AMD Zambezi news, info, fans !

[slaveondope]

Something I can add. Friday MicroCenter employee confirmed to me he has BD in store. We are getting close for sure, now someone pull a fast one at MC and lets see some numbers!

[FlanK3r]

Maybe more weak platforms, H67, H61 with 1333 RAM, honestly, I don't know slaveondope

no...its diferent between boards, if u used ASUS LGA1155 boards with auto setings, your turbo with 2600k (or 2500k etc) is in every moment at maximum boost (3800 MHz with 2600k)

[PatRaceTin]

Something I can add. Friday MicroCenter employee confirmed to me he has BD in store. We are getting close for sure, now someone pull a fast one at MC and lets see some numbers!

Wow

[TESKATLIPOKA]

Oliverda

Please share us the official description of one x86 core.

For BD core it should be L1I, L1ITLB, fetch, prediction, prefetch, decode, L2, L2TLB, IMAC, FMAC, FPU sceduler, some buffers, ALU, AGU, instruction retire, L/S, L1D, L1DTLB, int. scheduler, write coalescing cache.
Actually you should share with us why you think (ALU+AGU+retire+L/S+L1D+L1DTLB+int. scheduler) is a core? Because only these parts are doubled the rest is shared.

[xVeinx]

It's interesting that all the motherboards will require a bios update (according to the Microcenter leak).

[Olivon]

no...its diferent between boards, if u used ASUS LGA1155 boards with auto setings, your turbo with 2600k (or 2500k etc) is in every moment at maximum boost (3800 MHz with 2600k)

Sorry Flanker but I don't understand what you're saying

Passmark CPU test uses 8T and I doubt that Asus boards go 3800MHz on all core during all the test, K or non-K don't make a difference, processors react the same whatever brand of board is used

[liberato87]

http://www.overclock.net/15136591-post6839.html

[Manicdan]

^if ram is in slot 3 and 4, i dont think its using dual channel. but i could be wrong

[xVeinx]

The floating point scores look appropriate, but the integer scores don't look right at all. It's not that it's beaten by the 2600k, but that it's destroyed by the 2600k (and likely just about any other processor, including last gen). There must be a reason that bios updates are going to be doled out come release (assuming that has anything to do with it, which is just speculation on my part)...

EDIT:
<- Needs coffee!
I wasn't reading the graphs correctly. Integer is better then previous phenoms, but not to the degree I expected. I still expect better integer performance come release, but we'll see.

[cesariuth]

What's AGESA version will be needed for bulldozer? 0.0.9.2 version? or new one?

[TESKATLIPOKA]

cesariuth there is already a new one but who knows if its final.

[informal]

There will be (supposedly) final agesa release just prior to the launch. Whether it will affect the performance or not is anybody's guess. Oh wait, AMD knows at least .

[Olivon]

http://www.overclock.net/15136591-post6839.html

2600K results coming from an OCed CPU for sure (~4500MHz), @ default stock BIOS, I got ~9150pts :

[FlanK3r]

it is not bad for FX.... 8700points vs 9150

[The Stilt]

What's AGESA version will be needed for bulldozer? 0.0.9.2 version? or new one?

Most of the manufacturers still use V0.0.9.2 in their current (official) builds.
Asus, Biostar & Gigabyte atleast have beta bioses which use V1.1.0.0 version.
Official versions with the latest AGESA should appear within few days.

However the older AGESA versions work just fine.

Or so I´ve heard..

[Olivon]

it is not bad for FX.... 8700points vs 9150

I just hope that's not coming from an OCed processor, if you look at liberato87 screen, 2600K seems runing @stock, but in fact, it's not.

Likely many/most of the 2600K submissions are overclocked hints the spread from 2600.

Seems slaveondope was right regarding OC results mixin' with stock results. My apologizes dope machine

[BeepBeep2]

And it's not like BD "8 core" will work on a cheap mb 100$ either.
In overclocking it will need at least an 150$, 10phase VRM motheboard.
Memory it's not an issue for SB-E you buy 4*2gb modules.

8 cores will OC on $100 boards...
M5A97 $90, ASRock 970 Extreme3 is $90..., Gigabyte 970A-UD3 $100-110, ASUS M5A99X Evo $130

10 phase VRM? You're joking, right O_o

You are very misinformed.

...
About the passmark screens, yeah...people like to do things like that. PassMark is very intel biased to start out with...

[gosh]

About this:

http://library.madeinpresse.fr/samples/MPqY2Vg2v45Z-f

Besides, as we announced already in our previous issue, if they can sometimes compete with Sandy Bridge in the applications of rough calculations, the results are in video games very far behind. Only overclockers (and fanboys) will find them a great interest given their predisposition in this area.

If Bulldozer performs best in Battlefield 3 it will probably be the CPU gamers choose
Look at this picture:

http://gamegpu.ru/Action-/-FPS-/-TPS...-test-GPU.html

[Mechanical Man]

About this:

http://library.madeinpresse.fr/samples/MPqY2Vg2v45Z-f



If Bulldozer performs best in Battlefield 3 it will probably be the CPU gamers choose
Look at this picture:
...

That pic tells . Test must be made at level where ppl play at. For example my 1100T is using only four (4) cores in BF3.

[gosh]

That pic tells . Test must be made at level where ppl play at. For example my 1100T is using only four (4) cores in BF3.

Do you have DX10 GPU?

[GenTarkin]

About this:

http://library.madeinpresse.fr/samples/MPqY2Vg2v45Z-f



If Bulldozer performs best in Battlefield 3 it will probably be the CPU gamers choose
Look at this picture:

http://gamegpu.ru/Action-/-FPS-/-TPS...-test-GPU.html

MY GOD! .. STUPID!! WHY do websites insist on showcasing a systems performance at resolutions NOONE plays at!....who gives a at these levels...this is so ing annoying.
Its so damn misleading to potential shoppers - its ridiculous!
Needs to be done at a resolution people play at! At least 1680x1050 and above...I mean come on!

[gosh]

MY GOD! .. STUPID!! WHY do websites insist on showcasing a systems performance at resolutions NOONE plays at!....who gives a at these levels...this is so annoying.
Its so damn misleading to potential shoppers - its ridiculous!
Needs to be done at a resolution people play at! At least 1680x1050 and above...I mean come on!

This slide isn't about performance, it's about scaling on the CPU. As you can see Battlefield scales to at least 6 cores, probably 8

[BeepBeep2]

GenTarkin simply does not hold the intelligence needed to know that 640x480 is to remove the GPU bottleneck.

[GenTarkin]

This slide isn't about performance, it's about scaling on the CPU. As you can see Battlefield scales to at least 6 cores, probably 8

Ah...hehe well that makes sense, thanks for pointing that out. I just got pretty upset cuz that seems to be the trend w/ many reviews nowadays.
Thats cool to know.
I can say that my i3 550 @ 4ghz is WAY above the reflected 26 fps on ultra quality w/ my 6950 @ default clocks.
In fact, I have monitored my GPU usage in ultra quality and it constantly at 99.9x% usage...which tells me my i3 550 aint bottlenecking the GPU.
Also my CPU shows on average 4 threads between 50-80% throughout game.
This is at resolution of 1680x1050 =P
So, its pretty damn balanced on my setup =)
Also, there was a comparison done elsewhere on nvidia vs ati in BC2 and how their latest offerings perform w/ different core counts. ATi is a lot less sensitive to lack of cores whereas nvidia suffered heavily with less cores.

[GenTarkin]

GenTarkin simply does not hold the intelligence needed to know that 640x480 is to remove the GPU bottleneck.

No, I know that for sure....its just pointless to test that way nowadays.. if your audience is pure gamers.