AMD Processors
Decrease font size
Increase font size
Topic Title: Is it possible to make simple SocketF -->S940 adapter ?
Topic Summary:
Created On: 02/12/2007 03:31 AM
Status: Read Only
Linear : Threading : Single : Branch
<< 1 2 3 4 Previous Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 02/20/2007 09:38 PM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

quote:

Originally posted by: mckennma
Talk to them about signing a NDA. I have them at many companies I test with.



Don't you find it ironic that one can have data for just about everything else, but has to sign an NDA to get data for a socket that is supposed to be _open_ for third parties ?

At least this is how I have understood AMD's statement regarding open platforms etc...

Well, I'll ask about pacting with a devil, but I doubt that they'll accept NDA signed by non-organisation... /rolleyes.gif" style="vertical-align:middle" emoid=":rolleyes:" border="0" alt="rolleyes.gif" />
 02/20/2007 10:48 PM
User is offline View Users Profile Print this message

Author Icon
zeppelinrox
Senior Member

Posts: 8484
Joined: 11/30/2005

man, just BUY A MOBO.. geezus

you may believe the saying, a penny saved is a penny earned.

but time is priceless, and you can never get it back once you waste it all on some nonsensical notion akin to trying to make water flow uphill

-------------------------


Overclocking Help and Links . Why DRM, BluRay and HD-DVD Blow Great tools-> OC Bible v1.52 and Guidemania v1.2
 02/20/2007 11:34 PM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

quote:

Originally posted by: brankob
Don't you find it ironic that one can have data for just about everything else, but has to sign an NDA to get data for a socket that is supposed to be _open_ for third parties ?

At least this is how I have understood AMD's statement regarding open platforms etc...

Well, I'll ask about pacting with a devil, but I doubt that they'll accept NDA signed by non-organisation... /rolleyes.gif" style="vertical-align:middle" emoid=":rolleyes:" border="0" alt="rolleyes.gif" />



Information about that should be protected. Otherwise, Intel and AMD would know what each other are doing. Just buy a board like the rest of us do. Design information is always protected.

-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/21/2007 12:05 AM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

quote:

Originally posted by: mckennma
Information about that should be protected. Otherwise, Intel and AMD would know what each other are doing. Just buy a board like the rest of us do. Design information is always protected.




That doesn't make much sense. AMD has declared Socket F open for third party products.

That means that Intel could get to that data.

Even if that wasn't the case, Intel would have the data it needs in 30 min flat if they really needed to.

They could just get a few boards and analyze signals. They already have much of the data and would need to pick basically just timings and pinout...
 02/21/2007 09:06 AM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

quote:

Originally posted by: brankob
That doesn't make much sense. AMD has declared Socket F open for third party products.

That means that Intel could get to that data.

Even if that wasn't the case, Intel would have the data it needs in 30 min flat if they really needed to.

They could just get a few boards and analyze signals. They already have much of the data and would need to pick basically just timings and pinout...



You don't understand hardware and software design. You need to realize you lack the education and experience to even get to the first step of designing your device. It is not as simple as soldering a few wires to match up. Your idea is flawed.

I have designed motherboards in the lab. If you knew how to do this, you would realize what you want to do will not work. I think enough knowledgable people on the site and told you the same thing.

They can't get design data. Only data on code to get it to function with software. You don't need pin outs, gate layout, wafer configuration to write software. They get a list of API calls for the CPU for programming. Not how the CPU was developed and advanced information on gate design and wafer layouts. They also get the information required for the BIOS to function with hardware APIs.

If it was so easy to do, a company would have already designed it for you to buy. It is not worth you time and effort to even continue thinking about it. Spend your time at work and studying in school. Get a degree in engineering.

-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/21/2007 01:23 PM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

quote:

Originally posted by: mckennma
You don't understand hardware and software design. You need to realize you lack the education and experience to even get to the first step of designing your device. It is not as simple as soldering a few wires to match up. Your idea is flawed.

I have designed motherboards in the lab. If you knew how to do this, you would realize what you want to do will not work. I think enough knowledgable people on the site and told you the same thing.

They can't get design data. Only data on code to get it to function with software. You don't need pin outs, gate layout, wafer configuration to write software. They get a list of API calls for the CPU for programming. Not how the CPU was developed and advanced information on gate design and wafer layouts. They also get the information required for the BIOS to function with hardware APIs.

If it was so easy to do, a company would have already designed it for you to buy. It is not worth you time and effort to even continue thinking about it. Spend your time at work and studying in school. Get a degree in engineering.




At this point I have to say that I am dissapointed with AMD in general. One would expect that company that shelled a few $1Gig for ATI wouldn't have troubles maintaining decent forum and site for developers.

After registering at developer.amd.com, I couldn't shake off feeling that I'm really on Amway's network 21 or somesuch beast. It isn't clear what is AMD's intention with it. Are they after stimulating HW and SW development for their chips or they really want to develop "byu,byu,byu!" reflex ?

They are trying to push "machines with special, friendly prices for developers" there, which actually wouldn't be so bad if there were good offers and if the rest of the site was useful, but there is really no relevant documentation for new chips. One can get some data on old chips, but very little about new ones.

For Socket F Optys there is only "Power and Thermal data sheet" which contains about as much data as you could get for a lightbulb.

All that AMD's "Fusion" song and dance seems stupid. They declared some socket as an "open" invent faggotistic name for anticipated third-party cooperation and then _closed_ public acces to the data.

How can this be "Fusion" if old, "closed" sockets had been publicly available and new one, which is supposed to be "open", is under NDA ?
Wouldn't "Fission" (like in "We are saying bye,bye, you're on yer own") be more appropriate name for it ?

And this forum is probably one of more useless of all. It is slow, inflexible, ugly and once more _sloooow_. And that is before anything is said about its content, which is more or less unuseable. Some "technical articles" on about same level of technical expertise as average TV car commercial and a bunch of pretty bleak posts. I can't say that I have ever seen any useable bit of info that I haven't see elsewhere already.

I have found it through google,s earching for the solution of the S2885's problem with new AGP cards. derGhostrider has published some (unsuccesfull, but with noble intentions) effort to correct this. I have kept hoping that someone from AMD will shed some light on it. After getting a bunch of irrelevant and false technical answers and a round of "BuyBuyBuy" routine I kept hoping that I could evntually learn something here. Like some nice optimisation tricks with NUMA, VT, Linux etc etc etc. But nothing like that ever came up.

It's perfectly undestandable that people have ordinary problems and that they expect to be able to solve them and that such forum has to deal with that. But there should be some activity in "advanced stuff" questions or this forum has no real meaning, since all this has been asked and answered elsewhere _MANY_ times.

But if the forum is short in advanced stuff matter, it certainly isn't in quantity of beer gut philosophers "technical advisors". Quantity of their opinions ovewrhelmes any possibly remotely useful bit of info that might be there.

When I open the topic with inquiry for some data, I do expect answers to be more or less on topic. I don't mind occasional philosophy excursions and "opinions", but please learn when to stop. Thank you for opinions, they have been noted, but now I would like to see some useful post with requested data from someone...
 02/21/2007 01:36 PM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

Oh, one more tidbit:

I don't undestand your reasoning. I have requested Datasheet for Socket F.
AMD says it's under NDA.

You said that this is undestandable, since they wan't to fend off Intel.

I said this is nonsense, since:

1. Under "Fusion" socket F is practically declared open for third parties. If the socket is open and Intel wants to be a third party for some product, what's to keep them from getting the data ? Or are you sayng that "Fusion" doesn't mean "open" like open for anyone that wants to participate ?

2. Datasheet contains data about electrical characteristics and timings. If Intel had important goal to get to that data, tehy could couple a decent logic analyzer on the chip and get that info. Even pinout could be ripped, since most of the signals are from standard interfaces (HT links, memory links, which comprise probably 95% or more of all signal pins).

You mentioned some internal stuff registers, but those are not really relevant for datasheet, plus mosto of them will have to be publicaly known since Linux is an open source.
Some stuff _might_ stay in the hands of BIOS developers ( like some special configuration registers etc) but that has nothing to do with Datasheet, at least compared to AMD's datasheet for S940 Optys...
 02/21/2007 03:51 PM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

quote:

Originally posted by: brankob
Oh, one more tidbit:

I don't undestand your reasoning. I have requested Datasheet for Socket F.
AMD says it's under NDA.

You said that this is undestandable, since they wan't to fend off Intel.

I said this is nonsense, since:

1. Under "Fusion" socket F is practically declared open for third parties. If the socket is open and Intel wants to be a third party for some product, what's to keep them from getting the data ? Or are you sayng that "Fusion" doesn't mean "open" like open for anyone that wants to participate ?

2. Datasheet contains data about electrical characteristics and timings. If Intel had important goal to get to that data, tehy could couple a decent logic analyzer on the chip and get that info. Even pinout could be ripped, since most of the signals are from standard interfaces (HT links, memory links, which comprise probably 95% or more of all signal pins).

You mentioned some internal stuff registers, but those are not really relevant for datasheet, plus mosto of them will have to be publicaly known since Linux is an open source.
Some stuff _might_ stay in the hands of BIOS developers ( like some special configuration registers etc) but that has nothing to do with Datasheet, at least compared to AMD's datasheet for S940 Optys...



It does not matter since you probably don't qualify for a NDA. So you will never find out. Call them about getting a NDA or close the call. Don't ramble on what is fair and not fair. What is open source and protected by NDA? I really have nothing else to say on it. It is what it is. Sun, IBM, and Intel will do the same thing. Just like Microsoft charging $100K + $25K for testing/qualifying each signed driver for Vista. The NDA signed developers site is better than normal web site. You also have engineers on hand to answer questions at a huge fee. Hardware developers can pay $25,000+ for each project NDA. I don't discuss the projects I have NDAs on. I don't even disclose who I have them with.

Reverse engineering is done to see how it works. Patents prevent you from using the information in your designs. Manufacturers of CPUs patent gate technology designs and other key parts to protect their investments. If you use the information and are caught, you can pay millions in patent fees and court costs. Look at RAMBus suing and getting sued.


-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/21/2007 06:50 PM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

quote:

Originally posted by: mckennma
It does not matter since you probably don't qualify for a NDA. So you will never find out. Call them about getting a NDA or close the call. Don't ramble on what is fair and not fair.



Don't fog the issue. I was not rambling about AMD's NDA. If they don't want to release the info, well it is their property, so tehy can do whatever they want with it.

The difference between their words and actions is starting to bother me, that is all.


quote:

Originally posted by: Unknown
What is open source and protected by NDA? I really have nothing else to say on it. It is what it is. Sun, IBM, and Intel will do the same thing.



1. They are not all the same. I can download practically all I could need about PowerPC from IBM. I might not be able to get some documents about the Cell, but i can't get it on the free market as a separate component anyway.
For just about anything else from Motorola (now Freescale) and IBM I can get the data without problems.


2. Well, Linux, xBSD etc might be insignificant on desktop ( or lately not even there) but they are sure not insignificant on the server platform. If what you are saying is true ( which is very probably NOT) then this would preclude platforms use with anything nonproprietary. I expect that some obscure register might be "off the record", but nothing substantial for the kernel or OS. And probably not even that. S940 platform seems to be quite thoroughly documented.

quote:

Originally posted by: Unknown
Just like Microsoft charging $100K + $25K for testing/qualifying each signed driver for Vista. The NDA signed developers site is better than normal web site. You also have engineers on hand to answer questions at a huge fee. Hardware developers can pay $25,000+ for each project NDA. I don't discuss the projects I have NDAs on. I don't even disclose who I have them with.


Don't be modest. You forgot to mention special, silenced Walther PPK, Lotus, outfitted for MI-6 agent and Mr. Q's technical support.

Seriously, if Datasheet is so deadly in wrong hands, why is then so freely available for sockets S939 and S940 ? Why is data on power sequencing available for Socket F ?
Who would be expected to have a use for it without the rest of Datasheet ?

My guess is that not all bits for Socket F are finalized, especially those concerning fast HT links and that some timings there might change.



quote:

Originally posted by: Unknown
Reverse engineering is done to see how it works. Patents prevent you from using the information in your designs. Manufacturers of CPUs patent gate technology designs and other key parts to protect their investments. If you use the information and are caught, you can pay millions in patent fees and court costs. Look at RAMBus suing and getting sued.


Which, of course, have _nothing_ to do with datasheet in question.


 02/21/2007 10:33 PM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

quote:

Originally posted by: brankob
Don't fog the issue. I was not rambling about AMD's NDA. If they don't want to release the info, well it is their property, so tehy can do whatever they want with it.

The difference between their words and actions is starting to bother me, that is all.
1. They are not all the same. I can download practically all I could need about PowerPC from IBM. I might not be able to get some documents about the Cell, but i can't get it on the free market as a separate component anyway.
For just about anything else from Motorola (now Freescale) and IBM I can get the data without problems.
2. Well, Linux, xBSD etc might be insignificant on desktop ( or lately not even there) but they are sure not insignificant on the server platform. If what you are saying is true ( which is very probably NOT) then this would preclude platforms use with anything nonproprietary. I expect that some obscure register might be "off the record", but nothing substantial for the kernel or OS. And probably not even that. S940 platform seems to be quite thoroughly documented.
Don't be modest. You forgot to mention special, silenced Walther PPK, Lotus, outfitted for MI-6 agent and Mr. Q's technical support.

Seriously, if Datasheet is so deadly in wrong hands, why is then so freely available for sockets S939 and S940 ? Why is data on power sequencing available for Socket F ?
Who would be expected to have a use for it without the rest of Datasheet ?

My guess is that not all bits for Socket F are finalized, especially those concerning fast HT links and that some timings there might change.
Which, of course, have _nothing_ to do with datasheet in question.



Linux and PowerPC? Both are free and not really seen in industry.

Sig 9mm.

I beta test to make things better and find problems to help in development. Many times my ideas are added to the project. I get features I want. In return, I get better software and know the problems upfront.

What is stopping a NDA from misuse? You will get sued. You will go to jail and lose everything. They don't screw around with it. No company will ever hire you breaking a NDA. Look at the idiot who stoled the formula for Coke and try to sell it to Pepsi. Pepsi turned him in and he and two others are facing 5 years + millions in fines and court costs. Companies don't tolerate thieves trying to sell their competition secrets. Because it is not good business.

All I want to do it use my K8W and K8WE. I don't need to invent a way to save a few hundred dollars on upgrading a old board / BIOS to work with a new CPU. I upgraded to K8WE and 280s and kept my RAM.

-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/21/2007 11:36 PM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

And I want to learn something at every oportunity .

Not necesarrily by building the thing, but I don't consider having a thought about as a sin per se.

If idea isn't viable, who cares. It's just an idea.
Some gained knowledge about inner register workings on opteron might very well benefit me in the future, be it directly or indirectly.

Same goes for HT links etc. They are used more and more. Even some MIPS components use them for a several years now ( pmc-sierra comes to mind), so it wouldn't hurt to know something more about the standard or get some hands-on experience.

And if it doesn't work, who cares ? I wouldn't consider worth pursuing the idea that is almost certainly bound to work anyway, since there is not much new in it.

One way or the other those S2885s will lose their value totally in some few months, when Barcelona comes out. Who cared what happens to them then ?
 02/21/2007 11:40 PM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

quote:

Originally posted by: brankob
And I want to learn something at every oportunity .

Not necesarrily by building the thing, but I don't consider having a thought about as a sin per se.

If idea isn't viable, who cares. It's just an idea.
Some gained knowledge about inner register workings on opteron might very well benefit me in the future, be it directly or indirectly.

Same goes for HT links etc. They are used more and more. Even some MIPS components use them for a several years now ( pmc-sierra comes to mind), so it wouldn't hurt to know something more about the standard or get some hands-on experience.

And if it doesn't work, who cares ? I wouldn't consider worth pursuing the idea that is almost certainly bound to work anyway, since there is not much new in it.

One way or the other those S2885s will lose their value totally in some few months, when Barcelona comes out. Who cared what happens to them then ?



As an engineer, you must determine the best practice to complete a task.

My K8W and 265s run very fast and handle Adobe Photoshop CS2, Sony VMS, etc. It is far from obsolete. 35,000 MIPs and 15,000 MFlops with 8.4 GB/s memory bandwidth is better than most computers.

-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/21/2007 11:53 PM
User is offline View Users Profile Print this message

Author Icon
zeppelinrox
Senior Member

Posts: 8484
Joined: 11/30/2005

quote:

Originally posted by: Unknown All that AMD's "Fusion" song and dance seems stupid. They declared some socket as an "open" invent faggotistic name for anticipated third-party cooperation and then _closed_ public acces to the data. How can this be "Fusion" if old, "closed" sockets had been publicly available and new one, which is supposed to be "open", is under NDA ?Wouldn't "Fission" (like in "We are saying bye,bye, you're on yer own") be more appropriate name for it ?
And this forum is probably one of more useless of all. It is slow, inflexible, ugly and once more _sloooow_. And that is before anything is said about its content, which is more or less unuseable. Some "technical articles" on about same level of technical expertise as average TV car commercial and a bunch of pretty bleak posts. I can't say that I have ever seen any useable bit of info that I haven't see elsewhere already.
-----------------------
When I open the topic with inquiry for some data, I do expect answers to be more or less on topic. I don't mind occasional philosophy excursions and "opinions", but please learn when to stop. Thank you for opinions, they have been noted, but now I would like to see some useful post with requested data from someone...

first of all...
on topic.. question is.. "Is it possible to make simple SocketF -->S940 adapter ?"
answer is...
a Big FAT NOOOOOOO!!!

now, you had your rant.. here's mine.

who are you? who do you think you are? you may think you are the centre of the universe, but last i checked, the universe did NOT revolve around you, not even this planet.

you are an cheap, stubborn, self absorbed child without the common sense of solving a problem in the quickest and easiest manner, when considering time spent vs dollars. Time is money. and the Time you would waste has ALOT more worth than buying the PROPER HARDWARE.
and to top it off, 100% IGNORANT. and i mean ignorant in a bad way.

this whole thread should be wiped out so that nobody else wastes their time on your drivel

-------------------------


Overclocking Help and Links . Why DRM, BluRay and HD-DVD Blow Great tools-> OC Bible v1.52 and Guidemania v1.2
 02/22/2007 12:33 AM
User is offline View Users Profile Print this message

Author Icon
brankob
Member

Posts: 146
Joined: 09/11/2006

quote:

Originally posted by: zeppelinrox
first of all...
on topic.. question is.. "Is it possible to make simple SocketF -->S940 adapter ?"
answer is...
a Big FAT NOOOOOOO!!!



FYI, the answer is not no. It's more like "it's not worth trying- at least not if your end goal is result per se and not knowledge gained."

quote:

Originally posted by: Unknown
you are an cheap, stubborn, self absorbed child without the common sense of solving a problem in the quickest and easiest manner, when considering time spent vs dollars.



Actually, I'm more in the "middle aged a**hole with his own company" part of the spectrum. I publish a journal or two ( electronics, automation, that sort of stuff).
I did some HW design and microcontroller SW in the past and still do it occasionally.
I like to stay informed and have some experience.

It is a big help if you have some hands-on experience even if you write articles etc.
Equipment is usually not a problem. I have decent equipment, worth several $10K, but of course that is in the peanuts class for serious high-level work.

Still, being a journalist, I can have access to top end equipment when I need it (more or less).

quote:

Originally posted by: Unknown
Time is money. and the Time you would waste has ALOT more worth than buying the PROPER HARDWARE.



Not if you use it to learn something valuable. If you are just a trained monkey which likes to think about itself as competent computer user, then you are probably right.

 02/22/2007 08:57 AM
User is offline View Users Profile Print this message

Author Icon
mckennma
Senior Member

Posts: 2989
Joined: 06/01/2005

Zeppelinrox,

I think this was a bit much.

Learn how to use current hardware and troubleshoot it. Not design a component that won't work properly with the hardware you want to attach to old technology.

quote:

Originally posted by: Unknown I did some HW design and microcontroller SW in the past and still do it occasionally.


I did also. I could fix most electronics as a teen. I started out of high school working for a network company doing token ring and Novell. I went to engineering school to learn how it works and how to be an engineer. You learn proper engineering methods You don't reinvent the wheel. You don't try to invent a square wheel. You learn better ways of spending your time and upgrading hardware. I designed a custom motherboard for my workstation with all the features I wanted on-board. It was incredible. It would have been $4000 to prototype. It took over 300 hours to design and simulate. I never built it.

My workstation was built to see how far I could push technology. I incorporated server grade SCA enclosures in a configuration not tried by Supermicro. I could almost replace my Buick LuCerne selling my workstation. I proved obsolence is when a computer no longer functions properly with today's software. Performance wise it is faster than most file servers I work on. Many have copied its design for CAD and A/V.

Learn your craft as a journalist instead of trying to play Mr. Scott.

-------------------------
Tyan Thunder K8WE
Dual AMD 280 Opterons
8GB NUMA enabled DDR333 2.5-3-3-7 RAM
PCI-X SCSI RAID
http://69.14.190.80
 02/22/2007 11:26 AM
User is offline View Users Profile Print this message

Author Icon
zeppelinrox
Senior Member

Posts: 8484
Joined: 11/30/2005

quote:

Originally posted by: brankob
FYI, the answer is not no. It's more like "it's not worth trying- at least not if your end goal is result per se and not knowledge gained."
Actually, I'm more in the "middle aged a**hole with his own company" part of the spectrum. I publish a journal or two ( electronics, automation, that sort of stuff).
I did some HW design and microcontroller SW in the past and still do it occasionally.
I like to stay informed and have some experience.

It is a big help if you have some hands-on experience even if you write articles etc.
Equipment is usually not a problem. I have decent equipment, worth several $10K, but of course that is in the peanuts class for serious high-level work.

Still, being a journalist, I can have access to top end equipment when I need it (more or less).
Not if you use it to learn something valuable. If you are just a trained monkey which likes to think about itself as competent computer user, then you are probably right.


sorry for over-reacting.
if it was possible, it's definitely not worth it looking at it from a monetary angle. but if it was possible, somebody would have made such an adapter by now.. even if it was for an intel machine, if it could have been done it would have been done.

since you have your own company, then obviously you can write off the cost of whatever equipment you buy and pay less taxes so even then you don't incur the full cost.

but if the whole point is to try to learn something through tinkering, that's a whole different thing... just don't forget to have a backup computer in place so that you don't have any downtime.

-------------------------


Overclocking Help and Links . Why DRM, BluRay and HD-DVD Blow Great tools-> OC Bible v1.52 and Guidemania v1.2
Statistics
112018 users are registered to the AMD Processors forum.
There are currently 0 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.



Contact AMD Terms and Conditions ©2007 Advanced Micro Devices, Inc. Privacy Trademark information