By: David Kanter (dkanter.delete@this.realworldtech.com), May 16, 2006 8:44 am
Room: Moderated Discussions
Hey Guys,
I'm sitting in a keynote from AMD's Chuck Moore, and here is information on the next generation i.e. K8L.
0. Native quad core
1. Hypertransport up to 5.2GT/s
2. Better coherency
3. Private L2, shared L3 cache that scales up.
4. Separate power planes and pstates for north bridge and CPU
5. 128b FPUs - see 14,15
6. 48b virtual/physical addressing and 1GB pages
7. Support for DDR2, eventually DDR3
8. Support for FBD1 and 2 eventually
9. I/O virtualization and nested page tables
10. Memory mirroring, data poisoning, HT retry protocol support
11. 32B instead of 16B ifetch
12. Indirect branch predictors
13. OOO load execution - similar to memory disambiguation
14. 2x 128b SSE units
15. 2x 128b SSE LDs/cycle
16. Several new instructions
Coprocessors:
media processing
JVM/CLR acceleration
TOE, XML or SSL processing
Anyway, ours news posting is somewhat broken, so I might not get to turn this into news. But you heard it here first : )
DK
I'm sitting in a keynote from AMD's Chuck Moore, and here is information on the next generation i.e. K8L.
0. Native quad core
1. Hypertransport up to 5.2GT/s
2. Better coherency
3. Private L2, shared L3 cache that scales up.
4. Separate power planes and pstates for north bridge and CPU
5. 128b FPUs - see 14,15
6. 48b virtual/physical addressing and 1GB pages
7. Support for DDR2, eventually DDR3
8. Support for FBD1 and 2 eventually
9. I/O virtualization and nested page tables
10. Memory mirroring, data poisoning, HT retry protocol support
11. 32B instead of 16B ifetch
12. Indirect branch predictors
13. OOO load execution - similar to memory disambiguation
14. 2x 128b SSE units
15. 2x 128b SSE LDs/cycle
16. Several new instructions
Coprocessors:
media processing
JVM/CLR acceleration
TOE, XML or SSL processing
Anyway, ours news posting is somewhat broken, so I might not get to turn this into news. But you heard it here first : )
DK
| Topic | Posted By | Date |
|---|---|---|
| AMD's Next Gen MPU | David Kanter | 2006/05/16 08:44 AM |
| Great stuff. What is its ETA? (NT) | Paul DeMone | 2006/05/16 09:05 AM |
| Early 2007 | David Kanter | 2006/05/16 01:30 PM |
| Early 2007 | mas | 2006/05/16 01:44 PM |
| K8L FPUs | David Kanter | 2006/05/16 02:48 PM |
| K8L FPUs | mas | 2006/05/16 03:15 PM |
| K8L FPUs | Anil Maliyekkel | 2006/05/16 06:54 PM |
| More like late 2007,trust me ;-) (NT) | Latex | 2006/05/16 02:58 PM |
| Do they have silicon yet? | Paul DeMone | 2006/05/16 04:47 PM |
| Do they have silicon yet? | curious george bush | 2006/05/16 06:25 PM |
| Do they have silicon yet? | LauraBush | 2006/05/17 04:48 AM |
| Doubt it | savantu | 2006/05/18 03:08 AM |
| Doubt it | David Kanter | 2006/05/18 11:23 AM |
| Doubt it | savantu | 2006/05/18 11:39 PM |
| Doubt it | David Kanter | 2006/05/19 12:06 AM |
| Doubt it | Dresdenboy | 2006/05/19 04:00 AM |
| Doubt it | Dresdenboy | 2006/05/18 02:17 PM |
| Doubt it | anonymous | 2006/05/18 04:38 PM |
| Doubt it | savantu | 2006/05/18 11:41 PM |
| Don't be silly | David Kanter | 2006/05/19 12:09 AM |
| Doubt it | Dresdenboy | 2006/05/19 12:18 AM |
| Doubt it | Vincent Diepeveen | 2006/06/06 07:32 AM |
| Doubt it | David Kanter | 2006/06/06 10:05 AM |
| Execution core logics | Vincent Diepeveen | 2006/06/07 01:40 AM |
| Do they have silicon yet? | Alberto | 2006/05/17 04:57 AM |
| QC on 90nm would be a no brainer, definitely 65nm (NT) | Daniel Bizó | 2006/05/17 05:09 AM |
| He! like a 90nm Q3/2006 Montecito? (NT) | Alberto | 2006/05/17 05:55 AM |
| Totally irrelevant and out of context | Daniel Bizó | 2006/05/17 06:22 AM |
| I think not, yet QC K8 is on 65nm (link) | Alberto | 2006/05/17 06:49 AM |
| I think not, yet QC K8 is on 65nm (link) | Daniel Bizó | 2006/05/17 07:14 AM |
| I think not, yet QC K8 is on 65nm (link) | Alberto | 2006/05/17 07:39 AM |
| I think not, yet QC K8 is on 65nm (link) | Hans de Vries | 2006/05/17 10:33 PM |
| Do you remember AMD's DC die plot? | Dresdenboy | 2006/05/18 02:41 AM |
| Single threaded performance | ttaylor | 2006/06/02 01:16 PM |
| Single threaded performance | savantu | 2006/06/03 02:51 AM |
| AMD's Next Gen MPU | mas | 2006/05/16 09:15 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 09:32 AM |
| AMD's Next Gen MPU | mas | 2006/05/16 09:41 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 09:46 AM |
| Magnitude of difference | Daniel Bizó | 2006/05/17 12:59 AM |
| Magnitude of difference | David Kanter | 2006/05/17 08:29 PM |
| AMD's Next Gen MPU | Hans de Vries | 2006/05/17 10:26 PM |
| AMD's Next Gen MPU | Temp | 2006/05/16 09:25 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 09:34 AM |
| AMD's Next Gen MPU | Temp | 2006/05/16 09:41 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 08:47 PM |
| AMD's Next Gen MPU | Temp | 2006/05/16 10:50 PM |
| AMD's Next Gen MPU | fastpathguru | 2006/05/16 09:52 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 09:54 AM |
| AMD's Next Gen MPU | fastpathguru | 2006/05/16 10:08 AM |
| AMD's Next Gen MPU | savantu | 2006/05/16 09:58 AM |
| AMD's Next Gen MPU | Troll | 2006/05/17 06:51 AM |
| AMD's Next Gen MPU | Hannibal | 2006/05/17 10:14 AM |
| AMD's Next Gen MPU | mas | 2006/05/16 10:56 AM |
| AMD's Next Gen MPU | PiedPiper | 2006/05/16 11:08 AM |
| AMD's Next Gen MPU | fastpathguru | 2006/05/16 08:07 PM |
| AMD's Next Gen MPU | PiedPiper | 2006/05/16 09:21 PM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 11:05 PM |
| AMD's Next Gen MPU | M.Isobe | 2006/05/17 12:04 AM |
| AMD's Next Gen MPU | rwessel | 2006/05/17 01:43 AM |
| AMD's Next Gen MPU | Aaron Spink | 2006/05/17 12:12 PM |
| AMD's Next Gen MPU | rwessel | 2006/05/17 01:58 PM |
| AMD's Next Gen MPU | slim | 2006/05/18 08:06 AM |
| AMD's Next Gen MPU | William Campbell | 2006/05/17 02:29 AM |
| AMD's Next Gen MPU | Kava | 2006/05/17 10:06 AM |
| Yes, you'll see an article when it's over (NT) | David Kanter | 2006/05/18 12:18 AM |
| AMD's Next Gen MPU | Kava | 2006/05/16 11:25 AM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 03:04 PM |
| AMD's Next Gen MPU | fastpathguru | 2006/05/16 03:28 PM |
| AMD's Next Gen MPU | anonymous | 2006/05/16 04:24 PM |
| AMD's Next Gen MPU | anonymous | 2006/05/16 04:26 PM |
| AMD's Next Gen MPU | Kava | 2006/05/16 04:37 PM |
| AMD's Next Gen MPU | David Kanter | 2006/05/16 08:48 PM |
| XML acceleration | Rob Thorpe | 2006/05/17 01:17 AM |
| XML acceleration | Dan Lawrance | 2006/05/17 05:47 AM |
| AMD's Next Gen MPU | n | 2006/05/16 12:03 PM |
| AMD's Next Gen MPU | Dresdenboy | 2006/05/16 02:50 PM |
| AMD's Next Gen MPU | Dan Downs | 2006/05/16 12:57 PM |
| AMD's Next Gen MPU | Kava | 2006/05/16 03:18 PM |
| AMD's Next Gen MPU | Doug Siebert | 2006/05/16 05:58 PM |
| AMD's Next Gen MPU | Kava | 2006/05/16 08:00 PM |
| AMD's Next Gen MPU | Doug Siebert | 2006/05/16 08:13 PM |
| AMD's Next Gen MPU | Troll | 2006/05/17 06:56 AM |
| AMD's Next Gen MPU | Dipen | 2006/05/16 03:34 PM |
| AMD's Next Gen MPU | Vincent Diepeveen | 2006/06/03 09:46 PM |
| AMD's Next Gen MPU | Michael S | 2006/06/04 07:42 AM |
| AMD's Next Gen MPU | David Kanter | 2006/06/04 02:45 PM |
| AMD's Next Gen MPU | mas | 2006/06/05 02:34 AM |
| AMD's Next Gen MPU | David Kanter | 2006/06/05 09:00 AM |
| AMD's Next Gen MPU | anonymous | 2006/06/05 09:30 AM |
| AMD's Next Gen MPU | David Kanter | 2006/06/05 09:54 AM |
| AMD's Next Gen MPU | savantu | 2006/06/05 11:27 PM |
| AMD's Next Gen MPU | Howard Chu | 2006/06/06 12:13 AM |
| AMD's Next Gen MPU | Doug Siebert | 2006/06/05 11:55 AM |
| AMD's Next Gen MPU | mas | 2006/06/05 04:44 PM |
| AMD's Next Gen MPU | Doug Siebert | 2006/06/05 07:23 PM |
| AMD's Next Gen MPU | Jerry A. | 2006/06/05 10:55 PM |
| AMD's Next Gen MPU | David Kanter | 2006/06/06 02:09 PM |
| AMD's Next Gen MPU | Mr. Anonymous | 2006/06/06 06:15 PM |


