TSpire.com. AMD 2439 SE OPTERON SIX-CORE 2.8 GHz 9 MB CACHE HT3 D0 105 W SOCKET F CPU PROCESSOR
Recently viewed products
-
AMD 2439 SE OPTERON SIX-CORE 2.8 GHz 9 MB CACHE HT3 D0 105 W SOCKET F CPU PROCESSOR
Categories
-
CPUs
-
CPUs:VRM Module
-
CPUs:Quad-Core
-
CPUs:INTEL:Six-Core
-
CPUs:Dual-Core
-
CPUs:Eight-Core
-
CPUs:Twelve-Core
-
CPUs:INTEL:Dual-Core
-
CPUs:INTEL:Quad-Core
-
Intel i7
-
CPUs:Pentium III
-
CPUs:Six-Core
-
CPUs:Pentium II
-
CPUs:Pentium IV
-
CPUs:Sixteen-Core
-
Old CPUs
-
Intel Core LGA 2066
-
-
Video Cards
-
Thin Client
-
Dock / Docking Station / Port Replicator
-
Power Supply
-
Controller
-
Computer Hardware
-
Parts of Server
Manufacturers
- ORACLE
- IBM
- 3COM
- ADAPTEC
- APACER
View all
-
Home
/ -
CPUs
/ -
CPUs:Six-Core
/ -
AMD 2439 SE OPTERON SIX-CORE 2. 8 GHz 9 MB CACHE HT3 D0 105 W SOCKET F CPU PROCESSOR
Manufacturer:
AMD
SKU: 440
Part number: 2439 SE
Availability: In stock
-
Extended Warranty
* -
TSpire.com 6 Months Warranty
1 Year Extended Warranty
[+$172.08]
2 Year Extended Warranty
[+$219.66]
3 Year Extended Warranty
[+$307. 08]
4 Year Extended Warranty
[+$374.58]
5 Year Extended Warranty
[+$469.08]
Product description
Shop 371-4682 -Sun Microsystems 2.10GHz 6MB L3 Cache 6-Core AMD Opteron 2439 Processor – Echno Ltd TA ComputingParts.co.uk
Sun Microsystems
-
Product SKU:
371-4682
- Description
- General Information
- Shipping & Returns
371-4682 -Sun Microsystems 2. 10GHz 6MB L3 Cache 6-Core AMD Opteron 2439 Processor
371-4682 -Sun Microsystems 2.10GHz 6MB L3 Cache 6-Core AMD Opteron 2439 Processor
- Please note that Apple, Compaq, Cisco, Dell, IBM, HP, Maxtor, Fujitsu, Seagate, Sun or any item once processed can not be cancelled or returned for credit. There may be a restocking fee of minimum £ 20 or up to 20% on certain products which differs from product to product.
- For Memory Products: Ordered part may be shipped Micron, Samsung, Crucial, Hynix or any other manufacturers with 100% Compatibility.
- Please note that most of the major manufacturers (e.g. HP, IBM, Compaq, Dell) part numbers have established corresponding spare part numbers (spare P/N#), or field replacement part numbers (fru #’s) for their items. When you receive the part it may display this Spare P/N# or FRU # on it. You can easily reference this part number as a valid replacement part for your ordered item at any of the manufacturer websites. If you need any assistance in this regard please call our Sales Department at +44 (020) 3350 1222.
- This part # has a 30 days limited exchange warranty for replacement only with the same part. Some drives may have 90 days warranty which you can confirm. The product may have other warranties like manufacturer, service, or extended warranty that you can confirm with the sales department.
- All hardware sold are bulk equipment with no drivers and cd included. If you need a driver or CD installation for a particular item please call to confirm that we have it.
Shipping
- All orders require time for processing. Guaranteed shipping services are only guaranteed for the transit time. This does not include the time it takes to process your order.
- ComputingParts.co.uk will quote all destination charge prior once we have confirmed order. We are responsible for transporting goods to door after you have made the payment.
- Our online shipping calculator is based to calculate shipping for memory, hard drive and other items that are less than 5 lbs. It may not calculate the correct shipping cost of items that are more than 5 lbs. Please do call us to confirm before you place the order for items that are heavier in shipping. If you do place the order by mistake you may receive an email or call from our customer service department with the exact shipping charges. Some of the items that come in this category are laptops, servers, racks, motherboards.
Returns
- Under the distance selling regulations, you (the consumer) have the right to cancel your order within 28 days from the day after the date of arrival of your order. We will accept returns within 28 days of receipt of your order if you change your mind or no longer want the item as long as the item is returned in a salable condition and in its original packaging.
- We can offer an exchange or a credit note towards your next purchase or refund your money dependent on circumstances. Please note that for the return of non-faulty items, refund amount received (if any) will be based on our current sale price of the item and not your original purchase price. Items returned to us due to a manufacturer’s fault later than 28 days after the item was delivered will be eligible for a repair or direct replacement. Where a direct replacement cannot be offered, or a repair is beyond economical repair we will offer an exchange or credit voucher towards any other product on the website.
- Please note that if order is returned after 28 days for any other reason than it is faulty and the buyer has requested a refund, we reserve the right to charge a 20% restocking fee.
- Your 30-day right of cancellation does not apply to any software products or any consumable products which (by their very nature) cannot be returned e.g. Ink cartridges, ink refill kits.
- Carriage charges for returned goods will be incurred by the customer. Carriage charges will be applied to any order that is returned and that was shipped on a free delivery. In order to accept good(s) as being faulty, we must be able to replicate the described fault.
- Before returning any good(s) to us, you must contact us by email to notify us so that we can issue you instructions. After contacting us by e-mail your enquiry will be processed and you will be allocated a returns form and given instructions on how to return good(s). Please note that we aim to process all returned good(s) within 48 hours of receipt, but returns / refunds can take up to 7 working days to be processed.
- We strongly advise all customers to return their good(s) by recorded delivery. It is your responsibility to ensure that good(s) are packaged sufficiently prior to returning them. It is the buyer’s responsibility to ensure that all returned good(s) reach us safely.
- This cancellation and return policy does not affect your legal rights.
Supermicro Motherboard H8DI3+-F, AMD Opteron 2439 SE
Copyright 2006-2014 Standard Performance Evaluation Corporation
Supermicro
Motherboard H8DI3+-F, AMD Opteron 2439 SE
SPECint®_rate2006 =
213
SPECint_rate_base2006 =
167
CPU2006 license: | 001176 | Test date: | Sep-2009 |
---|---|---|---|
Hardware Availability: | Jun-2009 | ||
Tested by: | Supermicro | Software Availability: | Apr-2009 |
Hardware | |
---|---|
CPU Name: | AMD Opteron 2439 SE |
CPU Characteristics: | |
CPU MHz: | 2800 |
FPU: | Integrated |
CPU(s) enabled: | 12 cores, 2 chips, 6 cores/chip |
CPU(s) orderable: | 1,2 chips |
Primary Cache: | 64 KB I + 64 KB D on chip per core |
Secondary Cache: | 512 KB I+D on chip per core |
L3 Cache: | 6 MB I+D on chip per chip |
Other Cache: | None |
Memory: | 32 GB (8 x 4 GB, DDR2-800, CL5, Reg, Dual Rank) |
Disk Subsystem: | 1 x 320 GB SATA, 7200 RPM |
Other Hardware: | None |
Software | |
---|---|
Operating System: | Red Hat Enterprise Linux Server release 5. 3, Advanced Platform, Kernel 2.6.18-128.el5 |
Compiler: | PGI Server Complete Version 8.0 x86 Open64 4.2.2 Compiler Suite (from AMD) |
Auto Parallel: | No |
File System: | ext3 |
System State: | Run level 3 (Full multiuser with network) |
Base Pointers: | 32/64-bit |
Peak Pointers: | 32/64-bit |
Other Software: | binutils 2.18 SmartHeap 8.1 32-bit Library for Linux |
Benchmark | Base | Peak | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | |
Results appear in the order in which they were run. Bold underlined text indicates a median measurement. | ||||||||||||||
400.perlbench | 12 | 591 | 198 | 599 | 196 | 600 | 195 | 12 | 471 | 249 | 461 | 254 | 461 | 255 |
401.bzip2 | 12 | 823 | 141 | 826 | 140 | 834 | 139 | 12 | 795 | 146 | 788 | 147 | 784 | 148 |
403.gcc | 12 | 735 | 132 | 745 | 130 | 743 | 130 | 12 | 592 | 163 | 589 | 164 | 595 | 162 |
429.mcf | 12 | 853 | 128 | 859 | 127 | 858 | 128 | 12 | 542 | 202 | 545 | 201 | 542 | 202 |
445. gobmk | 12 | 654 | 193 | 654 | 192 | 655 | 192 | 12 | 566 | 223 | 565 | 223 | 565 | 223 |
456.hmmer | 12 | 460 | 243 | 467 | 240 | 467 | 240 | 12 | 315 | 355 | 318 | 352 | 318 | 352 |
458.sjeng | 12 | 772 | 188 | 772 | 188 | 770 | 188 | 12 | 701 | 207 | 703 | 206 | 702 | 207 |
462.libquantum | 12 | 1486 | 167 | 1439 | 173 | 1451 | 171 | 12 | 533 | 467 | 534 | 466 | 539 | 461 |
464. h364ref | 12 | 892 | 298 | 894 | 297 | 897 | 296 | 12 | 858 | 310 | 861 | 308 | 856 | 310 |
471.omnetpp | 12 | 727 | 103 | 720 | 104 | 713 | 105 | 12 | 707 | 106 | 712 | 105 | 718 | 104 |
473.astar | 12 | 745 | 113 | 739 | 114 | 737 | 114 | 12 | 609 | 138 | 610 | 138 | 612 | 138 |
483.xalancbmk | 12 | 410 | 202 | 410 | 202 | 409 | 202 | 12 | 399 | 207 | 396 | 209 | 398 | 208 |
The config file option 'submit' was used. 'numactl' was used to bind copies to the cores. See the configuration file for details.
'ulimit -s unlimited' was used to set environment stack size 'ulimit -l 2097152' was used to set environment locked pages in memory limit Set vm/nr_hugepages=5400 in /etc/sysctl.conf mount -t hugetlbfs nodev /mnt/hugepages
Environment variables set by runspec before the start of the run: HUGETLB_LIMIT = "450" LD_LIBRARY_PATH = "/usr/cpu2006/amd0905is-libs/64:/usr/cpu2006/amd0905is-libs/32" PGI_HUGE_PAGES = "450" The x86 Open64 Compiler Suite is only available from (and supported by) AMD at http://developer.amd.com/cpu/open64 System was tested in an open environment. ATX power supply 865W, PWS-865-PQ was used, [ 2 8-pin (+12V), and 24-pin are provided ] Product description can be obtained at: http://www.supermicro.com/Aplus/motherboard/Opteron2000/SR56x0/H8DI3+-F.cfm
C benchmarks:
opencc |
C++ benchmarks:
openCC |
400. perlbench: |
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX_X64 |
401.bzip2: | -DSPEC_CPU_LP64 |
403.gcc: | -DSPEC_CPU_LP64 |
429.mcf: | -DSPEC_CPU_LP64 |
445.gobmk: | -DSPEC_CPU_LP64 |
456.hmmer: | -DSPEC_CPU_LP64 |
458.sjeng: | -DSPEC_CPU_LP64 |
462.libquantum: |
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX |
464.h364ref: | -DSPEC_CPU_LP64 |
483.xalancbmk: | -DSPEC_CPU_LINUX |
C benchmarks:
-march=barcelona -Ofast -CG:local_sched_alg=1 -HP:bdt=2m:heap=2m |
C++ benchmarks:
-march=barcelona -Ofast -m32 -INLINE:aggressive=on -L/root/work/libraries/SmartHeap-8. 1/lib -lsmartheap |
C benchmarks (except as noted below):
opencc | |
456.hmmer: | pgcc |
C++ benchmarks (except as noted below):
openCC | |
473.astar: | pgcpp |
400.perlbench: |
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX_X64 |
401.bzip2: | -DSPEC_CPU_LP64 |
445.gobmk: | -DSPEC_CPU_LP64 |
456.hmmer: | -DSPEC_CPU_LP64 |
458.sjeng: | -DSPEC_CPU_LP64 |
462.libquantum: |
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX |
464.h364ref: | -DSPEC_CPU_LP64 |
483.xalancbmk: | -DSPEC_CPU_LINUX |
C benchmarks:
400. perlbench: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -IPA:plimit=20000 -LNO:opt=0 -OPT:unroll_times_max=8 -OPT:unroll_size=256 -OPT:unroll_level=2 -OPT:keep_ext=on -WOPT:if_conv=0 -CG:local_sched_alg=1 -CG:unroll_fb_req=on -HP:bdt=2m:heap=2m |
401.bzip2: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -OPT:alias=disjoint -OPT:unroll_size=0 -OPT:Ofast -OPT:goto=off -INLINE:aggressive=on -CG:local_sched_alg=1 -m3dnow -HP:bdt=2m:heap=2m |
403.gcc: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -Ofast -LNO:trip_count=256 -LNO:prefetch_ahead=10 -CG:cmp_peep=on -m32 -HP:bdt=2m:heap=2m -GRA:unspill=on |
429. mcf: |
-march=barcelona -O3 -ipa -INLINE:aggressive=on -CG:gcm=off -GRA:prioritize_by_density=on -m32 -HP:bdt=2m:heap=2m |
445.gobmk: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -OPT:alias=restrict -OPT:unroll_times_max=8 -OPT:unroll_size=256 -OPT:unroll_level=2 -OPT:keep_ext=on -ipa -IPA:plimit=750 -IPA:min_hotness=300 -IPA:pu_reorder=1 -LNO:prefetch=1 -LNO:ignore_feedback=off -CG:p2align=on -CG:unroll_fb_req=on -HP:bdt=2m:heap=2m |
456.hmmer: |
-fastsse -Mvect=partial -Munroll=n:8 -Msmartalloc=huge -Msafeptr -Mprefetch=t0 -Mfprelaxed -Mipa=const -Mipa=ptr -Mipa=arg -Mipa=inline -tp shanghai-64 -Bstatic_pgi |
458.sjeng: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -ipa -LNO:ignore_feedback=off -LNO:full_unroll=10 -LNO:fusion=0 -LNO:fission=2 -IPA:pu_reorder=2 -CG:ptr_load_use=0 -OPT:unroll_times_max=8 -INLINE:aggressive=on -HP:bdt=2m:heap=2m |
462. libquantum: |
-march=barcelona -Ofast -LNO:pf2=0 -CG:gcm=off -CG:use_prefetchnta=on -CG:cmp_peep=on -WOPT:aggstr=0 -HP:bdt=2m:heap=2m -OPT:alias=disjoint -INLINE:aggressive=on -IPA:space=1000 -IPA:plimit=20000 |
464.h364ref: |
-march=barcelona -fb_create fbdata(pass 1) -fb_opt fbdata(pass 2) -O3 -IPA:plimit=20000 -OPT:alias=disjoint -LNO:prefetch=0 -CG:ptr_load_use=0 -CG:push_pop_int_saved_regs=off -HP:bdt=2m:heap=2m |
C++ benchmarks:
471.omnetpp: |
-march=barcelona -Ofast -CG:gcm=off -INLINE:aggressive=on -OPT:alias=disjoint -WOPT:if_conv=0 -m32 -L/root/work/libraries/SmartHeap-8.1/lib -lsmartheap |
473.astar: |
-Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -Mipa=inline:6(pass 2) -fastsse -O4 -Msmartalloc=huge -Msafeptr=global -Mfprelaxed —zc_eh -tp shanghai-32 -Bstatic_pgi |
483. xalancbmk: |
-march=barcelona -Ofast -INLINE:aggressive=on -m32 -CG:cmp_peep=on -GRA:unspill=on -TENV:frame_pointer=off -L/root/work/libraries/SmartHeap-8.1/lib -lsmartheap |
C benchmarks:
456.hmmer: | -Mipa=jobs:4 |
C++ benchmarks:
473.astar: | -Mipa=jobs:4(pass 2) |
AMD Opteron — frwiki.wiki
Opteron is AMD’s trademark for x86 microprocessors designed for workstations and computer servers. It appeared on with the K8 architecture Sledgehammer kernel. In 2013, most of the opterons manufactured by AMD were single-core and then dual-core. They are borrowed from the K10 founder’s architecture. Today, optons can have up to 16 cores.
Like the Athlon 64 and Phenom, the Opteron has an integrated memory controller and HyperTransport bus.
In this area, AMD Opteron is located opposite the Xeon from the founder of Intel .
Summary
-
1 Numbering
- 1.1 Single and Dual Opteron — XYY
- 1.2 Second and third generation Opteron — XYZZ / ZYXX
-
2 Opterons (130 — 90 nm)
- 2.1 Sledgehammer
- 2.2 Venus (1yy)
- 2.3 Troy (2y), Athens (8y)
-
3 Dual core opteron (90nm)
- 3.1 Denmark (1yy)
- 3.2 Italy (2y), Egypt (8y)
-
4 Opteron — second generation (90 nm)
- 4.1 Santa Ana (12yy)
- 4.2 Santa Rosa (22yy, 82yy)
-
5 Opteron — Third generation (65-45nm)
-
5.1 Platform
- 5.1.1 Fiorano
- 5.1.2 Maranello
-
5.2 Range
- 5.2.1 Barcelona (23y, 83y)
- 5.2.2 Shanghai — Istanbul
- 5.2.3 Magny-Cours
-
5.1 Platform
-
6 32 nm
- 6. 1 Valencia
- 6.2 Interlagos
- 7 External links
- 8 Notes and references
Numbering
Single and Dual Opteron — XYY
The numbering system is very simple. It refers to two positions, and the values follow each other in even numbers (10, 12, 14 …) for single-core. The advent of the dual-core Opteron changes the sequence of versions following each other from 5 to 5 (60, 65, 70…).
- X indicates the number of processors
100: single processor
200: dual processor
800: up to 8 processors
- YY indicates the performance level of the series
EE (Energy Efficient): Low Power Version (30 W)Opteron second and third generation — XYZZ / ZYXX
The arrival of the second generation Opteron is accompanied by a nomenclature change that adds a third reference: socket (i.e. processor generation). This new numbering seems to be part of a longer-term perspective. Finally, he notes the return of even sequences (52, 54, 56…). The term HE remains, to which the name SE is added.
- x / z indicates the number of processors
1000: one CP
2000: two CP
8000: up to 8 CP- Y indicates a generation of socket
-20097 -300: a socket f +
- ZZ / XX indicates the performance level in the
series
SE: high performance version (120 to 125 W)
HE (high efficiency): low power version (65/68 W)Opteron (130 — 90 nm)
Single-core socket 939 Opteron, contrary to popular belief, does not have the same die as AMD Athlon 64 3700+, 4000+ and FX-55/57 from San Diego. Indeed, even if the L2 cache size is identical to the San Diego core, the Venus core also has Direct Connect technology and, above all, a 3-way Hyper Transport controller instead of a one-way controller. 3700+ and 4000+ and 2 channels for FX-57. In addition, it should be noted that Direct Connect, in addition to the capabilities in multi-processor solutions (processors communicate directly with each other, bypassing the chipset), allows you to offer better I / O control and more activity. case of dual-core processors in optimized dual-core applications.
On socket 939 Opteron (and socket 940, of course), Direct Connect takes control from the memory controller in the CPU on a single 128-bit channel instead of 2 64-bit channels for socket 939 Athlon64 in a dual-channel configuration.
Sledgehammer
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 100/200/800 series x50 1 2.40 GHz 2×64 KB 1 MB CG 1.5V 89W 940 x48 1 2. 20 GHz 2×64 KB 1 MB C0-CG 1.5V 89 W 940 x46 1 2.00 GHz 2×64 KB 1 MB C0-CG 1.5V 89W 940 x44 1 1.80 GHz 2×64 KB 1 MB B3
C0-CG1.55V
1.5V84.7 W
82.1 W940 x42 1 1.60 GHz 2×64 KB 1 MB B3
C0-CG1. 55V
1.5V84.7 W
82.1 W940 x40 1 1.40 GHz 2×64 KB 1 MB B3
C0-CG1.55V
1.5V84.7 W
82.1 W940 Opteron Series 100/200/800 HE x46 OH 1 2.00 GHz 2×64 KB 1 MB CG 1.3V 55W 940 Opteron Series 100/200/800 EE x40 EE 1 1. 40 GHz 2×64 KB 1 MB CG 1.15V 30W 940 Venus (1yy)
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 100 Series 156 1 3.0 GHz 2×64 KB 1 MB E4 1.30 — 1.35 V 104W 939 154 1 2.8 GHz 2×64 KB 1 MB E4 1. 35 — 1.40 V 104W 939 152 1 2.6 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 92.6 W
104 W940
939150 1 2.4 GHz 2×64 KB 1 MB E4
1.35 — 1.40 V89W
85.3W940
939148 1 2.2 GHz 2×64 KB 1 MB E4
1.35 — 1.40 V89W
85.3W940
939146 1 2. 0 GHz 2×64 KB 1 MB E4
1.35 — 1.40 V89W
67W940
939144 1 1.8 GHz 2×64 KB 1 MB E4
1.35 — 1.40 V89W
67W940
939142 1 1.6 GHz 2×64 KB 1 MB E4 1.30 — 1.35 V 89W 940 Opteron 100 Series HE 148 1 2. 2 GHz 2×64 KB 1MB E4 1.35 — 1.40 V 54.7W 940 Troy (2y), Athens (8y)
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron Series 200/800 x56 1 3.0 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 92.6 W 940 x54 1 2.8 GHz 2×64 KB 1 MB E4 1. 35 — 1.40 V 92.6 W 940 x52 1 2.6 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 92.6 W 940 x50 1 2.4 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 85.3W 940 x48 1 2.2 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 85.3W 940 x46 1 2.0 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 85.3W 940 x44 1 1. 8 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 85.3W 940 x42 1 1.6 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 85.3W 940 Opteron 200/800 series HE x50 OH 1 2.4 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 54.7W 940 x48 OH 1 2.2 GHz 2×64 KB 1 MB E4 1.35 — 1.40 V 54.7W 940 x46 OH 1 2.0 GHz 2×64 KB 1 MB E4 1. 35 — 1.40 V 54.7W 940 Dual Core Opteron (90nm)
Denmark (1yy)
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 100 Series 185 2 2.6 GHz 4×64 KB 2 × 1 MB E6 1.35 — 1.40 V 110W 939 / 940 180 2 2.4 GHz 4×64 KB 2 × 1 MB E6 1.35 — 1.40 V 110W 939 / 940 175 2 2. 2 GHz 4×64 KB 2 × 1 MB E6 1.35 — 1.40 V 110W 939 / 940 170 2 2.0 GHz 4×64 KB 2 × 1 MB E6 1.35 — 1.40 V 110W 939 / 940 165 2 1.8 GHz 4×64 KB 2 × 1 MB E6 1.35 — 1.40 V 110W 939 / 940 160 2 1.6 GHz 4×64 KB 2 × 1 MB E6 1. 35 — 1.40 V 110W 939 / 940 Italy (2yy), Egypt (8yy)
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron Series 200/800 x90 2 2.80 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 x85 2 2.60 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 x80 2 2. 40 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 x75 2 2.20 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 x70 2 2.00 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 x65 2 1.80 GHz 4×64 KB 2 × 1 MB E6 1.30 — 1.35 V 95W 940 Opteron 200/800 series HE x75 OH 2 2. 20 GHz 4×64 KB 2 × 1 MB E6 1.15 — 1.20 V 55W 940 x70 OH 2 2.00 GHz 4×64 KB 2 × 1 MB E6 1.15 — 1.20 V 55W 940 x65 OH 2 1.80 GHz 4×64 KB 2 × 1 MB E6 1.15 — 1.20 V 55 W 940 x60 OH 2 1.60 GHz 4×64 KB 2 × 1 MB E6 1.15 — 1.20 V 55W 940 Opteron — second generation (90 nm)
DDR2 memory management
Santa Ana (12y)
Model Nb. hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 1000 SE Series 1222SE 2 3.0 GHz 2 x (64 + 64) KB 2 x 1 MB F2 — F3 1.35 — 1.40 V 125W AM2 1220SE 2 2.8 GHz 2 x (64 + 64) KB 2 x 1 MB F2 — F3 1.35 — 1.40 V 125W AM2 Opteron 1000 Series 1222 2 3. 0 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 1220 2 2.8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 1218 2 2.6 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 1216 2 2.4 GHz 4×64 KB 2 × 1 MB F2 — F3 1. 30 — 1.35 V 103W AM2 1214 2 2.2 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 1212 2 2.0 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 1210 2 1.8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 103W AM2 Opteron 1000 Series HE 1218ET 2 2. 6 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1.25 V 65 W AM2 1216ET 2 2.4 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1.25 V 65 W AM2 1214ET 2 2.2 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1.25 V 65 W AM2 1212ET 2 2.0 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1. 25 V 65 W AM2 1210ET 2 1.8 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1.25 V 65 W AM2 Santa Rosa (22 years old, 82 years old)
Model Nb. Hearts Frequency L1 cache L2 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 2000/8000 SE Series x224 SE 2 3.2 GHz 4×64 KB 2 × 1 MB F3 1. 325 — 1.35 — 1.375 — 1.40 V 120W 2000 MT/s F x222 SE 2 3.0 GHz 4×64 KB 2 × 1 MB F3 1.325 — 1.375 V 120W 2000 MT/s F x220 SE 2 2.8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.325 — 1.375 V 120W 2000 MT/s F Opteron 2000/8000 series x222 2 3.0 GHz 4×64 KB 2 × 1 MB F3 1.30 — 1.35 V 95W 2000 MT/s F x220 2 2. 8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F x218 2 2.6 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F x216 2 2.4 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F x214 2 2.2 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F x212 2 2. 0 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F 2210 2 1.8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.30 — 1.35 V 95W 2000 MT/s F Opteron 2000/8000 series HE x218 OH 2 2.6 GHz 4×64 KB 2 × 1 MB F3 1.20 — 1.25 V 68W 2000 MT/s F x216 OH 2 2.4 GHz 4×64 KB 2 × 1 MB F2 — F3 1.20 — 1.25 V 68W 2000 MT/s F x214 OH 2 2. 2 GHz 4×64 KB 2 × 1 MB F2 — F3 1.20 — 1.25 V 68W 2000 MT/s F x212 OH 2 2.0 GHz 4×64 KB 2 × 1 MB F2 — F3 1.20 — 1.25 V 68W 2000 MT/s F 2210ET 2 1.8 GHz 4×64 KB 2 × 1 MB F2 — F3 1.20 — 1.25 V 68W 2000 MT/s F Opteron — Third generation (65-45nm)
Commercialization of the K10 architecture marks the arrival of 3- Generation Opteron. It features a quad-core structure and L3 cache. The first versions, like Phenoms, were prone to TLB bugs. The B3 Opteron Barcelona revision was accompanied by new HE models with a lower TDP.
Platform
Fiorano
AMD has taken advantage of the launch of its new SR56x0 chipsets to introduce a 100% AMD platform. Thus, the Fiorano platform is organized around the SR56x0 northbridge and SB5100 southbridge, and is dedicated only to socket F compatible opterons. Three chip models are planned: AMD SR5650, SR5670, and SR5690. They can support up to 42 PCI Express 2.0 lanes, together with two, their number can be increased to 84 lanes. But above all, they provide PCI-Express 2.0 and HyperTransport 3.0 bus management. They also include Advanced Virtualization Technology (AMD-Vi), which now allows for peripheral device virtualization (IOMMU technology, equivalent to Intel’s VT-d technology). This platform is not new, although it offers many improvements, but, above all, it allows you to fill the gap in anticipation of the next Opterons for the G34 socket.
Maranello
Range
Barcelona (23y, 83y)
Model Nb. Hearts Frequency L1 cache L2 cache L3 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 2000/8000 SE Series x360 SE 4 2.5 GHz 4×128 KB 4×512 KB 2MB Bi 2 F+ x358 SE 4 2.4 GHz 4×128 KB 4×512 KB 2MB Bi 2 F+ Opteron 2000/8000 series x356 4 2. 3 GHz 4×128 KB 4×512 KB 2MB B3 1.2V 115 W F+ x350 4 2.0 GHz 4×128 KB 4×512 KB 2MB Bi 2 1.2V 95W F+ x347 4 1.9 GHz 4×128 KB 4×512 KB 2MB Bi 2 1.2V 95W F+ Opteron 2000/8000 series HE x347 OH 4 1. 9 GHz 4×128 KB 4×512 KB 2MB B3
B2
1.15V55W
68WF+
x346 OH 4 1.8 GHz 4×128 KB 4×512 KB 2MB B3
B2
1.15V55W
68WF+
x344 OH 4 1.7 GHz 4×128 KB 4×512 KB 2MB B3
B2
1.15V55W
68WF+
Shanghai — Istanbul
Model Hearts Frequency Hidden Revision Voltage TDP Hypertransport Connector Help Marketing Heart Memory controller L1 L2 L3 Opteron 8 / 24xx SE 8439SE
2439SE6 2. 8 GHz 2.2 GHz 6×128 KB 6×512 KB 6 MB D0 1.3V 137W 4.8 GHz F OS8439YDS6DGN
OS2439YDS6DGNOpteron 8 / 24xx HE 8425 NOT
2425 NOT6 2.1 GHz 2.2 GHz 6×128 KB 6×512 KB 6 MB D0 1.3V 79W 4.8 GHz F OS8425PDS6DGN
OS2425PDS6DGN2423ET 6 2.0 GHz 2.2 GHz 6×128 KB 6×512 KB 6 MB D0 1. 3V 79W 4.8 GHz F OS2423PDS6DGN Opteron 8 / 24xx EE 2419EE 6 1.8 GHz 2.0 GHz 6×128 KB 6×512 KB 6 MB D0 1.125V 60 W 4.8 GHz F OS2419NBS6DGN Opteron 8 / 24xx 8435
24356 2.4 GHz 2.2 GHz 6×128 KB 6×512 KB 6 MB D0 1.3V 115 W 4. 8 GHz F OS8435WJS6DGN
OS2435WJS6DGN8431
24316 2.4 GHz 2.2 GHz 6×128 KB 6×512 KB 6 MB D0 1.3V 115 W 4.8 GHz F OS8431WJS6DGN
OS2431WJS6DGN2427 6 2.2 GHz 2.2 GHz 6 × 128 KB 6×512 KB 6 MB D0 1.3V 115 W 4. 8 GHz F OS2427WJS6DGN Test List for Opteron Istanbul
- 2435
- (en) X-bit Labs .
Magny-Cours
Magny-Cours is the first model with a dodeca core (with 12 cores), however the cores are not their own, but are the result of the assembly of two crystals, which is the first, because Istanbul AMD always preferred to be interested in one. main control. In addition, this new model is designed to consume no more than one Istanbul core thanks to lower frequencies and consumption optimization. Due to its configuration, it will use a new socket called G34 (1974 pins) that will make up the next Maranello platform with the following AMD 800 chipsets. Even though it uses 12 million L3 cache, 2 million is dedicated to HT-assistance .
Model Hearts Frequency L1 cache L2 cache L3 cache Revision Voltage TDP Hypertransport Connector Marketing Opteron 6xxx 2 × 6 1.7 to 2.3 GHz 12×128 KB 12×512 KB 2 × 6 MB 65 to 115 W 3.2 GHz G34 1- cut. 2010 2×4 1.8 to 2.4 GHz 8×128 KB 8×512 KB 2×6 MB 65 to 137 W 3.2 GHz G34 1- cut. Opteron 146-846: September 9, 2003, Opteron 246: August 5, 2003 - ↑ a and b Opteron 252 — 254: 68 W and 92.6 W
- ↑ Opteron 250: 68W and 85.3W
- ↑ a b c and d Opteron 265 — 270 — 275 — 280: 68 W or 95 W
- ↑ (fr) Frederick. AMD: New 55W TDP Opterons at Clubic May 13, 2008
- ↑ Florian Vieru. AMD Unveils New Server Chipsets at PCWorld.fr September 22, 2009
- ↑ Florian Vieru. AMD Magny-Cours, 12 cores since early 2010? at PCWorld, August 25, 2009
AMD
Processors
Current x86-64 - AMD FX
- Athlon X4
- Ryzen
- Epic
Abandoned Up to x86 - Am9080
- Am2900 (ru)
x86-16 (16bit) Am286 x86-32 (32 bit) - Am386
- Am486
- Am5x86
- K5
- K6
- K6-2
- K6-III
- Duron
- Athlon
- Geode
x86-64 / AMD64 (64bit) - Mobile Athlon 64
- Sempron
-
Athlon 64
- Athlon Neo
- Athlon 64FX
- Athlon X2
-
Name
- II
- Athlon II
- Turion
- Opteron
- Merger
Other - Am29000
- Au1
Microarchitectures Primary - K5
- K6
- K7
- K8
- K10
-
Bulldozer
- Piledriver (ru)
- Steamroller
- Excavator
-
Zen
- Zen 2 (ru)
Low consumption - Alchemy
- Lynx
- Jaguar (en)
- Puma
ARM K12 ( inch ) Sockets - 7
- Super 7
- B
- 563
- S1
- 754
- 939
- 940
- AM2
- AM2 +
- F
- F+
- AM3
- G34
- C32
- AM3 +
- FM1
- FM2
- AM4
- TR4
wikipedia.org/wiki/Special:CentralAutoLogin/start?type=1×1″ alt=»» title=»»>
Page not found — Technical City
Page not found found — Technical City
We did not find such a page on our site: /ru/cpu/rating%3fpg%3d13
Popular video card comparisons
GeForce RTX
3060 Ti
or
GeForce RTX
3060
GeForce GTX
1050 Ti
or
GeForce GTX
1650
GeForce RTX
2060
or
GeForce RTX
3050 8GB
GeForce GTX
1660 Super
or
GeForce RTX
3050 8GB
GeForce RTX
3060 Ti
or
GeForce RTX
3070
GeForce RTX
2060 Super
or
GeForce RTX
3060
Popular graphics cards
GeForce GTX
1050 Ti
GeForce RTX
3090 Ti
Radeon RX
Vega 7
GeForce GTX
1650
GeForce RTX
3060
GeForce GTX
1060 6GB
Popular
processor comparisons
Ryzen 5
5600X
or
Core i5
12400F
Core i5
1135G7
or
Ryzen 5
5500U
Ryzen 5
3600
or
Core i5
10400F
Ryzen 5
5600X
or
Ryzen 5
5600G
Ryzen 5
3600
or
Ryzen 5
5600X
Ryzen 7
3700X
or
Ryzen 5
5600X
Popular processors
Ryzen 5
5500U
EPYC
7h22
Core i3
1115G4
Core i5
1135G7
Ryzen 5
3500U
Ryzen 3
5300U
The main events of the past week.
Issue 90 / Analytics
3DNews Analytics Events. Data. Forecasts The main events of the past week. Release…
The most interesting news
In this issue: the official presentation of economical and «overclocked» 6-core AMD Opteron processors; preparation for the release of the Ibex-Peak platform; optimistic trends in the information technology industry; successful space flight Falcon 1 with satellite on board
We will start the next issue of the weekly news digest with the news that opened the last week. Namely, since the release by AMD of several AMD Opteron server processors with six cores at once. For the first time, six-cores were introduced by AMD at the very beginning of June 2009.year, and the presentation took place much earlier than planned. The company’s next step was an attempt to strengthen its position in the server market by releasing products for various categories of computing systems, both for cost-effective servers with an emphasis on energy efficiency, and high-performance machines.
The main feature of the three processor models should be their reduced power consumption — new products should be the most efficient products in terms of such a parameter as performance per watt of power consumption. In this case, we are talking about the models 8425 HE, 2425 HE and 2423 HE, the reduced power consumption of which is achieved by reducing the operating frequency to 2.1 GHz (for the 2423 HE model, the clock frequency is 2.0 GHz, which, however, does not affect average power consumption). Here it is necessary to note the following fact — AMD operates with such an indicator as the average power consumption of the processor (ACP), instead of the more usual maximum power consumption (TDP). For these CPU models, the ACP value does not exceed 55 watts, but do not forget that the maximum heat dissipation can be much higher than the indicated figures. As a result, it is still impossible to confidently talk about the advantage of AMD or Intel server processors — the end user must choose the most suitable option based on their own preferences and features of specific computing systems.
According to the developers, the new AMD processors are characterized by an 18% increase in specific performance, that is, computing power in terms of one watt of power consumption.
This means that AMD customers looking to build a power efficient computing system should look to the AMD Opteron six-core HE series. The developers claim that the novelties will significantly reduce the power consumption of finished computing systems. If platforms based on previously released six-core processors consumed about 270 W, then switching to economical CPU options can reduce consumption to 211 W (data are valid at 100% processor load).
Where performance comes first, customers are encouraged to purchase two other six-core Opteron processors. We are talking about models 8439 SE and 2439 SE, the main feature of which is the clock frequency increased to 2.8 GHz. A side effect of «overclocking» processors to such a high frequency was a noticeable increase in their power consumption — the average power consumption rose to 105 watts. In terms of performance, compared to quad-core processors, the computing power of the new AMD Opteron SE series has increased immediately by 50%.
Featured Opteron six-core server processors:
Model Features Cost 8439 SE 2.8 GHz, ACP — 105 Watts $2649 8425 HE 2.1 GHz, ACP — 55 Watts $1514 2439 SE 2.8 GHz, ACP — 105 Watts $1019 2425 HE 2.1 GHz, ACP — 55 Watts $523 2423 HE 2. 0 GHz, ACP — 55 Watts $455 Now let’s pay attention to AMD’s main rival in the microprocessor market — Intel. Here, first of all, we note the imminent appearance of a new platform for personal computers — Ibex-Peak, which includes quad-core Lynnfield processors and the Intel P55 system logic for them. According to information that appeared this week, the manufacturer expects to officially present both new items in the period from the eighth to the eleventh of September. The first CPUs based on the Lynnfield core will be the mid-price model Core i5 750 with a clock speed of 2.66 GHz, and two productive models Core i7 860 (2.8 GHz) and Core i7 870 (2.93 GHz). All models are designed for installation in the LGA 1156 socket, are equipped with an 8 MB L3 cache, as well as an integrated DDR3 RAM controller. The two older models will boast support for HyperThreading technology, while the Core i5 750 model will not have such functionality.
Of course, the release of new central processors of the corresponding motherboards that support working with them will turn out to be pointless. But you shouldn’t worry about such a scenario — motherboard manufacturers are increasingly announcing products based on the Intel P55 system logic. This week, Gigabyte was noted, showing the world the GA-P55M-UD4 motherboard of the Micro-ATX form factor (/news/pervaya_plata_formata_micro_atx_na_baze_intel_p55/), as well as Biostar, which talked about a whole line of T-Power motherboards, which included T models -Series T5 XE, T-Series TP55 XE, and T-Power I55. Thus, we can talk about the readiness of motherboard manufacturers to immediately release a whole range of different products based on the Intel P55 chipset — from compact Micro-ATX form factor motherboards for small personal computers to «top» products for high-performance computing systems.
However, information about the imminent appearance of the Ibex-Peak platform is not the only good news coming from Intel. The second positive moment was the publication of financial results for the second quarter of this year. It would seem that the results for the reporting period should not please — Intel ended the second quarter with a cash loss of $ 398 million, while for the same period last year, the profit of the world’s leading chipmaker amounted to $1.6 billion. But you need to remember the significant fine imposed on Intel by the decision of the European Commission. Recall that earlier the European Commission found Intel guilty of unfair competition and abuse of its dominant position in the market of central processors (/news/intel_priznana_vinovnoi_v_nechestnoi_konkurentsii/). The amount of the fine, which amounted to just under one and a half billion US dollars, is deducted from the company’s income, materially distorting the financial results. If not for the payment of the specified amount, Intel’s profit in the second quarter of this year would have amounted to quite a solid amount of one billion US dollars. And this result would be better than expected and predicted by analysts.
For the general public, it also means that the industry is gradually emerging from the crisis. As another example pointing to the process of recovery of the IT industry, we can consider the report of analysts on the personal computer market in the second quarter of 2009of the year. It turns out that sales of personal computers, although they fell relative to the same period last year, but only slightly. Sales volumes declined by only 3 percent, while many analysts had predicted a drop of more than 6 percent compared to last year. If we talk about the state of affairs of specific manufacturers, then the second quarter was best completed by HP, which is also the leader in the PC market. She managed not only to maintain her leading position — the company owns about 20% of the entire PC market, but even increased the volume of product deliveries by 3.6%, compared with the results of last year. However, Acer’s business is also going uphill — an increase in sales of personal computers amounted to 23. 7% at once. However, even this was not enough to get ahead of Dell, which occupies second place in the list of leading PC manufacturers today. But the latter was seriously affected by the crisis — in the second quarter of 2009year, the decline in sales of computers amounted to 13.7%. The reason for this result can be considered the company’s emphasis on corporate clients, who, during the unfavorable economic situation in the world, do not dare to expand and update the fleet of personal computers.
We propose to complete the release of the weekly news digest with the most important event for the space industry — the launch of an artificial satellite into the Earth’s orbit using the world’s first private rocket. We are talking about the Falcon 1 rocket, created by SpaceX specialists. In total, this is the fifth launch of the rocket — the previous ones ended in failure and the loss of Falcon 1 due to various problems. At the same time, during the third launch, research satellites were also lost, which the rocket was supposed to put into Earth orbit. The fourth launch, which turned out to be the first successful launch of the Falcon 1 into space, did not provide for the presence of a payload on board the rocket — there was only a ballast to simulate it. And so, July 13, 2009A Falcon 1 rocket launched the Malaysian satellite RazakSAT into Earth orbit, designed to receive HD photographs of the territory of Malaysia.
SpaceX’s main goal is to develop a rocket that would allow a payload to be lifted into Earth orbit at a relatively low cost — this is the main advantage of Falcon rockets. And it is precisely due to this that the company expects to compete with state structures of space powers, mainly the United States and Russia. For SpaceX, the successful launch of the Falcon 1 with a satellite on board is also an excellent advertisement for its services and a guarantee that the next launches will be successful.