AMD's Phenom Unveiled: A Somber Farewell to K8
by Anand Lal Shimpi on November 19, 2007 1:25 AM EST- Posted in
- CPUs
It's surreal isn't it? Is this how you pictured it? With forty-three days left in the year, AMD is finally letting us publish benchmarks of its long awaited Phenom microprocessor. The successor to K8, AMD's most successful micro-architecture to date, and the cornerstone of AMD's desktop microprocessor business for 2008: Phenom is here.
But shouldn't there be fireworks? Where's the catchy title? The Star Wars references were bound to continue right? Why were there no benchmarks before today, why are the next several pages going to be such a surprise?
AMD had been doing such a great job of opening the kimono as its employees liked to say, giving us a great amount of detail on Barcelona, Phenom and even the company's plans for 2008 - 2009. The closer we got to Phenom's official launch however, the quieter AMD got.
We were beginning to worry, and for a while there it seemed like Phenom wouldn't even come out this year. At the last minute, plans solidified, and we received our first Socket-AM2+ motherboard, with our first official Phenom sample. What a beautiful sight it was:
These chips are launching today, with availability promised by the end of the week. Phenom today is going to be all quad-core only, you'll see dual and triple-core parts in 2008 but for now this is what we get.
The architecture remains mostly unchanged from what we've reported on in the past. This is an evolutionary upgrade to K8 and we've already dedicated many pages to explaining exactly what's new. If you need a refresher, we suggest heading back to our older articles on the topic.
The Long Road to Phenom
Ever wonder why we didn't have an early look at Phenom like we did for every Core 2 processor before the embargo lifted? Not only are CPUs scarce, but AMD itself didn't really know what would be launching until the last moment.
At first Phenom was going to launch at either 2.8GHz or 2.6GHz; then we got word that it would be either 2.6GHz or 2.4GHz. A week ago the story was 2.4GHz and lower, then a few days ago we got the final launch frequencies: 2.2GHz and 2.3GHz.
Then there's the pricing; at 2.2GHz the Phenom 9500 will set you back $251, and at 2.3GHz you'd have to part with $283 (that extra 100MHz is pricey but tastes oh so good).
The problem is, and I hate to ruin the surprise here, Phenom isn't faster than Intel's Core 2 Quad clock for clock. In other words, a 2.3GHz Phenom 9600 will set you back at least $283 and it's slower than a 2.4Ghz Core 2 Quad Q6600, which will only cost you $269. And you were wondering why this review wasn't called The Return of the Jedi.
AMD couldn't simply get enough quantities of the Phenom at 2.4GHz to have a sizable launch this year (not to mention a late discovery of a TLB error in the chips), and the company was committed to delivering Phenom before the holiday buying season as these are tough times and simply waiting to introduce its first quad-core desktop parts was just not an option. Rather than paper launch a 2.4GHz part, AMD chose to go with more modest frequencies, promising faster, more competitive chips in Q1 2008. It's not the best PR story in the world, but it's the honest truth.
Two more quad-core Phenoms will come out in Q1: the 9900 and 9700, clocked at 2.6GHz and 2.4GHz respectively. The Phenom 9900 will be priced below $350 while the 9700 will be a sub-$300 part. As you can probably guess, the introduction of those two will push down the pricing of the 9600 and 9500, which will help Phenom be a bit more competitive.
It's worth mentioning that in the 11th hour AMD decided to introduce a multiplier-unlocked version of the Phenom 9600 sometime this year that will be priced at the same $283 mark. Whether or not it's called a Black Edition is yet to be determined.
Facebook
Twitter
RSS
124 Comments
View All Comments
Hellrazor0628 - Thursday, January 3, 2008 - link
Well I think that intel got the processing but i realy dont think at the first place that the phenom is ready they need time and money to be able to get rev. in good working and debuged intel had that money and time they realy took their time befor shoing up with core2 wonld say tow years almost. Amd had already a small part of the market enven whene they give the best performance for the price even compaired to the best intel. To say it I realy was hoping the truth native quad core phenom would be better proccesing too but in ther other hand there are a lot of technogie that need to be looking at about amd that poeple sould take a look at about the phenom that is key to all amd cpu that people are too stupid to look at and understand. like power comp. wtf man there is a bus and a memory controler my nvidie chip set coul burn and egg and it only have to run the pci and pci-x.hoelder - Wednesday, December 12, 2007 - link
I remember to have to save food of my mouth to buy the first ill conceived Pentium or the 486. How Intel set those prices evades me. They maximized their profits with no competition. Yes I know Intel produces a faster chip, if faster is the right word. However, when it comes down to competition and the markets, AMD is the strategically right choice. Unless of course you think that the Walmart buy cheap toys from China idea is your consumer ideology then you should stick with Intel, that actually is going like Nixon to China or you believe in the consumer choices should keep competition open.mpholland - Friday, November 30, 2007 - link
Maybe AMD just had to release these early to make a little capital this year. Personally I think that it is good that AMD let people know SOMETHING is close. I am hoping that with a little tweaking AMD and some MB partners can get performance up a little and be competitive with Intel sometime in 2008. I have seen just simple driver tweaks work wonders on other hardware, maybe just a little more time can help.Clauzii - Thursday, November 22, 2007 - link
--Clauzii - Thursday, November 22, 2007 - link
By reading through the benchmarks, where a single core (of a quad) is compared to all four cores running, it looks like the 8 core version of the "Phenom" would scale even better than the four-core one. The Barcelona btw. also shows this behavior, by having one core being just a little faster than the Opteron core, but the four running in tandem scales very well.Because of the not-so-good MHz numbers, it might not take AMD to new glorious heights for now, but when (soon?) 8 cores arive, AMD MIGHT be able to do better, because of their better core-scaling factor.
Looks like they HAVE to do something like gluing together two Phenoms to at least do 8-cores before intel, and before intel gets TOO fast for AMDs liking and the ability to catch up slips away.
Unless AMD already is working on a true 8-core design, which would probably scale even better than a glued one. And by incooperating knowledge on multithreading from ATIs designing of GPUs they might be able to do something even more serious in the future.
But for know, intel is still in the lead.
praeses - Thursday, November 22, 2007 - link
I really wish that AMD never went down the road of L3 cache for these processors. As the majority of applications still used today in the desktop/workstation market are going to be only one or two cores, the shared cache itself probably causes more of a hindrance.Personally I would have liked to see 128k L1 and 1MB of L2 for the higher models, and simply the 512K L2 for the lower models. The tweaks to the individual cores would almost enable them to catch up clock per clock with Intel without this L3 cache latency getting in the way, and that way powering down the individual cores would also power down all the cache they would be using as well. I realize that routing the L2 cache in larger quantities is trickier and consumes more die space than L3 but they should also be able to gain significantly cost measures in those produced without L3 and be able to compete better in the $180 or so market.
Granted if a single application was single threaded and the only one taxing the system while taking advantage of all the L3 at once, and the other 3 cores were sleeping, it would be a slight disadvatage, but that's an extreme situation.
WorkIsAFullTimeHobby - Thursday, November 22, 2007 - link
I think Anandtech power consumption graphs are way off. Phenom power consumption sould be compared to Penryn power consumption plus NSB power consumption. Does any body see any mention of this fact and do the graphs properly account for this?Phenom effectively has the north side memory controller bus built in. After looking at the Architecture now I know why Intel is always trying to increase MB bus speed. They only have one external bus to feed and comminicate between all the CPU's.
redzo - Wednesday, November 21, 2007 - link
After all of this being said i have only a few words for you:- AMD's faith it's in the hands of its ability to cut down the prices even more.
- Cheaper by 13% it is just not enough ! ! ! unless they cut down the prices even more they'll loose more customers!
- Of course that they can trust in their marketing strategy( true quad core ), but not for long: It's PERFORMANCE and PRICE/PERFORMANCE that matters and not STYLE or FASHION.
jwizmo - Wednesday, November 21, 2007 - link
One thing to keep in mind here is that the Quad-Core AMD chip is a 64-Bit part, like it's predecessor. To perform these tests without giving it a chance in 64-Bit mode is not a fair comparison. I would like to see some subset of these same tests with Windows Vista Ultimate 64-Bit. The drivers should be out there for that configuration. Let's see how it performs in that mode, I don't think there will be much competition from Intel then. When everything finally goes 64-Bit AMD will have the advantage, since they got in at the ground floor.newuser2 - Wednesday, November 21, 2007 - link
I was checking holidays buyers guide and then I noticed that in the tests you were using for intel a motherboard that costs $335 and DDR3 memory which is $580 (I think I must be wrong here), while for AMD you used a $180 board and DDR2 memory for $121. That would be comparing a $915 platform vs $301 one, am I correct? Don't misunderstand me, I just think I didn't understand what you used to test what.