New Phenom – Same as the Old Phenom?
Phenom II, with an expected release of January ’09, is right around the corner. And boy, does AMD need it. The original Phenom has been unsuccessful at besting Intel’s Core 2 Duo line, and with Core i7 now in the mix, AMD has a significant performance disadvantage that must be countered. Phenom II is to be that counter, but the relative lack of coverage regarding Phenom II in contrast to the hype surrounding Core i7 makes it easy to believe that Phenom II will simply be just another AMD processor that fails to meet the performance gauntlet laid down by Intel’s quicker products.
Part of this lack of hype is probably because Phenom II will be relatively similar to the original Phenom architecture. Intel’s Core i7 represented a major change in the way Intel approached chip design, and as a result there was plenty to talk about. Phenom II, on the other hand, uses the same basic architecture as Phenom. Another contributing factor is that AMD seems to have kept a fairly tight lid on Phenom II specifications so far. Why is anyone’s guess, but at this point during the design of Core i7 there were already multiple articles available which detailed the changes between Core i7 and Core 2 Duo. In fact, Anandtech had previews of Core i7 available as far back as June.
But there is still reason to be excited about Phenom, not the least of which is that Phenom’s architecture, at its base, is actually very modern. The development of Core i7 showed that, in many ways, Intel was playing catch-up to AMD. Ideas such as the Quick Path Interconnect (similar to AMD’s Hyper Transport), and placing the memory controller on the processor itself, are features already offered by AMD. Also, Phenom was built from the ground up as a Quad-Core processor, unlike Core 2 Quads, which were basically two Core 2 Duos fused onto a single chip. In other words, the ground-work for an amazing processor is already there. AMD just needs to back it up with faster clock speeds and better power efficiency.
The Future is Shanghai
Information about Phenom II may be sketchy at his point, but information about Shanghai is not. This is key, because unlike Intel, which tends to lead with consumer product and follow up with server platforms later, AMD follows a reverse pattern, releasing server-based products first. What this means is that Shanghai Opertrons are basically Phenom IIs, in the same way that Core 2s are basically Xeons. Or at least, that is the theory.
One of the largest changes made with Shanghai is the expansion of overall cache to 8MB, 6MB of which is L3 cache. AMD uses exclusive cache, which means each level of cache is responsbile for its own data. In contrast, the cache on Intel processors in inclusive, which means that the data on higher levels of cache are replicated on the L3 cache. That means all of Shanghai’s 6MB L3 cache can be utilized, though it can cause snooping, where one core starts looking for things in another core’s cache. Another interesting feature is SmartFetch, which dumps the L2 and L1 cache associated with a core into the L3 cache when that core is not needed. This overcomes a traditional problem with Phenom’s exclusive cache system – if the data in a cache isn’t replicated in the L3 cache, then a core can’t be put completely to sleep, because the cache associated with that core has to remain active. Other improvements include the move to the faster HyperTransport 3 interconnect (although that is not enabled in the earliest Shanghai Opterons) and an increase in supported memory frequency from 667Mhz to 800Mhz.
The results are positive, but not overwhelmingly so. Early reviews of Shanghai Opterons are, of course, focused on server applications, and as such they do not provide a completely accurate picture of how the Phenom II will perform. That said, the results show that Shanghai Opterons are slightly behind to even with the performance of Core 2 Xeons on a clock-for-clock basis. They do best with programs that take advantage of their superior architecture, and fall behind in benchmarks that are bound by compute power. This is in line with what we’ve come to expect from Phenoms.
Other Early Indicators
The relaitve inability to hit clock speeds above the 3Ghz mark has plagued Phenom since its release. Though Phenom chips can offer quad-core advantages for less than the price of Intel’s Core 2 Quads, this benefit is obscured by the lack of software that takes advantage of multiple cores. Most programs used for basic, day-to-day tasks use only a single core, and as such the maximum performance that can be squeezed from a single core is still just as important as the number of cores on a processor.
AMD recently lifted hopes that Phenom II may be able to provide impressive clock speeds by showing off a number of over-clocked Phenom II chips. These apparently chips reached speeds of up to 4Ghz on air and over 6Ghz on liquid nitrogen. This says nothing about the performance-per-clock, of course, and it also does not indicate that Phenom II chips will be available at speeds anywhere near 4Ghz. However, it does show that Phenom II chips are reaching overclocking potential that is roughly equal to what one could expect from 45nm Core 2 processors. Since Core 2 processors can be found at speeds slightly above 3Ghz, it isn’t unreasonable to expect that Phenom II chips, which appear to overclock comparably, may also reach stock speeds slightly above the 3Ghz mark. This would be major boon for AMD.
This possibility is reinforced by recent reports that the first Phenom II processors will be rated at 2.8Ghz and 3.0Ghz.
On the Lookout
So far, it is impossible to say if the Phenom II can truly upset Intel’s Core 2 Quad line-up. Early results seem to suggest that Phenom II will only match the performance of the Core 2 Quad. However, AMD will likely price their offerings below Intel’s current Core 2 Quad products. This could put Intel’s Core 2 Quad in an uncomfortable position, pinned in by the better price of AMD’s processors and by the better performance of their own Core i7 architecture.
This article is the first in a series about the Phenom II processor. As more information becomes available concerning the Phenom II, look to this series for more in-depth information regarding architecture changes, performance-per-clock improvements, and pricing.
This post is part of the series: Phenom II Preview
- The Phenom II Part 1: Look to Shanghai
- The Phenom II Part 2: Pricing Preview
- The Phenom II Part 3: It Has Arrived
- The Phenom II Part 4: Platform Comparisons