What is Digg?143 Comments
- inactive, on 10/12/2007, -16/+75"Oh man I want one so badly"
oh come on, the article is rediculous, the conroe is intels next gen chip, which isn't going to be released for another 6 months, until we see AMD's offerings how can they have the performance crown back? can you call it unbiased when it says things like: "While we're still comparing to Socket-939 and only using RD480, it does seem very unlikely that AMD would be able to make up this much of a deficit with Socket-AM2 and RD580."
I MEAN COME ON, how are we supposed to respect andantech when they say crap like that? let me translate:
"we are comparing last gen amd stuff with next gen intel stuff, but we don't think AMD's next gen offerings will be able to compete either"
how the hell is that not biased? why not say: "intels set the bar high, can AMD pull it off ?"
i'm forced to ask did anyone recieve a kickback for this "article"? - geminitojanus, on 10/12/2007, -6/+29I can understand being upset, but AMD's next offering isn't expected to be 40% faster; a jump of more than 15% inside the same archetecture is almost unheard of. Keep in mind the AM2 chips aren't expected to be a major update to the AMD64 line. Support for DDR2 ram and a change in the heatsink/fan unit are the biggest publicized changes, and this next generation chip from AMD will be released at about the same time Conroe is.
Also keep in mind that they did overclock the chip they were running it against to try to emulate the difference the new chips would make. While not perfect, I believe they will probably be fairly accurate; Anand is a fairly smart, straight forward and relatively unbiased guy. But when running against these chips, the Conroe chip saw 10-40% gains over it.
I'm all for AMD doing well, but when Intel does well, we have to pay attention as well. Being a fan of both companies is the all-around better idea (since both companies competing against each other means lower prices for all). - BugMeNot2, on 10/12/2007, -5/+25Is the FX-60 worth $1000?
- geminitojanus, on 10/12/2007, -3/+22You know, you can preach it all you want, but until the ink dries on the benchmarks people just won't listen.
I've been evangelizing the PM marchetecure for years now, I've been calling in all my chips to Intel, to motherboard manufacturers, I've been writing letters to anyone that will open them saying "Get this damned product on the mainline, and do it before everyone buys up Pentium 4s and blows their budget".
The "Core" archetecure is lightyears ahead of where Intel has been, and they're still being very careful with it, as evidenced by the fact it's not shipping yet (but the truth is, if they're demonstrating the chip, the design probably taped out 6 months ago and production has hit at least 50%). These numbers have just affirmed what I thought ages ago when I bought my first laptop with the Banias processor.
This is Innovation. And now it's time for AMD to up the ante. - LaughingMan11, on 10/12/2007, -2/+17It's still a valid comparison because the FX-60, even though it's several months old, is widely considered the fastest desktop CPU that exists.
You've got to have a baseline of comparison, but they picked the FX-60 because it has been proven as a fast chip.
No one is saying that AMD isn't going to fire back. That's the fun in innovation and competition between two companies. This is not a knockout punch by Intel... this is just them reminding AMD and everybody else that they are still in this fight. This is good for everybody. - LaughingMan11, on 10/12/2007, -5/+19There's a very good reason for that... Intel simply doesn't make a chipset that is compatible with the same slower DDR ram that the AMD uses... and AMD doesn't support DDR2 memory on their on-die memory controllers just yet.
The purpose of the test is to show the fastest AMD chip you can get (overclocked even) with the fastest memory you can buy for that chip compared to the same fastest for Intel. If AMD doesn't support the faster memory, that's a flaw for AMD.
The benchmark here is full system performance, which includes RAM. - LaughingMan11, on 10/12/2007, -2/+15I think the point is that this isn't *just* a newer, better, processor that's incrementally better than what came before it... this IS the quantum leap in technology that you are seeing.
This Intel chip runs at 2.66 Ghz but spanks Intel's old chips running close to 4Ghz on the old Netburst architecture. That signals a massive sea change in Intel's microarchitecture design. - panique, on 10/12/2007, -10/+20I can't wait to buy my new Mac Tower later this year with a pair of Conroe's. That is going to be sweet.
- w0rd, on 10/12/2007, -5/+15To the other 2 children posts here: Considering that he said "Price to performance ratio is almost always better with AMD" see the "almost always" part? He didn't claim anything, he made a generalization.
- teamparadox, on 10/12/2007, -7/+15So they are benching a few month old AMD cpu against an Intel cpu that wont be out for a few months? Well how does that mean Intel took the crown? Who's to say AMD isnt going to fire back at the same time with something even better? I call BS on this.
- mikm, on 10/12/2007, -3/+11AMD has traditionally beat Intel (in gaming benchmarks) regardless of the memory speed difference.
- matt.rubin, on 10/12/2007, -0/+8another reason why h.264 sucks and we should all stick to divx it takes 20 min to convert from mpg for 2 hours on my 3200+ and ripping DVDs to divx is slightly faster than real time :)
Apple sux oncee again - TheKillDoctor, on 10/12/2007, -1/+9This is fantastic which ever way you want to cut it.
My money is on a nice competitive price reduction for the FX-60.
As long as the customer (me) can buy these chips cheaper, I'm all for competition.
That's why I'm chip agnostic.
Speed can come later... - FRAGaLOT, on 10/12/2007, -2/+9FX-60 is barely two months old.. not several.
- nebunezzar, on 10/12/2007, -2/+9I'd have to agree there. That's an awesome leap for Intel, but just how much is this new hotness gonna cost? At least now it's some real performance competition.
- SuperSloth, on 10/12/2007, -2/+9I want the best performance, or the best performance for a reasonable price. I don't care who made it.
Brand loyalty makes you look stupid. It was stupid when Intel fanboys touted the P4 after it became clear that AMD had better processors. It was stupid when nVidia fanboys touted the FX series when it became clear that the 9000 ATI series was better. It was stupid when ATI fanboys touted the X000 series when it became clear that the 7000 nVidia series was better. And it's stupid here, too. - Brennan, on 10/12/2007, -4/+11No acttually, it didn't. DDR2 667 isn't really faster then DDR400, especially at the timings they were running. The AMD machine most likely had more memory bandwidth available.
- Asher, on 10/12/2007, -0/+6I'm an AMD fanboy, I've always been an AMD fanboy, and I'm able to admit it. Wow though, I'm quite impressed, I'd like to think that AMD will be able to pull off a ~40% performance increase in the next sixth months, but I'm just not sure.
- LaughingMan11, on 10/12/2007, -1/+7It is not a lame comparison because AMD will not release a new microarchitecture until 2007. AMD doesn't have any plans to significantly change their chips this year except for adding faster memory and a new socket. They will still be 90nm all this year.
There will potentially be up to 6 months after this chip is released until AMD releases something else.
When AMD shows a prototype of their next generation chip, anandtech will compare that new AMD chip to Conroe, which will be out in the market by then... and you won't catch anyone calling it LAME. So why is it lame when Intel does it?
You need a baseline to do a comparison, and the FX-60 is a baseline because it's the fastest chip on the market you can buy. If you're not going to compare it with that, then what's your alternative? Make numbers up? just not do a comparison? - Woknblues, on 10/12/2007, -1/+7I just want performance at a cheap price. I want intel and amd to duke it out over and over again. I don't care who wins, I just want to pay less for my processor, and have it do what I want it to do. I don't understand why people are "pro-amd" or "pro-intel". Unless you own stock in the company, why would you care? The best value for your dollar should be your part time favorite until something else comes along. They made the greatest buggy whips known to man one year before the "horseless carriage" was invented. Who the hell cares about buggy whips now? I currently use AMD and am happy, but if Intel brings out something of better value to me, I will jump all over it.
- geminitojanus, on 10/12/2007, -1/+6Seeing as the entirity of the next generation from Intel was built around the Pentium M, which up until recently had the IPC/W crown (until the Core Duo came out), these chips will likely pump these kind of speeds while staying under roughly 60-80W (this is an assumption; the Core Duo's operating window is between 3W and 47W depending on the model, so it's safe to assume that Conroe will not go over double the maximum thermal production of the Core Duo).
The AMD64 chips run up to about 140W.
(for more information, check sandpile.org; they have the most complete non-NDA information on these chips available). - ZachPruckowski, on 10/12/2007, -2/+7E6700, a 2.67GHz 4MB cache - $529
E6600, a 2.40GHz CPU 4MB cache - $315
E6400 at 2.13GHz with 2MB cache - $240
E6300, at 1.86GHz, 2MB of cache - $210
Those are from http://www.theinquirer.net/?article=29504
They all have a 1066 MHz bus. - kkapoor, on 10/12/2007, -5/+10That's rubbish, Anandtech usually tells it like it is. Most of the time it has priased AMD to the detriment of Intel.
- geminitojanus, on 10/12/2007, -0/+5Conroe is a desktop chip. There's no way it's going in an Apple laptop with a TDP of 60-80W.
Expect instead one at the heart of the -Power-Ma.. err.. MacMacs? - odysseus, on 10/12/2007, -2/+7Intel's CTO actually said "Israeli engineers saved Intel’s butt with the Pentium M." [which Conroe/Woodcrest/Merom are based on.] - http://www.theinquirer.net/?article=30111
Thank you Israel! (Now I have something excellent to replace my P3 with.) - geminitojanus, on 10/12/2007, -3/+8"The problem here is that Intel and AMD take very different stances. Intel's "64" bit technology right now is just a 32 bit processor emulating 64 bits. AMD's path is a true 64 bit processor that can emulate 32 bits."
-1, Inaccurate. AMDs approach is to take a 32-bit processor and extend it to 64-bit by adding a few registers and 64-bit instructions to the core. Intel's approach is to take a 32-bit processor and extend it to 64-bit by adding a few registers and 64-bit instructions to the core. Both the AMD64 chips and the EM64T chips go about adding 64-bit instructions the same way, though the AMD64 implementation is much more mature (seeing as they invented the technology).
It seems like this argument is crossing the earlier argument against the Pentium D processors, which are simply 2 Pentium 4 processors stamped on the same die, with a bit of logic in between. AMDs implementation of dual core technology from the beginning was an integrated approach (and now Intel has caught up and surpassed AMDs approach with cache sharing and such). I love AMD, but it's time to up the ante. - 9mmCensor, on 10/12/2007, -0/+5Intel could buy AMD now. It however wont happen. There are two main CPU manufactures, when there is one, there will be a total monopoly.
- kemche, on 10/12/2007, -4/+9Sum up the Review from Anand...
"Going into IDF we expected to see a good showing from Conroe, but leaving IDF, well, now we just can't wait to have it."
KG - LaughingMan11, on 10/12/2007, -0/+5@ conorryan
They did mention power. http://anandtech.com/tradeshows/showdoc.aspx?i=2712
"Up to 40% Improvement in Performance, up to 40% Decrease in Power"
... compared to Pentium D
65W TDP Power, which is a FAR cry from the 100W TDP that the old Netburst designs used to hit. - weiran, on 10/12/2007, -6/+11How is this test conclusive at ALL? It was setup by Intel, not an independent reviewer. All they got was someone from the outside to do the actual testing. HOWEVER, do you trust all the figures that Apple and Microsoft give you about their performance? It's hard to believe there isn't any spin from Intel on this, even though the reviewer is from Anandtech.
My respect and opinion from Anandtech has dropped significantly now, especially with phrases such as "it does seem very unlikely that AMD would be able to make up this much of a deficit with Socket-AM2 and RD580" even though these technologies are around six months away!
Not until I see final products being tested independently side by side will I believe any of Intel's hype. - opnotic, on 10/12/2007, -5/+10Oh God. It just hit me. The reason there has been so much Intel Fanboi stuff lately is because of all the Mac Heads. Ugh. Here is a tip for the newbies joining the X86 crowd. Intel is sometimes up.. AMD is sometimes up. It's been a while (a long while) since Intel has been the one up... but I'm sure they will be up again.
- CatFood, on 10/12/2007, -1/+6The new Intel micro-architecture has a very short pipeline, and is actually comparable to AMD. That means lower frequencies, lower power consumption, and as the result lower heat dissipation. My guess is heat is very comparable to AMD.
- Arramol, on 10/12/2007, -3/+8I'll hold out on writing AMD off until I see their next-gen processor architecture. Still, interesting to see these results, even if Intel did set it up themselves.
- BassCadet, on 10/12/2007, -2/+6Where in the article did it say this Conroe chip will come out at around $700?
$700 is the domain of the FX and Extreme chips. Surely this new Core technology trickles down to the mainstream CPU market ($400 or less)? I've never paid more than ~$500 for a CPU because the market on premium CPU's is so volatile. - Brecourt_Manor, on 10/12/2007, -16/+20results are impressive, yet I will still stick with AMD for desktop systems.
Price to performance ratio is almost always better with AMD - saleens281, on 10/12/2007, -0/+4I just want to make this comment to everyone talking about anandtech being biased. PLEASE do a little investigation before making such well... outright ridiculous claims. Anandtech is the LEAST biased site I have ever had the pleasure of reading. Hell, all of the servers they run currently are opteron chips. They were one of the first to tout how well AMD was doing, and have stood by their side from the get go. If he says we'll be lucky to see a 15% gain from AM2 you can bet your bottom that he has good reason to say so. While I think his reviews can draw on a little much at times, if you want someone who is unbiased and a technical genius, that site would be the place to go.
Sorry for the public service announcement, it just needed to be said with some of the fanboi uneducated responses here. - paulchu, on 10/12/2007, -1/+5yea, conroe's not available for another 6 months, but that 40% increase in FPS means that 2.8GHz will have to be clocked at around 4GHz to catch up.... AM2 will have more memory available to it, but I doubt memory will make that big of a jumpin framerates or media encoding, not to mention that Conroe isn't fully optimized...
this is scary.... crazy scary....
time to sell your AMD stock! - bshankle, on 10/12/2007, -1/+5I bet H.264 will kick Conroe's ass.
Read more here: http://bruceshankle.blogspot.com/2006/03/h264-will-kick-your-cpus-ass.html
-Bruce - gotamd, on 10/12/2007, -0/+4I'm a big AMD fan (see username) but the Conroe seems to have smacked the FX-60 *really* badly. Ouch!
We'll just have to wait and see what's available from AMD when Intel finally releases the Conroe. - CaptMorgan09, on 10/12/2007, -0/+4Another thing that people should understand is that this is early silicon they are demonstrating. I'm sure there will be a couple more spins before final production units are out, which usually means a more stable and efficient design.
- geminitojanus, on 10/12/2007, -1/+4"Intel people set up the AMD machine overclocked? Really or is it underclocked?"
Well, Anand was allowed access to it to do their own benchmarks (check the "Update:" messages throughout the article), and they took pictures of the AMD box booting to insure they weren't cheating. A quote from Anand: "As far as we could tell, there was nothing fishy going on with the benchmarks or the install. Both systems were clean and used the latest versions of all of the drivers (the ATI graphics driver was modified to recognize the Conroe CPU but that driver was loaded on both AMD and Intel systems).
Intel told us to expect an average performance advantage of around 20% across all benchmarks, some will obviously be higher and some will be lower. Honestly it doesn't make sense for Intel to rig anything here since we'll be able to test it ourselves in a handful of months. We won't say it's impossible as anything can happen, but we couldn't find anything suspicious about the setups." - CaptMorgan09, on 10/12/2007, -1/+4But there are benchmarks out there comparing current Core Duo mobile chips against current AMD chips and the AMD chips aren't much faster. An extremely low watt mobile chip which was not far behind a desktop chip had to tell you that Intel was going to be coming out with something very impressive.
It's good to see that Intel finally figured out that it's not all about the Ghz like AMD has been saying the past 3-4 years. - iluvdrbonner, on 10/12/2007, -1/+4Did they measure the heat differences? I know this has always been an issue for Intel while not so much with the X2's. I am interested to see those benchmarks. Correct me if I just missed them though.
Willy - geminitojanus, on 10/12/2007, -1/+4That would be an inference you're drawing. The AM2 chips aren't likely to make up a 10-40% deficit, this is a fact. It's difficult for processors to grow that much within one core design. This is the nature of the industry. The main changes that are going into the AMD chips with the AM2 are a redesign of the heatsink mount and DDR2 support. The only problem is, memory is hardly a bottleneck to AMD processors; they've already got superior bussing and lower latency onboard memory controllers. What Conroe has on the AMD chip is raw power. Since the AM2 socket isn't likely to add any raw power, their statement holds weight.
However, AMD could fire back with something that's entirely not on the radar (as Intel did with Yonah).
Kinda funny how things make sense when you stop voting for companies out of your emotions and start applying logic. - scotticus, on 10/12/2007, -1/+4It's going to be two years before reasonable people can afford to buy this chip. We'll see what's available when that time comes. Until then, it's nice to see that performance keeps improving.
- sixister, on 10/12/2007, -3/+6My precious AMD-loving sentiments are being shattered :(
- Elranzer, on 10/12/2007, -0/+2"I wonder how a Conroe core will affect the battery life of a MacBook Pro...."
We're still wonmdering how the Core Duo affects the battery life... - gamefoo21, on 10/12/2007, -0/+2WoW... the Amdroids are bad on here...
Conroe is running on engineering class boards... the AMD is running on what almost everyone agree's is the best 939 board maker out there.
Yet I keep hearing fanboy *****. Intel has 65nm and they have it down, AMD hasn't even started the transition yet. They have fully functioning Conroe's and already running stable at 2.8ghz. Intel is going to slam Amd this time around. AM2 brings nothing to the market but support for DDR2... Intel still supports the 478 platform after all this time with the LGA. Conroe is going to be an LGA775 chip... The only reason it doesn't work on current boards is the PWM setups won't handle it.
If your going to flame Intel atleast do it with some intelligence and reason. - Herolint, on 10/12/2007, -6/+8Intel set this up themselves didn't they? Intel has a history of tweaking things to make themselves look better. If they can't do that, they reverse-tweak against AMD. I am friends with several ex-Intel engineers who have all admitted that while at Intel they worked on certain Windows DLLs to make them either pro-Intel or anti-AMD (or both if possible).
I think Intel is one of the lamest companies out there so I'm sticking with AMD. - lilrabbit129, on 10/12/2007, -1/+3That's the point. These are preview tests. No independent company can do them because the chip hasn't been released. You always take the benches with a grain of salt, but still they look promising.
-
Show 51 - 100 of 143 discussions
Digg is coming to a city (and computer) near you! Check out all the details on our 
© Digg Inc. 2009
— Content created and posted by Digg users is