Shopping Resources
Shop AMD United States
Shop AMD Canada
Shop ATI
Technology
AMD64 Solutions
Multi-Core
Systems Management
Your PC Experience
AMD LIVE!
Better by Design
AMD and Windows Vista®
Products
AMD Opteron™
Server
Workstation
AMD Turion™
AMD Phenom™
AMD Athlon™
AMD Sempron™
AMD Chipsets
Processor-In-a-Box
Articles & Reviews
Older Products
Processor Pricing
Packaging and Development
Packaging Technology
ROHS Compliance
Resources for:
Channel Partners
Software Developers
Software Partners
Ask AMD Knowledgebase
Processor Support Forums

Server
Shop AMD United States
Shop ATI
Partners
AMD Opteron™ Related Links
Press Releases
Partner Press Releases
Awards
Power Day Event
Success Stories
AMD Processor Support Forum

Understanding Third-Generation AMD Opteron™ Processor Model Numbers

Third-Generation AMD Opteron™ processors are identified by a four digit model number, ZYXX, where:

Z—indicates the maximum scalability of the processor.
1000 Series = 1-way servers and workstations
2000 Series = Up to 2-way servers and workstations
8000 Series = Up to 8-way servers and workstations

Y—indicates socket generation. This digit is always ‘3’ for Third-Generation AMD Opteron processors for Socket F (1207) and Socket AM2.
1000 Series in Socket AM2 = Models 13XX
2000 Series in Socket F (1207) = Model 23XX
8000 Series in Socket F (1207) = Model 83XX

XX—indicates relative performance within the series.
An AMD Opteron Processor Model 2350 is higher performing than an AMD Opteron™ Processor Model 2347, etc.
XX values above 40 indicate a Third-Generation Quad-Core AMD Opteron processor
XX values 40 and below indicate a Third-Generation Dual-Core AMD Opteron processor (available in 1H’08)

HE—indicates a 68W low-power AMD Opteron™ processor.
Performance of a low-power AMD Opteron processor HE model number, ZYXX, is equal to the performance of the same model number standard-power version. For example, an AMD Opteron processor Model 2347 HE offers the same performance as a standard wattage AMD Opteron processor Model 2347.

The chart below further explains the differences between AMD Opteron™ processor Series and Model Numbers.

Series 1000 Series 2000 Series 8000 Series
Scalability 1-way Up to 2-way Up to 8-way
Socket Socket AM2 Socket F (1207) Socket F (1207)
Performance-Optimized SE Models  
Wattage   105W 105W
Frequency Model Numbers
2.5GHz - Model 2360 SE* Model 8360 SE*
2.4GHz - Model 2358 SE* Model 8358 SE*
Standard Power Models  
Wattage 75W 75W 75W
Frequency Model Numbers
Quad-Core 2.0GHz - Model 2350 Model 8350
Quad-Core 2.1GHz Model 1352* Model 2352 -
Quad-Core 2.2GHz Model 1354* Model 2354 Model 8354
Quad-Core 2.3GHz Model 1356* Model 2356 Model 8356
Low-Power Models HE  
Wattage   55W 55W
HE Frequency Model Numbers
Quad-Core 1.7GHz - Model 2344 HE* -
Quad-Core 1.8GHz - Model 2346 HE* Model 8346 HE*
Quad-Core 1.9GHz - Model 2347 HE* Model 8347 HE*
Direct Connect Architecture Yes Yes Yes
AMD Virtualization™ (AMD-V™) Yes Yes Yes
Rapid Virtualization Indexing Yes Yes Yes
Tagged TLB support Yes Yes Yes
Device Exclusion Vector (DEV) Yes Yes Yes
AMD PowerNow!™ technology Yes Yes Yes
Independent Dynamic Core Technology Yes Yes Yes
Supported power states Up to 5 Up to 5 Up to 5
CoolCore™ Technology Yes Yes Yes
Dual Dynamic Power Management (DDPM) Supported Supported Supported
Integrated DDR2 memory controller Yes Yes Yes
AMD Memory Optimizer Technology Yes Yes Yes
DDR2 Memory type supported Unbuffered Registered Registered
DDR2 Memory controller width 128-bit 128-bit 128-bit
DDR2 Memory Max Frequency DDR2-800 DDR2-667 DDR2-667
DDR2 Memory Max DIMM support/CPU 4 @ DDR2-800 8 @ DDR2-533 8 @ DDR2-533
On-Line spare RAS support Yes Yes Yes
ECC DRAM protection Yes Yes Yes
Advanced ECC protection Yes Yes Yes
HyperTransport™ technology Yes Yes Yes
HyperTransport technology links (total/coherent) 1/0 3/1 3/3
HyperTransport technology link width 16 bits x 16 bits 16 bits x 16 bits 16 bits x 16 bits
HyperTransport bus frequency Up to 2GHz 1GHz 1GHz
AMD64 Yes Yes Yes
Simultaneous 32 & 64-bit computing Yes Yes Yes
L1 Cache Size (data/instruction) 64KB/64KB 64KB/64KB 64KB/64KB
L1 Cache associativity (data/instruction) 2-way/2-way 2-way/2-way 2-way/2-way
L2 Cache Size (dedicated cache per core) 512KB 512KB 512KB
L2 Cache associativity (data/instruction) 8-way/8-way 8-way/8-way 8-way/8-way
L3 Cache Size (shared cache) 2MB 2MB 2MB
L3 Cache associativity (data) 32-way 32-way 32-way
Pipeline stages (integer/floating point) 12/17 12/17 12/17
L1/L2 data cache protection ECC ECC ECC
L1/L2 instruction cache protection Parity Parity Parity
Global History Counter Entries 16K 16K 16K
L1 TLB entries 4KB pages (data/instruction) 48/32 48/32 48/32
L1 TLB associativity 4KB pages (data/instruction) Full/Full Full/Full Full/Full
L1 TLB entries 2/4MB pages (data/instruction) 48/16 48/16 48/16
L1 TLB associativity 2/4MB pages (data/instruction) Full/Full Full/Full Full/Full
L2 TLB entries 4KB pages (data/instruction) 512/512 512/512 512/512
L2 TLB associativity 4KB pages (data/instruction) 4-way/4-way 4-way/4-way 4-way/4-way
L2 TLB entries 2/4MB pages (data) 128 128 128
L2 TLB associativity 2/4MB pages (data) 2-way 2-way 2-way
AMD Wide Floating-Point Accelerator Yes Yes Yes
SIMD Instruction Support SSE, SSE2, SSE3, SSE4A SSE, SSE2, SSE3, SSE4A SSE, SSE2, SSE3, SSE4A
Process 65 nanometer SOI 65 nanometer SOI 65 nanometer SOI
Manufactured In Fab 36, Dresden, Germany Fab 36, Dresden, Germany Fab 36, Dresden, Germany

*Availability planned for late Q2’08


©2008 Advanced Micro Devices, Inc.    |    Contact AMD    |    Terms and Conditions    |    Privacy    |    Trademark information    |    Site Map