AMD Zambezi news, info, fans !

[undone]

I swear, none of this still makes sense, particularly after reflecting upon these prices from a Microcenter employee:

http://www.overclock.net/amd-general...ate-specs.html

Price is the same as shopblt, about $280 for fx8150, looks like final price.

edit:
prices for comparison
2600k $279.99
1100T $189.99
fx8150 $279.99

same as 2600k, 50% more expensive than 1100T, make sense?

[PatRaceTin]

280 look high for me

[spursindonesia]

Things just don't add up, they don't compute ...... confused, literally

[Dr. Vodka]

Dunno man, AMD just can't flop so bad... they can't possibly have spent 5+ years designing this new uarch just to release a piece of crap... Just 2 more days for the real thing to show up. I can't help but compare this to the HD4850/70 times, where word was that RV770 would have 480sp.. yet it had 800sp. Get what I mean? There must be something AMD is keeping for themselves. Why would they even bother asking U$S 280 for their top model when it can't compete with the 2600k? That doesn't add up.

But yeah, that preview going around and that guy who was benching his new BD rig over the net and the rest of the leaks/rumors aren't of much help, they all point to the same conclusion. I'm not expecting a "miracle" anymore... Anyway, I'll just wait till the 12th to see the real deal.

[PerryR]

same as 2600k, 50% more expensive than 1100T, make sense?

By making "sense," I meant the price vs performance ratio; particularly if this processor performs as indicated by these early benches.

That's what doesn't make sense.

Why would they even bother asking U$S 280 for their top model when it can't compete with the 2600k? That doesn't add up.

Or 2500k.

[james333]

For anyone thinking BD will be worse than Thuban, think of VW bringing out the next generation Golf GTI with the 1973 110bhp 1.6 8V engine Instead of 300bhp TFSI engine...... really?..... would any company do that?

[zalbard]

For anyone thinking BD will be worse than Thuban, think of VW bringing out the next generation Golf GTI with the 1973 110bhp 1.6 8V engine Instead of 300bhp TFSI engine...... really?..... would any company do that?

It doesn't lose to Thuban in all the applications, just some.

[sin0822]

yes not in all programs, in many its faster. I think in realworld performance this CPU is superior, but in benches like superpi and wprime it falls short, maybe its how the arch works, i hates benchmark. Superpi will cause BDOS with this CPu, even after wprime passed.

[Nintendork]

Dunno man, AMD just can't flop so bad... they can't possibly have spent 5+ years designing this new uarch just to release a piece of crap... Just 2 more days for the real thing to show up. I can't help but compare this to the HD4850/70 times, where word was that RV770 would have 480sp.. yet it had 800sp. Get what I mean? There must be something AMD is keeping for themselves. Why would they even bother asking U$S 280 for their top model when it can't compete with the 2600k? That doesn't add up.

But yeah, that preview going around and that guy who was benching his new BD rig over the net and the rest of the leaks/rumors aren't of much help, they all point to the same conclusion. I'm not expecting a "miracle" anymore... Anyway, I'll just wait till the 12th to see the real deal.

My signature could be mythical xD.

[duron]

2 days to reality

http://lenzfire.com/2011/10/amd-bull...edium=facebook

[AAbenson]

what i really dont understand is if the Bulldozer cpus are really that bad then why was Cray so eager to use these chips in its new supercomputers?

[Leeghoofd]

Did the Lab51 guys have access to all this stuff that JF-AMD pointed out in his FAQ?
Not sure if OS optimize and app with the latest flags would make a big difference, but who knows.

Q. I saw a benchmark on xyz website. Is that how bulldozer will perform?A. No. Nothing posted before launch will be representative of actual performance. To get actual performance, you need:

Final production silicon
Final processor microcode
Final system BIOS
Final OS optimizaitons
Final drivers
An app compiled with the latest flags
A person who understands the app and configures the test properly

Without these things (and there are probably more), you cannot get an accurate benchmark. Any extrapolation of a crappy benchmark gives you a crappy estimate of actual performance. Period.

1) you can't suspect reviewers to only test programs that are adapted for this CPU generation, comparisons have to be made with previous generation. Speaking for myself, my little testsuite has been like this and I will add some tests once in a while. But complete change my way of testing, dunno..
2) thinking the guys at Lab501 have enough experience with software, setting up soft/hardware.
3) Final silicon remark, most of the reviewers/powerusers just got the CPUs not even a week ago to start testing.
3) If final drivers/Bios are not ready now, then something horribly went wrong in the time schedule...and the board manufacturers should be ***whooped
4) Never forget first impressions count... it's D-day tomorrow... after such a long wait...


For the prices : new product, higher priced, will prolly stabilise in a month or so...

[bamtan2]

3dmark06
FX-8150 + 1333MHz RAM = 17612
FX-8150 + 1866MHz RAM = 17699

well, at least now we know that fast memory is as useless with bulldozer as it is with sandy bridge.

[savantu]

what i really dont understand is if the Bulldozer cpus are really that bad then why was Cray so eager to use these chips in its new supercomputers?

Well, Cray won the deal on those supercomputers when AMD probably didn't even had silicon in hand. Cray has to supply at a certain date a certain number of FLOPs. On each, you have a penalty for missing the target. Penalty is ussually in the millions $. Cray commited based on AMD assurances ( contract between Cray and AMD, AMD at the time could not forsee things will be that s**y ).

With BD failling, Cray still had to deliver on time. From what I've read, AMD had delivered tens of thousands of Interlagos already to Cray for the deal. It didn't matter they were probably power hungry and couldn't clock very high ( older steppings ), Cray needed the 1st to work reliable even if slower just to make the deadline. As for the performance penalties, AMD will cough up the difference. I wouldn't be surprised if Cray got the CPUs for free.

Why are people interested in Interlagos ? Because in HPC, AMD carved a niche for having the highest Flops/$ with Magny Cours. BD was supposed to improve on that with FMA. So to a few customers Interlagos makes sense. The question is whether they will keep some faith in AMD or ask for Intel powered systems in the future from Cray considering this is the 2nd major screwup after Barcelona ( Cray promised 2.5Ghz Barcelonas and had to ship with 2GHz TLB bugged ones ).

[Leeghoofd]

well, at least now we know that fast memory is as useless with bulldozer as it is with sandy bridge.

Really ? Does SB not work better with high speed rams ? Plus 3Dmark06 is not a good sign of ram speed scaling... other tests do .... For gaming indeed you are right, the gain is minimal on SB. But other apps can get a nice boost...

[imamage]

2 days to reality

http://lenzfire.com/2011/10/amd-bull...edium=facebook

I hate to say but I don't want to face the reality

[bamtan2]

Price is the same as shopblt, about $280 for fx8150, looks like final price.

edit:
prices for comparison
2600k $279.99
1100T $189.99
fx8150 $279.99

same as 2600k, 50% more expensive than 1100T, make sense?

those prices are simply not correct. the amd guy said they'd be under $250. all the benchmarks indicate bulldozer is much worse than 2600k. that also means it has to be under $250.

just because a store charges whatever they want on a new product today doesn't mean that'll be the price 3 days or a week from now

and if bulldozer bellyflops, the price will go down below MSRP. 8120 could easily fall from $220 to $200 this month.

whatever is the lowest 2500k price you've seen, expect that or less to be the 8120 price.

[AAbenson]

i see,thank you for clearing this up.
maybe AMD should start giving for free the FX8150 to anyone who buys a Radeon 6950/70/90

[savantu]

Did the Lab51 guys have access to all this stuff that JF-AMD pointed out in his FAQ?

AMD marketing was doing damage control while they were busy tunning the ecosystem to get a few % more.

Not sure if OS optimize and app with the latest flags would make a big difference, but who knows.

Q. I saw a benchmark on xyz website. Is that how bulldozer will perform?A. No. Nothing posted before launch will be representative of actual performance. To get actual performance, you need:

Final production silicon
Final processor microcode
Final system BIOS
Final OS optimizaitons
Final drivers

Launch is tomorrow. Who's fault is it if the whole chain isn't ready ? It's like buying the latest BMW and the sales guy will tell you " Well, we promised 300bhp and 40mpg, but since the latest software stack for the ECU isn't ready, you'll get only 150 bhp and 20 mpg until sometime next year when we'll fix it. Peace man." Imagine this to happen to you.

An app compiled with the latest flags

Yeah, who cares about x86 compatibility. Let's just ask all the SW vendors to recompile their apps for the new CPU with 10^-27 market share.

A person who understands the app and configures the test properly

That to me equals sleazy tactics. "Configure the test properly" = choose the settings and scenarious which will put BD in the best light.
Rumours has it that AMD will force reviewers to use predefined SW and limit their resolution choices to very high details where CPU differences are masked by the GPU bottleneck.

Without these things (and there are probably more), you cannot get an accurate benchmark. Any extrapolation of a crappy benchmark gives you a crappy estimate of actual performance. Period.

You're average Joe user will make sure he has the latest silicon, latest microcode, latest BIOS, all the OS patches, drivers are latest, beta if possible, ask the ISV to recompile all his apps starting with Angry Birds and will read the 300 page BD optimization handbook that AMD will supply for free with each CPU. Only then, he'll start using his PC for his day to day activities and be completely satisfied.

[duron]

I hate to say but I don't want to face the reality

reality hurts
BUT anyhow let us know when they are available in HK coz i'm still gonna grab one for souvenir

[Negaterium]

+1 savantu. Nice sum of facts, fanboys take note and see reality.

#SGS2

[JumpingJack]

Well, Cray won the deal on those supercomputers when AMD probably didn't even had silicon in hand. Cray has to supply at a certain date a certain number of FLOPs. On each, you have a penalty for missing the target. Penalty is ussually in the millions $. Cray commited based on AMD assurances ( contract between Cray and AMD, AMD at the time could not forsee things will be that s**y ).

Well, considering Cray has had issues with supply in the past with AMD, I think they most likely wiggled themselves some buffer before committing on contracts related to delivery/timing of systems featuring BD, they had been bitten once before. http://www.itjungle.com/tlb/tlb061207-story07.html

[B.A.T]

An norwegian E-tailer searchengine has slipped the price on FX 8150.



1956,- NOK = 337$

in comparison

2500K: 1490,- NOK = 257$

2600K: 2115,- OK = 364$

1100T: 1347,- NOK = 232$

[Smartidiot89]

That to me equals sleazy tactics. "Configure the test properly" = choose the settings and scenarious which will put BD in the best light.
Rumours has it that AMD will force reviewers to use predefined SW and limit their resolution choices to very high details where CPU differences are masked by the GPU bottleneck.

Very interesting. There is a very clear difference between a recommendation and being forced to do something

I think I speak for most reviewers when not many, if any will divert from their regular testing methodology.

[savantu]

Well, considering Cray has had issues with supply in the past with AMD, I think they most likely wiggled themselves some buffer before committing on contracts related to delivery/timing of systems featuring BD, they had been bitten once before. http://www.itjungle.com/tlb/tlb061207-story07.html

You can wiggle a few months, but not 2-3 quarters. The customers aren't stupid, they've seen too the marketing slides from AMD, they've done their homework. If they were told Interlagos will be H1 ( Q2 ) , then rest assured they didn't expect to get it in early Q4. On HPC supercomputers, researchers apply for time slices, there is a lot of preparation going on even before the computer is installed. Imagine going back and saying " Hello everyone, we're delaying everything by 6 months ".

How likely are those customers to order again AMD based systems and even more interesting, how likely is Cray to bet high end deals on Opteron CPUs ? What about IBM, HP, Dell in the commercial sector ? Imagine production and banking systems. The factory is ready, the server is MIA or they're forced in the interim to use previous generation CPUs. Imagine the penalties and the extra downtime for upgrading once Interlagos is available. If you find the answer to these, AMD's server market share of <6% suddenly is explained.