Sign in with
Sign up | Sign in
Your question

The truth about AMD Benchmarks

Last response: in CPUs
Share
June 11, 2007 10:22:51 AM

Quote:

For a while we had lost confidence in AMD, like many of you had as well, and although AMD's position in the market hasn't changed we are more confident now that it can actually bounce back from this



Quote:

We know that Barcelona works and runs benchmarks, as we saw back at AMD in May. But the demos that AMD ran were on its own motherboards, not on motherboards from its partners. AMD's partners just recently received their first Barcelona samples, and as expected, the boards require some heavy BIOS work before the new chips will even work.

The motherboard we tested on had minimal HT functionality and wouldn't run at memory speeds faster than DDR2-667; most video cards wouldn't even work in the motherboard. Memory performance was just atrocious on the system, but the motherboard manufacturer we were working with attributed this to a BIOS issue that it expects to be fixed in the very near future.

In the end, performance was absolutely terrible. We're beginning to understand why AMD didn't let us test Barcelona last month. It's not that AMD is waiting to surprise Intel; it's that the platform just isn't ready.




Quote:

None of the partners we talked to are really impressed with Barcelona yet, but then again none of the true potential of the chip and chipsets have been realized due to CPU stepping issues, BIOS maturity, and generally low clock speeds. AMD simply hasn't gotten the process under control yet and after hearing our friends at the motherboard companies talk, AMD is close to near/total panic mode right now.

The general consensus is that until AMD gets to at least 2.0 - 2.3GHz, we will not see any significant volumes. Alternatively, AMD may delay the launch until September, with volume hitting the channel and partners by November

Linakge!

More about : truth amd benchmarks

June 11, 2007 10:41:16 AM

BARCELONA WILL LAUNCH ON TIME
\
June 11, 2007 10:53:15 AM

Quote:
BARCELONA WILL LAUNCH ON TIME
\
link?
Related resources
June 11, 2007 11:43:50 AM

lol, them pictures always pop up.. can someone like me to one of them uncle sam pictures that has him pointing and it saying "it is rude to point" or something to that effect.
a b à CPUs
June 11, 2007 12:28:53 PM

Anand must have finally decided that bashing AMD with FUD (like is frequently done here by Tom's crew) isn't a good idea. And I applaud Anandtech for some frank commentary.

If Barcelona doesn't launch then kiss goodbye future innovation.

Intel will then respond by increasing prices and you won't see any new innovation in silicon for ages.

Vanilla will be the only flavour of icecream fullstop !!

If your lucky they might release some lower speed half disabled Penryn - remember they make more money on 45's as opposed to 65's providing the same percentages of batches are good. Plus they can later "enable" some of the core logic as a means of demonstrating better performance ... to make you want to upgrade through lies and deception, rather than true innovation born from competition.

That will be the only reason Intel release 45's if AMD goes belly up.

I hope AMD survive - they have been great innovators ... if not so good at managing their money in the short term.

Remember that with their volume they couldn't truly compete with Intel anyway ... they don't produce enough silicon from their fabs. Intel wipes the floor in terms of real sales ... which are predominantly lower end parts, plus their integrated graphics and chipset divisions are truly huge.

The single die Quad is a big ask ... even Intel baulked at it ... and acknowledged this at Computex.

We need competion - all fanboi rubbish aside.

If it wasn't for the Athlon scaling well, and the A64 mem innovation, then we would still be driving P3's or P4 heaters !!!

Intel wouldn't then have got those Israeli geniouses to respond with the Core2 lineup - an admirable comeback indeed.

At the end of the day we benefit fom competition ...

Any new CPU advances must be offset by decent power savings too... we still havn't seen much of that yet.

Please buy something AMD and keep the dream alive ... heh heh.

No ... this isn't Hector ... heh heh ... :) 
a b à CPUs
June 11, 2007 12:55:00 PM

I see your really busy too ...

You might want to rephrase your last sentence ... the grammar is just sad.
June 11, 2007 1:08:54 PM

Quote:
Anand must have finally decided that bashing AMD with FUD (like is frequently done here by Tom's crew) isn't a good idea. And I applaud Anandtech for some frank commentary.

If Barcelona doesn't launch then kiss goodbye future innovation.

Intel will then respond by increasing prices and you won't see any new innovation in silicon for ages.

Vanilla will be the only flavour of icecream fullstop !!

If your lucky they might release some lower speed half disabled Penryn - remember they make more money on 45's as opposed to 65's providing the same percentages of batches are good. Plus they can later "enable" some of the core logic as a means of demonstrating better performance ... to make you want to upgrade through lies and deception, rather than true innovation born from competition.

That will be the only reason Intel release 45's if AMD goes belly up.

I hope AMD survive - they have been great innovators ... if not so good at managing their money in the short term.

Remember that with their volume they couldn't truly compete with Intel anyway ... they don't produce enough silicon from their fabs. Intel wipes the floor in terms of real sales ... which are predominantly lower end parts, plus their integrated graphics and chipset divisions are truly huge.

The single die Quad is a big ask ... even Intel baulked at it ... and acknowledged this at Computex.

We need competion - all fanboi rubbish aside.

If it wasn't for the Athlon scaling well, and the A64 mem innovation, then we would still be driving P3's or P4 heaters !!!

Intel wouldn't then have got those Israeli geniouses to respond with the Core2 lineup - an admirable comeback indeed.

At the end of the day we benefit fom competition ...

Any new CPU advances must be offset by decent power savings too... we still havn't seen much of that yet.

Please buy something AMD and keep the dream alive ... heh heh.

No ... this isn't Hector ... heh heh ... :) 



Admittedly it was a positive spin on a negative development. I think what;s happening is a problem of the 65nm - 90nm mix. Brisbane and Turion are at 65nm but Opteron is not. Also, there are still Windsors out there which means that Fab 30 is almost obsolete at 90nm. It can't help with margins as OEMs will want the cooler 65nm chips.

There are only 3 SKUs for Brisbane presently as again Fab 30 has to put out mainly 90nm right now. Hopefully that little influx of cash will help them ramp at least 1/2 by the end of the summer.

That way Fab 36 can run more lines of Barcelona derivatives. It's a tough time for AMD but as I noted in other posts, AM2 was looking really bad right up to the launch last year. It's real problem was high latency not lack of perf.

It looks a lot like a YMMV case since some people reported that the chips could go up to 2.5GHz OC. He is also implying that it won't really be as effective until 2.6GHz. But that's because Opteron and C2Q are at 3GHz.

But the main thing was platform not as much CPU problems ( other than clocks). That's why I thought they would go with some of the high end desktop parts first to stop that bleeding. Opteron still owns 4P.


I guess it's interesting as to why they didn't shrink Opteron. That would have probably gotten them to 3.4GHz as Opteron can do 3GHz at 120W.

I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.

Anyway,

ALL HAIL THE DUOPOLY!!
June 11, 2007 2:05:40 PM

Quote:

I guess it's interesting as to why they didn't shrink Opteron. That would have probably gotten them to 3.4GHz as Opteron can do 3GHz at 120W.


First of all, AMD's 65nm has problems scaling at clock speeds nearing 3GHz. Secondly, the silicon on insulator technique doesn't scale well with higher freqencies. Last but not least, a 3.4GHz Opteron (if they were able to somehow make one) would be priced way too high and the heat output would take us back to the NetBurst days.

Quote:
I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.


Hm...yeah...

I just wish that they started the Barcelona project a year earlier, or as soon as K8 launched. It was only a matter of time until Intel would drop NetBurst.
June 11, 2007 2:17:17 PM

Quote:
I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.

Anyway,

ALL HAIL THE DUOPOLY!!


YO!!!!!

You know guys, I'm starting to love the NEW BaronMatrix.

Bad sentence aside, your reflexion are less on the "fanboy" side these days and I really like that. For me it's OK to prefer one of both company, but not to the point of getting blind! It seems to me you just got there.

While I actually prefer Intel in general (hope we're still friend :wink: ), I know and wish AMD to thrive so that innovation doesn't stop. Like you say; "All hail the duopoly".

Well said!
a b à CPUs
a b À AMD
a b V Motherboard
June 11, 2007 2:21:15 PM

Quote:
I see your really busy too ...

You might want to rephrase your last sentence ... the grammar is just sad.


I see you don't know the difference between your and you're. Sorry man, you're not qualified to judge his grammar skills :p 
a c 110 à CPUs
a b À AMD
a b V Motherboard
June 11, 2007 2:25:32 PM

Quote:
I see your really busy too ...

You might want to rephrase your last sentence ... the grammar is just sad.


I see you don't know the difference between your and you're. Sorry man, you're not qualified to judge his grammar skills :p 

I'm not gonna get involved in this except to say:

il qui habite en les maisons de verre ne doit pas lancer de pierre
June 11, 2007 2:32:29 PM

Quote:

First of all, AMD's 65nm has problems scaling at clock speeds nearing 3GHz. Secondly, the silicon on insulator technique doesn't scale well with higher freqencies. Last but not least, a 3.4GHz Opteron (if they were able to somehow make one) would be priced way too high and the heat output would take us back to the NetBurst days.

Well said. In addition scaling opteron means additional cost. Scaling a product, K8 Opteron, that is soon to be replaced by another, rarely makes sense. Well, if they had known about the delays and difficulties before they might have done it, but it´s too late for that now.
June 11, 2007 2:38:09 PM

To be honest, I don't keep up with AMD technology much. Being I have family that work at Intel, I'm somewhat impartial to them.

AMD's made some big mistakes in the past(like their lack of temperature control that Tom's Hardware did a very nice video on). Granted, Intel's done their share of screw ups too, but Intel hasn't launched a product that was designed to inherently fail if some dummy installed the heat sink wrong. This has been one of my biggest beefs with AMD. That's pretty bad design if you ask me. Intel's FPU Div bug from the Pentium sure, there's that small percentage of people that could be goofed with the bug, but it didn't result in you having to either attempt to RMA the chip or buy a new one. AMD has been very much an enthusiast market, and so they were much more likely to encounter some poor schmuck that burned up their chip cause they didn't know better.

So, with that disclaimer aside...

I have been buying Intel chips, even when I knew the P4s were slower and more expensive than AMD's alternative. AMD has done some great things with their products. They have innovated some great things and I do applaud them for that. HT technology is a very good idea, and I wish Intel would adopt a system like AMD where you weren't buying a new MB every time you wanted a new chip. AMD did good with the AM and AM2 setups. Ive bought 5 motherboards in the last 3 years, and my buddy who has been AMD has bought only 2. Big savings in the money department there.

We all have to realize that AMD is still merging with ATI and that takes alot of work. They've gotten to change their entire outlook on their future products because of ATI's technology. AMD is going through a tough time right now, much worse than Intel when they were getting a whooping with their Netburst technology, but if they are smart(which they have proved they can be because they've survived Intel's slaughter before) they'll bounce back. So what if Barcelona comes out and really sucks. AMD will surely come out with better successors that compete with Intel's market lead. All companies have their ups and downs. Most companies don't want to admit when they're having a down time, but it's showing through for AMD right now. I'm glad AMD is a company built to survive, because without AMD Intel wouldn't be trying to build better CPUs and so forth. So although AMD didn't build Intel's great Core2, they did build it indirectly by building a better product than Intel had for quite a while.

The table has turned and is favoring Intel right now. So the 2 big questions are:

1. Will it last 6 months or 2 years?
2. Who do you think is gonna have the next big innovation?

I bet it's gonna last more than 6 months, just because AMD got too complacent with their lead and didn't expect Core 2 to be such a killer processor. I also bet AMD will be providing the next big innovation because their desperation will cause them to create great new products(just like people always say humans are at their best when things are at their worst).
June 11, 2007 2:44:42 PM

Quote:
BARCELONA WILL LAUNCH ON TIME
\


haha :D  i loved that guy, was so funny.
June 11, 2007 3:22:16 PM

Quote:

I guess it's interesting as to why they didn't shrink Opteron. That would have probably gotten them to 3.4GHz as Opteron can do 3GHz at 120W.


First of all, AMD's 65nm has problems scaling at clock speeds nearing 3GHz. Secondly, the silicon on insulator technique doesn't scale well with higher freqencies. Last but not least, a 3.4GHz Opteron (if they were able to somehow make one) would be priced way too high and the heat output would take us back to the NetBurst days.

Quote:
I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.


Hm...yeah...

I just wish that they started the Barcelona project a year earlier, or as soon as K8 launched. It was only a matter of time until Intel would drop NetBurst.


First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks. AMD CAN'T stop making say the 6000+ and Opteron x80 as they can't just turn off Fab 30.

If they take the 65nm up to 3GHz then they have overlapping PR Ratings at 65nm and 90nm.

Since 65nm is much smaller why not limit the clockspeed to try and help margins.
a c 110 à CPUs
a b À AMD
a b V Motherboard
June 11, 2007 3:31:16 PM

Just my humble ignorant opinion but AM2 is where AMD fooked-up. DDR2 brought little if nothing to the table for AMD because of the need to reconfigure the IMC. Precious resources and time were spent on AM2 that could have been devoted to Barcelona, AIT integration/chipset development, Fusion, etc.

AMD is now caught in the middle of DDR2/DDR3 HT2/HT3 migration. They should screw us all and advance DDR3/HT3, new chipsets, Fusion, etc and forget about the *plus*. Call it Socket F-U.

AMD can from time to time lead the market - and from time to time drive the market (IMC, x86-64, Fusion, etc) but there is no way they can continuously drive Chipzilla. Ain't gonna happen!

Long Live s939/40! (still where the value in AMD lies . . . :)  )

((...he that lives in the glass house must not launch rock...))
June 11, 2007 3:37:49 PM

Quote:

Since 65nm is much smaller why not limit the clockspeed to try and help margins.


wow thats something i wouldn't have thought of
June 11, 2007 3:52:23 PM

Quote:
I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.

Anyway,

ALL HAIL THE DUOPOLY!!


YO!!!!!

You know guys, I'm starting to love the NEW BaronMatrix.

Bad sentence aside, your reflexion are less on the "fanboy" side these days and I really like that. For me it's OK to prefer one of both company, but not to the point of getting blind! It seems to me you just got there.

While I actually prefer Intel in general (hope we're still friend :wink: ), I know and wish AMD to thrive so that innovation doesn't stop. Like you say; "All hail the duopoly".

Well said!


Why bring up a simple thing like a period vs a comma? There is nothing different about how I post only who is around to respond. Maybe now we can have unbiased discussion about both companies.

Normally by now the FUD patrol would have crapped all over the thread with off-topic posts.

I don't really prefer AMD, I just have bad memories of the K6 days when my machine ran constantly without a problem yet Intel lived on the phrase "not stable enough."
June 11, 2007 4:08:42 PM

Quote:
BARCELONA WILL LAUNCH ON TIME
\


Can anyone provide me the history of this picture? I think I know why it's being used, but I'm not sure... I believe it's the Iraqi whatever minister who said that they were winning the war just as American tanks began rolling into Baghad right behind him. Is that it?
June 11, 2007 4:17:42 PM

Quote:


There is nothing different about how I post only who is around to respond. Maybe now we can have unbiased discussion about both companies.

...
I don't really prefer AMD, ...


Just when there are feint signs you are talking less crap, you crap out a new big load of BS :lol: 
June 11, 2007 4:20:42 PM

AM2 on the desktop was probably not required. However in the server world is was. You can drop a little in latency if you improve bandwidth which DDR2 did.
The biggest problem is A64 and Opteron are pretty much the same die wise so with their current model you change one line you will have to change the other.
June 11, 2007 4:53:53 PM

Yeah, he was their minister of information or something... the joke is as you said: He was sitting there saying that coalition forces were being driven back into Saudi Arabia while the World's TVs were jammed with videos of our air power blowing the shit out of everything that they could find.


Back on topic: I really dont know what to make of the whole state of AMD right now. I dont understand the real technical details like why they are having problems with clock speeds on their smaller dies. It seems to me that a quad-core/single-die chip would not bring any significant increases to the table if you still are running an OS that doesnt even utilize 2 cores fully, let alone 4. I know that AMD has their lab in Europe that is supposed to be working on software/hardware integration, but if Micro$oft isn't involved or helping, what's the upside? I think that it's just such a waste of effort and money to market and release a product that can't be utilized to it's capacity because the software won't utilize it for another 5 years or more.
Now, I know some will say: "But the software will come." Bullshit! HOW LONG HAVE WE WAITED FOR 64 BIT APPLICATIONS/GAMES TO GO MAINSTREAM?!!?!?!? You have more potential locked away in the hardware that you have now than will come out "new" in the next 2 years. I can guarantee that everyone who reads this has a 64 bit processor, but maybe 5 of you use WIN 64. Why? Because it's not supported on the back-end. How long ago did they release the 64 bit cpu's? Hell, I bought my Athlon64 2000+ back in 2000. Why don't they force 64-bit computing down everyone's throats like they did Crapta on new computers? Hell, why did they even bother making a 32-bit version of it? Oh, yeah... backwards compatibility with old products and software from when I owned a Gravis Ultrasound and Doom was the shit. Why not force your "partners" to make 64 bit games, drivers, and applications available? I don't know anything, but I bet that some uber-nerdle will tell me why... :roll:
Seems to me that all this forward hardware development is great and AMD has some great things in the works (IF it works and WHEN it works are 2 totally seperate arguments). I think that instead of shrinking the dies and making the same products (only smaller), they should work on software integration and enable us to use the true potential that the chips that we have now.
Oh, wait... that would kill their sales. Nevermind :lol: 
June 11, 2007 4:56:56 PM

Quote:
First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks. AMD CAN'T stop making say the 6000+ and Opteron x80 as they can't just turn off Fab 30.


Hm... Good point.

But I swaer I read an article that discussed the reasons why AMD has so much problems hitting 3GHz and beyond.

I'll post the link here if I can find it. :) 
June 11, 2007 5:44:30 PM

Quote:
First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks. AMD CAN'T stop making say the 6000+ and Opteron x80 as they can't just turn off Fab 30.


Hm... Good point.

But I swaer I read an article that discussed the reasons why AMD has so much problems hitting 3GHz and beyond.

I'll post the link here if I can find it. :) 

I think they use SOI but in a different process... if it where at all possible im sure AMD would also hit 4.7Ghz, to say that its possible with AMD's process is crazy the engineers would be tripping over themselves to hit that speed.
June 11, 2007 5:59:21 PM

Quote:
First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks. AMD CAN'T stop making say the 6000+ and Opteron x80 as they can't just turn off Fab 30.


Hm... Good point.

But I swaer I read an article that discussed the reasons why AMD has so much problems hitting 3GHz and beyond.

I'll post the link here if I can find it. :) 


I'd say it was because Brisbane is a shrink and the original design was at 130nm. Of course they will have problems with high clockspeed. I think that the native design of Barcelona will get Kuma up to 3.4 or so. By next year I would expect to see even the quad cores that high.

AMD has historically been very good at improving the clock speed and power of their chips. I don't see this being different.
June 11, 2007 6:03:39 PM

Quote:

First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks.

It´s a little more complicated than that. If it weren´t you´d use an IBM processor right now.


Quote:

If they take the 65nm up to 3GHz then they have overlapping PR Ratings at 65nm and 90nm.

Well, they did just that with the 2x 3600 processors. There are 65nm and 90nm versions. Overlapping PR ratings (which became totally meaningless anyway) aren´t a valid argument.

Quote:

Since 65nm is much smaller why not limit the clockspeed to try and help margins.

Nice but why not sell better performing parts for more money? If a 65nm part clocks up to 3Ghz without going supernova why not sell it as just that and make a nice profit? Well unless AMD isn´t about profit anymore - i had my doubt in the past and recent launches and anouncements made me think of AMD as a charity organisation.
June 11, 2007 6:11:17 PM

Quote:
Just my humble ignorant opinion but AM2 is where AMD fooked-up. DDR2 brought little if nothing to the table for AMD because of the need to reconfigure the IMC. Precious resources and time were spent on AM2 that could have been devoted to Barcelona, AIT integration/chipset development, Fusion, etc.

AMD is now caught in the middle of DDR2/DDR3 HT2/HT3 migration. They should screw us all and advance DDR3/HT3, new chipsets, Fusion, etc and forget about the *plus*. Call it Socket F-U.


That is indeed a good and valid point. AMD gained nothing from DDR2. The memory bandwidth increased, performance didn´t. The same will happen with DDR3. I´m starting to believe that Intel is switching memory again just to hit AMD where it hurts. DDR3 will have even worse latencies and if AMD doesn´t come up with something really smart, it will offer no benefit again.
June 11, 2007 6:18:29 PM

Quote:

Back on topic: I really dont know what to make of the whole state of AMD right now. I dont understand the real technical details like why they are having problems with clock speeds on their smaller dies. It seems to me that a quad-core/single-die chip would not bring any significant increases to the table if you still are running an OS that doesnt even utilize 2 cores fully, let alone 4. I know that AMD has their lab in Europe that is supposed to be working on software/hardware integration, but if Micro$oft isn't involved or helping, what's the upside? I think that it's just such a waste of effort and money to market and release a product that can't be utilized to it's capacity because the software won't utilize it for another 5 years or more.
Now, I know some will say: "But the software will come." Bullshit! HOW LONG HAVE WE WAITED FOR 64 BIT APPLICATIONS/GAMES TO GO MAINSTREAM?!!?!?!? You have more potential locked away in the hardware that you have now than will come out "new" in the next 2 years. I can guarantee that everyone who reads this has a 64 bit processor, but maybe 5 of you use WIN 64. Why? Because it's not supported on the back-end. How long ago did they release the 64 bit cpu's? Hell, I bought my Athlon64 2000+ back in 2000. Why don't they force 64-bit computing down everyone's throats like they did Crapta on new computers? Hell, why did they even bother making a 32-bit version of it? Oh, yeah... backwards compatibility with old products and software from when I owned a Gravis Ultrasound and Doom was the ****. Why not force your "partners" to make 64 bit games, drivers, and applications available? I don't know anything, but I bet that some uber-nerdle will tell me why... :roll:
Seems to me that all this forward hardware development is great and AMD has some great things in the works (IF it works and WHEN it works are 2 totally seperate arguments). I think that instead of shrinking the dies and making the same products (only smaller), they should work on software integration and enable us to use the true potential that the chips that we have now.
Oh, wait... that would kill their sales. Nevermind :lol: 


:lol: 

Sorry but your rant was quite amusing. I think you´re an advertisment victim. 64 bit is present and is actually used. People tend to forget where it is coming from. The server space. There it is used and there it belongs. AMD has always developed server CPUs and the dirt that´s been falling off the tables turned into desktop CPUs. It was a nice idea and AMD advertised it´s 64 bit as the holy grail of computing. In fact the advertisment took off so well, intel had to react. Now every processor can use 64 bit even though it´s only really used in the server space. I have no doubt that it will take another 5 years before 64 bit is mainstream.
June 11, 2007 6:53:52 PM

Quote:
I see your really busy too ...

You might want to rephrase your last sentence ... the grammar is just sad.


I see you don't know the difference between your and you're. Sorry man, you're not qualified to judge his grammar skills :p 

I'm not gonna get involved in this except to say:

il qui habite en les maisons de verre ne doit pas lancer de pierre

[flame]

WiseCracker...I don't know where to begin with your French, it was awful. While the general meaning got out, it caused my eyes to bleed. So before you go commenting on a grammatical error post, please check your grammar or just stop speaking a language you can't speak. It should have read: "Celui qui habite dans une maison de verre ne doit pas lancer des pierres”

[/flame]
June 11, 2007 7:17:43 PM

Quote:

I guess it's interesting as to why they didn't shrink Opteron. That would have probably gotten them to 3.4GHz as Opteron can do 3GHz at 120W.


First of all, AMD's 65nm has problems scaling at clock speeds nearing 3GHz. Secondly, the silicon on insulator technique doesn't scale well with higher freqencies. Last but not least, a 3.4GHz Opteron (if they were able to somehow make one) would be priced way too high and the heat output would take us back to the NetBurst days.

Quote:
I still have confidence in AMD as a CPU company. Maybe not as a marketing company but fortunately they can outsource that.


Hm...yeah...

I just wish that they started the Barcelona project a year earlier, or as soon as K8 launched. It was only a matter of time until Intel would drop NetBurst.


First IBM uses SOI and Power 6 is at 4.7GHz so it can scale with clocks. AMD CAN'T stop making say the 6000+ and Opteron x80 as they can't just turn off Fab 30.

If they take the 65nm up to 3GHz then they have overlapping PR Ratings at 65nm and 90nm.

Since 65nm is much smaller why not limit the clockspeed to try and help margins.

infact, If I remember correctly.. IBM IS USING Ge-SOI..
germanium-silicon on insulator, thats the only way they get higher speeds..
amd still doesnt have that technology, someone correct me if im wrong.
a c 110 à CPUs
a b À AMD
a b V Motherboard
June 11, 2007 7:48:07 PM

I got you're point . . . :lol: 
June 11, 2007 7:51:25 PM

Quote:

Why bring up a simple thing like a period vs a comma? There is nothing different about how I post only who is around to respond. Maybe now we can have unbiased discussion about both companies.

Normally by now the FUD patrol would have crapped all over the thread with off-topic posts.

I don't really prefer AMD, I just have bad memories of the K6 days when my machine ran constantly without a problem yet Intel lived on the phrase "not stable enough."


Baron,

To be honest with you.

The only reason you are now toning down your comments is because you were asked to. You know it I know it and others here should know it.

The reason the FUD patrol (people spreading the truth about processors not just one side as you have done) have not been around is mainly because the likes of you. So well done to you!

As for you (don't really prefer AMD)!! I do not know who you are trying to convince of that yourself or others! You have a very clear and direct Bias towards AMD and always have.

As far as the benchmarks go... As soon as Benchmarks are released with near retail stepping/SOI and they are confirmed by multiple third parties they will all be useless.

The reason I say this is because it is difficult to discern any real information without the ability to trend available data. This throws out outliers which could skew the information to a negative (in the case of Barcelona an Intel preferred site providing not so great/possibly skewed stats). In the case of a positive skew (AMDZone for instance). This would create cause for dismissal of either too far positive or too far negative which usually leaves a larger more accurate data set. The same should be said for the release of the Penryn.

Sorry, I know my comment above is old news...

As soon as we start to see information we can gather it and trend it to identify both price and performance without the skew.

So yeah there is my FUD for the day!!

There are people here still, that will always want for accurate information.

I hope your new found posting skills can help to provide that. I hope your bias can be set back, for truth to be set forward.

Good luck with your new persona.

For the record "I hope Barcelona smokes the C2D/C2Q arch".
June 11, 2007 8:14:18 PM

8O you got me :lol: 
June 11, 2007 9:26:05 PM

Quote:

Why bring up a simple thing like a period vs a comma? There is nothing different about how I post only who is around to respond. Maybe now we can have unbiased discussion about both companies.

Normally by now the FUD patrol would have crapped all over the thread with off-topic posts.

I don't really prefer AMD, I just have bad memories of the K6 days when my machine ran constantly without a problem yet Intel lived on the phrase "not stable enough."


Baron,

To be honest with you.

The only reason you are now toning down your comments is because you were asked to. You know it I know it and others here should know it.

The reason the FUD patrol (people spreading the truth about processors not just one side as you have done) have not been around is mainly because the likes of you. So well done to you!

As for you (don't really prefer AMD)!! I do not know who you are trying to convince of that yourself or others! You have a very clear and direct Bias towards AMD and always have.

As far as the benchmarks go... As soon as Benchmarks are released with near retail stepping/SOI and they are confirmed by multiple third parties they will all be useless.

The reason I say this is because it is difficult to discern any real information without the ability to trend available data. This throws out outliers which could skew the information to a negative (in the case of Barcelona an Intel preferred site providing not so great/possibly skewed stats). In the case of a positive skew (AMDZone for instance). This would create cause for dismissal of either too far positive or too far negative which usually leaves a larger more accurate data set. The same should be said for the release of the Penryn.

Sorry, I know my comment above is old news...

As soon as we start to see information we can gather it and trend it to identify both price and performance without the skew.

So yeah there is my FUD for the day!!

There are people here still, that will always want for accurate information.

I hope your new found posting skills can help to provide that. I hope your bias can be set back, for truth to be set forward.

Good luck with your new persona.

For the record "I hope Barcelona smokes the C2D/C2Q arch".


You could have left that comment out. That's the kind of comment that causes me to be toxic. I told you why I post the way I do and you say "Nooo, that's not it," as if you read my mind.
June 11, 2007 10:07:08 PM

Nice to see ya Ches
June 11, 2007 10:43:53 PM

Quote:
AMD's made some big mistakes in the past(like their lack of temperature control that Tom's Hardware did a very nice video on). Granted, Intel's done their share of screw ups too, but Intel hasn't launched a product that was designed to inherently fail if some dummy installed the heat sink wrong. This has been one of my biggest beefs with AMD. That's pretty bad design if you ask me. Intel's FPU Div bug from the Pentium sure, there's that small percentage of people that could be goofed with the bug, but it didn't result in you having to either attempt to RMA the chip or buy a new one. AMD has been very much an enthusiast market, and so they were much more likely to encounter some poor schmuck that burned up their chip cause they didn't know better.


Intel had exactly the same problem as AMD with CPUs that would die if you removed the heatsink which is exactly why they added the thermal protection to their chips. AMD did the same BUT the difference was that motherboard support didn't come along straight away, when the THG video was published AMD sorted it out pretty quickly.

Do you not remember Intel being first with CPUs with the die exposed (FC-PGA on the P3), just so your system builder could break lumps off his CPU core if he wasn't careful installing it, where Intel lead AMD followed!

The whole Netburst architecture was bad strategy, letting the marketing department dictate what the product was going to be produced technically flawed products....
June 12, 2007 1:02:41 AM

Hey, Turpit

Nice to see you too.

I will not abandon this forum for the sake of one person or the many for that matter.

These forums still need a voice of truth such as yourself.

Folks here have shown a FULL bias toward one tech or the other (they know who they are). It seems that on an enthusiast site we would show bias toward performance no matter which badge it carries.

The reason people started flaming/correcting/contradiction here is because of direct mis-information shown from direct bias. People were correcting things and that is wrong. Because we can not have correct information :roll: :roll: if is discredits their particular badge/brand.

Fact is Intel currently holds the performance crown (from trended data) you may eat that as you like or dislike but it is true.

Fact is Barcelona may change that (I hope it does "from a brood member 8O 8O" )

Fact Barcelona benchies so far have been few and far between with very little collaborative data.
a b à CPUs
June 12, 2007 1:30:05 AM

Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL BULLSH!T!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.
June 12, 2007 2:01:56 AM

Quote:
Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL ********!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.



As always, I would just like to point out that there have been no actual K10 benchmarks yet. Penryn for that fact of the matter either. When Anands, or Toms, or some other 3rd party gets their hands on these CPUs, and their hands are not tied by NDAs, then we will have real benchmarks.
a b à CPUs
June 12, 2007 2:03:37 AM

Quote:
Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL ********!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.



As always, I would just like to point out that there have been no actual K10 benchmarks yet. Penryn for that fact of the matter either. When Anands, or Toms, or some other 3rd party gets their hands on these CPUs, and their hands are not tied by NDAs, then we will have real benchmarks.

AMD peeps stated that the performance increase is 40%
Click ME!
June 12, 2007 2:11:21 AM

Quote:
Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL ********!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.



As always, I would just like to point out that there have been no actual K10 benchmarks yet. Penryn for that fact of the matter either. When Anands, or Toms, or some other 3rd party gets their hands on these CPUs, and their hands are not tied by NDAs, then we will have real benchmarks.

AMD peeps stated that the performance increase is 40%
Click ME!

I know they did. But thats not a benchmark. Thats a claim. A claim and $1.25 will get you a coffee.

The POV fiasco was not a benchmark.
The comptuex AMD dog & pony show was not benchmarks
What Intel has been doing with Penryn are not benchmarks (though far closer than anything from AMD)

They are all markerting and PR gimicks.

We just have to wait and see, and in the meantime, try to keep the false rumors and cleverly disguised benchmark wannabes supressed
June 12, 2007 2:32:52 AM

Since when did $1.25 get you coffee?
a b à CPUs
June 12, 2007 2:35:20 AM

Quote:
Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL ********!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.



As always, I would just like to point out that there have been no actual K10 benchmarks yet. Penryn for that fact of the matter either. When Anands, or Toms, or some other 3rd party gets their hands on these CPUs, and their hands are not tied by NDAs, then we will have real benchmarks.

AMD peeps stated that the performance increase is 40%
Click ME!

I know they did. But thats not a benchmark. Thats a claim. A claim and $1.25 will get you a coffee.

The POV fiasco was not a benchmark.
The comptuex AMD dog & pony show was not benchmarks
What Intel has been doing with Penryn are not benchmarks (though far closer than anything from AMD)

They are all markerting and PR gimicks.

We just have to wait and see, and in the meantime, try to keep the false rumors and cleverly disguised benchmark wannabes supressed

We don't supress anything. That's absurd man.. that's what DaSickNinja tries to do and it's wrong. I've been in this scene for far longer then he and many others who claim we should be doing that. WE'VE always participated in open debates and speculation. It's what makes this entire scene amusing.

The POV ray test DOES in fact give us an idea of performance. But now as we see it was not a realistic idea due to supposed chipset issues. Leaving open more debate and speculation. It's fun. In the end we'll have our answer... but till then it's good to discuss techology openly without feeling that we should fear some supposed superior group of know-nothings raining down on our parade.

I was one of the first ones who crusaded against Extremism (fanboyism) in general... as I've been involved in this scene since the VERY begining (was one of the first waterblock designers at Procooling.com). It's FUN!

Now to get back on track.. what we can assertain from this is that AMD hasn't a clue how Barcelona will perform since now they've confessed to never having actually been able to benchmark it properly due to issues. So it was a random claim someone took by using theoritical means.

As such we're back to square one.. :) 
June 12, 2007 2:37:43 AM

Excuse me? Just where do you get off insulting me?
June 12, 2007 2:48:03 AM

Quote:
Then it has to be said.. if AMD were having these issues since the beginning then were did the "We will outperform Kentsfield by up to 40%" claim come from?

TOTAL ********!!!!

Was a lie to begin with. Truth is they don't know since they have yet to be able to test it.



As always, I would just like to point out that there have been no actual K10 benchmarks yet. Penryn for that fact of the matter either. When Anands, or Toms, or some other 3rd party gets their hands on these CPUs, and their hands are not tied by NDAs, then we will have real benchmarks.

AMD peeps stated that the performance increase is 40%
Click ME!

I know they did. But thats not a benchmark. Thats a claim. A claim and $1.25 will get you a coffee.

The POV fiasco was not a benchmark.
The comptuex AMD dog & pony show was not benchmarks
What Intel has been doing with Penryn are not benchmarks (though far closer than anything from AMD)

They are all markerting and PR gimicks.

We just have to wait and see, and in the meantime, try to keep the false rumors and cleverly disguised benchmark wannabes supressed

We don't supress anything. That's absurd man.. that's what DaSickNinja tries to do and it's wrong. I've been in this scene for far longer then he and many others who claim we should be doing that. WE'VE always participated in open debates and speculation. It's what makes this entire scene amusing.

The POV ray test DOES in fact give us an idea of performance. But now as we see it was not a realistic idea due to supposed chipset issues. Leaving open more debate and speculation. It's fun. In the end we'll have our answer... but till then it's good to discuss techology openly without feeling that we should fear some supposed superior group of know-nothings raining down on our parade.

I was one of the first ones who crusaded against Extremism (fanboyism) in general... as I've been involved in this scene since the VERY begining (was one of the first waterblock designers at Procooling.com). It's FUN!

Now to get back on track.. what we can assertain from this is that AMD hasn't a clue how Barcelona will perform since now they've confessed to never having actually been able to benchmark it properly due to issues. So it was a random claim someone took by using theoritical means.

As such we're back to square one.. :) 

I disagree. Without an extensive amount of assumption and guestimation, the POV runs give nothing.

I dont mean to "rain" on anyones parade, but remember, 65nm, AM2, and R600 all turned into routs for AMD. Why? Because fanboys latched onto BS info, and promoted it out of context and out of proportion to the point where people set their expectations to the level of the fanboy BS, not the actual claims from AMD, and when the products appeared many people were sorely disappointed because 65nm didnt "beat" C2D, and AM2 didnt perform any better than 939 (not to mention the memory issues) and R600 failed miserably to beat Nvidia. Spectulation can be fun, but speculation based on facts, not Henri Richards frantic scramblings to bolster AMDs shrinking marketshare, or Intels "this is what penryn does" All that does is mislead, and always, it misleads to believing products will perfrom better than they actually wind up doing.

Back to the track, I agree they have no idea. Which means no one else does, and only further supports that the POV and comptuex stuff was meaningless.
June 12, 2007 3:07:08 AM

Quote:
Hell, I bought my Athlon64 2000+ back in 2000.


I would just like to point out that this so inaccurate that it's borderline absurd.
June 12, 2007 3:29:59 AM

Arg, I'm sure Amd will find a way out...Don't want Amd to die out you know? Lets just trust their abilities, like before, and stop talking about the future cause no one can precisely predict the future...I think... :?
June 12, 2007 7:24:02 PM

Aren't we all forgetting the important point from Anandtech, which was AMD is really nowhere near ready to make any sort of claim about performence, still. The indication is, they may either come to market with a possibly lower frequency product than expected or have to delay the launch to get the products out they said they would.

Considering their bold claims, the current landscape does not look so good. They are not holding back on releasing benchmarks to avoid giving Intel room to counter, they simply are not there yet. That does not mean they won't be, but I wouldn't go betting the house on it right now.
a b à CPUs
June 13, 2007 12:21:04 PM

I was very impressed with Anand's article.

It really struck a chord with me because he brought some real perspective on a few things.

I have 8 or so PC's here at home - mostly AMD64 gaming boxes and a 3100+ lapptop - but I also have some Intel PC's too ... A server, a core2duo laptop and a P3 Firewall.

Iv'e bought whatever was the best at the time ... at work I have a shitload of work PC's ... presently all D's. Not very happy with the heat and power they suck ... but the choice was one of procurement policy for the company ... not mine.

The 939 / AM2 debarcle really messed up my upgrade path for 4 of the machines and that pisses me off.

I guess no different from the 945 people who couldn't upgrade to Core2 without a new mobo.

I try to avoid the fanboi baiting but if pushed I am firmly in the Baron's camp ...

Iv'e been following the usuall stuff from the sites close to AMD and it seems they are messing with possibly final silicon now ... B2 or so? I lost the link to that yesterday so if anyone cares to post it I'd appreciate it.

The waiting ... close to the gestation period for an elephant ... heh heh.

Apolgies for the previous post ..I wrote in haste too ... took the bait from that moron. I see the Canadian's all fired up ... your all related anyway arn't you?
!