Roadmap/Calendar for both Intel/AMD........

[Mephistopheles]

<A HREF="http://freespace.virgin.net/m.warner/Roadmap2005.htm" target="_new">Quite rich in information. </A>You can also try "previous" for a 2004 roadmap.

Intel seems hopelessly defenseless against a massive onslaught of 90nm Athlon 64s... like 4000+, 4200+...

Actually, it's only by Q3 2005 (with Smithie and Montecito ultra-high-end) that they might stand a chance again, according to that roadmap, if they don't play any other cards... The roadmap is pretty thin on Intel info, anyway... What is interesting is that Intel plans 65nm still in 2005... And dual-core all the way in every field (Centrino/Sonoma, Desktop, IA64...)...

AMD's roadmap seems clean, well polished and quite impressive. Intel will probably be quite embarassed in the next 6-9 months.

Also interesting is that the successor to montecito seems to utilize a new type of bus, called "asynchronous serial FSB", whatever that is... It's currently under development by the ex-alpha team, and if Intel plans Xeon/Itanium socket compatibility, this thing may get more detailed within a few years.

I'd certainly expect the ex-alpha team to be able to devise an impressive and innovative bus architecture (hypertransport anyone?) and saving some face (what's left) for Intel... Maybe in a few years...

In any case, meanwhile, AMD will certainly grab a very strong hold of the performance crown, judging from that roadmap.

My next computer will definitely be an A64-based one, that's for sure!! <P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/16/04 01:54 AM.</EM></FONT></P>

 

[DonnieDarko]

So much ahead. I've never seen a list of so many different processor cores to be released in one year.

Watch out for the <b><font color=red>bloody</font color=red></b> Fanboys!

AMD64 2800+ :: MSI Neo-Fis2r :: 512mb Kingmax ddr400 :: Sapphire 9800pro 128mb :: 10K WD Raptor

 

[BeyRevRa]

Here I'll just save PeteRoy having to voice his ignorant opinions:

"Don't you worry about Intel they know what they are doing"

Anyway, looks impressive, gonna be a "fun" year. I was gonna do some mild upgrading around christmas time but I think I will wait to see how things unfold.

AMD 64 3400+
MSI K8N Neo Platinum
1 GB Kingston HyperX PC3200
ATI AIW 9600XT
WD Raptor 74GB

 

[Mephistopheles]

I think Intel doesn't have a clue as to what they're doing... Oh, no, wait, I think I'm damn sure they don't. That's why their roadmap is vague, they're <i>really, really</i> shuffling their deck of cards... and looking for a trump card to play.

If they do manage to pull good cards out of their hats, they'll probably try to deploy them ASAP; necessity is the mother of invention (and DNA analysis proves with 99.9% certainty that insanity is the father! nice one, slvr)...

Meanwhile, AMD is leaving intel in the dust. They don't have to worry about the dust, though - the way things are going, they'll even be behind the dust cloud!

 

[DonnieDarko]

I think intel's main problem is they are trying to push new, but unnecessary, technologies. Right now, ddr2 isn't really needed. BTX layouts only cause people to buy new cases. To fast of a jump to 90nm, which caused alot of problems. The list goes on and on.

Watch out for the <b><font color=red>bloody</font color=red></b> Fanboys!

AMD64 2800+ :: MSI Neo-Fis2r :: 512mb Kingmax ddr400 :: Sapphire 9800pro 128mb :: 10K WD Raptor

Addicted, finally.

 

[DonnieDarko]

I'm not trying to say that these wiln't be nesscessary in the future. But right now i don't think it's needed/ready.

Watch out for the <b><font color=red>bloody</font color=red></b> Fanboys!

AMD64 2800+ :: MSI Neo-Fis2r :: 512mb Kingmax ddr400 :: Sapphire 9800pro 128mb :: 10K WD Raptor

Addicted, finally.

 

[Mephistopheles]

Indeed, DDR2-400 is waste, DDR2-533 is nearly useless, and only DDR2-667 is actually interesting.

And intel tried to introduce too much too fast and all at once (PCIe, DDR2, LGA775, BTX)...

 

[SUPERPOPE]

I asked the same questions in my WELCOME BACK THREAD "INTEL GLORY LOST"

I think INTEL is a ship adrift right now...albiet with treasure filled coffers....

i guess INTEL is like...we can do nothing for a year and AMD is still no real threat

 

[phial]

Also interesting is that the successor to montecito seems to utilize a new type of bus, called "asynchronous serial FSB", whatever that is... It's currently under development by the ex-alpha team, and if Intel plans Xeon/Itanium socket compatibility, this thing may get more detailed within a few years.

serial usually refers to be full duplex no? as you said, "HT anyone?"

at least all this information tells me that upgrading wont be nessessary for another 8-12 months or so

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>

Brand name whores are stupid!

 

[P4Man]

>"asynchronous serial FSB"

Serial would be just that, instead of using a parallel bus (64 bit wide in the case of P4, 128 bit for Itanium), you serialize data like most modern busses today (PCI-E, HT,..). This allows much easier designs, and requires less pins/traces to achieve similar bandwith. About time Intel ditches their parallel approach, cause it gets very hard to design motherboards that for instance implement 2 cpu sockets with a 128 bit 400 MHz bus (Itanium). In order for a parallel bus to work, each trace has to work exactly as fast, meaning traces need to be exactly as long. Good luck doing that with a 128 bit design at 533 MHz!

I'm at a loss as well though, what they mean with "asynchronous" in this context. It could refer to a different speed in each direction (remember nVidia's nForce3 150), or it could refer to the fact the bus won't run in synch with cpu (which is rather obvious) or something else.

Anyway, its not nearly as telling as the fact it will be serial, and therefore, brandnew. This however, also implies future Itaniums will not be socket compatible (no one here will care), but also since Xeons will share the same socket and therefore bus that future Xeons will not be socket compatible (again, not really a loss: who would want a slowpoke shared 400 Mhz bus in 2007 ?). However, it would only make sense to use this same bus for Pentiums as well then.. which means.. S775 might not live that long either.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[phial]

.. S775 might not live that long either.

****GASP****

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>

Brand name whores are stupid!

 

[P4Man]

>****GASP****

Hmm.. if true, not such a big deal IMHO. 2007 is still a few years from here, not too mention a S775 board bought today is not likely to have such a long life anyway since it won't ever support 1066 fsb (at least not officially). Furthermore maybe desktop chips will be slower to migrate to this new bus, so I wouldn't loose too much sleep over it.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[Mephistopheles]

I also considered that Xeon/Itanium socket compatibility might be compromised by Itanium using a serial bus... But instead of completely negating the compatibility as you initially considered, I was more optimistic and immediately assumed that the serial bus was going to migrate to xeons. They also need it desperately.

...and eventually, I'd expect the serial bus to also make its way into desktops, where this new bus would make the least difference, unlike in IA64/Xeon systems.

 

[peteroy]

How can you say AMD roadmap looks better than Intel.

Suddenly you all are chips engineers knowing what's good and bad for a cpu design.

From that roadmap you can see that Intel is going to release more new stuff than AMD.

And no one can predict the future, and don't claim to know that Intel doesn't know what to do unless you work at Intel.

===========================
<A HREF="http://www.clancas.net" target="_new">clan CHAOS</A><P ID="edit"><FONT SIZE=-1><EM>Edited by PeteRoy on 08/16/04 09:15 PM.</EM></FONT></P>

 

[juin]

There some much mistake is this roadmap.There CPU are directly from is imagination.

i need to change useur name.

 

[DonnieDarko]

How can you say AMD roadmap looks better than Intel.

Suddenly you all are chips engineers knowing what's good and bad for a cpu design

Because we base our opinions on facts, not fanboyism, like yourself.

Watch out for the <b><font color=red>bloody</font color=red></b> Fanboys!

AMD64 2800+ :: MSI Neo-Fis2r :: 512mb Kingmax ddr400 :: Sapphire 9800pro 128mb :: 10K WD Raptor

Addicted, finally.

 

[phial]

i was being sarcastic

and also was poking fun at Kenavit

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>

Brand name whores are stupid!

 

[kal326]

Ya no joke they tried to push new and unnecessary technologies. On top of the fact that you basically need to cut your whole system to run a Socket 775 system, you would really want to wait anyway because the new 1066fsb chips are right around the corner anyway. Maybe with the change to 1066 we might see some reason to switch over to socket 775, since there has been little reason to switch so far. But im sure the changes are needed for some greater good to be had in the next years time. Just like when everybody thought Intel lost their f'n minds when the first P4's where slower then the P3's of the day. Once Intel got all the bells and whistles up and going with HyperThreading and twice the speed of the original P4's all was good. The general public just had to deal with the teething problems associated with the fact that said progress spanned three differant cores, 3 differant fsb speeds, 3 major chipsets, 3 differant memory types, and two differant sockets. And to think they still call it the "P4". Now intel is onto its 4th major chipset change, 4th memory type, and 3rd socket, and soon to be 4th fsb speed.

Abit IC7
2.4c @ 3.0
1024 PC4000 2.5-4-4-7
GF FX5900 @ 455/955
2x120 Seagate SATA Raid0
3DMark03: 6414
Thats not a bug, its an undocumented feature....