Overclock.net › Forums › Components › Keyboards › USB vs PS/2 Keyboard: latency, bandwidth -> inputs per second?
New Posts  All Forums:Forum Nav:

USB vs PS/2 Keyboard: latency, bandwidth -> inputs per second? - Page 10

post #91 of 131
Quote:
Originally Posted by dan_k View Post
There is no argument, it's personal preference. Why did you buy your "special keyboard", why not just use the ordinary cheap one? -- So, you say 125 keys per second is "enough"? Even though I gave you example where it definitively is not enough?


Some people can play games @ 30fps, some find even 60fps is "too flickery" for them, there is no argument, if you do not care, that's fine, just go do it elsewhere. What is the point of coming into someone else's house and complain about the TV channel they are watching? Let us be, or join the discussion by actually making an argument, explain your reasoning so we know if you have understood everything correctly and are not making unreasonable assumptions, ok?
I can tell when a game drops down to 57fps from 60, and I haven't been able to tell any difference a/b'ing a keyboard ps/2 vs usb. Except for the 6 key limitation that is. I can hit it on the THPS games after they added the walk function, without that I can only go up to the 6 keys, which works fine on USB.
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
post #92 of 131
you know, I've never actually tried to see if I could notice the difference. I've never used a USB keyboard on my desktop that I can remember. Not really because of any sort of PS/2 superiority, but because I always figured "why waste a USB port when I've got a dedicated keyboard port right there?" But that was mostly from the PIII/P4 days, when you only had between 2-4 ports.
Server
(11 items)
 
  
CPUMotherboardRAMHard Drive
Intel Core i5-3470 ASRock Z77 Extreme6 16 GB G.Skill Sniper DDR3-1600 1TB WD Caviar Black 
Hard DriveCoolingOSCase
4TB WD Caviar Red Cooler Master Hyper 212+ VMWare ESXi NZXT Crafted Series Tempest 410 
OtherOtherOther
LSI 9280-16i4e RAID Card Intel I350 Quad Port Gigabit NIC Intel Pro/1000 PT Dual Port Gigabit NIC 
  hide details  
Reply
Server
(11 items)
 
  
CPUMotherboardRAMHard Drive
Intel Core i5-3470 ASRock Z77 Extreme6 16 GB G.Skill Sniper DDR3-1600 1TB WD Caviar Black 
Hard DriveCoolingOSCase
4TB WD Caviar Red Cooler Master Hyper 212+ VMWare ESXi NZXT Crafted Series Tempest 410 
OtherOtherOther
LSI 9280-16i4e RAID Card Intel I350 Quad Port Gigabit NIC Intel Pro/1000 PT Dual Port Gigabit NIC 
  hide details  
Reply
post #93 of 131
Thread Starter 
Quote:
Originally Posted by Crazy9000 View Post
I can tell when a game drops down to 57fps from 60, and I haven't been able to tell any difference a/b'ing a keyboard ps/2 vs usb. Except for the 6 key limitation that is. I can hit it on the THPS games after they added the walk function, without that I can only go up to the 6 keys, which works fine on USB.
Try tapping "jump+right" in Donkey Kong. Here, on my keyboards, it either only jumps up, or it just moves right a little a bit, rarely both. After that, assign the the same key for both "jump" and "right", then you can try again and see how it is actually supposed to jump AND move in the same time - "jump sideways". Although, this may very well be application or driver's fault.


USB could potentially outperform PS/2.

a.) Number of packets per second ~ 1000
b.) Number of bytes/keys per packet ~ 8, 10, 32 ..?

If you look at some of those custom keyboard encoders like I-PAC you may see they advertise to be able to send 32 bytes per packet, so in some theory, that USB keyboard would be able to communicate 1000x32 = 32,000 keys per second, way better than PS/2, in theory.

However, all evidence, so far, tells me that keyboard controller & keyboard buffer, to which this data is ultimately forwarded at some point, are still "legacy", and only capable of performing at PS/2 ratings, as before. So, I really doubt USB keyboards can *successfully* plug more than 1000 keys per second into keyboard buffer, regardless of their potential ability to do so, which in my opinion would only make problems like buffer overflow and such. -- Anyhow, would you mind performing some test on those keyboards of yours and share the results with us?

* USB sniffer diagnostic software, free to try:
http://www.topshareware.com/USBTrace-transfer-42419.htm

* Passmark keyboard test
http://www.passmark.com/products/keytest.htm



Here is example of using USB sniffer. It starts with four packets when I pressed [Q+W+E+R] in the same time, I hold them down for a while, and after that red line is when I release them all in the same time. You can see that even though it uses 8 bytes per packet, it still sends as many packets as keys were pressed, one-by-one, 125 keys per second max. -- Perhaps only a new/better driver could make this keyboard perform 1000 keys per second, or maybe even more, and maybe not, who knows.
Edited by dan_k - 10/7/10 at 6:42am
post #94 of 131
Thread Starter 
Quote:
Originally Posted by Manyak View Post
you know, I've never actually tried to see if I could notice the difference. I've never used a USB keyboard on my desktop that I can remember. Not really because of any sort of PS/2 superiority, but because I always figured "why waste a USB port when I've got a dedicated keyboard port right there?" But that was mostly from the PIII/P4 days, when you only had between 2-4 ports.
Yes, things become worse if there are more than one USB device.

I'm really only talking about the numbers, I do not even expect to notice anything with 99% of the games. I just simply want to know that when I waste 5 hours trying to break a record in Donkey Kong that my game was legitimate and genuine as much as possible, so when I fail to jump over that barrel I want to blame only myself, not the keyboard, drivers, or emulator.
post #95 of 131
Quote:
Originally Posted by dan_k View Post
Try tapping "jump+right" in Donkey Kong. Here, on my keyboards, it either only jumps up, or it just moves right a little a bit, rarely both. After that, assign the the same key for both "jump" and "right", then you can try again and see how it is actually supposed to jump AND move in the same time - "jump sideways". Although, this may very well be application or driver's fault.
What keyboard are you using? I just tried a flash version of donkey kong, and was able to jump any direction in the same move without any problems on my realforce via USB.
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
post #96 of 131
Thread Starter 
Quote:
Originally Posted by Crazy9000 View Post
What keyboard are you using? I just tried a flash version of donkey kong, and was able to jump any direction in the same move without any problems on my realforce via USB.
Any keyboard I tried give similar results, USB and PS/2 alike, laptop or desktop all the same, which is why I suspect there is lousy input implementation in MAME and why I am trying to get rid of it all, replace with direct lpt driver and hopefully see the change. -- Perhaps there is a lesson to be learned from that flash Kong, please give me a link so I can do the test on the same game and then we will be able to compare and draw some conclusions.


Do you know if your USB keyboard can send multiple keys per packet?

What minimum lag you get between key presses in "Passmark keyboard test"?
post #97 of 131
I just went with the first result for "donkey kong flash" on google lol

http://www.cyberiapc.com/flashgames/donkeykong.htm

I'm at work now so I'll have to wait to do keyboard tests. If you tell me exactly what you want me to run I'll do that when I get home. I use the 55g Topre Realforce that's on elitekeyboards, I don't believe it does anything unusual with packets. It does support NKRO (if plugged into ps/2 with adapter) if that makes any difference.
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
post #98 of 131
Thread Starter 
Quote:
Originally Posted by Crazy9000 View Post
I just went with the first result for "donkey kong flash" on google lol

http://www.cyberiapc.com/flashgames/donkeykong.htm
That was interesting, thank you. Sometimes I could do it 7-8 times out of 10, and sometimes only 2-3 times out of ten. Then I went back to MAME, did some more testing and found out it makes quite a difference what keys are used and where are they located from each other. So, I had to think of some more precise test as having to use two hands introduced a lot of human error.

This new test requires input keys can be redefined, so we can choose two keys that are next to each other, like Q+W or DOWN+RIGHT arrows keys. Here are some new interesting discoveries:

* when I assign keys for 'right' and 'jump' to Q and W,
then I can do it 9-10 times out of 10

* when I assign keys for 'right' and 'jump' to RIGHT and DOWN arrow keys,
then I can do it only 1-2 times out of 10


Now, I can see only one explanation for this to be so, and that is because the PS/2 packets for arrow keys are actually made of two scan codes, where most of the other keys are described with only one scan code.

And, since this happens equally with USB and PS/2 keyboards of mine, then I conclude USB keyboard protocol is indeed only an outer layer around good old (legacy) keyboard buffer.

However, what is confusing about it, is that it turns out even PS/2 actually has quite a problem of delivering even only four scan codes per frame @ 60fps, while USB keyboard surprisingly can actually deliver 2 keys per frame without a problem, every time just like PS/2, even though with 125Hz polling rate it should not be able to do it so easily.


Quote:
I'm at work now so I'll have to wait to do keyboard tests. If you tell me exactly what you want me to run I'll do that when I get home. I use the 55g Topre Realforce that's on elitekeyboards, I don't believe it does anything unusual with packets. It does support NKRO (if plugged into ps/2 with adapter) if that makes any difference.
That would be most welcome.


A.) * Passmark keyboard test
http://www.passmark.com/products/keytest.htm

1. Press [Q+W+E+R] in the same time, or try banging on the keyboard as fast as you can.

2. In any case see what is the minimum number you can get for "lag time", the time between key presses.




B.) * USB sniffer diagnostic software, free to try:
http://www.topshareware.com/USBTrace-transfer-42419.htm

1. Start session

2. Press THREE buttons in the same time [Q+W+E]
...hold ~2 seconds

3. Release all three buttons in the same time
...wait ~5 seconds

4. Press FOUR buttons in the same time [Q+W+E+R]
...hold ~2 seconds

5. Release all four buttons in the same time

6. Take a screen capture you can upload here

* Maybe also apply filters, so to only show "IN" packets with "DATA".



...hopefully you will have fun doing it too
Edited by dan_k - 10/7/10 at 11:17pm
post #99 of 131
Quote:
Originally Posted by lmnop View Post
MOAR GOLD

She could do with less gold and more ass.

Quote:
Originally Posted by De-Zant View Post
I don't understand why people prefer USB over PS/2 for ANYTHING else than hotswapping...

USB polls 1000hz... PS/2 is interrupt based, making it very much more than 1000hz....

PS/2 has NKRO. My keyboard can hit any key and have it register . I can press ALL the keys at the same time and have them register. My keyboard is PS/2....

And YOUR USB keyboard has a maximum of 6 alphanumeric keys and 4 modifiers at a time. However, it's rare for a keyboard to support this; most keyboards only support less than 3 keys at a time at certain key combinations....

The fact of the matter is: PS/2 wins for everything except hotswapping...
Yep.

Still, very few people are going to actually notice the difference in use.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #100 of 131
Alright, here you go:
LL
LL
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
SUPERPWN
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 7820x @ 4.7 GHz Asrock x299 OC Formula MSI GTX 1080 Aero 32GB Gskill DDR4-4000 
Hard DriveOSMonitorMonitor
4TB Seagate Windows 8 x64 Overlord x270 OC HP ZR27 
MonitorKeyboardPowerAudio
Asus PG279Q G-Tune Topre Realforce 800w Benchmark Dac2 | Audeze LCD-X 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Keyboards
Overclock.net › Forums › Components › Keyboards › USB vs PS/2 Keyboard: latency, bandwidth -> inputs per second?