KeyRah test!

[Photon]

I'd like to keep this separate from my Keyrah 'question' thread, since I did some testing and read some muttering about Keyrah performance in Pinball Fantasies.

Software used: Asm-One, Gravity-Force, Pinball Dreams (all in the line of duty of course!! Hehe.)

Setup: Core2Duo+ATi 4800 Series, latest WinUAE (with gfx_backbuffers=0 and 50Hz on a CRT), two USB keyboards.


At first when I compared the keyboards, I wasn't quite sure which was the best. In Gravity-Force you can instantly tell any delay - for the sound card, by boosting to land on a platform and checking the bounce sounds vs. the visual bounce of the spaceship, and for the keyboard delay, the delay between pressing Enter and the shot ejecting from it.

I found the sound delay negligible, but the fire delay less so. And I have an integrated sound card. You can set it to buffer size 1 without side effect, but even so, that means the keyboard delay was 'bad for both'. I couldn't tell them apart, except in AsmOne where I could type faster by nature (less typos), my fingers having been honed by an Amiga keyboard since time immemorial. Or something

Now, I have a freaky but sane setup, ie. 'some cheap USB hub' with a single USB lead going into a walk-in closet that serves as computer room.

And when I tested the hub with my good optical mouse and mouserate.exe, the report rates varied from 83 to 1037 Hz. Not good.

So I tested both USB keyboards in my 'dedicated 5 meter USB lead' which goes to a USB port overclocked to a stable 500Hz (normally intended for a mouse since I used to play online FPS), and the difference was obvious. Both keyboards performed MUCH better.


Conclusion: any USB keyboard sucks in a hub (whether it's my cheap König one or one built into some device like a screen), and Keyrah is no worse than any USB keyboard.

I did find that Keyrah had a tendency to 'repeat' a key pressed on one of my PCs; but since it was hotplugged back and forth in an emu I'm not sure it's Keyrah's fault. Could be, I guess.


I don't really understand why Jens went from a simple serial to parallel converter and a (translation) ROM to PS/2 instead of USB, but I guess it's a part of his master plan and to do with auto-configuration which is not possible on PS/2. But ofc for gaming performance, there is no subtitute for a ZERO millisecond "instant" PS/2 keyboard, even if USB lag is lowered to 1/500 second.

This is *not* a 'laying down the law' test at all; I've no way of testing the actual actuator->visual feedback delay. It's just a 'how it feels' quick user test. I would like input from the #Longplays guys, especially those who favor or disfavor the Keyrah as game input devices, as [tight/frustrating] games usually test the input->feedback loop the best.

I did set a highscore in 'Steel Wheel' with one of the keyboards - an A1200 keyboard via Keyrah. But I don't know if that counts, it was a measly 19M or so from 'some ADF'.


Since I want the absolute best for WinUAE on the occasion I use it instead of a real Amiga, I would also like to hear from those interested in making a PS/2 converter for A500 'spring' type keyboards (either from the ribbon cable point or controller pins point), and only because I hold a personal opinion these were the best made for the Amigas.

(ps. For anyone using an LCD monitor, I would instantly recommend the Keyrah! Since a 2-8ms delay over the delay built into your display will hardly be noticable.)

[_ThEcRoW]

I don't know, but wasn't the keyrah advertised as performing like a normal usb keyboard?. I was planning on getting one, but if you say it has a delay, then it's no good for a keyboard.

[Photon]

The (long) post says that all USB keyboards have a delay compared to PS/2 keyboards, that I could not tell any difference between the Keyrah and my previous USB keyboard, and that I would be interested in making a PS/2 keyboard adapter for A500 keyboard --> PC.