AMD posts Barcelona benchmark numbers

[pxc]

No, Barcelona does not use Z-RAM.

CPU development cycles are years long.

 

[RamonGTP]

Show me where I've been proved wrong. Quote it. Please. Copy and paste.... If you can do that you get a cookie

I said AMD has been in exactly this same position before and provided examples in the past where they were. The counter argument is that Anand says it aint so... OK......

You provided examples based on YOUR memory and YOUR conspiracy theories, not real examples and definitely not proof. If you're going to demand proof from someone, you should hold yourself to those same standards.

 

[LstBrunnenG]

Regardless, the statement is false as shown above by many different sites and, frankly, my own experience with the K7. Do I get a cookie now?

Even though Duby may never concede defeat, you win all my cookies.

 

[duby229]

The benches just dont show what you conclusion is. You conclude that the K7 took the crown on launch but the benches show that it was back and forth with the P3. The only benches with a clear lead are 3dnow tests. Which in the real worlld means absolutely nothing.... Zip zero... nothing.... On all of those sites you linked to the data just doesnt support your conclusions...

I'll stick to what I said. AMD has never taken a clear lead on a launch product. Not with the K6, otr the K7 or the K8. They all faired well, and did take the lead later on, but not on the launch product.

Besides your making an argument that doesnt matter anyway. Your looking at paid reviews. These companies were --PAID-- to make the review they made. Same as all other reviews. You see the thing yopu all know just as much as I do is that if a site gives a bad review on a product they will likely --never-- get to review products from that companie again. You all know thta just as well as I do, and I shouldnt have to say anything about it... It's common knowledge. The fact thta you refuse to accommodate this is just silly.

The thing you --should-- be looking at is what enthusiasts, like us were saying at that time. Start looking in the forums. Look at Anands forums from those days. Look at Toms forums from those days. Look at Hard's forums from those daysand you'll see that people are saying --now-- the --exact same things they were saying then-- No difference what-so-ever. Things are exactly the same they were....

 

[dexvx]

Besides your making an argument that doesnt matter anyway. Your looking at paid reviews. These companies were --PAID-- to make the review they made. Same as all other reviews. You see the thing yopu all know just as much as I do is that if a site gives a bad review on a product they will likely --never-- get to review products from that companie again.

The problem is that your relying on Anandtech to prove your opinions. 90% of Anands reviews were paid for and have an expected outcome. Very few data released by Anand can be viewed as trustworthy, or accurate. Things just didnt happen the way Anand says they did. Anand has been an AMD fanboy untill about 2005, when he screwed AMD on some marketing gimmick. AMD stopped paying him, and he is now an Intel fanboy.

This is all common knowledge though. He always artificially favored AMD untill about 2005, and and now he is artificially favoring Intel.

AMD X2-3800+ (August 2005) Review
http://anandtech.com/showdoc.aspx?i=2484&p=13

Pentium 955XE (December 30, 2005) Review
http://anandtech.com/showdoc.aspx?i=2658&p=14

Pentium 965XE (March, 2006) Review
http://anandtech.com/showdoc.aspx?i=2725&p=10

Notable Quotes from Anand:

We really didn't want to see AMD become a more expensive CPU manufacturer, and with the X2 3800+, we finally have a more sensibly priced dual core option. The choice is clear - the Athlon 64 X2 3800+ is better in every way than the Pentium D 830. For Intel's sake in the enthusiast community, Conroe had better be very competitive next year - because ever since Prescott, the Pentium 4 has been an utter disappointment.

If you absolutely must upgrade today, the Athlon 64 X2 is still probably going to be a better bang for your buck...

It is pretty much a toss-up at this point, but we'd recommend sticking with AMD for now and re-evaluating Intel's offerings when Conroe arrives. If all goes well, we will have a cooler running, faster processor with Conroe that may provide some even tougher competition for AMD's Athlon 64 X2.

But if you absolutely must spend $1000+ on a CPU today, what should you do? While the 965 has become a lot more competitive with the FX-60, our overall nod still goes to AMD. The 965 does have its strengths in 3dsmax and video encoding, so if those are areas in which you are partcularly looking for stellar performance, then the recommendation obviously swings to Intel.

So yea, if by notable fanboy Anand, you mean that ever since mid-2006 when Conroe was out and Anand started recommended Core2 over X2, than you're right.

 

[Citizen86]

Duby, your powers of logic are astounding! Here all this time I thought that review sites were paid to show buyers what was worth their money. If review sites only handed out gold stars I don't think they would be around very long. I'm not saying to trust every review, but you're basicallly trying to discredit every professionally done review?!

NO COOKIE FOR YOU!!!

 

[jimmyb]

Your looking at paid reviews. These companies were --PAID-- to make the review they made. Same as all other reviews. You see the thing yopu all know just as much as I do is that if a site gives a bad review on a product they will likely --never-- get to review products from that companie again.

Many reviewers have a strict policy of not accepting products and payments from companies. They will go out and purchase the actual product themselves.


Anyway, my impression of the K7 and K8 launches was that there was quite a lot of excitement and fairly high expectations regarding their performance. It's pretty much the opposite of how it is now.

I really hope Barcelona does prove all this negative press wrong though.

 

[BigDH01]

The benches just dont show what you conclusion is. You conclude that the K7 took the crown on launch but the benches show that it was back and forth with the P3. The only benches with a clear lead are 3dnow tests. Which in the real worlld means absolutely nothing.... Zip zero... nothing.... On all of those sites you linked to the data just doesnt support your conclusions...

I'll stick to what I said. AMD has never taken a clear lead on a launch product. Not with the K6, otr the K7 or the K8. They all faired well, and did take the lead later on, but not on the launch product.

Besides your making an argument that doesnt matter anyway. Your looking at paid reviews. These companies were --PAID-- to make the review they made. Same as all other reviews. You see the thing yopu all know just as much as I do is that if a site gives a bad review on a product they will likely --never-- get to review products from that companie again. You all know thta just as well as I do, and I shouldnt have to say anything about it... It's common knowledge. The fact thta you refuse to accommodate this is just silly.

The thing you --should-- be looking at is what enthusiasts, like us were saying at that time. Start looking in the forums. Look at Anands forums from those days. Look at Toms forums from those days. Look at Hard's forums from those daysand you'll see that people are saying --now-- the --exact same things they were saying then-- No difference what-so-ever. Things are exactly the same they were....

LOL, you must be joking. To quote Aces

Depending on the application, the Athlon is 10-40% faster. It wins in almost every single benchmark, but it really shines in creative applications.

From Damage at Ars

At the same time, it's faster clock-for-clock than Intel's Pentium III--in some cases, a whole lot faster. Taken together with its 200MHz EV6 bus and the AMD 750 chipset, AMD's claims that this is a true next-generation x86 platform seem entirely warranted.

From FastSite at X-bit

Today our tests prove that AMD Athlon is the fastest x86 processor with progressive architecture. Intel is absolutely unable to withstand AMD's success, because Katmai architecture has almost exhausted its potential while AMD can increase the frequencies of its newest chips without facing any serious hardships.

Those aren't the conclusions of people that thought the K7 and P3 were going back and forth. That's a clear victory for AMD and it happened at release. Did you even look at those sites? You can't just claim something is true when all of the evidence points to the contrary. There's being stubborn and then there's outright lying.

You know, I had some stuff here about you just swallowing your pride but why bother? Continue to believe what you believe. I never thought I'd see the day where I'd be arguing with an AMD fanboy (no less) that the K7 was a kickass chip when it was released. This is just silly.

 

[duby229]

LOL, you must be joking. To quote Aces



From Damage at Ars



From FastSite at X-bit



Those aren't the conclusions of people that thought the K7 and P3 were going back and forth. That's a clear victory for AMD and it happened at release. Did you even look at those sites? You can't just claim something is true when all of the evidence points to the contrary. There's being stubborn and then there's outright lying.

You know, I had some stuff here about you just swallowing your pride but why bother? Continue to believe what you believe. I never thought I'd see the day where I'd be arguing with an AMD fanboy (no less) that the K7 was a kickass chip when it was released. This is just silly.

All of the evidence shows that the 93 won some, and the K7 won some.... The conclusions made were paid for. If you want to know what people thought at those times, then go back and read the respective forums. If your not willing to look at what real people with real opinions had at that time, then this conversation is over.

 

[duby229]

Duby, your powers of logic are astounding! Here all this time I thought that review sites were paid to show buyers what was worth their money. If review sites only handed out gold stars I don't think they would be around very long. I'm not saying to trust every review, but you're basicallly trying to discredit every professionally done review?!

NO COOKIE FOR YOU!!!

I'm not discrediting reviews at all. I think reviews are handy in determining performance of specific loads. They are not at all usefull for comparisons purposes. This is a whole nother thread though so I'll stop right there....

Instead I'll just say that the data collected is usefell. The type of data collected is almost always selected based on what will give the best scores on what hardware... Fore example you read a C2D review and SuperPI is always there.... You read a X2 review and SuperPI is --never-- there.... Those benches that have been done do provide usefull information. The conclusions made by the review is almost never useful.

 

[BigDH01]

All of the evidence shows that the 93 won some, and the K7 won some.... The conclusions made were paid for. If you want to know what people thought at those times, then go back and read the respective forums. If your not willing to look at what real people with real opinions had at that time, then this conversation is over.

This conversation was over before it began. Like most conversations with you, it goes you making false claims, you getting destroyed by evidence, you making conspiracy theories while refuting all evidence on the basis that it doesn't agree with your version of reality, then you dismissing anyone that disagrees. I don't even know why I reply to you anymore. At least others are able to see your MO in action.

 

[duby229]

This conversation was over before it began. Like most conversations with you, it goes you making false claims, you getting destroyed by evidence, you making conspiracy theories while refuting all evidence on the basis that it doesn't agree with your version of reality, then you dismissing anyone that disagrees. I don't even know why I reply to you anymore. At least others are able to see your MO in action.

The evidence shows that I was right. If you beleive the conclusions made then that us your problem. If you want to solve your problem then you can get back to the point and look at what people were saying at that toime, and realized that people were saying --exactly-- the same things they are saying now.

Look at the popular forums from those days. If you dont want to do that then you fail....

 

[Manny Calavera]

The evidence shows that I was right.

Deny,deny,deny !! It works well doesnt it ?


What must it be like,to go through life singing 'LA la la' at the top of ones lungs,while having your ears firmly plugged up,and 3 inch thick rose colored glasses on ?

 

[pxc]

Like most conversations with you, it goes you making false claims, you getting destroyed by evidence, you making conspiracy theories while refuting all evidence on the basis that it doesn't agree with your version of reality, then you dismissing anyone that disagrees. I don't even know why I reply to you anymore.

Everyone sees that. It's pointless to argue with a guy who just makes up his own 180 degree reality. I pretty much just chuckle at him nowadays.

 

[Citizen86]

I'm not discrediting reviews at all. I think reviews are handy in determining performance of specific loads. They are not at all usefull for comparisons purposes. This is a whole nother thread though so I'll stop right there....

Instead I'll just say that the data collected is usefell. The type of data collected is almost always selected based on what will give the best scores on what hardware... Fore example you read a C2D review and SuperPI is always there.... You read a X2 review and SuperPI is --never-- there.... Those benches that have been done do provide usefull information. The conclusions made by the review is almost never useful.

Unless it's a professional review where the same programs are used on ALL CPU's tested. I was willing to give you the benefit of the doubt, but everything that BigDH01 and everyone else here is saying is completely true.

Fore example you read a C2D review and SuperPI is always there.... You read a X2 review and SuperPI is --never-- there....

Well it's time for you to show some proof. I went to Anandtech to see if what you were saying is true. Look at the tests:

E4300 Review

X2 BE-2350 Review

Notice how not all the tests are EXACTLY the same, but pretty similar, in regards to the programs and such they use? Biased, not really....

And I LOVE how you said that Anandtech were AMD Fanboys until Conroe, now they're Intel fanboys because they were getting paid by AMD but now getting paid by Intel... OR!!!!!! OR!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Because Conroe is a better processor and they show that through TESTING.

Anyone can come on here and make whatever comments they want, without proof and denying anyone elses proof. You are VERY good at that Duby.

 

[RamonGTP]

And I LOVE how you said that Anandtech were AMD Fanboys until Conroe, now they're Intel fanboys because they were getting paid by AMD but now getting paid by Intel... OR!!!!!! OR!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Because Conroe is a better processor and they show that through TESTING.

Pffffft... NOOOOO... That would make WAAAYYYY too much sense for it to be true!

 

[brucedeluxe169]

discuss

 

[visaris]

For example you read a C2D review and SuperPI is always there.... You read a X2 review and SuperPI is --never-- there.... Those benches that have been done do provide useful information. The conclusions made by the review is almost never useful.

I agree. I have even read reviews where the text in the conclusion is the exact opposite of what the graphs say, and that isn't even getting into selective benchmarking/graphing. You are 100% spot on. Take the data as a couple of data points, and drop the text as worthless.

You are also correct about the bad press at the K7 and K8 launches. I think too many people here are relative newcomers to the x86 enthusiast space and simply don't remember. Further, the ones that are new to the scene don't seem to want to spend the time needed to get a good grasp of all the bad press AMD was getting at those times. Remember, why spend 30 min doing some Google searches when you can just bash people without it (yes, that is sarcasm)?

 

[APOLLO]

You are also correct about the bad press at the K7 and K8 launches. I think too many people here are relative newcomers to the x86 enthusiast space and simply don't remember. Further, the ones that are new to the scene don't seem to want to spend the time needed to get a good grasp of all the bad press AMD was getting at those times. Remember, why spend 30 min doing some Google searches when you can just bash people without it (yes, that is sarcasm)?

Yes, he has been correct on the past negative publicity. You're right about forum members' memories being somewhat short. My experience goes back to the early 90s, and even though AMD is facing a steep uphill climb in a very tough year, they have been through this before several times, to a lesser or greater extent.

What is amazing about AMD is their phoenix-like capability of rising out of the ashes of despair. They arguably accomplish their greatest feats of engineering and marketing turnaround at the 11th hour.

Whatever the case may be, some good news might be arriving soon, if this thread is any indication. [fingers crossed]

 

[DemonDiablo]

Whatever the case may be, some good news might be arriving soon, if this thread is any indication. [fingers crossed]

LOL are you kidding me? We already went over that in this thread I believe. For a second there I thought you have some new enlightening information. Not some stuff that everyone and there mama already knows about. Try to keep up chief.

 

[Manny Calavera]

Whatever the case may be, some good news might be arriving soon, if this thread is any indication. [fingers crossed]

I read that thread and see no positivity in it at all,where is it ?? So the fools in R&D are optimists ? That and that AMDati are buying Intel Xeon servers

Well,one could also intimate thiers a fool born every minute.

 

[Citizen86]

That and that AMDati are buying Intel Xeon servers

LOL!!!! I don't take that as good news at all..... Poor Barcelona, sounds like it's pretty doomed when it's on creators don't think it's good enough....

You'd think they could throw in some dual-core dual-socket Opterons as servers?

 

[LstBrunnenG]

Look at the popular forums from those days. If you dont want to do that then you fail....

I remember those days quite well. I remember Intel fanboys clinging desprately to the hope that Prescott would save them - since newer has to be faster, right? I remember AMD fanboys shouting down everyone else in a new build thread saying you were a complete moron if you didn't get an A64, while the lone voice in the wilderness was patiently reminding people that the Northwoods (and later Prescotts) were still good at video encoding. Fewer still reminded people that the A64's "dominance" in gaming was a few bare percentage points, while everyone else was acting as though the P4 was to the A64 as Intel Extreme Graphics were to the 9800 Pros of the day.

I'd glaldy go back and dredge up your sentiments of the day, but you didn't join this forum until two years after I did, in May of 2005, on the eve of the X2 launch if I remember correctly. I hardly believe you would have assumed the role of lone voice in the wilderness if you were here back in 2004. Or would you have risen to the Pentium 4's defense, and asked people to stop warning new users away from it like it was barely able to run an operating system? Or would you join the scores of others blindly preaching K8's dominance? I think what we can see of your record, from 2005 on, speaks for itself.

 

[RamonGTP]

It's pointless to post any type of evidence to someone who is so fargone from reality that he makes up his own. It's already been done, and you come up with your conspiracy theories. Until you snap back into reality, it's just a wasted effort becase all you'll do is start running off at the mouth again with more BS.

Oh and I'm curious... Where is your "evidence"? You ask people to post links to back up their claim, and they do. Then you refuse to believe it (see above paragraph) I don't see anything from you but talk though. I wonder why that is?

I'll indulge you though... From Anandtech

We first took a look at the upcoming AMD K7 slightly before this year's E3 Expo, as fate would have it, around the same time as the release of the K6, just two years ago. The performance of the K7, then clocked at 500MHz, was on par with a Pentium III 500. The system bus operated at a unique 125MHz, and AGP support was very limited. Needless to say, we were not impressed.
Click to Enlarge


The second encounter resulted in more positive results, in all but one test, the K7 550 managed to beat the Pentium III 500. The one test that resulted in a considerably lower score on the part of the K7 was Ziff Davis' FPU Mark, luckily the problem turned out to be related to a hardware bug with the CPU, one that was later fixed. The important thing to keep in mind at that point was that AMD had actually produced a chip, albeit still in beta stages, that could compete with a Pentium III, clock for clock, across the board, in games and business applications. The potential was there, but could AMD do it?

AMD promised a June release of the K7, and in June AMD approached us about reviewing the K7 now clocked at 600MHz and ready for action. Without refusing the NDAs were signed, and the system was in our hands. The results? Well, you're about to find out.

There you have it... A pre-release review of the K7 with FAVORABLE results... What conspiracy theories will you pull out of your bag of tricks now?

Edited for the sake of being politically correct.

 

[BinarySynapse]

nevermind, posted on impulse, bad judgement call on my part. disregard.

 

[APOLLO]

LOL are you kidding me? We already went over that in this thread I believe. For a second there I thought you have some new enlightening information. Not some stuff that everyone and there mama already knows about. Try to keep up chief.

I read that thread and see no positivity in it at all,where is it ?? So the fools in R&D are optimists ? That and that AMDati are buying Intel Xeon servers

Well,one could also intimate thiers a fool born every minute.

http://www.theinquirer.net/default.aspx?article=41256

 

[DemonDiablo]

I'm mildly impressed. In the end AMD is still releasing just pictures though. No information on these products, no benchmarks, no nothing. Im a realist not a optimist. I need some hard hitting information to get drawn into the hype not f'ing pictures. I hate it when news sits say, "Breaking news! Look at our 20 pages worth of PICTURES!!!!!!!"

 

[Manny Calavera]

I'm mildly impressed. In the end AMD is still releasing just pictures though. No information on these products, no benchmarks, no nothing. Im a realist not a optimist. I need some hard hitting information to get drawn into the hype not f'ing pictures. I hate it when news sits say, "Breaking news! Look at our 20 pages worth of PICTURES!!!!!!!"

Thank you for saving me the time,and putting it so well DD



More pics,another PR stunt...

 

[vraa]

Just curious
if someone says that they can't hit 3ghz
Yet they show a picture of them their processors at 3ghz
Isn't that technically by definition NOT a PR stunt?

 

[BinarySynapse]

Just curious
if someone says that they can't hit 3ghz
Yet they show a picture of them their processors at 3ghz
Isn't that technically by definition NOT a PR stunt?

It's not really being said they didn't hit 3GHz with this system. But how do we know it's not a hand picked extremely low yeild processor? Something they'd be willing to show off saying "See we can hit 3GHz (btw this is one of only four that can and it won't be on shelves for several months while we work out the issues)." I'm all for AMD getting their act together and kicking ass so that Intel has to come back and kick AMD's ass. This whole "We estimate--" or "See look at the pictures of --" bullshit is tiresome. They need to bring out a system and let the review jockees play with it to their hearts desire. Then and only then will I be satisfied enough with AMD's pre-launch situation. As it is now, it's still the samw "We'll have to wait and see."

 

[Snowdog]

Even if this is a cherry picked unit, it is running on air, looks like stock cooling. They are only planning low speed realeases initially. Plenty of time sort out fab issues. This is the first genuine good AMD news in a while.

Now if they would only let out some real benches.

 

[visaris]

Now if they would only let out some real benches.

Slide 45 is as close as AMD will give us. This slide was also represented by actual systems present on demo during the meeting. A number of tech writers commented on watching the benchmarks run on the 2GHz Barcelona. So far, this the only halfway credible bench we have access to.


Sorry to the site hosting this image I'm hot-linking to. I don't have time to host it myself right now:
http://www.iian.ibeam.com/events/th...0070726084721294707/default_large/Slide45.JPG

Edit: A link is more considerate than image embedding.

 

[Cannon_Fodder]

I have been watching the [H]ard threads for some time now, but some of the posts here just made me laugh. Some of you are the type, if you where dying of thirst you would complain about the glass I gave you water in.

Do I think that AMD is going to be the Intel killer with this new processor? No

Do we need to give AMD credit for what they have done? Yes

Don't dig the grave before the processor even comes out. Wait for the benches and then you can do the “I told you so” or the “Wow I was wrong”. Until then all we are doing is talking out of our ass.

 

[HardLiner]

Is AMD coming out with 3Ghz quad core processors this year?

 

[APOLLO]

Is AMD coming out with 3Ghz quad core processors this year?

Highly doubtful, but if they do, it would be a engineering and production triumph in light of the industry's atmosphere of doubt surrounding AMD's capabilities. The recent 3.0GHz operational demonstration was just that, a demonstration for PR purposes. FWIW, it did however, assuage some fears of AMD's current standings with regards to its yield potential. Hopefully, that potential will translate to available products in the near future.

 

[Leo`]

I have a dual 1207 board so this means 8 cores for me