X-bit labs

Started by: Mr. BonBon | Date 08/17/07

Comments: 47 | Last Comment:  11/23/07

[1-20 | 21-30]

1. A lot of hype sorrounds the AMD products recently. Barcelona will be a big dissapointment thats for sure. Just like their 2900XT.

AMD is teh new SUCK.
VIVA LA INTEL VIVA LA NVIDIA

2. Mr. BonBon has always been teh suck. Word.

3. This article only shows that the real processors are done by AMD.

Really? Why?
Because Core architecture is based on PIII? Or because in reality their quad core is just two dual cores slapped together?
What I'm trying to say is... WHO CARES?
I don't care if its based on Pentium 1 or slapped together with duck tape, as long as it has excellent performance, price, and power consumption and destroys anything AMD comes up with, we're good.

we're good?!

Are you Intel?

You removed all the words from my mouth without me saying anything, very GOOD!!!

You like technology that only needs one or two engineers to duct tape one thing and sell one for the price of two and say its really cheap… fine with me.

I prefer innovation, technologies advance, complex designs and state-of-the-art architecture.

Core architecture is based on an old simple design, which even today works very well through process tweaks, delivering unprecedented performance, unmatched power efficiency at an unbeatable price.

Go ahead a pickup a 3.0 GHz dual core Intel chip that destroys anything AMD has to offer for $280. If that ain't cheap, I dunno what is.
At this point, AMD can only hope to match Intel clock for clock, and even if they do, they won't be able to compete in terms of raw MHz. Core processors, upon their release, were able to cruise in the 4.0GHz range with proper cooling. On the other hand, Barcelona is to be launched at 1.9-2.2GHz? Good luck.

Well it's better to have a "duct taped" chip that can actually be purchased cheap and works fine instead of your "I prefer innovation, technologies advance, complex designs and state-of-the-art architecture" that can only be seen on paper or in internet techsite editorials.

Until AMD releases the actual product and the benchmarks are published, these are all just a bunch of mumbo jumbo that only looks good on paper.

Benchmarks?!
Is it all about speed?
Did you read the article?
Did you see any benchmark in the article?
Did you see any performance test in the article?

maybe you should learn how to read?!? or you can't understand mumbo jumbo?!?

>>AMD has elected not to supply samples of this product to reviewers, and, well, you can probably guess why that might be. Why trigger a wave of reviews showing the competition to be faster?

>>>Benchmarks?!
Is it all about speed?
Did you read the article?
Did you see any benchmark in the article?
Did you see any performance test in the article?

State of the art architecture?

Do you mean 128-bit SSE units done by Merom first? Dedicated stack engine introduced by Banias? Load ahead of stores also in Merom first? Indirect branch prediction introduced by Banias? Merom's still better prefetching and caching?

These "innovations" were all done by Intel first. Just like x64 was done by AMD first. They each borrow/copy/steal off each because that is just the direction that x86 is heading. I'm not an expert, but other "innovations" on both sides are more architecture specific so are easier to implement on one side like macro-op fusion for Intel or dual 128-bit reads for AMD so are hardly a key end all advantage of one over the other. Things like unaligned loading are definitely innovative for AMD, while other things like SSE4a will depend on adoption. Both Intel and AMD are making changes to their virtualization implementation with AMD in K10 and Intel already making improvements in the latest G0 Core 2 stepping and additional changes with Penryn.

I don't think it's fair to dismiss either AMD or Intel of incapable of making "real processors". Even in the case of dual die processors, AMD's 4x4 solution is more or less elegent depending on your point of view and it's already been reported that AMD will produce a dual die 8-core chip joined by HT links instead of an FSB.

4. Clock to clock, and dollar to dollar will be good enough for some.

5. Oh my, bashing AMD seems to be the new trend lately.Something Mr. BonBon is doing here. But you know, assuming something will suck just because you heard about that somewhere is pretty stupid... time will tell but I am sure K10 will not suck. Maybe it will not overtake Penryn, maybe it will blow it away, lets wait and see.

6. Hm. I always thought each core would have been able to adjust its V-core independently. So I was wrong.
p.s. The conclusions are the only part of the article I read. :|

Mr.B's dad, thanks for the laugh :D

7. I wonder what two scenarios fastpunk conveniently forgot to mention, take off your green tinted glasses.

Can't wait to see some benchmarks going to head to head with Intel.

8. contents aside, i was kind of amazed by the poor quality of the writing, missing words, poorly formulated sentences.... some parts almost look like someone took a chinese article and put it through a translation engine

as for the contents, its nice to see some insight into what AMD is doing with barcelona (IT IS NOT K10!!), but untill the benches fall, all this means pretty much nothing

9. 2006...

Where AMD used to have two design sites, now the sites total five – two in Austin, one in Sunnyvale, one in Boston and one in India, said Chuck Moore, AMD senior fellow, who delivered the opening keynote address at the In-Stat Spring Processor Forum here Tuesday [5/17/2006].

“We do have multiple generations under design now,” he said.

AMD said the new architecture would be available in the first half of 2007.



2007...

" AMD promises to introduce its new quad-core processors with K10 micro-architecture in the end of August – beginning of September this year. "

10. And what two scenarios have I so conveniently forgot to mention Brendan?

11. That green tint is darker then I imagined fastpunk, maybe it won't overtake Penryn and or it will be blown away by Penryn.

Agreed it just drives me crazy all the fanboy that post on this site. Just to let you Intel fanboys know if your beloved company manages kills CPU prices will go up. I am really hoping for AMD's survival sake that K10 not only be able to beat Conroe but also match or beat Penryn.

It is good for us consumers when there is competition between different tech companies. Oh and fastpunk you are just as big a fanboy as Mr. BonBon. I am 100% certain that if AMD manages to destroy Intel then they too would jack up their prices. After all why would either company sell their products at great prices if they had no competition?

12. Hell yeah. Maybe will be the no 1 again:D.

13. Joz says:

this thing looks like its gona suck for overclockers ane other hardcore people. imo ,

1. the way it looks like k10 clocks each core up/down is gona make overclocking a bitch, unless it can be disabled.

2. it looks like the branch prediction is gona be low...not a good thing. I think they should work on that more then dynamic core clocking.

3. I have concluded, that the K10, may be able to hold out IPC wise agianst the current Conore...but I doubt its gona be the killer that AMD hopes for. if im wrong, im wrong, but AMD, either get it into the market and 3rd party testers, or I suggest you work on improving the K8 like Intel improved the on the Pentium III to make Core.

AMD sucks no matter what they say about K10.

14. Brendan, mate. Read my post again, but slowly this time so you can think about what you are reading. I said: "Maybe it will NOT overtake Penryn, maybe it will blow it away, lets wait and see." So that means that K10 could be better than Penryn, or it could be worst. Saying that makes me a fanboy? Maybe your glasses are a bit too dark...

I'm sorry I called you a fanboy I failed to read your post correctly. :{ What Stubborn Hud wrote fills me with dread. If what he says turns out to be true how long would AMD last before going under?

15. wow i've never seen so much FUD and outright bs posted after an article. Its sad that some of you actually believe the crap you are posting.

16. Why does it strangely feels like R600 launch all over again?

Late to the party and no benchmark leaks, problems with ramping up chips with high enough clock to beat the competition, and resorting to launch a product that cannot beat the competitiions top performer instantly

Yea I have seen that picture of that rig AMD showed off with what they claim was a barcelona quad core but unless they have those kinda chips in volumes its not gonna help them keeping people away from buying intel quads that are alreary running at high clocks/can be overclocked over 3GHZ.

Remember back when Conroe was about to launch? Intel leaked benchies of their new design all over the net. Yes that is right when you are about to have a winner on the market you do that kind of thing to win over enthusiasts and keep them from buying your comptetion's product, specially when you have been beaten down by the competition for a good while and need to win the customers'/enthusiaists' trust back. Kinda like what Intel had to put their P4's end of life bad reputation behind them and like how now AMD needs to do the same to win over potential custs.

Dont you find it strange that AMD is not leaking barcelona's great performance numbers even when they are awfully close to the release date? At least for the equally clocked Barcelona vs Core 2 Qs?

To me seems like dejavu. Exactly like R600. So much hope and high expectations but in the end it couldnt score "a sound and clear win" although much was promised and much trust everyone had because of the architectural achievments its design had.

But I am not certainly hoping for something like that to happen. Last chance AMD dont waste it. Though I guess not much can be changed at this stage before the initial launch.

You must have been sleeping under a rock. AMD released benchmarks of a 2.33ghz Barcy against a 2.66ghz Clovertown. And it beat clovertown solidly as well. Google is your friend.

PS: I'm sorry that you had to write a 300 word long pile of crap.

LOL! PWNED!1!

Yes...->AMD

why did 90% of my post vainsh?

I know about that benchmark, but that was retracted by AMD themselves cuz it was inaccurate and misleading that is why I didnt mention it. Guess youre the one living under a rock and talking crap now. lol

Yes you heard it right AMD themselves accepted that the benchmark was misleading and innacurate and took it down. So there is still no real benchies around.

"AMD gave us some simulated benchmarks that showed the 2.6 GHz Quad-Core AMD Opteron smashing the Intel Xeon 5355 to bits in both SPECint and SPECfp.

Unfortunately (for AMD), that was quickly debunked by George Ou of ZDNet as misleading and deceptive. You can read more about it here and here. The issue drew a lot of outrage in the online tech community that saw it as an attempt by AMD to hoodwink the press in less-developed countries. After all, AMD had already retracted the results 10 days prior to the press briefing. If that was their intent, then the borderless Internet sure put paid to those aspirations."

"AMD has requested that we remove the benchmark results as they are outdated and inaccurate. They have promised us updated results soon, so keep an eye out for these results. With the launch looming so close, we think they will be doing everything they can to have the results out ASAP." (no benchmark results after that)

http://techarp.com/showarticle.aspx?artno=429&pgno=0
http://www.techarp.com/showarticle.aspx?artno=424&pgno=6
http://blogs.zdnet.com/Ou/?p=567

Good Day!

That and AMD has waited so long the X5355 isn't the top Xeon DP anymore, the 3GHz X5365 is. Plus not long after Barcelona launches for Opteron, Intel will be releasing Penryn based Xeon DPs up to 3.16GHz in addition to the new Stoakley chipset which they promis will alleviate some of the memory bandwidth and latency shown with the Benseley FB-DIMM implementation. And Intel has already demoed a 3.2GHz 1600MHz FSB Penryn based quad core for Xeon DP that is waiting in the wings depending on how good Barcelona really is.

17. Re:Myself: Post vanished or something >_>

Yes...->AMD

18. Re:Myself:twice> my post keeps vanishing..
Yes...->AMD, AMD released their own benchmarks, THOSE MEAN NOTHING. Remember Intel's hype over its Core 2's? Every one was skeptical untill 3rd parties proved that the Athlon X2 had to be 400-600mhz faster to equal the Core 2? And then showed how overclockable the Core 2's where and ten days latter people where shooting up all over the place with 3+ghz'ed E6600's and 3.5ghz X6800's and other overclocks. And since almost any Core 2 can hit 3+ghz on air.

When AMD can get k10 to 3rd parties who can prove AMD's own scores, THEN I will pay some attention.

19. Fastpunk, my bad I retard. Lions of Zion I just want some trusted review sites to get their hands on AMD's latest and greatest and compare it to Intel best whomever wins is the champ for a few months .

20. Great Article.. Keep up the good work..

JIGAR

[1-20 | 21-30]