What's the future of SocketA?

Phuz

Diamond Member
Jul 15, 2000
4,349
0
0
Hmmmmmmm? :)

Will my A7V be able to enjoy CPU's like the Mustang, or better? Or am I being to optomistic?

Thanks guys.
 

Dark4ng3l

Diamond Member
Sep 17, 2000
5,061
1
0
read it around the forums and in a few articles I cant remember exacly where but I read it at least 3+ times
 

SSP

Lifer
Oct 11, 1999
17,727
0
0
I hope they'll last longer then the Slot A's (with out having to upgrade to different chipsets/mobos).
 

medic

Diamond Member
Oct 9, 1999
5,160
0
0
Someone correct me, but I have heard....

All SocketA's

Mustang SocketA 1MB-4MB L2 (Servers/Workstations/High-end Desktop)
Palomino(Mustang core) SocketA 512KB core (Desktop)
Morgan (Mustang core) SocketA 64-128KB core (Value-line Replaces the Duron in the Spring, most likely launched with 128KB L2)
Corvette(Mobile) Mustang core SocketA 256KB-1MB core
T-Bird and Duron obviously.
 

Cosmic_Horror

Golden Member
Oct 10, 1999
1,500
0
0
if the mustang core is significantly different we maybe forced to use a new chipset, which means a new motherboard. :-( (hey it happened from the classic athlons to the thunderbird!)

 

Noriaki

Lifer
Jun 3, 2000
13,640
1
71
I believe i've read that AMD intends to make all socketA backwards and forwards compatible.

Duron, TBird, Morgan, Palimino, and Mustang all use the same form factor.

There will be newer socketA boards with DDR and ATA100 and the like, but any socketA chip shoudl work in any socketA board.
 

tigger80

Golden Member
Feb 15, 2000
1,198
0
0
Go AMD, lets just hope it doesn't become another Intel once it takes over the server market
 

andylawcc

Lifer
Mar 9, 2000
18,183
3
81
no, Mustang Core is still based on the K7 architecture, (same pin)
it will be enchanced in terms of amount of cache and most likely its better branch predictor which I have no clue about. micron size is still .18 which makes me doubt it will go beyond 1.7 gig.

however, I am not optimistic that KT133 (ie: your precious KT7 and A7V) may able to run a Mustang Core CPU. that's what a KT133A will do. this will be like old Socket370 PPGA mobo can't run FCPGA. it sucks.
 

Jumpem

Lifer
Sep 21, 2000
10,757
3
81
The fuss about needing new boards doesn't really bother me. I always get a new MB when I grab a new processor!
 

Phuz

Diamond Member
Jul 15, 2000
4,349
0
0
Wicked!
And don't forget, the BX was only supposed to run a lower end chip, but they can take the new Copermines.. right?

 

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
The only problem I can forsee is that unless AMD has a way to override their own muiltiplier rules(like what Intel does with the Celeron), they'll have to jump to 266FSB to hit speeds past 1.25ghz. In that case, you'll have to use a different chipset(like switching from LX to BX) in order to use most of the new CPUs at their right speed.
 
Sep 5, 2000
102
0
76
Amd changed the multplier rules for Socket 7 the 2x = 6x in the K6III's and the K6II's over 500. I've put a K6III in an old Asus motherboard set the jumper to 2x and got 400mhz (6x66=396) for my sister's computer. So AMD could change the same way for the socket A's it would be wired in the CPU. Now whether or not they do is another story.
------------------------------------------------------
If you're too open minded, your brains will fall out.
 

andylawcc

Lifer
Mar 9, 2000
18,183
3
81
Virge: can KT133 support 133 DDR fsb or 266 fsb then? I mean, that's what the Name KT133 means right? or it just supports 133 RAM?
 

Alfred

Member
Oct 13, 1999
126
0
0
i hope amd is smart enough to re-map the multiplier for 13, 13.5 and so on... so that i can keep my a7v instead of buying new mobo.

Alfred
 

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
The KT133 isn't stable very often over 110mhz(220DDR), period. Otherwise, we'd find the CPU's strong point by multipliers, then pump up the FSB.:Q So, I'd have to say the 133 part is ment to be the max mem speed.:(
 

Sunner

Elite Member
Oct 9, 1999
11,641
0
76
Noxipoo, are you sure there's a "hard" limit to the KT133 at 1.25 GHz, and not just that there are no higher multipliers, as is the case with the BX and very high multipliers?