Rumour: Bulldozer 50% Faster than Core i7 and Phenom II.

[eternalone]

There are several new benchmarks emerged for the upcoming AMD Bulldozer which show quite awesome results:

http://news.mydrivers.com/1/160/160285.htm

For example at SYSMARK 2007 the Zambezi is 154% faster than a Phenom II X4 965. And that already with a clock of 3.2 GHz.

It also tells about an additional trace cache and 4 KB of Level-0 cache as well as that the Turbo will push up to 1.6 GHz to the base frequency.

Woah is this real?

 

[Accord99]

Isn't this the infamous 2010 April Fools joke that got JF-AMD into some trouble?

http://www.xtremesystems.org/forums/showpost.php?p=4319403&postcount=72

This was incredibly uncool.

I have had to spend a lot of my day cleaning this mess up with sales and customers. It is not inconceivable that I will have to spend months replying to emails...

 

[Voo]

There are several new benchmarks emerged for the upcoming AMD Bulldozer which show quite awesome results:

http://news.mydrivers.com/1/160/160285.htm

For example at SYSMARK 2007 the Zambezi is 154% faster than a Phenom II X4 965. And that already with a clock of 3.2 GHz.

It also tells about an additional trace cache and 4 KB of Level-0 cache as well as that the Turbo will push up to 1.6 GHz to the base frequency.

Is posting 1 year old april fools on the 30th march, considered funny?

 

[HW2050Plus]

Isn't this the infamous 2010 April Fools joke that got JF-AMD into some trouble?

http://www.xtremesystems.org/forums/showpost.php?p=4319403&postcount=72

You are right. I checked the numbers. They are the same as from April 2010. So it's fake.

 

[HW2050Plus]

Is posting 1 year old april fools on the 30th march, considered funny?

I don't find that funny too. You can make a April fools but then PLEASE on 1st of April and not two days before ...

And then even reuse an old one ...

 

[Idontcare]

Is posting 1 year old april fools on the 30th march, considered funny?

The timestamp says 2010...you can google for, and read, last year's April fool articles any day of the year.

 

[HW2050Plus]

Here is an article about interpreting the OpenBenchmark early results for the Interlagos engineering sample, it's from David Kanter:

http://www.realworldtech.com/page.cfm?ArticleID=RWT033011040021

 

[JFAMD]

What david should have said is you can't interpret them because a.) that might not even be real data and b.) if it is those chips are not representative of the real capabilities of the processor.

 

[podspi]

JFAMD, you might just want to stay away from any tech forums tomorrow... :biggrin:

I might do the same...

 

[PreferLinux]

Actually, anyone can play April Fools jokes (on the likes of here!) for ~3850 hours: starting 15.75 hours ago (Chatham Islands, NZ at UTC +13.75 hrs at midnight), to 2234 hours time (at UTC -12 hours at mid-daymidnight [Just found out may places let it go til then, rather than mid-day.])!

 

[IntelUser2000]

Actually, anyone can play April Fools jokes (on the likes of here!) for ~38 hours: starting 15.75 hours ago (Chatham Islands, NZ at UTC +13.75 hrs at midnight), to 22 hours time (at UTC -12 hours at mid-day)!

So if a company wants the public to pay serious attention to a product, and are about to release something that's going to exceed the expectations, they shouldn't release it few days before or after April 1st.

Why do we have such a seriously messed up tradition again? Corporations should blame April Fools for missing critical timelines and therefore sales expectations.

 

[Phynaz]

So if a company wants the public to pay serious attention to a product, and are about to release something that's going to exceed the expectations, they shouldn't release it few days before or after April 1st.

Why do we have such a seriously messed up tradition again? Corporations should blame April Fools for missing critical timelines and therefore sales expectations.

Interestingly the IBM PS/2 was launched April 1st 1987.

 

[Cerb]

Interestingly the IBM PS/2 was launched April 1st 1987.

Hah! x86 for the masses was an April Fools joke by IBM. Little did they know at the time...

 

[AtenRa]

I dont know if this have been posted before,

http://emm.msi.com/display.php?List=43&N=1029

MSI’s AMD 800 series mainboards offer the easiest upgrade trajectory for AMD’s future AM3+ CPUs. By offering compatibility with a simple BIOS upgrade end-users are guaranteed future compatibility on their current mainboard. This allows end-users to start with a simple dual-core processor today and scale up to an eight core (octa-core) processor when the higher performance is required by todays and future games and applications.

1. 890FXA-GD70
2. 890FXA-GD65
3. 890GXM-G65
4. 880GMA-E55
5. 880GMA-E35
6. 880GMS-E35
7. 870A-G54
8. 870A-G54H
9. 870A-G46
10. 760GM-P33

Now MSI says that those Mobos are Bulldozer ready.

 

[Ancalagon44]

Last time I bought an MSI motherboard, they never bothered with a bios release to add AM2+ capability. Despite the fact that Asus and Gigabyte did. So, not for me, I dont trust MSI after that.

 

[podspi]

I was planning on buying an AM3+ board and slapping my Thuban in it -- but I think I'm just going to wait and see. We still don't know what the difference is between the 8xx and 9xx chipsets are (or if there is any) and what all these mobo manufacturers are talking about...

 

[Morg.]

Last time I bought an MSI motherboard, they never bothered with a bios release to add AM2+ capability. Despite the fact that Asus and Gigabyte did. So, not for me, I dont trust MSI after that.

Last timeS i bought MSI motherboardS (two of em), both died after a year without any overclocking or tweaking whatsoever.

 

[eflat123]

I was planning on buying an AM3+ board and slapping my Thuban in it -- but I think I'm just going to wait and see. We still don't know what the difference is between the 8xx and 9xx chipsets are (or if there is any) and what all these mobo manufacturers are talking about...

Does anybody know this?

 

[dma0991]

Does anybody know this?

So far there have been rumors about 9xx series chipsets having SLi capabilities as opposed to the 8xx series chipsets that does not support SLi.

 

[Idontcare]

So far there have been rumors about 9xx series chipsets having SLi capabilities as opposed to the 8xx series chipsets that does not support SLi.

I thought nowadays being SLI compatible merely means the mobo is licensed as a SLI-compatible mobo and then Nvidia's SLI drivers get the magic DRM key to work on that mobo?

Why would AMD need a new chipset in order for SLI to work if crossfire already functions on the existing chipset?

 

[busydude]

I thought nowadays being SLI compatible merely means the mobo is licensed as a SLI-compatible mobo and then Nvidia's SLI drivers get the magic DRM key to work on that mobo?

Why would AMD need a new chipset in order for SLI to work if crossfire already functions on the existing chipset?

Nvidia can give SLI DRM key to any AMD chipset, but maybe they only chose 9XX series of chipsets.

Additionally, that leaked slide was made by Nvidia and not AMD.(Judging by the style of the slide)

 

[grimpr]

I was planning on buying an AM3+ board and slapping my Thuban in it -- but I think I'm just going to wait and see. We still don't know what the difference is between the 8xx and 9xx chipsets are (or if there is any) and what all these mobo manufacturers are talking about...

Hypertransport 3.1 for starters.

 

[podspi]

Hypertransport 3.1 for starters.

That's the only thing I've heard so far as well. Depending on which thread and which forum you're in, it is either a dealbreaker or not important at all :biggrin:

Barring catastrophic hardware failure, I'll probably just wait. Not like I'm getting anonymous BD ES shipments like some people are claiming

 

[dma0991]

I thought nowadays being SLI compatible merely means the mobo is licensed as a SLI-compatible mobo and then Nvidia's SLI drivers get the magic DRM key to work on that mobo?

Why would AMD need a new chipset in order for SLI to work if crossfire already functions on the existing chipset?

The technical part I am not too sure but if AMD can sell a new product that can SLi they can use that as a reason to get consumers to change their motherboards to the 9xx chipsets.

If they were to enable the 8xx chipsets just with a DRM key then the 9xx boards would not sell. Sort of like there is nothing to distinct the 9xx chipsets from the 8xx chipsets in terms of features.

 

[The J]

All of these "AM3+ Ready" motherboards require a new revision of the board, right, since the physical sockets differ by one pin? Has there been any indication as to when we'll see these new boards?

My concern is that people could buy up these boards thinking a BD will drop in only to find out they got the older, unsupported rev of the board instead.