To get about the same clock, you need to pay $325.
The Core-i7 980X is made on a smaller process node, 32 instead of 45 nm. The two processors aren't easily compared, the cores in the Core have half the L2 cache of the Phenom (256 vs. 512KB) and the shared L3 is twice as big (12 vs. 6MB).
As for the pricing, according to Wikipedia there's just this one unlocked Core desktop part, vs. a couple of dozen Xeon server parts and 8 4 core dual processor parts. It sure looks like to me like Intel is selling all the server parts it can and is throwing out this one desktop version for the enthusiasts with a price to match.
Anyone have an idea of what the different L2 and L3 cache tradeoffs will mean in practice?
For the intel the L2 cache (and the L1 of course) are shared for both the hyperthreads, so each hyperthread would get even less than 256K. On AMD, however each thread gets the full 512K. The L3 latency is much larger than the L2 (~40 cycles IIRC) so the smaller L2 would limit the performance of a cache hungry (but not too hungry: see next paragraph) application.
One place where the larger L3 is a win is on low parallelism, memory hungry loads -- they have a much larger L3 all to themselves.
You can't compare Xeons to Phenoms, you need to be looking at Opterons for that. AMD recently released their 12-core Opterons (4 different models) and they have 3 8-core and 5 6-core Opterons as well (9 if you distinguish between 2-socket and 8-Socket versions).
I'm only comparing them as a business decision WRT allocating fab lines. I'm assuming the profit margins are higher on server chips and the high end enthusiast desktop ones and proposing this as an explanation for Intel's product mix. There's got to be some reasons Intel is only selling one model of the latter while selling a couple dozen of the former with this microarchitecture and on this process node.
Speaking generally, lower latency to main memory is better than more cache.
On a generation of one platform I dealt with, main memory latency and bandwidth (faster processor design, on-board memory controller, RDRAM) were on par with the L3 cache from the previous platform generation.
To give people a rough idea, the 1055T X6 scores 5960 points on average in the PassMark benchmark while the i7 980X scores 10165.
As always, benchmarks are not useful for deciding how your application will perform on a certain CPU and you should do some tests for yourself, but considering the huge difference in this synthetic benchmark, I'd say some people would be willing to pay difference.
Or you could just spend it on a second AMD machine. Intel certainly leads on raw power but that's certainly not the only consideration...is the Intel brand worth 3x the price after you normalize for performance? I think AMD will sell a ton of these.
The power consumption of two AMD machines over their lifespans, compared to that one Intel machine, might be an issue though I've not run any calculations.
The article is missing the elephant in the room - the 6 core Phenoms are intended to compete against quad core i7s. Nehalem multithreading gives an incredible boost for multithreaded apps, usually 50%+ even for heavy tasks like rendering.
On top of that, Intel is way ahead in the IPC game. I don't know what the lay of the land is currently, but back when it was a choice between 45nm Phenom X4 and 45nm Core i7 9x0 chips, in some circumstances a Core i7 thread performed the same as a Phenom core.
We're also now in a leapfrogging situation, where year by year this may change. AMD is one process node behind Intel; I'm not sure how to count microarchitectures, but ignoring processor interconnects (which Intel has only just started to address by copying the AMD approach) if you consider Core to be more advanced than K8 and Nehalem more than K10 then AMD is also one microarchitecture behind.
I found it really odd how much the MC scores increased relative to the Xeon in the one Linux benchmark they ran. Maybe there's something about the that Windows handles NUMA that doesn't agree with AMD's system?
The difference in price between the two processors is basically equal to the rest of the components in a decent system. Intel is going to have to drastically cut their prices down soon. More and more, it's the GPU that bounds the system, CPUs have long ago surpassed the needs of most users...and I'm not just talking about people who only use office and email.
On per-chip performance, if price is disregarded, Intel still wins, but AMD has since long given better performance per $, and is a far better choice for the average consumer not keeping their machine at a constant 100% cpu use, not requiring the best available performance for something akin to a single work-horse type machine.
To get about the same clock, you need to pay $325.
The Core-i7 980X is made on a smaller process node, 32 instead of 45 nm. The two processors aren't easily compared, the cores in the Core have half the L2 cache of the Phenom (256 vs. 512KB) and the shared L3 is twice as big (12 vs. 6MB).
As for the pricing, according to Wikipedia there's just this one unlocked Core desktop part, vs. a couple of dozen Xeon server parts and 8 4 core dual processor parts. It sure looks like to me like Intel is selling all the server parts it can and is throwing out this one desktop version for the enthusiasts with a price to match.
Anyone have an idea of what the different L2 and L3 cache tradeoffs will mean in practice?