDMI on the same monitor. I don't have lag testing equipment, but I do have a camera and video editing software, so I can break it down frame by frame to see if there is a visual difference.

 

The issue is trying to have them play simultaneously, so I was wondering if it were possible to splice one SNES or Genesis controller to having two connectors, so I can play both consoles simultaneously with one controller.

 

Is that splicing a controller like that feasible?

[Keatah]

The way to measure total lag in a system is to hook an LED to the fire button and mount it next to the monitor. Press the button, the LED light comes on instantly. Then wait for the video image to "catch up". Count the frames or fraction of frames till the ship fires. The higher the speed camera, the better. 60fps minimum.

 

For bonus points you can use an LED counter/timer with 1/100th second resolution.

Edited January 11, 2016 by Keatah

[thetallguy24]

The way to measure total lag in a system is to hook an LED to the fire button and mount it next to the monitor. Press the button, the LED light comes on instantly. Then wait for the video image to "catch up". Count the frames or fraction of frames till the ship fires. The higher the speed camera, the better. 60fps minimum.

 

For bonus points you can use an LED counter/timer with 1/100th second resolution.

Won't the LED need a power source or will the controller suffice?

[Keatah]

Ideally you'd use a sense line and a separate power source for the LED. My setup uses a transistor and resistor to isolate the LED from the controller/console circuit. All depends on what system you're measuring.

[thetallguy24]

Ideally you'd use a sense line and a separate power source for the LED. My setup uses a transistor and resistor to isolate the LED from the controller/console circuit. All depends on what system you're measuring.

 

Well I am trying to create a visual reference of the lag difference with the equipment I have. I have a couple spare controllers lying around and wanted to see if I could play lets say SMB simultaneously on both consoles using one controller. Record it at 60fps, then show in slow motion the visual difference in lag.

[teh_lurv]

Just use games with a demo attract mode, no need to mess around with controllers.

[thetallguy24]

Just use games with a demo attract mode, no need to mess around with controllers.

 

How does that allow for testing input lag when there aren't any inputs taking place?

[teh_lurv]

 

How does that allow for testing input lag when there aren't any inputs taking place?

 

Input lag is the time delay between when the video signal being sent from the console to when it appears on the screen. The "input lag" phrase describes the noticeable effect of that delay; you feel a lag from your controller input to the action on the screen. But that lag is always happening, even if you're not interacting with the console.

 

Rather than mess around with games, I'd recommend heading over to JunkerHQ and try out the 240p test suite homebrew. It has a frame display counter you can use to compare input lag between displays. This is a pic I took using the suite to measure the input lag on my HDTV (I split the composite output to a CRT TV.) The difference between displays is 3 frames, so the input lag on my HDTV is about 50ms.

 

Edited January 14, 2016 by teh_lurv

[Reaperman]

 

Input lag is the time delay between when the video signal being sent from the console to when it appears on the screen.

 

In this case, it seems to be about the delay in the software emulation used by the retron 5, as it compares to the delay of original hardware+scaling.

 

My questions on it are:

Do LEDs really come on instantly (enough), or do they take some ms to fully light? If so, some controllers already have LEDs (capcom power stick comes to mind).

Which scaler is going to be used?

 

It also might be interesting to figure out the latency of a corded vs wireless controller on the retron 5--if it does the usual 'wireless thing' of having a blinking LED on it.

 

It'd be pretty interesting to see how this one pans out. Retron 5 is fairly modern emulation, and I'd really like to know how close it is.

Edited January 15, 2016 by Reaperman

[goldenegg]

After a long round of testing last night, I believe I know what's causing the lag on the Retron 5.

Using a Nvidia Shield Tablet, running the latest version of RetroArch, I was able to reproduce the same lag. For best performance, Threaded Video should be disabled and Hard GPU Sync should be enabled (with Hard GPU Sync Frames set to 0). If you enable Threaded Video and disable Hard GPU Sync, you end up with the same lag experienced with the Retron 5.

I've reached out to Hyperkin about this. If these settings are in fact set incorrectly, i'm hoping that can be addressed in the next firmware update.