AMD's answer to the i7?

Review sites on internet has not been that important before.
It is easy to see what type of result one review wants just checking the type of applications they are using. It is easy to create a test that show one processor stronger just by selecting "the right" applications.
Why do you think that there are so many review sites that is using CrossfireX or three SLI testing i7?
Most sites are using applications that few are using or are using on regular basis.
Databases is very important, registry in windows is used by almost all applications and that is one internal database. How fast Windows starts is also something that is important for most users, when did you see tests that was checking something like that?
Most review sites are driven by commercials, you probably know that there is a fight regarding Intel in EU because Intel has tried to make it beneficial for those that avoid selling AMD

Way to duck the questions....but nevermind.
 
Btw Kyle, I'm sorry if I miss it, but I didn't see any power consumption test on the Core i7 from [H] yet.

Was nothing to report. With a multi meter and a kill-a-watt you would not be able to tell Core i7 at 3.2GHz from Core 2 Quad at 3.2GHz. Sort of uneventful. We did mention it in our articles.
 
I was just explaining one reason why AMD could avoid sending processors to review sites. If you like Intel you will of course don't understand that reason.
Regarding this site so if you criticize (comparing AMD to Intel) Intel in the Intel area you will be warned, but in the AMD area it is possible for Intel fanboys to bash AMD as much as they want.


You know, you did not explain shit. Your fanboy posting is not needed here. Your account is revoked.

HardOCP_FAIL_250-Forum.jpg
 
HAMMER TIME!

I fucking love it when he does that. I just hope it's never me! Oh, and I'm totally stealing the graphic..it's my new background.

You're ban quote is being added to my sig line. Classic [H] attitude. I love this place.
 
i couldnt even follow what he was saying... :/

Glad to know I wasn't the only one totally confused by kassler's rambling. :eek:

Deneb will be lucky to match current C2Q.

Bulldozer hopefully will be the answer to Core i7. Only time will tell though...competition would be nice to cause price wars. :D
 
Glad to know I wasn't the only one totally confused by kassler's rambling. :eek:

Deneb will be lucky to match current C2Q.

Bulldozer hopefully will be the answer to Core i7. Only time will tell though...competition would be nice to cause price wars. :D


Agreed.

Agreed.

Agreed, but I think it will slip to 2011, or at least that is what I am hearing from inside AMD.
 
He was doing the same nonsense on THG's boards last year.
 
2011? :eek:

If that's the case what'll we see between now and then other then Deneb? I thought it was supposed to be Bulldozer? :confused: :confused:
Um, nothing? ;) Deneb is pretty much it for 2009, unless you are waiting for low end MCM Fusion chips.

On the server front, there are the DDR3 Shanghai refresh next year, "San Paolo" (Istanbul, 6 cores) in early 2010 and "Magny-Cours" (MCM 12 core) some time after that.

Deneb's successor probably isn't coming until Bulldozer, unless AMD makes some higher end desktop chip based on Istanbul (pretty likely), but that's still a Shanghai derived chip. On the plus side, if AMD does release a 6 core desktop chip in 2010, that would probably be competitive with the 45nm 4 core i7.
 
What about Fusion? Isn't that supposed to help on the gaming side of things? Sorry if this is a noob question. I'm not entirely familiar with AMD and their road map.
 
Fusion is a part with a low end graphics chip in the same package as the dual core CPU (i.e. 2 separate chips in one CPU package). Good for lower power, not good for any serious gaming performance.
 
the bad part about being a realist and a AMD fanboi is that I have to admit to Kyle and Dan probably being right. But I can hope. ATI managed to pull a rabbit out of there hat this round. maybe it will rub off on AMD.

Speaking of ATI I can make this argument about 4800 vs the GTX200 statements above, you can argue performance matching and whatnot all you want from a business point of view the 4800 slaughter the GTX200 line. they can make money, good money, on their cards and still strangle Nvidia. The did not have to match the performance here to win.
 
In AMD's great last quarter, they reported $47 million operating income on $385 million in sales in the graphics unit. nvidia, in a terrible quarter (the one that just ended), reported: $61.7 million operating income on $898 in sales. The gorilla is still big.
 
Its only a rumor...
But....

The Shanghai chips coming our way are the A1 batch.

Those chips seem to be clocked at 2.7-3.0Ghz with an avg overclock of 3.6 without getting high up into the voltage.

There are plans for another stepping that should be released in March. The hope is that, that stepping will allow for desktop processors with clocks up to 3.6Ghz and possibly a FX branded option.

The next desktop quad core release beyond that is scheduled for late 4th quarter 2009, but not much is being said about that release.

If anyone is looking for pre-release benchmarks... Hardspell has them.


Speculation:
I am not claiming any special knowledge, but will say that the 45mn AMD CPU's being released are going to be a decent low to mid range sell.
 
Its only a rumor...
But....

The Shanghai chips coming our way are the A1 batch.

Those chips seem to be clocked at 2.7-3.0Ghz with an avg overclock of 3.6 without getting high up into the voltage.

There are plans for another stepping that should be released in March. The hope is that, that stepping will allow for desktop processors with clocks up to 3.6Ghz and possibly a FX branded option.

The next desktop quad core release beyond that is scheduled for late 4th quarter 2009, but not much is being said about that release.

A1 batch of rev C NOW revision D is on the table of the 45nm chips that will bring in High K that will be the i7 killer yes i know its fudzilla but heres to hope http://www.fudzilla.com/index.php?option=com_content&task=view&id=6986&Itemid=35

"In 2009, AMD plans to migrate to the 45nm K10.5 Rev D and this new process will still be based on Silicon on insulator, but this time they will use a high-K process. AMD calls these cores Hydra and it will enable up to eight cores with 6MB shared L3 cache and more L2 per core."

mmm NATIVE 8 CORE AMD chip
 
High-K may be the 4.0Ghz on air FX chips that AMD wants to produce.

Even so, 4.4Ghz Phenom would just about equal the i7 965 3.2Ghz chip.
 
High-K may be the 4.0Ghz on air FX chips that AMD wants to produce.

Even so, 4.4Ghz Phenom would just about equal the i7 965 3.2Ghz chip.

with in 2 weeks we will know where shanghai and deneb stand i think you are nuts speculating that 4.4ghz will be needed to beat a i7 perhaps a Barcelona would need 4.4 ghz to match an i7 shanghai will need a lot less
 
I wonder if Kyle or Dan will be able to get their hands on a Shanghai chip for review?

Could Kyle ask AMD for one?
 
I thought that SOI got worse and worse with shrinks down from 90nm? If so, why do 45nm SOI at all?
 
High-K may be the 4.0Ghz on air FX chips that AMD wants to produce.

Even so, 4.4Ghz Phenom would just about equal the i7 965 3.2Ghz chip.

Might be a bit premature to make assertions about speeds since nobody has benched a 45mn Phenom. Your probably right-on with the FX assesment........high-k and additional tweaks to the core, Q3-Q4 2009 would push the clocks to point that a high-clocked FX would be doable. 32mn probably the end of 2010 with Bulldozer following in 2011......dunno.:D
 
perhaps a Barcelona would need 4.4 ghz to match an i7 shanghai will need a lot less

How do you know that Shanghai will need a lot less? How do you know that Shanghai will be faster than Barcelona by any significant amount altogether for that matter?
 
I wonder if Kyle or Dan will be able to get their hands on a Shanghai chip for review?

Could Kyle ask AMD for one?

I would not expect to see coverage till almost Christmas.
 
Its only a rumor...
But....

There are plans for another stepping that should be released in March. The hope is that, that stepping will allow for desktop processors with clocks up to 3.6Ghz and possibly a FX branded option.

Same story I got with Phenom.....I am not overexpecting this time.
 
i think Deneb will be somewhere b/w Core2 65nm and Core2 45nm in regards to performance. Hopefully, the power usage goes down significantly. I use AMD only so i look forward to their release.
 
you know after reading this whole thread.. even though the deneb core will not match the i7.. the 3 advantages the deneb will have over the i7.. 1 if you already own an am2+ motherboard you dont need to buy a new socket sized motherboard.. 2 if you already own ddr2 memory you dont need to buy ddr3.. and 3.. they will probably be dirt cheap compared to the i7.. for right now thats what the deneb has going for its self.. but there better be a huge performance gain when they start releasing the am3 versions of the deneb otherwise amd will be screwed...

and yes i always buy amd before intel.. but im a realist at the same time..
 
you know after reading this whole thread.. even though the deneb core will not match the i7.. the 3 advantages the deneb will have over the i7.. 1 if you already own an am2+ motherboard you dont need to buy a new socket sized motherboard.. 2 if you already own ddr2 memory you dont need to buy ddr3.. and 3.. they will probably be dirt cheap compared to the i7.. for right now thats what the deneb has going for its self.. but there better be a huge performance gain when they start releasing the am3 versions of the deneb otherwise amd will be screwed...

and yes i always buy amd before intel.. but im a realist at the same time..

i agree though right now amd has a server window and they need to exploit it asap as well for all the companies that have amd opterons in the racks replacing chips and keeping server hardware = more economical than new server heh we wil lsee soon sooon!
 
but there better be a huge performance gain when they start releasing the am3 versions of the deneb otherwise amd will be screwed...

There won't be a performance gain going to AM3. The chips will be the same, so any benefits will be limited to those gained from the faster DDR3 memory, which will most likely be minimal. The Phenom architecture was obsolete before it was even released, and there's nothing that AMD can do to it that will make it reach the efficiency of Nehalem, and probably not even Yorkfield for that matter.
 
Whoopy doo, there was no performance gain going to AM2 from S939 either, well except for a marginal clock speed bump and a few more cores added... :rolleyes:

Btw some of you guys really need to watch the end of 2008 Formula 1 Brazilian GP, you really shouldn't cheer up unless it is really over because you don't want to be like these Ferrari guys.
 
i agree though right now amd has a server window and they need to exploit it asap as well for all the companies that have amd opterons in the racks replacing chips and keeping server hardware = more economical than new server heh we wil lsee soon sooon!

Absolutely right on! You only need to look at current economic conditions to realize a chip swap and bios upgrade is looking really good to the business sector.:D
 
Absolutely right on! You only need to look at current economic conditions to realize a chip swap and bios upgrade is looking really good to the business sector.:D

In regard to the server market it is my experience that companies won't look into replacing processors until the machines are already 3 to 5 years old at which point they usually end up replacing the entire server. Like the consumer level counterparts, computers in the business sector tend to only get drive and memory upgrades during their operational lifetime. The only time in the server world where that isn't the case is in regard to adding processors with more cores. If they can, and there is a need to do so, they will sometimes do it. but don't think that most companies upgrade CPUs in their server systems because they don't.
 
In regard to the server market it is my experience that companies won't look into replacing processors until the machines are already 3 to 5 years old at which point they usually end up replacing the entire server. Like the consumer level counterparts, computers in the business sector tend to only get drive and memory upgrades for the most part. The only time in the server world where that isn't the case is in regard to adding processors with more cores. If they can, and there is a need to do so, they will sometimes do it. but don't think that most companies upgrade CPUs in their server systems because they don't.

true dat...
 
He was doing the same nonsense on THG's boards last year.

He's been doing it here for quite some time aswell.
I've reported quite a few of his posts. I hope that's the last we see of him here.
 
so the outlook next year REV D quads will reign in the 1h with soi+high k and 32nm in the 2h so this time next year we should have 32nm
You're dreaming even beyond AMD's fantasy schedule. AMD does not have 32nm or 45nm Hi-K on any roadmaps through 2010, nor any test chips (usually SRAM cells) that would precede either one by at least a year.
 
The Phenom architecture was obsolete before it was even released

The architecture it not obsolete. It would be obsolete if there was a new architecture from AMD or the Phenom architecture wasn't in use any longer - but it's neither of those things. Something is not obsolete just because it isn't the best/fastest at what it does.
 
http://www.theinquirer.net/gb/inquirer/news/2008/11/11/amd-announces-move-32nm dropping the fabs saved amd they are trying their best to keep up with intels process tech so the outlook next year REV D quads will reign in the 1h with soi+high k and 32nm in the 2h so this time next year we should have 32nm amd chips all i can say is they better not pull a Barcelona on us!!

This sounds highly unlikely.

You're dreaming even beyond AMD's fantasy schedule. AMD does not have 32nm or 45nm Hi-K on any roadmaps through 2010, nor any test chips (usually SRAM cells) that would precede either one by at least a year.

Agreed.

I think there is a lot of wishful thinking from AMD fans on their future but let's face facts. AMD has barely ever turned a profit in the past 20 years and when they have it was never by much. They are in debt up to their eye brows and nothing they are doing right now is going to change that. Bulldozer is really their only hope and that's a ways off. Once Intel gets' server versions of Nehalem out the door things are only going to get worse for AMD.
 
This sounds highly unlikely.



Agreed.

I think there is a lot of wishful thinking from AMD fans on their future but let's face facts. AMD has barely ever turned a profit in the past 20 years and when they have it was never by much. They are in debt up to their eye brows and nothing they are doing right now is going to change that. Bulldozer is really their only hope and that's a ways off. Once Intel gets' server versions of Nehalem out the door things are only going to get worse for AMD.

Showing bias and preference as a fan is one thing, doing it as a moderator is surely another.
 
Showing bias and preference as a fan is one thing, doing it as a moderator is surely another.

When Dan_D post here I think everyone else here reads his post as his own personal opinion...except you?
His post is logical valid...we are borderline ad hominem here...
 
Back
Top