Phenom 9700, AMD's 1st Quad-Core CPU

[Evilonigiri]

Indeed. Intel is not like the same company, they are better now. We must hope that translates into better computing for the dollar.

Better? They are the same as before! They still monopolize and sell expensive (extreme priced at $1470?) cpus. Better in technology is what you mean.

 

[techgeek]

Better? They are the same as before! They still monopolize and sell expensive (extreme priced at $1470?) cpus. Better in technology is what you mean.

Whoa whoa whoa, hold up there. How do you see that they are a monopoly? That is still trying to be decided in a US court. As far as I can see AMD is still selling processors and has been for sometime. If that's the case, I have a difficult time in seeing how Intel is a monopoly. Not only that but up until Q2 2006, AMD was selling a tremendous amount of A64's and Opterons. As for extreme priced CPU's, what was the FX line? I seem to remember them coming in over $1200. The performance leader gets to set whatever price they want for their very best. As long as there is someone out there that will buy them at that price, there will always be a CPU at that price. If Intel or AMD didn't sell a single processor at that price, there wouldn't be one at that price. It's a simple case of market driven supply and demand.

If you want to get into shady business practices, don't just look at Intel. I'm sure if you get digging around in any huge corporation (including AMD) you would find things that said corporation wouldn't want public.

One other thing to think about when discussing this anti-trust suit. The US government has to be really careful with what they do if they in fact find Intel guilty. Intel is a huge exporter. They contribute a huge amount towards the overall GDP of the US. Hurting Intel by fining them heavily or breaking them up could result in the equivalent of cutting their own nose off to spite their face. The US economy is hurting right now, this isn't the time to hurt a critical exporter. I also know that these things don't officially factor into the decision, but don't think for one minute that the US government isn't thinking about it anyway.

 

[onestar]

True. Better is an ambiguous term when the entire spectrum of operations is concerned. Intel is obviously better at producing processors that consumers want, even though we might ask ourselves at what eventual expense that will come to mean for said consumers.

 

[dustbunny]

I'm just amazed that in most benchmarks even Intel's first "double cheeseburger" quad-core schools Phenom in all it's native glory.

I find it ironic and highly amusing that AMD mocked Intel publicly with the "quad core for dummies" stunt.

 

[dmacfour]

I bought an AM2 motherboard just before the core 2 duo came out, and was alarmed at my poor choice. these super cheap processors give me a little hope.

 

[mactronix]

Stop Quoting the THG article.
It was clueless.

The 45nm parts will use LESS power than current Chips.
THG got their chip to 3.8Ghz and maxed out at 110w.
At 3.2Ghz, it stayed below 80watts.

Other Reviews show the Penryns hitting 4.0Ghz and still using minimal power.

Hmm Stop quoting the THG article. So how do you sugest i ask a question related to it if i dont let people know what im talking about ? just what are you sugesting we start using telepathy?
Did you actually read what i posted or did you just scan it and see the article mentioned and assume?
You quoted it as much as i did yourself.
Mactronix

 

[xrider]

I haven't seen a product launch this good since New Coke.

Now thats funny.How about apples 1984 launch.

 

[xrider]

I hope AMD pulls a rabbit out of it's hat next year, I certainly wouldn't want to pay intel prices without any good competition

WOW.Lets see how much AMD priced there cpu in there hay day 2005
Model Clock Speed L2 Cache Size Price (USD)
Athlon 64 X2 4200+ 2.2 GHz 512 KB $537
Athlon 64 X2 4400+ 2.2 GHz 1 MB $581
Athlon 64 X2 4600+ 2.4 GHz 512 KB $803
Athlon 64 X2 4800+ 2.4 GHz 1 MB $1001

So what do we get out of this when your on top you can charge a premium for your product. when you don't have the best you have to lower your price. This is just simple economics

 

[xrider]

.Phenom being 40% faster
.The true Quadcore (not glued together)
. Barcy and phenom being d.o.a
. Priceless

 

[Crank]

The reason that I will use the Penryn in my next gaming rig is because of its overclocking headroom. I can use my cooling systems from an older rig, so it won't be a problem. I just wish that I could get my hands on the custom drivers that XtremeSystems use to run SLI on Intel chipsets. All the same, cranking it up is more fun than running any kind of stock rig. It's like modding your bike or guitars.

 

[JerryC]

Does ANYBODY actually think thats necessary for AMD?

These chips (phenoms...or pheni perhaps?) have just been released, so early versions are seemingly pricey. The price will likely fall quite alot quite soon for the low end models weve already seen. The manufacturing isnt the greatest yet, but AMD need to put something on the market now to generate revenue and iron out manufacturing problems on the higher clocked parts as they go along.

Yeah they are slower than quad core intel chips on either 65 or 45 nm clock for clock, but if the price is right it doesnt mean squat basically. Dont forget unlocked black box editions are coming out with no premium attached, which will very likely be a big seller.

Also this is the first stepping. In time cache size might increase along with reduced heat output at higher clocks, and I should think cache latency might come down, as I have read its actually quite high. AMD have been selling record amounts of the old X2 chips, so I wouldnt jump to the conclusion that Phenom will be a flop just quite yet.

The problem isnt the price, its the performance. The Phenom doesnt even beat several of the existing AMD dual core offerings. Not to mention the fact that it gets completely owned by most everything in the current intel generation. Intel is starting to put out their new generation of chips and is already making plans for their switch to the next generation chips. AMD needs to get off their butt and do something to recapture market share. Putting out a CPU that nobody but AMD zealots will want isnt the answer.

They have to do something to trouncy on Intel. Maybe they should focus on being the first to market with an 8 or 12 core cpu.

 

[pogsnet]

What about this thread?

 

[dark41]

I waited to see benchmarks of Phenom before buying quads. I've just placed an order for a dozen Q6600s (not many as really no one really cares about quad cores yet, we sell 10x the amount of C2Ds). Whether running at stock speeds or overclocking, it's obvious that the Q6600 is a better buy at this point.

I pretty much expected Phenom to bomb when compared to C2Q, but not this badly. I was hoping it would at least be faster than existing AMD X2. A sad day indeed for AMD.

 

[Atolsammeek]

One thing I did see is Amd not working on Being a leader. It must make people who are so gunho on Amd really Mad. But I see it as this. Go For the leader not the Follower. It dont matter if it AMD or INTEL.

If you rate them as in Cars. AMD is a Saturn cheap and reliable.

Intel is a sports car Like Porsh and Ford Mustangs. Fast and high cost.

They both have there pros and Cons. But we all have to deal with it.

 

[DaDane]

AMD the better platform choice!? = What a load of crap (intel is atm surperior in every way, check out the specs of x38)

Intel processors will also run on the old platform (P35 - X38), and on the new one.. Quad-core Intel runs even on my 2 year old Asus mobo just with the latest bios...

AMD increased the number of instructions pr. clock but it is even slower than their own former model Athlon 64 x2

Currently AMD'S top model Quad core dosent even get near Intel's lowest quad model = Seems to me like this is the final breath of AMD

Intel curently rocks the CPU market, cant figure out why AMD even bother launching such a crappy slow CPU which no one can use and no one will ever by
(unless you are really really stupid..)

Just an opinion.... ;-)

 

[Reynod]

This is funny the Intel fanbois are upset that "Tom" wrote an article that seems pro-AMD.

I bagged the hell out of Toms a while back saying they were pro-Intel ... then I did some research ... and felt pretty stupid.

Do you feel just as stupid now??

People write the articles ... Tom left a while ago. There is no "Tom" who controls the site via a magic wand now.

Toms is good because you get a fair bit of free reign ... providing you don't go psycho.

I also appreciate the interesting comments ... you know ... the technical ones that seem pretty absent right now??

They are few and far between at present with all of the flaming, fanboism, and "p3ssing" competitions going on.

Anyone like to comment about the L3 cache which is runnng a bit slow, and has latency issues ... getting choked as a consequence ... or maybe something else interesting?

Something the B3 spin is likely to address?

Anyone with a brain feel free to comment ... or is this one another thread gone down in flames?

 

[cnumartyr]

Sure. I'd love the B3 to be great. Unfortunately looking at some thermals that were posted it seems that the current leakage is insane above 2.8 GHz.. Let's get some phase change so we can break 3.0 GHz! (Exagerration.. because apparently you take everything on this board literally).

There is no Tom? THE WORLD IS ENDING!

Here's a technical comment. The Penryn is faster than the B2 clock for clock. Will probably still be faster than the B3 clock for clock. The Q9300 will retail in the high 200s to low 300s and will have a 2.5 GHz clock speed. Will AMD still be competitive?

 

[turpit]

This is funny the Intel fanbois are upset that "Tom" wrote an article that seems pro-AMD.

I bagged the hell out of Toms a while back saying they were pro-Intel ... then I did some research ... and felt pretty stupid.

Do you feel just as stupid now??

People write the articles ... Tom left a while ago. There is no "Tom" who controls the site via a magic wand now.

Toms is good because you get a fair bit of free reign ... providing you don't go psycho.

I also appreciate the interesting comments ... you know ... the technical ones that seem pretty absent right now??

They are few and far between at present with all of the flaming, fanboism, and "p3ssing" competitions going on.

Anyone like to comment about the L3 cache which is runnng a bit slow, and has latency issues ... getting choked as a consequence ... or maybe something else interesting?

Something the B3 spin is likely to address?

Anyone with a brain feel free to comment ... or is this one another thread gone down in flames?

Good on you Reynod....finally, someone who made it beyond just walking in and reading a ew threads and jumping to an assumption. THG has always been biased towards performance or value, but not a manufacturer. Had anyone bothered reading Pats post, they would have noted his explanation of the test team being in a rush to get the review done. With all the NDA's AMD laid out and the controled testing, can anyone really blame them? Yes, their prices were off, and their conclusions were based on that, not any bias towards AMD. As a result their conclusion appeared biased, however, consider that had that pricing that they used been accurate, their conclusions would also have been accurate. It was a error, and it is being worked on to be rectified/prevented in the future

And yes, THG forums is very open. I wish people would appreciate that freedom, unlike certain other forums where you must share the same opinions to post. Here, you can post your opinions without fear of having them deleted, or being banned, as long as your dont cross a few lines...advertising, lying, or excessive personal attacks...thats it...you can be a total flaming fanboy..just so long as you avoid being a total flaming arse.

 

[jimmysmitty]

Reynod, I look at it this way. AMD had 1.5 years to get Phenom up and running and 1.5 years is pleanty of time to stop talking and showing and fixing the L3 cache error. The Vice president was talking about how the L3 cache was so great. This was a year ago but obviously something happened between then and now.

The B3 stepping might help the L3 error but I doubt it is going to drop their thermals to compete with Yorkfeild and we have only seen what the top of the line QX9650 can do in terms of efficiency compared to the Phenom. Once the low end Q series comes out in January we will be able to see what a Yorkfeild CPU in the same price range will do.

Now for some technical aspects of it. The Q9650 that review sites had are C0 steppings. By the time the lower end Q series does come out we might be at the same B3 that the Q6600 Kentsfeild was at and then a few months later release a G0 like stepping.

We have to also remember that while AMD gets Phenom to a B3 stepping Intel will also be working on that too. So unless the B3 stepping drops the thermals drastically then its just going to help the L3 problem but not get them to the 3GHz arena although they had one running on a small stock fan in a youtube video.

 

[yomamafor1]

Anyone like to comment about the L3 cache which is runnng a bit slow, and has latency issues ... getting choked as a consequence ... or maybe something else interesting?

Something the B3 spin is likely to address?

Anyone with a brain feel free to comment ... or is this one another thread gone down in flames?

In my humblest opinion, I think what we're seeing in Barcelona's L3 is a result of cache misses. Theoretically speaking, the smaller the cache, the higher chance cache misses will occur. K10's L3 cache is surprisingly small for it to be shared across four cores. With only 2Mb, each core gets roughly about 0.5Mb of L3. Although L3 can be dynamically adjusted, it is still too small. In Core's architecture, each die roughly gets about 0.5~6Mb of cache, and it is only shared across two cores. The work load of adjusting dynamically is a lot lighter than Barcelona's, where a cache needs to be dynamically adjusted to fit 4 cores.

In the event of cache misses, the core must wipe the affected area clean, and reload data from the main memory. This takes processing times. This is probably an explanation for Barcelona's latency.

Also, this is AMD's first attempt at shared cache. For Intel, the first generation of shared cache was already implemented back in Yonah. So Intel had a lot of experience compared to AMD. Therefore Intel's Core 2 does not suffer from excessive cache misses. Oppositely, not only this is AMD's first attempt, AMD also implemented it across four cores. This is probably one of the reasons for L3's lackluster performance.

I've been saying this for a while. From a "elegance" standpoint, Barcelona is nothing short of beauty. Compared to Nehelam's die shot, Barcelona is indeed more aesthetically pleasing. But from a design standpoint, Barcelona is nothing short of a disaster.

Now, will B3 fix this? No. The problems I described above are considered major design issues, and AMD cannot fix this with only a revision or two. The entire cache system may need to be redesigned to alleviate latency penalties. I've seen David Kanter from RWT talking about 8Mb of L3, but I've yet to seen it on AMD's roadmap. Increasing L3's size might alleviate the problem too.

As for the TLB errata, it has been rumored that it is more of a symptom, rather than a cause (which I'm more inclined to agree). In other words, this errata is a problem regarding manufacturing, rather than design. If it was a design issue, all K10 would be affected. But since only the higher clocked ones are affected (2.4Ghz +), it is more likely a physical errata. IMO, this is probably because Barcelona was manufactured on a process that's unfit for it. Just like Intel's Prescott, where the design simply overwhelmed the process technology at that time, AMD simply did not have the right process to manufacture such a big, and sophisticated die.

I'll scope out some links, and update this later.

 

[TechnologyCoordinator]

I don't think L3 would have to be divided like that, right? Let's say two different cores need access to the same cached information, wouldn't it just be cached once in L3?

 

[yomamafor1]

I don't think L3 would have to be divided like that, right? Let's say two different cores need access to the same cached information, wouldn't it just be cached once in L3?

It really depends on how AMD written their driver. If both cores need to access the same information, then yes, they would go to the same place. But for example, in Cinebench, where four cores render a different part of the image, the cache then needs to be divided to accommodate it.

Unfortunately, so far I don't have any technical documents to back this up, so consider this my speculation.

 

[Vitruvius]

So can someone tell me what happened to the OctoCore promise? Boards like the Asus L1N64 SLI WS were supposed to enable two AMD Quad cores to run together (4X4) without Opterons or server software. Like this review; http://www.goodgearguide.com.au/index.php/taxid;1776357743;pid;3740;pt;1

But these are not socket F processors. So what is going on? If Intel (via ASUS) puts two Quad cores (not Xeon) on one Mobo soon then AMD will have lost that race as well. What the heck?

 

[TechnologyCoordinator]

So can someone tell me what happened to the OctoCore promise? Boards like the Asus L1N64 SLI WS were supposed to enable two AMD Quad cores to run together (4X4) without Opterons or server software. Like this review; http://www.goodgearguide.com.au/index.php/taxid;1776357743;pid;3740;pt;1

But these are not socket F processors. So what is going on? If Intel (via ASUS) puts two Quad cores (not Xeon) on one Mobo soon then AMD will have lost that race as well. What the heck?

Good point! I remember all this "clear upgrade path" BS. Quad FX was nothing but a joke. Not that many people have use for 8 cores. If you need 8 cores, you probably are doing server stuff, so suck it up and get server level hardware with the ECC RAM.

 

[epsilon84]

Good point! I remember all this "clear upgrade path" BS. Quad FX was nothing but a joke. Not that many people have use for 8 cores. If you need 8 cores, you probably are doing server stuff, so suck it up and get server level hardware with the ECC RAM.

Have you actually come across anyone with a Quad FX setup? Seriously. I have honestly not seen ANYONE claim to have one running, which is rather odd since there are a lot of AMD fanboys out there. Baron pulling out after making all those promises about getting Quad FX tells you something, doesn't it...