Originally Posted by storm20200 Thank you soooooooooooo much bit-tech, this is most certainly going to be my motherboard :) I'm hoping it will be around the £100 mark but life tells me it will certainly be £120+ T_T but as for the 4 USB ports you talked about where are they? I could only see two? Maybe I'm blind but I need 2 for my HAF 932 case and 1 for a floppy/card reader drive.
I would expect the UD6 to be at least £150-160, while the UD5 slots in slightly beneath that, between £130-140. Bear in mind they also have to fit in 4 other mobos below that in the pricing structure. My guess in the UD4 will be the one closer to £120, with only the mATX boards being below £100
Originally Posted by storm20200 Thank you soooooooooooo much bit-tech, this is most certainly going to be my motherboard :) I'm hoping it will be around the £100 mark but life tells me it will certainly be £120+ T_T but as for the 4 USB ports you talked about where are they? I could only see two? Maybe I'm blind but I need 2 for my HAF 932 case and 1 for a floppy/card reader drive.
I would expect the UD6 to be at least £150-160, while the UD5 slots in slightly beneath that, between £130-140. Bear in mind they also have to fit in 4 other mobos below that in the pricing structure. My guess in the UD4 will be the one closer to £120, with only the mATX boards being below £100
Originally Posted by naokaji Why the **** did they put a plastic cap where the SB heatsink used to be in past dual chipset designs? doesn't make much sense to me.
It's aluminium, and to quote Gigabyte - more metal looks faster. It's a theory that works too - I've read people complain about small or not enough heatsinks there = slow before. It's simply psychological. You cant blame them listening to users.
Quote:
With the number of skus they are without a doubt going to release they should have at least one that does not have any ancient connectors.
Nope, they will all be there for P55, it's a mainstream product and emerging markets where the most money is to be made, demands them.
Originally Posted by Bindibadgi For this very reason, we still exclusively use PS2 keyboards in the labs. PS2 mice are long, long gone, but keyboards are still very useful until EFI FINALLY arrives.
There has got to be an easy way to get the nested quote, but I'm too lazy to do it the hard way.
I will try and explain what it takes to bring up the PS/2 port vs the USB port. but it'll be a quickie, so I don't plan on going in depth about PC architecture.
The problem is that the is a s LOT of stuff that needs to get configured for a PC to 'boot'. A modern PC is not a monolithic thing. It is insead a collection of a lot of smaller things. All of these smaller things have registers. These are like tiny pieces of memory (in some cases, they are memory) that need to have the correct numbers placed into them. And they can't be done in one shot, there is an order they must be done in.
The amount of work to get a PS/2 keyboard running is far less than what you need for USB. Even if the system did not have any RAM in it, you could still write software to operate a PS/2 keyboard. The same cannot be said for USB.
So here is the process for the two interfaces on an Intel chipset (since that is what I know) briefly, menaing there will be tech terms I don't explain. It's the amount of effort that is imporant, not the techno babble :)
For PS/2,
1) The LPC bridge needs to have its I/O decode ranges set up to program the Super I/O chip, as well as ports 60 and 64 for the keyboard.
2) (optional) The Super I/O needs to have the PS/2 keyboard controller turned on. But the chip can also be strapped to have the controller come up automatically so this is an optional set.
3) Communicate witht eh keyboard via I/O ports 60/64
For USB,
1) Get memory initialized so the PCI BARs can be programmed
2) Program the PCI BARs for the USB host controllers. All of them probally need to be programmed since the keyboard can be on any port and could be on either USB 1.1 or USB 2.0
3) Set up the USB stack
4) Enumerate the bus
5) if it is a USB HID, see if it is a keyboard and supports the keyboard boot protocol
6) set the keyboard to use the boot protocol
7) communicate with USB stack using the vendor defined interface
Now consider having to write all the assembly code needed for either system. Which one sounds esier to do and fix when it's broken? :)
@Bindi UEFI is out there, but not too much in the consumer space. There is some apprehension about how consumers will recieve it so it will probably creep in unannounced rather than be touted as a feature. But before that happens, OEMs and ODMs need to learn how to work with it first. I do not think I am overstating when I say it is a drastic change from developing 'legacy' BIOS.
@Bindi UEFI is out there, but not too much in the consumer space. There is some apprehension about how consumers will recieve it so it will probably creep in unannounced rather than be touted as a feature. But before that happens, OEMs and ODMs need to learn how to work with it first. I do not think I am overstating when I say it is a drastic change from developing 'legacy' BIOS.
Originally Posted by Splynncryth For PS/2,
1) The LPC bridge needs to have its I/O decode ranges set up to program the Super I/O chip, as well as ports 60 and 64 for the keyboard.
2) (optional) The Super I/O needs to have the PS/2 keyboard controller turned on. But the chip can also be strapped to have the controller come up automatically so this is an optional set.
3) Communicate witht eh keyboard via I/O ports 60/64
For USB,
1) Get memory initialized so the PCI BARs can be programmed
2) Program the PCI BARs for the USB host controllers. All of them probally need to be programmed since the keyboard can be on any port and could be on either USB 1.1 or USB 2.0
3) Set up the USB stack
4) Enumerate the bus
5) if it is a USB HID, see if it is a keyboard and supports the keyboard boot protocol
6) set the keyboard to use the boot protocol
7) communicate with USB stack using the vendor defined interface
Now consider having to write all the assembly code needed for either system. Which one sounds esier to do and fix when it's broken? :)
+Rep for info. :)
Quote:
Originally Posted by Bindibadgi For this very reason, we still exclusively use PS2 keyboards in the labs. PS2 mice are long, long gone, but keyboards are still very useful until EFI FINALLY arrives.
I was happy to see the end of PS/2 mice (although I'd been happier to see the death of serial mice before that! Horrible things...) but when PS/2 keyboards finally go, I'll have to get rid of my favourite keyboard. :( I've tried various replacements over the years, and none of them are as good.
the UD5 seen on this picture seems to have 6 memory slots, different than what BT have shown.
but it only supports 16GB, so i think its point is to allow use of 4x2GB for now, and upgrade by adding 2x4GB later in the white slots?
if there's no compatibility and overclocking issues with 6 memory slots, im sold on this 6 memory slot version! i've been waiting for 4GB DDR3 sticks for a long time. and this motherboard looks like a money saver
Originally Posted by wuyanxu the UD5 seen on this picture seems to have 6 memory slots, different than what BT have shown.
but it only supports 16GB, so i think its point is to allow use of 4x2GB for now, and upgrade by adding 2x4GB later in the white slots?
if there's no compatibility and overclocking issues with 6 memory slots, im sold on this 6 memory slot version! i've been waiting for 4GB DDR3 sticks for a long time. and this motherboard looks like a money saver
That's P57 - it has Intel flash soldered to the board under the memory slots. It's mislabeled.
It's still dual channel, but with an extra bank for more memory within the same channel.
There will be a limit to the density of memory supported by four sticks in one channel - there's only so much addressable space.
Looks nice and the combo SLI/Crossfire is a definite bonus. I do find having the EIDE connector available on just in case grounds pretty handy, especially when trying to grab data from an old drive when someone has "borrowed" your external adaptor. Grrrr....
PS/2 is useful as well - I don't mind needing an extra mouse but one keyboard on the desk at a time is sufficient and my KVM switch doesn't support USB.
Any word on which features the different models (i.e. UD6/UD5/etc.) will include/won't include? Be nice to know how 'cheap' I can go before losing important features. :)
Nice board though; might well be my next, barring any mess-ups.
Yes, it's out there. The MSI thing was a rather visible, bu there are other examples I know about as well as some rumors. One I found interesting is that the Wind netbook runs EFI, but I don't have one to mess with and see.
One of the goals in a number of the products I have worked on was to make sure everything behaved like legacy BIOS.
I have seen some of that come through Bit, but that stuff usually isn't aimed at the desktop space.
I could ramble on, but I think I have done enough thread hijacking for a bit.
Comments 26 to 43 of 43
ReplyWith the number of skus they are without a doubt going to release they should have at least one that does not have any ancient connectors.
But regardless of that, looks like the Gigabyte P55 boards are shaping up nicely.
I would expect the UD6 to be at least £150-160, while the UD5 slots in slightly beneath that, between £130-140. Bear in mind they also have to fit in 4 other mobos below that in the pricing structure. My guess in the UD4 will be the one closer to £120, with only the mATX boards being below £100
Sadface....
It's aluminium, and to quote Gigabyte - more metal looks faster. It's a theory that works too - I've read people complain about small or not enough heatsinks there = slow before. It's simply psychological. You cant blame them listening to users.
Nope, they will all be there for P55, it's a mainstream product and emerging markets where the most money is to be made, demands them.
There has got to be an easy way to get the nested quote, but I'm too lazy to do it the hard way.
I will try and explain what it takes to bring up the PS/2 port vs the USB port. but it'll be a quickie, so I don't plan on going in depth about PC architecture.
The problem is that the is a s LOT of stuff that needs to get configured for a PC to 'boot'. A modern PC is not a monolithic thing. It is insead a collection of a lot of smaller things. All of these smaller things have registers. These are like tiny pieces of memory (in some cases, they are memory) that need to have the correct numbers placed into them. And they can't be done in one shot, there is an order they must be done in.
The amount of work to get a PS/2 keyboard running is far less than what you need for USB. Even if the system did not have any RAM in it, you could still write software to operate a PS/2 keyboard. The same cannot be said for USB.
So here is the process for the two interfaces on an Intel chipset (since that is what I know) briefly, menaing there will be tech terms I don't explain. It's the amount of effort that is imporant, not the techno babble :)
For PS/2,
1) The LPC bridge needs to have its I/O decode ranges set up to program the Super I/O chip, as well as ports 60 and 64 for the keyboard.
2) (optional) The Super I/O needs to have the PS/2 keyboard controller turned on. But the chip can also be strapped to have the controller come up automatically so this is an optional set.
3) Communicate witht eh keyboard via I/O ports 60/64
For USB,
1) Get memory initialized so the PCI BARs can be programmed
2) Program the PCI BARs for the USB host controllers. All of them probally need to be programmed since the keyboard can be on any port and could be on either USB 1.1 or USB 2.0
3) Set up the USB stack
4) Enumerate the bus
5) if it is a USB HID, see if it is a keyboard and supports the keyboard boot protocol
6) set the keyboard to use the boot protocol
7) communicate with USB stack using the vendor defined interface
Now consider having to write all the assembly code needed for either system. Which one sounds esier to do and fix when it's broken? :)
@Bindi UEFI is out there, but not too much in the consumer space. There is some apprehension about how consumers will recieve it so it will probably creep in unannounced rather than be touted as a feature. But before that happens, OEMs and ODMs need to learn how to work with it first. I do not think I am overstating when I say it is a drastic change from developing 'legacy' BIOS.
http://www.bit-tech.net/hardware/motherboards/2009/01/11/msi-click-bios-evaluating-uefi-review/1
:) Already looked at it.
but it only supports 16GB, so i think its point is to allow use of 4x2GB for now, and upgrade by adding 2x4GB later in the white slots?
if there's no compatibility and overclocking issues with 6 memory slots, im sold on this 6 memory slot version! i've been waiting for 4GB DDR3 sticks for a long time. and this motherboard looks like a money saver
That's P57 - it has Intel flash soldered to the board under the memory slots. It's mislabeled.
It's still dual channel, but with an extra bank for more memory within the same channel.
There will be a limit to the density of memory supported by four sticks in one channel - there's only so much addressable space.
http://www.tomshardware.com/picturestory/513-p55-ud6-p55-ud4p-gigabyte.html
does anybody know when SATA 6 HDD's / SDD's appear ?
Braidwood flash socket.
surely there'd be no point for tiny flash memory if we are using SSD and have 16GB of RAM. (yes, i'll wait for 4GB sticks then upgrade)
PS/2 is useful as well - I don't mind needing an extra mouse but one keyboard on the desk at a time is sufficient and my KVM switch doesn't support USB.
Nice board though; might well be my next, barring any mess-ups.
Yes, it's out there. The MSI thing was a rather visible, bu there are other examples I know about as well as some rumors. One I found interesting is that the Wind netbook runs EFI, but I don't have one to mess with and see.
One of the goals in a number of the products I have worked on was to make sure everything behaved like legacy BIOS.
I have seen some of that come through Bit, but that stuff usually isn't aimed at the desktop space.
I could ramble on, but I think I have done enough thread hijacking for a bit.
-
« Previous
-
1
-
2
-
Next »
Discuss in the forums