40 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]

   #39. Posted at 12:02 PM on Jan 30th 2007 Edit   Reply

According to this piece, Intel was going to release this later but hurried with publicizing this to head off IBM's announcement for a PR brownie point

More info can be found about here:

http://arstechnica.com/news.ars/post/20070130-8732.html
collapse

   #37. Posted at 10:46 AM on Jan 30th 2007 Edit   Reply

Hans put up a very interesting post: http://www.siliconinvestor.com/readmsg.aspx?msgid=23229817

While SRAM shrunk as expected, it seems that logic is not. Very interesting look though! The 45 nm Penryn is still pretty small, and should pack a good whollup. But when looking at potential size, I don't know if AMD is going to be at much of a disadvantage?

Lots of new tech in Intel's 45 nm process, so I will be very curious to see how transistor speed and power usage is as compared to the older 65 nm process. Immersion Litho, while still a ways off for AMD, might be able to compress the geometries better than Intel's dry litho. Time will tell.
collapse

   #29. Posted at 02:45 PM on Jan 29th 2007 Edit   Reply

I'm not sure, but I believe that the following
[quote]Intel says its new process allows it to double transistor density, increase transistor switching speed by over 20%, and reduce transistor switching power by 30%.[/quote]

Should be
"...double transitor density and either increase transistor switching speed by over 20% or reduce transistor switching power by 30%".

Increasing the switching speed by 20% while simultaneously decreasing the leakage by 30% would be no small feat.
collapse

   #35. Posted at 08:19 PM on Jan 29th 2007 Edit   Reply

As a bit of interest, and partly cos I feel like deflating the massive bubble over the improved transistor switching speeds, how is this design helping dissipate the effects of wire delay?

You know, the thing that actually limits switching speed once we get past a certain size, forces everyone to go to massive pipeline lengths, and then forces them to insert "catchup" stages in their 30+ stage pipelines. Before realising that you can't cheat physics.

"Biggest change in transistor technology" my ass. Welcome news, but hardly as revolutionary as they say, especially as this stuff has been about and something to move to as long as I've been able to grow a beard (OK, so my actually growing a beard is debatable). I think I'd rather call it the Intel SOI.
collapse

   #33. Posted at 06:52 PM on Jan 29th 2007, Edited at 06:53 PM on Jan 29th 2007 Edit   Reply

Pretty good write up here...
Intel's Penryn Core
Putting the Metal Back into CMOS
(Author: MS, January 28, 2007)
http://www.lostcircuits.com/cpu/intel_penryn/
MS notes that the IBM/AMD alliance announced low K junctions late last year. AMD and IBM Detail Early Results Using Immersion and Ultra Low-K in 45NM Chips. There is a link to IBM's paper in the Lost Circuits link.

Quote from Lost Circuits Review...

According to IBM's Bernard Meyerson, their own new Hafnium-based transistors will compare like Ferraris to can openers introduced by Intel - whereas Dan Hutchinson (VLSI Research) claims that Intel's competitors "are going to wake up in terror".
collapse

   #5. Posted at 11:38 AM on Jan 29th 2007 Edit   Reply

Man Intel rolls with the tech punches again. Hopefully AMD can recover from that first upper cut it got dealt before Intel lands this right hook....
collapse

   #31. Posted at 03:13 PM on Jan 29th 2007 Edit   Reply

Maybe the only way for AMD to compete is if IBM Microelectronics bought them AND they got a huge chunk of cash from Intel for those nasty (illegal?) tactics Intel used on AMD in the past.
collapse

   #27. Posted at 02:15 PM on Jan 29th 2007 Edit   Reply

"minor electrical changes"... lovely.

Actually that's pretty par for the course, MB makers on their better-designed boards will be ok... and quite a few won't be.
collapse

   #26. Posted at 02:10 PM on Jan 29th 2007 Edit   Reply

Jesus, Intel is just mopping the floor with AMD. I really don't have high hopes with K8L, but I hope AMD can prove me wrong.
collapse

   #21. Posted at 01:22 PM on Jan 29th 2007 Edit   Reply

how much engineering intel is putting into future architectures. so far i've only been hearing about getting die shrinks faster than everyone else can. what's after core 2?
collapse
#21, Nehalem  :   (#22)  «

   #10. Posted at 12:13 PM on Jan 29th 2007 Edit   Reply

The fact that they have to make claims like "it even ran Office!" means that it's still not final and they have a bit of work to do. This is a preliminary statement of where they'll be in several months.

On the other hand, YAY progress! Keep it up.
collapse

   #1. Posted at 11:02 AM on Jan 29th 2007 Edit   Reply

With a bit of good fortune, maybe DDR2 prices will come down by the time this is released. I don't plan to upgrade from my current (DDR-based) system until they have.
collapse

   #2. Posted at 11:24 AM on Jan 29th 2007 Edit   Reply

DDR3 support would have been nice.
collapse

   #12. Posted at 12:31 PM on Jan 29th 2007, Edited at 12:40 PM on Jan 29th 2007 Edit   Reply

Overpriced and too late to the party. </anti-Shintai>
collapse

   #15. Posted at 01:07 PM on Jan 29th 2007 Edit   Reply

wow...just wow. Look what is happening on TR
collapse

   #11. Posted at 12:29 PM on Jan 29th 2007 Edit   Reply

The company plans to introduce Penryn-based processors in its desktop, mobile, and workstation/server lineups all in the second half of this year.

Actually, I think the current party-line is "starting production in H2/07".
collapse

   #4. Posted at 11:38 AM on Jan 29th 2007 Edit   Reply

Hmmmmmm, it looks like Nehalem is going to rock.
collapse
40 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]
 
Name/Password: / Remember
Reply to:
[click to clear]

[RED] [GREEN]
[BOLD]
[ITALIC] [STRIKE]
[UNDERLINE]

Notice: All posts should abide by the rules, please.
Note: Ctrl-Enter submits the post. (In IE)
DThread keys: Click on a reply to position the blue bar. 'A'/'Z' move it up/down.
Jazztags: (they MUST be closed)
    r{ red }r     g{ green }g     /[ italic ]/     *[ bold ]*
    _[ underline ]_     -[ strike ]-     s[ sample ]s     o[ spoiler ]o  q[ (QUOTE) ]q