Intel Tigerton and AMD Barcelona price list
Summary: Intel has launched the Xeon 7300 platform codenamed "Caneland" and the Core micro-architecture Xeon MP Processor codenamed "Tigerton". Since AMD is set to launch "Barcelona" quad-core Opterons next week and DailyTech leaked the Barcelona price list, we can now compare the prices for the new MP (multi-processor platform) CPUs.
Intel has launched the Xeon 7300 platform codenamed "Caneland" and the Core micro-architecture Xeon MP Processor codenamed "Tigerton". Since AMD is set to launch "Barcelona" quad-core Opterons next week and DailyTech leaked the Barcelona price list, we can now compare the prices for the new MP (multi-processor platform) CPUs.
*Note that all the unit prices listed below are for quantities of 1000 ** IBM X3950 M2 X4 chipset (DDR2-based) *** Unisys 32-socket chipset
| Intel Tigerton XEON MP price list (4, 8-16**, 32*** socket) | |||
| Model | Frequency | TDP | Price* |
| X7350 quad-core 8MB cache | 2.93 GHz | 130W | $2301 |
| L7345 quad-core 8MB cache | 1.86 GHz | 50W | $2301 |
| E7340 quad-core 8MB cache | 2.4 GHz | 80W | $1980 |
| E7330 quad-core 6MB cache | 2.4 GHz | 80W | $1391 |
| E7320 quad-core 4MB cache | 2.13 GHz | 80W | $1177 |
| E7220 dual-core 8MB cache | 2.93 GHz | 80W | $1177 |
| E7310 quad-core 4MB cache | 1.6 GHz | 80W | $856 |
| E7210 dual-core 8MB cache | 2.4 GHz | 80W | $856 |
| AMD Barcelona Opteron quad-core price list (4 and 8 socket) | |||
| Model | Frequency | TDP | Price* |
| 8350 | 2.0 GHz | 95W | $1004 |
| 8347 | 1.9 GHz | 95W | $774 |
| 8347 HE | 1.9 GHz | 68W | $861 |
| 8346 HE | 1.8 GHz | 68W | $688 |
[update 5:07PM - I should clarify this "win" I'm granting AMD on energy efficiency against is at the low-end. When you get to Intel's 80W E7340 2.4 GHz 8MB Cache model, you get a lot more work done which offsets the small power differential which changes the performance/watt metric significantly. [UPDATE 9/8/2007 - Furthermore, Sun's engineer explained that the Barcelona processors only support 4 DDR2-667 memory DIMMs because that's a limit of DDR2 technology UNLESS you're willing to drop down to DDR2-533 memory which Sun is not willing to do. This is why Sun's next 4-socket Intel server will have double the memory capacity of Sun's 4-socket AMD server so it isn't a black and white argument on DDR2 versus FBDIMM. This information I got isn't correct. Each Opteron chip has two channels which means any 4-socket Opteron server will natively support 32 DDR2-533 or 667 DIMMs. I could have sworn 2-socket servers had 16 DIMM slots which they do so I thought the info I got from the Sun engineer sounded funny at the time but I figured he would know better.] IBM went to a very expensive and exotic design with their X4 chipset to be able to utilize DDR2 at high capacities.
We also need to recognize the fact that the 4-socket market cares about performance and memory capacity first and foremost. Customers of 4-socket servers are willing to pay 4x the money for 2x the performance. Someone willing to pay a $10K to $20K premium up-front for just a little more performance isn't that worried about saving $100-$200 on power per year.]
We will not know exact performance numbers for Barcelona Opteron quad-core until hopefully next week. Of course when they do release performance numbers, I'll do a detailed price, watt, and performance shootout. AMD by all estimates will dominate memory-intensive HPC (High Performance Computing) but they will not likely threaten anything above the Intel E7320 on mainstream server performance until AMD can ramp the clock frequencies on Barcelona.
If you also want to see leaked prices for Barcelona dual-socket processors, see this post.
Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.
Talkback
what about 2XXX Barcelona
AMD by all estimates will dominate, I cant believe you said that! Make sure the PPW is with power saving ON
The link is there to DP prices
Ah TY
AMD silences quad-core critics
80 plus percent of the market doesn't care about SPECfp
the Present Opty beat intel PPW
Yes, but who doesn't pay for the cutting edge?
But on the other hand, they get to say that they owned an iPhone before anyone else.
IBM's solution costs more money, but from the energy and performance standpoint, costs are justified.
It's performance people want, ROI on power savings will take forever
the power supply wattage is what really consumes power
because the power supply is always ready to supply power on demand.
right
lets really think this over.
thats probably true
Not True
They don't know, otherwise Microsoft wouldn't be in the position of power that they are today. They didn't jump on the Opteron bandwagon as quickly as they should have and they pushed the Itanium, especially on the development side which is what the SCO Group was hopping up and down about. Not the work on it, but the fact that they were originally working on Project Monterey and then abandoned it for Linux on pretty much everything.
The same goes for SPECint_rate peak performance
Base values are slightly better for generalization, but it's also noteworthy that a different compiler sometimes has more impact on the results than the selected CPU.
In general, there are much better benchmarks to choose a server product on than SPECCPU. The latter is used mostly in one area: Marketing.
AMD refuses to quote base for their simulated marks
Sure
I've never ignored SPECfp_rate. I've always stated AMD dominates it
If anyone is ignoring something, it's your refusal to acknowledge the market size for HPC.
firstly and foremostly you're ignoring what I wrote
That's exactly the moot point. The SPECint suite does not contain a single classical server application. SPECint_rate does not reflect applications with large memory footprints well, which are run on many servers (databases). SPECint_rate intentionally avoids shared resources to measure raw throughput, while that sharing of resources is a central aspect of many server applications (networking, I/O, file servers, databases). Whoever submits SPECint_rate results, carefully avoids more than one context per core to achieve the best result, while such scenarios exist widely in server setups (virtualization, application servers).
The bottom line is, SPECint and its *_rate flavor are not very useful to estimate server performance across a broad range of applications and your repeated statement, that SPECint_rate2006 applies to 80% of the server market is plain wrong.
One example: The SPECint_rate comparison of a 2-way Opteron 2220 and a 2-way Xeon 5160 suggests a significant performance advantage for the Xeon, but both CPUs perform almost identically in similar setups for the SPECweb2005 and SAP-SD benchmarks. SPECint_rate fails to reflect the performance balance in two important server roles. QED.
secondly you can't refrain from turning this into an Intel vs. AMD debate
I did not make any assumptions about the market relevance of either benchmark. I actually expressed very clearly that SPECfp_rate as well as SPECint_rate are used for marketing by [b]both[/b] companies. Intel is favoring SPECint2006 (well, they used to favor SPECint2000, but they can hardly do that anymore), because their platform performs exceptionally well in that benchmark compared to the competition. They even published new results when they had a better compiler that boosted the numbers, which has next to zero real world relevance. They seldom publish SPECfp results at all. AMD on the other hand favors SPECfp, because their architecture has the right mix and setup of execution units to beat the competition here. That particular advantage is useless when no floating point instructions are executed. Still, they did not show SPECint numbers for "Barcelona" on their analyst day and it's not hard to imagine why.
I do not blame either company for those tactics. That's marketing. But you're either easily taken in by Intel's marketing or you're playing the naive as part of an agenda.
thirdly you're self-righteous and dishonest
That statement only comes as a side note in this blog post and often enough only in the talkback section when someone brings the topic up. In your infamous attack against AMDs simulated benchmarks, you had a couple of graphs showing SPECint_rate results and did not use a single word on SPECfp_rate. Even if you mention it elsewhere, it is always prejudiced (you do not mention SPECfp_rate2006 but HPC here) and accompanied by a comment to downplay its significance. In contrast to that, SPECint_rate is not only left unquestioned by you, but is often enough overrated as a synonym for general performance. Unfortunate as it is, there is no single benchmark to reflect that.
SPECFP Rate does not matter in this market segment
This any IT manager worth their salt can attest to.
There are exceptions in HPC Niches but it's not what moves MP Server purchases.
EOF
HUm