Build Your Own Arcade Controls Forum
Main => Main Forum => Topic started by: crashdmj on May 30, 2003, 12:51:40 pm
-
Is there any reason why a lot of people choose the IPac for USB over PS2, the only thing I can think of choosing PS2 is that you can leave your USB port open for gamepads.
-
PS/2 mode works (slightly) better than USB mode. A standard USB keyboard controller is supposed to only allow a max of 6 simultaneous inputs. Andy has worked around that somehow but I don't know to what extent. I have personally noticed some slight intermittent problems with the I-Pac in USB mode and none whatsoever in PS/2 mode.
Major advantage of USB mode is it's ability to be hotswapped. This is useful for a standalone hotrod style control panel for use on a typical desktop or laptop machine. You plug it in when you want to play and unplug when you don't. No need to shutdown and re-start like you should when attaching a PS/2 device.
If you are making a dedicated machine that will have the controller attached permanently then use the PS/2 mode.
-
I actually had to use PS/2 since I am running Advance Mame in Dos. I tried using the Ipac in USB mode and enabling the "USB Keyboard" option in Bios, but it didn't work properly (kept pulling one direction or the other).
-
Major advantage of USB mode is it's ability to be hotswapped. This is useful for a standalone hotrod style control panel for use on a typical desktop or laptop machine. You plug it in when you want to play and unplug when you don't. No need to shutdown and re-start like you should when attaching a PS/2 device.
I just want to add something to this. In the old days, hot-swapping a PS2 device was a real bad idea. The old keyboards actually used a lot more power, and there seemed to be less of it available on the port. A blown fuse on the mother board was a common occurance.
Eventually, and keep in mind that this was before USB, the PS2 port was turned to for supplying +5v to external devices. GPS's, joysticks, tablets and other devices had little passthroughs that allowed a device to "leech" it's power from the PS2 port. This was to help make a device as portable and easy to install as possible, and to save money by not needing to include dedicated power supplies.
This led to the PS2 port becoming more robust and less susceptible to damage on newer computers. Just to give an example of this, on my test machine (cheapo Emachines Celeron) I have hot swapped hundreds of times, and never had a problem.
I even had a dead short on a breadboarded project (which cooked a resistor on my project and stunk up the room) and even that didn't kill the port.
Of course, this is all anecdotal and if you have an old system, caution is in order. But newer systems (less than 3 years old) probably won't have much of a problem with a hot-swap on the PS2 port.
YMMV :)
RandyT
-
I actually had to use PS/2 since I am running Advance Mame in Dos. I tried using the Ipac in USB mode and enabling the "USB Keyboard" option in Bios, but it didn't work properly (kept pulling one direction or the other).
Actually, Andy replied to this in a thread and recommended disabling USB keyboard in BIOS and said it should work.
That said, the I-PAC has a limitation of 14 or 15 simultaneous keypresses in USB mode and unlimited in PS/2 mode.
And as Randy T said, I have a 6 or 7 year old computer and have hot-swapped the PS/2 port dozens of times with no ill effects.
-
I too have swapped my PS/2 port hundreds of times out of ignorance without the MoBo complaining even once. Almost all of that swapping has been fidgeting with the IPac.
I read recently that this was not advisable. However, after reading about how so many of you have done it without problems, I'm getting encouraged to have two separate mouse hacks for spinner and trackball and switch between them with a custom made selector switch.
_Iz-: You know what I'm talking about ;D
-
I too have swapped my PS/2 port hundreds of times out of ignorance without the MoBo complaining even once. Almost all of that swapping has been fidgeting with the IPac.
I read recently that this was not advisable. However, after reading about how so many of you have done it without problems, I'm getting encouraged to have two separate mouse hacks for spinner and trackball and switch between them with a custom made selector switch.
_Iz-: You know what I'm talking about ;D
Well, the earlier comments were about the PS/2 keyboard port, and I can't say whether the same thing applies to the PS/2 mouse port. ???
-
I originally ran my I-Pac in PS/2 mode but noticed when I changed it to USB it moved through the menu on Mame32 so much quicker when pulling down on the joystick which I was quite happy about because it took forever before in PS/2 mode. I have'nt noticed any differences while playing games though.
-
I originally ran my I-Pac in PS/2 mode but noticed when I changed it to USB it moved through the menu on Mame32 so much quicker when pulling down on the joystick which I was quite happy about because it took forever before in PS/2 mode. I have'nt noticed any differences while playing games though.
I'm guessing it enables KeyRepeat in USB mode and not in PS/2. I know the KeyWiz has this feature enabled even though it is PS/2 only, FWIW.
-
Well, the earlier comments were about the PS/2 keyboard port, and I can't say whether the same thing applies to the PS/2 mouse port. ???
Dough!!!
There goes that idea.
-
The method I discussed involved 1 mouse hack and using a switch to feed it with input from 2 seperate devices depending on the switch position. This would not require hotswapping the mouse as it is always connected and all devices have power at all times. You're just allowing the selection of input to feed the mouse pcb. With this method the computer doesn't have any idea anything has changed! No risk of frying the port.
Check this writeup. It's basically the same as I have done.
http://www.oscarcontrols.com/DPDTswitch.shtml
-
Just to make a silly point:
You most likely can hot swap in/out/off ps/2 keyboards while the machine is on (without hardware damage)... but windoze usually gets pissed and will make you reboot anyways =P
at least that's been my experience.
Rampy
-
Just to make a silly point:
You most likely can hot swap in/out/off ps/2 keyboards while the machine is on (without hardware damage)... but windoze usually gets pissed and will make you reboot anyways =P
at least that's been my experience.
Hey Rampy!
How do I disagree with you without ticking you off :) ?
Seriously though, I haven't found this to be the case. Mouse? Yes, but not the keyboard. I routinely hot swap keyboards and encoders half a dozen times at a sitting, and I don't have to reboot.
The only thing that might get weird is the LED lock indicators. They could possibly get out of sync.
But it could also just be your/my keyboard :)
RandyT
-
FYI. I have been a computer tech since before the PS/2 connector. ;) In all my years only one fuse has ever blown. And that one was from a keyboard that someone else had bent a pin on, and that pin went across 5v and GND. And I "fixed" that MB.
-
FYI. I have been a computer tech since before the PS/2 connector. ;) In all my years only one fuse has ever blown. And that one was from a keyboard that someone else had bent a pin on, and that pin went across 5v and GND. And I "fixed" that MB.
That's good news.
Does your comment apply to PS/2 in general including mice?
-
Just to make a silly point:
You most likely can hot swap in/out/off ps/2 keyboards while the machine is on (without hardware damage)... but windoze usually gets pissed and will make you reboot anyways =P
at least that's been my experience.
Hey Rampy!
How do I disagree with you without ticking you off :) ?
Seriously though, I haven't found this to be the case. Mouse? Yes, but not the keyboard. I routinely hot swap keyboards and encoders half a dozen times at a sitting, and I don't have to reboot.
The only thing that might get weird is the LED lock indicators. They could possibly get out of sync.
But it could also just be your/my keyboard :)
RandyT
How do I agree with RandyT without ticking anyone else off :-)
I too have swapped KB's many times and not had windows complain (Ooops . . . inserting foot ) Now that I think about it. These were AT keyboards (Large 5-pin DIN port). Anyways, I've never known windoze to care. I think I have seen the LED's get out of synch, tho'
FWIW!
-
FYI. I have been a computer tech since before the PS/2 connector. ;) In all my years only one fuse has ever blown. And that one was from a keyboard that someone else had bent a pin on, and that pin went across 5v and GND. And I "fixed" that MB.
That's good news.
Does your comment apply to PS/2 in general including mice?
I'm not a computer tech, but I've removed & replugged both keyboards on mice in PS/2 ports. I've have no problems with keyboards (used to fix a "stuck" keyboard by unplugging & replugging in an old computer I had).
Mice, on the other hand, are more picky. Unplugging and replugging the same mouse: no problems. Unplugging and plugging in a different mouse rarely worked for me; but if I replugged the mouse I unplugged in the first place, usually things would go back to normal. (Makes sense with all the different mouse drivers.)
Just my experiences. *shrug*
-
I have always hot-swapped PS/2 before but was surprised to find that on three new socket 7 motherboards with VIA chipsets doing this locks up the keyboard controller on the M/B every time and the keyboard and mouse stop working until you power off/on.
Actually the motherboard manufacturers have been trying to dump PS/2 for ages as it is an interface without any proper standard and is generally a pain. We are starting now to see micro-ATX boards with no PS/2 but I expect it will be around for a few more years yet on full size boards.
On the subject of USB vs PS/2: In Windows 98 Microsoft didnt implement USB properly so it may be slightly slower than PS/2 although it's probably not noticeable. The problem mentioned above in DOS is a different issue, the I-PAC should not be used in DOS in USB mode as this is relying on the PC BIOS for support, which is not up to the job when it comes to high speed gameplay.
USB should be super-fast in XP and W2K which have a proper implementation. This should be especially noticeable in games which need simultaneous button presses, as all buttons are sent in one "report" at the speed of the USB bus instead of being piped down the PS/2 interface in a relatively slow manner.
-
Actually the motherboard manufacturers have been trying to dump PS/2 for ages as it is an interface without any proper standard and is generally a pain.
Please explain this comment. I know the PS/2 interface pretty well and there certainly is a proper standard. IBM developed it and it is well documented. There is one thing you can always count on with a properly designed PS/2 device. You plug it in and it works.
We are starting now to see micro-ATX boards with no PS/2 but I expect it will be around for a few more years yet on full size boards.
PS/2 Keyboards and mice probably outnumber their USB counterparts by at least 1000 to one. MB manufacturers know that forcing the consumer to purchase a new keyboard and mouse in order to upgrade their system will result in a significant reduction of their potential market.
It has also been my observation that these "USB only" systems don't seem to sell very well. Many fall into the set-top box / internet appliance category and I routinely see them in large quantities in the surplus listings. On the other hand, the hottest selling boards of this type are the mini-ITX Via boards (even smaller than the mini-ATX) and they have the full complement of PS/2 ports. So it appears that size doesn't matter.
This, coupled with the fact that it costs them very little to implement the interface (the code could be implemented on the same chip as the one that controls the USB, if it isn't already) tells me that the PS/2 ports will be with us until the consumer deems it unnecessary. And considering that there are other PS/2 devices used in commercial settings such as bar-code readers, card swipes, etc., that may be a very long time.
....snip
This should be especially noticeable in games which need simultaneous button presses, as all buttons are sent in one "report" at the speed of the USB bus instead of being piped down the PS/2 interface in a relatively slow manner.
The USB keyboard is considered a "low-speed" device, as is the microcontroller used in devices that behave like one. The reports don't travel at the speed of the USB bus, but at the speed maintainable by the controller. Not that this speed is inadequate, but the statement you made here isn't true because the USB bus can handle speeds far in excess of these devices.
Also, USB is still serial by nature (Universal Serial Bus). The bits are piped down the line, one after another, just like the PS/2 or standard serial ports. Just because the bits are assembled into a "report" before sending doesn't mean it will have a positive effect on speed. In fact, it could take longer due to the extra processing required to do the report assembly, especially if two reports are used to circumvent the 6 simultaneous keypress limitation. But this will depend on a number of other factors as well.
But nonetheless, this "report assembly" is a step that the PS/2 interface doesn't need to perform.
Of course, for the most part, original arcade machines only dealt with the controls once every 1/60th of a second anyway, so all of this is probably moot ;).
RandyT
-
Randy, your post was too long to quote, but I disagree with a few things you say. For one, the writing is on the wall for the PS2 port. Sure it will be a few years, but more and more the keyboards and mice in the stores are USB (sometimes they come with a USB to PS2 converter). The consumers have already said that they don't want PS2 anymore. Or rather, getting rid of PS2 simplifies the attachment of devices to the PC, thus the *average* consumer benefits and is less scared of the beast. PS2 is going the way of the floppy drive (Both Dell and Gateway sell their PCs without floppy drives by default now).
Also, regarding your comment about the USB port suffering because of its serial architecture.....I don't know too much about this but perhaps you could explain why this is a limitation. We are just now moving from a parallel IDE bus to a serial one and with the first iteration we jump from a maximum of 133 MB/sec with parallel to 150 MB/sec with the new Serial ATA. They also have a lot of room to push that bandwidth up steadily throughout the next few years.
-
Sure it will be a few years, but more and more the keyboards and mice in the stores are USB (sometimes they come with a USB to PS2 converter). The consumers have already said that they don't want PS2 anymore. Or rather, getting rid of PS2 simplifies the attachment of devices to the PC, thus the *average* consumer benefits and is less scared of the beast.
It's ok to disagree, I do it all the time :)
I'm not sure which consumers you are talking about that have "said" this or where they speak collectively, other than with their wallets. But the single most common thing I hear with USB is "where do I plug this in?" To which I reply, "the square shaped port that is the same size as the plug". Then I hear, "Which one? There are 4 of them?". Of course then I say, "Any of them, they are all the same". Inevitably, the next comment is "I tried, and it doesn't fit".
This is because the symmetrical nature of the connector makes people believe that the moment it doesn't slide into the port that something is wrong. They can't easily tell by looking at it that it might be upside down and sometimes it can be tricky even when correctly oriented :).
BTW, I am an IT manager, and of the 5 people I have recently given USB devices to for their computers, this was the case with each individual to one extent or another. If it's easier for the "average" person, I haven't seen it, but that's just my experience.
On the other hand, the color coded port connectors seem to work pretty well for all the other ports. I have been consistantly surprised by how many "less than average" users were able to get everything hooked up properly when setting up complete systems with this color scheme implemented.
PS2 is going the way of the floppy drive (Both Dell and Gateway sell their PCs without floppy drives by default now).
The floppy drive is strictly a cost saving measure. When they virtually stop selling 1.44 meg floppy disks and drives (like they did with the 5 1/4"), then it would be a good indicator. You wouldn't believe how many times I have heard "You mean they expect me to burn a CD just to take home a Word document to work on?" Shortly after that I am told to make sure there is a floppy drive in the system, or order one and install it seperately.
Also, regarding your comment about the USB port suffering because of its serial architecture.....I don't know too much about this but perhaps you could explain why this is a limitation. We are just now moving from a parallel IDE bus to a serial one and with the first iteration we jump from a maximum of 133 MB/sec with parallel to 150 MB/sec with the new Serial ATA. They also have a lot of room to push that bandwidth up steadily throughout the next few years.
You are comparing apples and oranges and inferring that I said something I did not. There was a comparison being made between how a PS/2 interface operated and how a keyboard USB application operated. I merely pointed out the similarities, especially using the microcontrollers in question. If you re-read what I wrote, I stated that even normal USB is extremely fast. The microcontrollers are the limitation. Don't think for a second that Serial ATA will use the same class of hardware or OS drivers that are common to serial Keyboard applications. They are not even close :) .
RandyT